Advances in Advertising Research, Volume 2

Shintaro Okazaki (Ed.) Advances in Advertising Research (Vol. 2)

GABLER RESEARCH European Advertising Academy Executive Board Members: Sandra Diehl, University of Klagenfurt Robert Heath, University of Bath Tobias Langner, University of Wuppertal Peter Neijens, University of Amsterdam Shintaro Okazaki, Universidad Autónoma de Madrid Patrick de Pelsmacker, University of Antwerp Edith Smit, University of Amsterdam Ralf Terlutter, University of Klagenfurt

The objective of the European Advertising Academy (EAA) is to provide a professional association to academics and practitioners interested in advertising and its applications that will promote, disseminate and stimulate high quality research in the field.

Shintaro Okazaki (Ed.)

Advances in Advertising Research (Vol. 2) Breaking New Ground in Theory and Practice

RESEARCH

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.

1st Edition 2011 All rights reserved © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011 Editorial Office: Ute Wrasmann | Nicole Schweitzer Gabler Verlag is a brand of Springer Fachmedien. Springer Fachmedien is part of Springer Science+Business Media. www.gabler.de No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the copyright holder. Registered and/or industrial names, trade names, trade descriptions etc. cited in this publication are part of the law for trade-mark protection and may not be used free in any form or by any means even if this is not specifically marked. Umschlaggestaltung: KünkelLopka Medienentwicklung, Heidelberg Printed on acid-free paper Printed in Germany ISBN 978-3-8349-3134-4

Preface

It is my great pleasure to present Advances in Advertising Research Vol. II— Breaking new ground in Theory and Practice, published by the European Advertising Academy (EAA). This volume consists of extended versions of papers presented at the 9th ICORIA (International Conference on Research in Advertising) held at Universidad Autónoma de Madrid, Spain, from June 24-26, 2010. The conference was sponsored by the European Advertising Academy. The editor served as chairperson of the conference. The authors of this book chose the papers, which examine a wide-ranging view of issues that address how we can understand advertising and its role in our society. Both distinguished and junior scholars from around the globe contributed to this book. As the subtitle indicates, the book is intended to provide a unique opportunity to stimulate and generate new ideas in a diverse range of research topics. I believe that such efforts will help us, both directly and indirectly, to guide the future research of both present and emerging advertising scholars. The book has five sections. Part 1 examines one of the hottest topics, product placement and gaming; Part II addresses generational issues in advertising; Part III examines the interaction between social issues and advertising; Part IV examines issues related to corporate organization, publicity and reputation; Part V explores branding issues in advertising; Part VI explains the roles of media and agency; and, finally, Part VII scrutinizes advertising content, appeals and execution. All these chapters consider cutting-edge issues that offer important insights that advance our knowledge. This book would never have been possible without the willingness of the contributors who traveled to Madrid to participate in the 9th ICORIA, and later agreed to make generous extensions and refinements of their work. I owe them a great deal of gratitude, for making this volume so exceptional. I would like to thank the Yoshida Hideo Memorial Foundation (Tokyo, Japan) for their generous financial support of the publication of this book. A special mention of appreciation must go to the Spanish Ministry of Science and Innovation (National Plan for Research, Development and Innovation EC02008-

VI

Preface

01557), Universidad Autónoma de Madrid, and the KDDI Foundation (Tokyo, Japan), for their financial support for the 9th ICORIA. I am also indebted to my colleagues, Sara Campo, Javier Oubiña, and Natalia Rubio for their superb job of planning, coordinating, and implementing the numerous tasks that made the conference successful. Finally, my grateful appreciation goes also to my student employee, Anne Schmitz, who compiled and formatted the chapters. I strongly hope this volume will prove to be a useful, enjoyable, and stimulating reference for academics, practitioners, and graduate students in this field.

Shintaro Okazaki Editor Madrid, April 2011

The European Advertising Academy (EAA)

The objective of the European Advertising Academy (EAA) is to provide a professional association to academics and practitioners interested in advertising and its applications that will promote, disseminate and stimulate high quality research in the field. The association particularly serves as a meeting and communication forum for its members. It offers a network for the exchange of knowledge on an international level and constitutes a framework allowing for a better dissemination of information on research and teaching. The EAA is closely related to the yearly International Conference on Research in Advertising (ICORIA). The purpose of this conference is to create a forum where people studying advertising in the academic world could exchange ideas, and where they could meet with practitioners who have experience with advertising in the commercial world. Every natural person that I professionally concerned with or interested in research or teaching in the field of advertising is, irrespective of nationality, eligible to become a full member of the organization. The EAA was founded in 2005. The current board members are: Peter Neigens (president, University of Amsterdam), Ralf Terlutter (president elect, University of Klagenfurt), Sandra Diehl (treasurer, University of Klagenfurt), Robert Heath (School of Management, Bath), Shintaro Okazaki (Autonomous University of Madrid), Patrick De Pelsmacker (University of Antwerp), Edith Smit (information manager, University of Amsterdam) and Tobias Langner (University of Wuppertal).

For further information please visit our website: www.icoria.org

Table of Contents

Preface

I.

Product Placement and Gaming Enrique Bigné, Inés Küster, Asunción Hernández, Torpong Suemanotham and Natalia Vila Product Placement in Video Games as a Marketing Strategy: An Attempt to Analysis in Disney Company

II.

VI

3

Yves Van Vaerenbergh, Dieneke Van de Sompel, Neal Van Loock and Iris Vermeir The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes: Does the Attitude toward the Artist Matter?

21

Martin K. J. Waiguny and Ralf Terlutter Differences in Children’s Processing of Advergames and TV Commercials

35

Social Issues and Advertising Martine Lewi and Patrick De Pelsmacker Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

55

X

III.

Contents

Nieves García de Frutos and José Manuel Ortega Egea Environmental Information Influences on Pro-environmental Behavior Aimed at Fighting Climate Change: Main and Moderating Effects

75

María José Miquel-Romero and Franco Manuel Sancho-Esper The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption: Affective Responses and Advertising Effectiveness

91

Erlinde Cornelis, Leen Adams and Verolien Cauberghe The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages for a Fair Trade Campaign

107

Marlize Terblanche-Smit and Nic S. Terblanche The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention in HIV/AIDS Fear Appeal Advertising

119

Generation Katarina Panic, Verolien Cauberghe and Delphine Verhoye The (B)old and the Beautiful: Investigating the Preference of Senior Consumers Concerning (the Age of) Models Used in Advertising

133

Julia Spielvogel and Ralf Terlutter Children’s Advertising Literacy: Do BMI, Body Shape Perception, Self-Esteem and TV Exposure Matter?

145

XI

Contents

Andrea Leopold and Sandra Diehl The Relevance of Life Changing Situations for Media Usage and their Relevance as a Segmentation Strategy for Media Companies and Advertisers

IV.

V.

161

Social Interaction and Advertising Shu-Chuan Chu and Sara Kamal An Investigation of Social Media Usage, Brand Consciousness, and Purchase Intention towards Luxury Products among Millennials

179

Rafael Currás-Pérez, María José Miquel-Romero, Carla Ruiz-Mafé and Silvia Sanz-Blas The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behavior

191

Manuela Lopez and Maria Sicilia The Impact of e-WOM: Determinants of Influence

215

Shintaro Okazaki and María Ángeles NavarroBailón Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

231

Organization, Publicity and Reputation An-Sofie Claeys, Verolien Cauberghe and Patrick Vyncke Determinants of the Impact of Crises on Organizational Reputation: An Experimental Test of Crisis Communication Strategies and the Moderating Impact of Locus of Control

247

XII

VI.

Contents

Luisa Andreu, Anna S. Mattila and Joaquin Aldás Effects of Message Appeal when Communicating CSR Initiatives

261

Martin Eisend and Franziska Küster The Effectiveness of Publicity versus Advertising: A Meta-Analysis

277

Branding Sophie C. Boerman, Edith G. Smit and Lex van Meurs Attention Battle; the Abilities of Brand, Visual, and Text Characteristics of the Ad to Draw Attention versus the Diverting Power of the Direct Magazine Context

295

Marieke L. Fransen and Thomas J. L. Van Rompay Living Brands: Consumer Reactions toward Online Experienced-based Marketing Communication

311

María Ángeles Navarro-Bailón What is the best Communication Strategy (Integration vs. Repetition) to persuade Consumers? An Analysis across Levels of Brand Familiarity

325

Nathalie Dens, Patrick De Pelsmacker and Bianca Puttemans Text or Pictures? Effectiveness of Verbal Information and Visual Cues in Advertisements for New Brands versus Extensions

341

XIII

Contents

VII.

VIII.

Media and Agency Sheila Sasser, Scott Koslow and Mark Kilgour Assessing the Quality of Self-reported Measures and the Reliability of Empirical Findings: Exploring Creativity Differences across Worldwide Agency Creatives and Managers

359

Charles R. Taylor and George R. Franke Public Opinion towards Digital Billboards in the United States: An Analysis of Recent Polls

373

Edward C. Malthouse and Don Schultz The Response Surface Process for Optimal Allocation of Media Budgets

393

Shintaro Okazaki, Morikazu Hirose and Hairong Li QR Code Mobile Promotion: An Initial Inquiry

405

Advertising Content, Appeals and Execution Lotte M. Willemsen, Peter C. Neijens and Fred E. Bronner Perceived Expertise vs. Perceived Trustworthiness: The Suppressed Effect of Source Type on Review Attitude

423

Günther Silberer, Sascha Steinmann, Inga Blohm and Julia Vetter A Study on the Impact of TV Advertising Contacts During Dinner

437

XIV

Contents

Tobias Langner and Martin Eisend Effects of Celebrity Endorsers’ Attractiveness and Expertise on Brand Recall of Transformational and Informational Products

451

Sandra Diehl, Shintaro Okazaki, and Barbara Mueller A Procedure for the Development of Fictitious Advertisements in Multi-country Research: Crafting Messages with Hard-sell and Soft-sell Appeals

461

Ioannis G. Theodorakis and Christos Koritos Figuratively Bleeding or Just Bleeding? Exploring Consumers’ Personal Values and Emotions within Simple and Rhetorically Constructed Violent Ad Contexts!

473

Part I

Product Placement and Gaming

Product Placement in Video Games as a Marketing Strategy: An Attempt to Analysis in Disney Company

Enrique Bigné, University of Valencia, Spain Inés Küster, University of Valencia, Spain Asunción Hernández, University of Valencia, Spain Torpong Suemanotham, University of Valencia. Spain Natalia Vila, University of Valencia, Spain

1

Introduction

What is the best marketing strategy in video games? The literature is not unanimous in upholding the best strategy. Researchers investigating advertising effectiveness disagree on the value of product placement in video games and the methods firms should use with product placement strategies. The state of the question now must figure out the dichotomy between two perspectives: to identify the differences in the effectiveness of product placement in video games when using “traditional placement approaches” compared to the “branded entertainment approach”. The “traditional approach” to product placement in video games involves the product appearing as a visual object that is passive, used by characters and players or mentioned in the audio, which functions as an advertisement for the product (Shrum, 2004). The “branded entertainment approach” to product placement in video games involves integrating a proprietary character into the theme or plot, with the branded product anthropomorphised (Winkler and Buckner, 2006). “Branded entertainment” differs from the “traditional approach” because the plot of the entertainment is fully focused on the branded product. Because of the growth in popularity of video games, advertisers are increasingly attracted to the medium as a means of reaching targeted age demographics that include children, young adults and adults between the ages of 18 and 34. Advertising expenditures for product placement in video games is forecasted at $733 million by 2010, with the majority of the advertising aimed at the target market of adult males who are the most common users of video games (Gerdes, 2006).

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_1, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

4

Bigné, Küster, Hernández, Suemanotham and Vila

Nevertheless, while advertisers are increasingly attracted to video games, the effectiveness of the medium and placement strategy remains uncertain. Also, even some evidence suggests that product placement in video games can increase brand and product awareness, previous results support that theses kind of product placement strategies are not useful to motivate the final purchasing the product (Nelson et al., 2004; Winkler and Buckner, 2006). The general attitude of the consumer toward advertising may also influence the effectiveness of product placement in video games (Beirne, 2008). In addition, the effectiveness of “branded entertainment approaches” to developing video game content used by the large multimedia firms, has not been extensively investigated (Winkler and Buckner, 2006). However, and despite what is mentioned in the preceding paragraph, we can assert that the way in which the placement occurs in a video game may be more critical for achieving the desired outcomes of increased awareness and motivating purchase decisions than with product placement in other types of media. The case study will focus on the Walt Disney Corporation (Disney), which uses product cycling in branded entertainment types of video games. Disney was selected for the case study because it has created numerous proprietary characters that can be considered product franchises because the characters appear in many different media and types of products. Because of the increased popularity in video games, Disney has also produced video games incorporating these characters. The product placement strategy of the firm in its video games is part of its integrated advertising and marketing strategy (Gilbert, 2005). So, the research design used a case study approach, which is suitable for investigating research questions intended to provide exploratory or descriptive information about a topic (Yin, 2003). The purpose of this research is to investigate the perceptions of marketing staffs and consumers towards product placement in video games and the effectiveness of the placement. The purpose of the research is exploratory and intended to determine whether marketing staffs and consumers perceive differences in the effectiveness of traditional product placement strategies and branded entertainment.

3

The Product Placement in Marketing. What’s the Effectiveness in Video Games?

Product placement is traditionally defined as ‘a paid product message aimed at influencing movie (or television) audiences via the planned and unobtrusive entry of a branded product into a movie (or television programme)’ (Balasubramanian, 1994: 29). It is a composite type of messages involving visual, auditory (Newell et al., 2006), and plot components that can potentially

Product Placement in Video Games as a Marketing Strategy

5

overcome some of the problems associated with traditional advertising in media such as television in which the viewer can delete or avoid advertising with zipping or zapping (Kuhn et al., 2004). From this perspective, product placement is related to branding theory because it is intended to increase awareness of the product and knowledge about product attributes (Haugtvedt et al., 2005). The implication of branding theory for product placement is the opportunity for the consumer to gains greater awareness of the subjective product attributes. Linking the product with the consumers’ positive perception about the entertainment character’s use of the product increases the subjective knowledge of the consumers’ about the product’s non-functional attributes (Shrum, 2004). The consumer is generally aware of the persuasive purpose of an advertising message and can easily ignore the message or engage in mental counterarguments reducing the effectiveness of the persuasion (Panda, 2004). With product placement advertising, however, the consumer theoretically has greater difficulty ignoring the message because it is intertwined with an entertainment communication in which the consumer has interest. Actually, the development of new forms of entertainment media requires an extension of the definition to include video games and other forms of computermediated entertainment. A video game can be broadly defined as a game based on software played on an electronic device allowing the presentation of graphical imagery and the interaction of the player with the software (Wolf, 2008). When the player becomes fully engaged with the gaming experience, the video game is considered as an immersive virtual environment (IVE) (Grigorovici and Constantin 2004). Galician (2005) indicated that the level of brand immersion within the IVE can range from associations to illustrations about use of the product depending on the way the brand is integrated into the tasks required for game play. The perceived quality of the video game by the player can affect the general attitude towards the game, which is analogous to the attitude of individuals towards other entertainment products (Ozer et al., 2009). Russell (1999) proposed a three-tiered framework for assessing the way in which a product is placed in a video game. The first tier involves screen placement in which the brand or product is visible in the screen frame. The second tier uses script placement in which the audio content mentions the brand or product. The third tier involves plot placement in which the brand or product is incorporated into the plot. These placements are analogous to the way in which product placement occurs in film. Because of the nature of the video game, a fourth tier to the

6

Bigné, Küster, Hernández, Suemanotham and Vila

framework includes the degree of interaction between the game player and the brand. In effect, the video game adds an additional dimension to product placement by allowing the user to interact with the product, which theoretically increases engagement and awareness. While the traditional approach to product placement in video games is to embed the product in the game play, content or motif, it is also possible to create a game solely around a brand or a branded character. These types of video games are referred to as ‘branded entertainment’ and ‘advergames’ (Winkler and Buckner, 2006). These games are often provided to consumers without cost and are played on the internet, although sometimes the firm owning the brand makes a nominal charge for the product (Mathieson, 2005). Yang et al. (2006) identified an emerging practice in video games played on the internet or requiring a periodic internet connection of rotating product placement. In effect, the branded products placed in the background of the game can change over time as the promotional contracts with the game sponsors change. This type of approach has also been used to promote new gaming products in the run-up to launch based, with the relatively short life-cycle of video games requiring firms to continuously develop new gaming products.

3.1

Effectiveness of Product Placement in Video Games

The evidence from research investigating product placement in video games has produced mixed findings. Effectiveness for product placement in video games is defined in various ways, which may account for the mixed findings. It is defined in terms of improving brand equity or the favourable impression of the brand (Nelson et al., 2004). Effectiveness is also defined as prompting the consumer to take an action such as an information search or the purchase of the product (Beirne, 2008). The approaches to determining effectiveness of product placement in video games are similar to the approaches used to assess product placement effectiveness in other media (Galician, 2005; Haugtvedt et al., 2005). Several researches have been made in this sense (Beirne 2008, Yang et al., 2004; Nelson et al., 2004) with different and contradictory results. Research conducted by Lee and Faber (2007) determined that many different factors can influence brand awareness with product placement in video games. The position of visual placements in the game must be sufficiently prominent to attract the notice of the gamer when engaged in a high involvement task. Experienced players are more likely to observe and assimilate the product placement message because they are less engaged with gaming tasks. The findings also showed that the degree of continuity or relationship between the

Product Placement in Video Games as a Marketing Strategy

7

product and the game content was a critical factor for the gamer to become aware of the placement. Hernandez et al. (2004) also determined that product placements in video games involve a fantasy or a deviation from the simulated reality of the game had a negative effect on the player’s attitude toward the brand. Grigorovici and Constantin (2004) determined that the level of engagement of the video game player with the IVE influences brand recall, with greater involvement reducing recall. Brand preference tends to increase with greater player engagement. The finding of Grigorovici and Constantin (2004) suggest that a subliminal process occurs during high engagement video game play increasing brand preference without increasing awareness or knowledge about the brand. According to Mathieson (2005), firms offering branded entertainment products via the internet to fixed or mobile playing platforms measure effectiveness based on the number of downloads. This approach solely examines the reach of the advertising and branding effort, with a download considered a customer contact. In addition, ratings agencies such as Nielson and game producers such as Activision have established tracking methods for internet game players to obtain information about the degree of engagement with the game and the type of brand information they have been exposed to because of play. This approach for determining effectiveness adopts the premise that exposure to the product information will lead to greater brand awareness and a favourable image.

4

Research Questions and Objectives

In view of above, the objective of this qualitative research is to answer the main research question: RQ: What are the differences in the effectiveness between” traditional product placement in video games” and the “branded entertainment approach to product placement in video games”? The question of this study aims to address the differences between two types of products placement: a traditional strategy (Winkler and Buckner, 2006) commonly used by advertisers, and a new one based on the central role that brand plays in entertainment (Hudson and Hudson, 2006). To compare the degree of effectiveness between both types of product placement, it is necessary to asses a subsidiary research question: RQ1: What are the strategies used for product placement in video games?

8

Bigné, Küster, Hernández, Suemanotham and Vila

In this sense, for product placement in film media, placement strategies include visual placement allowing the viewer to observe the produce, mention of the brand or product, and use of the brand or product by a character (Russell, 2009). These traditional techniques have been used in video games, but may involve different approaches to factors such as prominence or repetition (Yang et al., 2006). Video games may include other types of placement strategies including branded entertainment in which the game is played using a branded product as a theme or character (Mathieson, 2005). The different types of strategies may have an influence on the effectiveness of product placement in video games. Additionally, a second subsidiary research question is: RQ2: What are the perceptions of marketing staffs and consumers of the effects of “traditional product placement strategies” versus “branded entertainment in video games” on the general brand awareness? Product placement has influences on brand awareness, although uncertain exists in research concerning the degree of the influence (Beine, 2008; Ivory and Kalyanaraman, 2007; Shrum, 2004). Brand awareness has been assessed through recall measures (Galician, 2005; Law and Braun, 2000). Variability exists, however, in the way in which recall is assessed, with some approaches examining independent recall of the placement with no secondary stimulus (Law and Braun, 2000) and other approaches examining recall following a stimulus (Lehu and Bressoud, 2007). The branded entertainment approach to product placement may result in differences in perceptions concerning awareness of the brand in addition to merely recalling or recognizing the brand after the entertainment product has been used. Also, and as a consequence of the main research question, a third subsidiary research question could be: RQ3: What are the perceptions of marketing staffs and consumers of the effects of “traditional product placement strategies” versus “branded entertainment in video games” on the awareness of specific brand attributes? This aspect of product placement has not been extensively examined in previous research. Winkler and Buckner (2006) as well as Yang et al. (2004) determined that product placement in video games increased brand knowledge, but did not examine the difference in effects between traditional placement and branded entertainment. The “branded entertainment approach” to product placement may provide the consumer with greater knowledge about the range of attributes the company desires to associate with a product.

Product Placement in Video Games as a Marketing Strategy

9

A final research question is: RQ4: How do demographic factors influence perception of the effectiveness of “traditional product placement strategies” versus “branded entertainment in video games”? Demographic factors can influence responsiveness and attitudes toward product placements in entertainment media (Galician, 2005). Culture can also influence attitudes towards product placement and the effectiveness of various placement strategies (Gould et al., 2005). The general theoretical model is based on the assumption that a relationship exists between product placement strategies in video games and the effectiveness of the placement for increasing brand recognition, awareness of the brand’s attributes, and a favourable impression of the brand. The theoretical model also presumes that many intervening factors can influence the relationship between product placement strategies and effectiveness such as the perceived entertainment value of the game, the degree of immersion in game play, and the thematic content of the game. As a result demographic factors function as moderating variables, influencing the relationship between product placement strategies and effectiveness.

5

Methodology

The selection of Disney for the case study was based on defined criteria and the information-oriented selection process recommended by Flyvbjerg (2006). The criteria included the use of product placement by the firm as a marketing strategy, the use of “branded entertainment approach” as part of the product placement approach, and sufficient video game products to allow consumers to be familiar with the content and product placement in the products. An additional criterion was the willingness of marketing staffs from the firm to provide data for the study. The information-oriented selection process sorts firms meeting the initial inclusion criteria based on a specific characteristic intended to provide the desired type of information. Because of its size and planned marketing procedures, Disney was considered a critical case in which conclusions about the use of product placement in the firm could be considered representative of product placement in the video gaming industry. The data collection consisted of (1) interviews with marketing staffs from the firm, (2) a focus group conducted with consumers of Disney video games, and (3) review of information available to the public from secondary sources about Disney video games. The interview is appropriate for data collection in qualitative research because it allows the interviewees to establish the direction taken when answering open-ended questions (Darlington and Scott, 2002). The

10

Bigné, Küster, Hernández, Suemanotham and Vila

focus group is also appropriate for gathering qualitative data because it allows the participants to discuss theoretical issues using the language and terms that construct meaning from their perspective (Darlington and Scott, 2002). Defined criteria were used to identify individuals to participate in the interview and focus group segments of the study. Because of the inclusion criteria, a purposive sampling approach using a snowball methodology was used to identify participants in the study (Gliner and Morgan, 2000). The selection process resulted in four interviewees agreeing to participate in the study and nine participants in the focus group. Table 1 and Table 2 shows some information related to the participants. Table 1. Demographic Information Interviewees

________________________________________________________________ Interviewee Number 1 2 3 4 ________________________________________________________________ Date of Interview 7/7/2009 7/7/2009 9/7/2009 12/7/2009 Position Marketing Product Production Marketing Team Manager Team Team Assistance DIMG DIMG DIMG DIMG Gender M M M M Age 37 41 29 33 Exper. at Disney 8 years 3 years 4 years 6 years ________________________________________________________________

The data analysis procedure involved synthesising the data obtained from the interviews, focus group, and secondary sources to create the case study. The primary data from the interviews and focus group was initially analysed using a thematic content approach. Content analysis is an appropriate method for analysis when the research design is qualitative and the purpose of the research is to obtain information about factors that may be causally related to the phenomenon under investigation (Creswell, 2007). The content analysis identified the themes and patterns within the data related to the research questions. It consisted of the three stages of open coding, axial coding and narrative as recommended by Daymon and Holloway (2002). The findings of the content analysis was combined with the secondary data to develop the case study of Disney’s use and perceived effectiveness of product placement and branded entertainment in video games. The reliability and validity of the data was established based on the trustworthiness criteria used in qualitative research as recommended by Daymon and Holloway (2002). Trustworthiness in

Product Placement in Video Games as a Marketing Strategy

11

qualitative research is composed of the elements of credibility and dependability of the methods, and applicability to the research problem, and ability to confirm the findings of the study. Table 2. Participants in the Focus Group

________________________________________________________ Participant Age Gender Occupation ________________________________________________________ 1 32 F Shop Worker 2 37 F County Council Employee 3 27 M Shop Worker 4 39 F Estate Agent 5 42 M IT specialist 6 23 M Student 7 29 F Junior School Teacher 8 36 F Shop owner 9 34 F Chef ________________________________________________________

5

Results and Discussion

The results of the study were organised in four sections (1) a description and content analysis of the interviewees from Disney staff, (2) a description and content analysis of the focus group participants, (3) the case study of Disney based on a synthesis of the data, and (4) the discussion linking the findings to the literature review. Only some results from fourth section are shown due to space limitations. In this sense, the findings suggest that Disney uses a subtle approach to product placement involving the use of branded characters rather than placing specific products into the background or incorporating the product into the game play. Based on the framework for assessing product placement developed by Russell (1999), Disney uses the most sophisticated product placement strategy by incorporating the product, which is the branded character, into the action of the game play. The effect is to create closer alignment between the entertainment product in the form of the video game and the branded product in the form of the character, which is the outcome of branded entertainment proposed by Hudson and Hudson (2006).

12

Bigné, Küster, Hernández, Suemanotham and Vila

The interviewees indicated that the current strategy is to use video games as an extension of a branded entertainment product based on the assumption that increasing the number of game players increases the general market awareness of the branded character, and therefore leads to sales of other products associated with the character. The results also suggest that the “branded entertainment approach” to product placement may be more effective for increasing brand awareness and a favourable impression of the brand than the more “traditional product placement approaches” of depicting a product in the background or the use or mention of a product by a character. Previous researchers have also found that carefully managed product placement in video games can increase brand awareness among gamers (Kuhn et al., 2004; Winkler and Buckner, 2006; Yang et al., 2004). The concern of many focus group participants that product placement in all entertainment media including video games has a stronger effect on children conforms to the findings of Calvert (2008). The participants suggested that children do not recognise a product placement incorporated into the theme or content of the game as advertising or marketing. The results further suggest that Disney does not excessively use product placement in video games (is not intrusive); with the participants in the focus group indicating that the use of branded characters in the game is appropriate. The findings also suggested that determining the effectiveness of product placement strategies in video games is complex because product placement interacts with other variables such as the degree of general engagement with the game. These results are similar to those of Grigorovici and Constantin (2004) who determined that the perceived entertainment value of the video game influences the gamers’ attitudes toward product placement. The participants of the focus group reached no conclusion concerning the perceived effect of the video game on the purchase intentions of the children playing the games. The participants also suggested that the purchase intentions are driven by the affinity of the child for the branded character, with any product associated with a likeable character considered desirable. The results provide evidence to refute the argument of Winkler and Buckner (2006) that branded entertainment should be considered an advergame under all situations. The interviewees and participants in the focus group considered the video games to be primarily an entertainment medium with some advertising or branding elements incorporated into the game. The participants in the focus group distinguished between a video game created around an existing product and a video game using a character created for entertainment purposes.

Product Placement in Video Games as a Marketing Strategy

6

13

Conclusions and Recommendations

The findings of the study supports the conclusion that the branded-entertainment approach to product placement in video games is more effective for achieving greater brand awareness and positive brand perceptions than the traditional product placement approach of placing the product into the content of the video game. The traditional approach to product placement in video games is to embed an audio or visual reference to the brand in the game or to have the player use a branded product to perform a task associated with game play (Russell, 1999). This approach appears to be a direct transfer of the techniques used in film, television and other passive entertainment media. In the brandedentertainment approach in video games, the entire game functions as a product placement, with the gamer interacting with the branded character (Winkler and Buckner, 2006). The findings of the study indicated that Disney, the producer of the video games using the branded entertainment approach, believed that it was more effective for increasing awareness among consumers of the attributes associated with both the branded character and the overall Disney brand. The findings of the study also indicated that consumers of the video games find the use of a branded character less objectionable than the use of a direct product placement as occurs in other entertainment media and video games that do not use branded entertainment. The findings of the study also support the conclusion that the strategy used for product placement in video games can have an influence on brand awareness. Prior research has determined that product placement in video games affects brand awareness because the gamer becomes immersed in play of the game and reduces mental barriers to processing advertising and brand information (Yang et al., 2006). The findings of the study suggested that branded entertainment in which the brand is the subject of all video game activities may be highly effective for indirectly or implicitly communicating information about brand attributes to the game player. The interviewees from Disney indicated that the approach was effective for communicating the attributes of both the character and the Disney brand. The participants in the focus group also indicated that they had a positive perception of the Disney brand because of its use of branded characters in video games. An additional conclusion of the study is that product placement in video games has the potential to generate negative brand equity if the placement is intrusive or perceived as a deceptive advertising practise. If the brand placement is incongruent with the content of the video game and if the brand is excessively repeated in the game, it contributes to the perception that the brand placement is intrusive (Hernandez et al., 2004). The perception of intrusiveness results in negative attitudes toward the brand and an ultimate reduction in brand equity. In

14

Bigné, Küster, Hernández, Suemanotham and Vila

a worst-case situation, gamers become so annoyed with the brand placement that they avoid purchasing the brand. The findings of the study, however, indicated that this type of problem with the perception of intrusiveness is less likely to occur with branded entertainment. The gamer has an expectation that the brand in the form of the character will be an integral aspect of game play. In effect, the gamer accepts that a character product will be part of the gaming activity before engaging in play. The study, however, did not provide sufficient information to determine the degree that this expectation of encountering a branded character during the course of play affects the way in which the gamer processes the implicit marketing message. Theoretically, consumers process product placement information differently from other forms of advertising (Shrum, 2004). The mechanisms that consumers used to process branded entertainment messages may also differ from the mechanisms used to process information conveyed by traditional product placements. The issue of greater acceptability among consumers of product placement in video games by using a branded entertainment approach is particularly relevant to large multinational firms that create video game products for international distribution. Cultural differences can have a significant influence on the perception of the appropriateness of product placement strategies (Stafford and Faber, 2005). As a result, the branded character approach with no other brands placed in the video game may provide multinational firms developing products for international sales with a general product placement strategy that is acceptable across cultures. Disney uses only a branded entertainment approach to product placement in video games and very sparingly places brands not already associated with the branded character into the video game. As a result, the firm increases the marketability of the product in all geographic regions where the video game is distributed. The conclusion that can be drawn from the findings is that a branded entertainment approach in which the only placement involves a branded character may be more effective for firms in international markets for increasing acceptance of the product than the traditional approach of video or audio insertions of a product into a video game. An additional conclusion of the study is that the demographics of the target market for a video game should influence the product placement strategy. The Disney video games are targeted at a child and young adolescent demographic. In theory, this demographic is more susceptible to marketing messages using product placement (Calvert, 2008). At the same time, video games aimed at this market have to ensure they conform with regulations in different nations concerning marketing to children as well as the perceptions of the adults that often make the purchases of video games. The findings of the study support the conclusion that the adults involved with video gaming with children find the

Product Placement in Video Games as a Marketing Strategy

15

branded entertainment approach an acceptable marketing method. Although playing the video game reinforces the interest of the child in the entertainment and other products related to the character, the adults perceive the marketing aspects as secondary or subordinate to the entertainment value. In addition, findings indicated that Disney designs its video games to market its branded characters to children in a manner that adults deem acceptable. The findings of the study also support the conclusion that the branded entertainment approach to product placement can be used to increase existing brand knowledge and to introduce the brand to consumers that have no prior knowledge of the brand. In theory, product placement in video games as well as other media functions as a means to communicate information about the attributes of the brand by associating the brand with the content of the game (Pardun and McKee, 2000). The purchase or internet play of a video game with a specific character depends to some degree on the prior awareness of the consumer of the attributes of the branded character. In most cases, the consumer is initially attracted to the branded entertainment video game because of a preexisting affinity for the branded character as well as the presumed entertainment value of the game. As a result, the consumer often has some level of knowledge and awareness of the attributes associated with the character. The video game functions to reinforce and build on an existing knowledge base. Ideally, the outcome is for the consumer to have a positive gaming experience, increasing awareness and affinity for the branded character. At the same time, the video game has to be designed to provide some information about character attributes to consumers that have no prior knowledge of the character. This dual process suggests that the branded entertainment approach to video game content is more complex than the traditional approaches in which products are merely placed into the game environment. As noted by the interviewees from Disney, the development of a video game using a branded character requires close coordination among software developers, game content planners, and marketing.

6.1

Limitations of the Research

The findings and conclusions of this research are limited by the use of the case study approach, which focused on the video game products produced by Disney. The Disney brand is well-established in the market for children’s and young adult entertainment products. As a result, the firm has a strong emphasis on protecting its existing brand equity and avoids using product placement strategies that the firm perceives as representing a risk to its brand equity. The findings and conclusions of the study may not reflect the attitude of consumers towards product placement and branded entertainment products of firms that do not have a well-established reputation in the marketplace. In addition, the

16

Bigné, Küster, Hernández, Suemanotham and Vila

conclusions and findings of the study may not reflect the perspectives of adult gamers toward product placement in video games. In addition, Disney does not use a range of different product placement strategies in video games, which limits the ability to assess the perception of differences among consumers of the product.

6.2

Recommendations for Future Research

While the findings and conclusions of this study indicated that the branded entertainment approach to product placement in video games is more effective in a children’s target market than the traditional approaches to product placement, it did not assess product placement strategies among other demographic groups. Future research should examine the specific attitudinal, awareness, and recall responses of consumers through a direct comparison of the different types of product placement strategies used in video games. This type of research should assess the response of consumers in carefully defined demographic segments to products placed in video games using various combinations of audio, visual and character use of the product to more carefully determine the differences between these product placement strategies and the branded entertainment product placement strategy. Future research should also examine the extent that branded entertainment approaches to product placement in video games can be linked to traditional product placement methods. The combination of branded characters and traditional product placement has been successful in film for creating brand awareness for the product when it is associated with the branded character and the product is relevant to the plot of the film. It is not certain, however, whether this same effect would occur with a video game.

7

References

Auty, S. & Lewis, C. (2004). Exploring children’s choices: The reminder effect of product placement. Psychology & Marketing, Vol. 21, No. 9, 699-716. Balasubramanian, S. (1994). Beyond advertising and publicity: hybrid messages and public policy issues. Journal of Advertising, Vol. 23, No. 4, 29-46. Beirne, M. (2008). Study: Gamers respond to ads. Brandweek, Vol. 49, No. 32. Available at:http://www.technologymarketing.com/bw/content_display/news-andfeatures/digital/e3ibd93dba87a9330a3ed295b7ba76e8e25. [Accessed 19 June 2009]. Brookey, R.A. & Westerfelhaus, R. (2005). The digital auteur: Branding identity on Monsters Inc. DVD. Western Journal of Communications, Vol. 69, No. 2, 109-128. Calvert, S. (2008). Children as consumers. The Future of Children, Vol. 18, No. 1, 205-232.

Product Placement in Video Games as a Marketing Strategy

17

Chiou, J. (2000). Antecedents and moderators of behavioural intention: Differences between U.S. and Taiwanese students. Genetic, Social and General Psychology Monographs, Vol. 26, No. 1, 105-124. Cowley, E. & Barron, C. (2008). When product placement goes wrong: The effects of program liking and placement prominence. Journal of Advertising, Vol. 37, No. 1, 89-98. Creswell, J.W. (2007). Qualitative inquiry and research design. Thousand Oaks CA: Sage Publications Croteau, D. & Hoynes. W. (2005). The business of media. Thousand Oaks CA: Pine Forge Press. Darlington, Y. & Scott, D. (2002). Qualitative research in practice. Crows Nest NSW: Allen & Unwin. Daymon, C. & Holloway, I. (2002). Qualitative research methods in public relations and marketing communications. London: Routledge. Flyvbjerg, B. (2006). Five misunderstandings about case study research. Qualitative Inquiry, Vol. 12, No. 2 Galician, M. (2005). Handbook of product placement in the mass media. Binghamton NH: Haworth Press. Gerdes, L. (2006). Ads People Play. Business Week, No. 401. Gilbert, J. (2005). Jazz, gestalt and the year ahead for marketers. Journal of Advertising Research, Vol. 45, No. 3, 294-295. Gliner, J. & Morgan, G. (2000). Research methods in applied settings. Mahwah NJ: Lawrence Erlbaum Associates. Grigorovici, D. & Constantin, C. (2004). Experiencing interactive advertising beyond rich media. Journal of Interactive Advertising, Vol. 5, No. 1. Available at: http://jiad.org/article53. [Accessed 12 July 2009]. Gould, S., Gupta, O. & Grabner-Krauter (2000). Product placement in movies: A cross-cultural analysis of American, French and Austrian consumer attitudes. Journal of Advertising, Vol. 29, No. 4, 41-58. Haugtvedt, C., Machleit, K. & Yalch, R. (2005). Online consumer psychology. New York: Routledge. Hernandez, M., Chapa, S., Minor, M., Maldonado, C., & Barranzuela, F. (2004). Hispanic attitudes toward advergames. Journal of Interactive Advertising, Vol. 5, No. 1. Available at: http://jiad.org/article57. [Accessed 13 July 2009]. Hill, L. (2007). From one franchise to another: CSI spawns video game series. Television Week, Vol. 26, No. 18. Homer, P.M. (2009). Product placements: The impact of placement type and repetition on attitude. Journal of Advertising, Vol.38, No. 3, 21-32. Hudson, S. & Hudson, D. (2006). Branded entertainment: A new advertising technique of product placement in disguise? Journal of Marketing Management, Vol. 22, nos. 5-6, 489-504. Ivory, J. & Kalyanaraman, S. (2007). The effects of technological advancement and violent content in video games on players’ feeling of presence, involvement, physiological arousal and aggression. Journal of Communications, Vol. 57, 532-555. Keller, K. (2003). Brand synthesis: The multidimensionality of brand knowledge. Journal of Consumer Research, Vol.29, No.4, 595-600. Kreuger, R.A. (1998). Analysing and reporting focus group findings. Thousand Oaks CA: Sage Publications. Kuhn, K., Love, A. & Pope, N. (2004). Brand placements in video and computer games: An overview and research questions. Available at: http://www98.griffith.edu.au/dspace/bitstream/10072/2325/1/26471_1.pdf. [Accessed 13 July 2009].

18

Bigné, Küster, Hernández, Suemanotham and Vila

Law, S. & Braun, K. (2000). I’ll have what she’s saying: Gauging the effect of product placement on viewers. Psychology and Marketing, Vol. 17, No. 12, 1059-1075. Lee, M. & Faber, R. (2007). Effects of product placement in on-line games on brand memory. Journal of Advertising, Vol. 36, No. 4, 75-90. Lenderman, M. (2006). Experience the message. New York: Carroll & Graf. Lehu, J.-M., & Bressoud, E. (2007). Viewers and brand placement in movies. La Londe Seminar in Marketing Communications and Consumer Behaviour, La Londe, 5-8 June. Available at: http://hal.archivesouvertes.fr/docs/00/34/45/35/PDF/Lehu_Bressoud_2007_Brand_PLacement_viewers_La_Londe .pdf. [Accessed 14 July 2009]. MacDonald, E. & Sharp, E. (2003). Management perceptions of the importance of brand awareness as an indicator of advertising effectiveness. Marketing Bulletin, Vol. 14, 1-11. Mathieson, R. (2005). Branding unbound. New York: Amacom. Milberg, S.J., Park, C.W. & McCarthy, M. (1997). Managing negative feedback effects associated with brand extension: The impact of alternative branding strategies. Journal of Consumer Psychology, Vol. 6, No. 2, 119-140. Nelson, M., Keum, H., & Yaros, R. (2004). Advertisement or adcreep: Game players attitude toward advertising and product placement in computer games. Journal of Interactive Advertising, Vol. 5, No. 1. Available at: http://www.jiad.org/article52. [Accessed 19 June 2009]. Newell, J., Salmon, C., & Chang, S. (2006). The hidden history of product placement. Journal of Broadcasting and Electronic Media, Vol. 50, No. 4, 575-603. Ozer, A., Argan, M., Suher, H., Akin, E., & Argan, M. (2009). Game Quality: A Study on Turkish Game Players. Available at: http://bildiri.anadolu.edu.tr/papers/bildirimakale/102_923y83.pdf. [Accessed 8 July, 2009]. Panda, T. (2006). Consumer responses to brand placement in films. South Asian Journal of Management, Vol. 11, No. 4, 7-25. Pardun, C. & McKee, K. (2000). Product placements as public relations. Public Relations Review, Vol. 25, No. 4, 481-493. Reijmersdal, E., Niejens, P., & Smit, E. (2007). Effects of television brand placement on brand image. Psychology and Marketing, Vol. 24, No. 5, 403-420. Rosini, J., Mash, M., & Rupp M. (2008). Hot licensing issues in the video game industry The Licensing Journal, (February), 14-23 Russell, C. (1998). Toward a framework for product placement: Theoretical propositions. In Alba, J. & Hutchinson, J. (Eds.) Advances in consumer research, Vol. 25. Provo UT: Association for Consumer Research. Sargent, J., Tickle, J., Beach, M., Dalton, M., & Ahearns, M. (2001). Brand appearances in contemporary cinema films and contribution to global marketing of cigarettes. The Lancet, Vol. 357, No. 9249, 29-32. Schultz, M., Antonini, Y., & Csaba, F. (2005). Corporate branding. New York: CBS Press. Shrum, L. (2004). The psychology of entertainment media. Mahwah NJ: Lawrence Erlbaum Associates. Stafford, M. & Faber, R. (2005). Advertising promotion and the new media. Armonk NY: M.E. Sharpe. Turner, C. (2004). Product placement of medical products. Journal of Promotional Management, Vol. 10, No. 1, 159-170. Vorhaus, M. (2009). Ad presence in console games has little effect on purchase. Advertising Age, Vol. 80, No. 5. Walsh, P., Kim, Y. & Ross, S. (2008). Brand recall and recognition: A comparison of television and sport video games as presentation modes. Sport Marketing Quarterly, Vol. 17, 201-208.

Product Placement in Video Games as a Marketing Strategy

19

Walt Disney Company. (2009). Annual report 2008. Available at: http://amedia.disney.go.com/investorrelations/annual_reports/WDC-AR-2008.pdf. [Accessed 18 June 2009]. Winkler, T. & Buckner, K. (2006). Receptiveness of gamers to embedded brand messages in advergames. Journal of Interactive Advertising, Vol.7, No. 1. Available at: http://jiad.org/article85. [Accessed 19 June 2009]. Wolf, M. (2008). The video game explosion. Westport CT: Greenwood Press. Yang, M., Roskos-Ewolden, D., Dinu, L. & Arpan, L. (2006). The effectiveness of in-game advertising. Journal of Advertising, Vol. 4, No. 4, 143-152. Yaveroglu, I. & Donthu, N. (2008). Advertising repetition and placement issues in online environments. Journal of Advertising, Vol. 37, No. 2, 31-43. Yin, R. (2003). Case study research: design and methods. Thousand Oaks CA: Sage Publications.

20

Bigné, Küster, Hernández, Suemanotham and Vila

Appendix A. Interview Questions 1. What is Disney’s approach to product placement in its entertainment products? 2. How does Disney’s approach to product placement in video games differ from product placement in its films or other entertainment products? 3. How does Disney use branded entertainment? (Explain: Branded entertainment refers to the use of a branded character as the central character of the story line or content in a video game.) 4. How does Disney assess the effectiveness of product placement and branded entertainment in video games? 5. What is the relationship between product placement in video games and consumer awareness of the Disney brand and its attributes? 6. In what ways can product placement in video games create negative attitudes towards brands among the players of the games? 7. What methods does Disney use to ensure that product placement and branded entertainment video games result in a positive attitude toward the Disney brand among players?

Appendix B. Focus Group Guidelines Moderator will initially define the meaning of the product placement and branded entertainment concept. The moderator will also ensure that the participants understand the concepts. 1. What is your attitude toward product placement in video games? 2. Do you have a different attitude towards product placement in video games than in films or television shows? 3. What type of product placement in video games do you find annoying? 4. Have you ever played a Disney video game online? If so, which ones? 5. What do you remember about your experiences playing these games? 6. Have you ever played a Disney video game offline? If so, which ones? 7. What do you remember about your experiences playing these games? 8. What is your attitude towards Disney games that are based on a Disney character such as Hannah Montana or Bolt? 9. If you played a Disney video game with children, do you think that the characters or products used in the game had an influence on their attitudes? 10. Did playing a Disney video game motivate you to purchase another Disney product related to the character in the video game?

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes: Does the Attitude toward the Artist Matter? Yves Van Vaerenbergh, University College Ghent, Belgium Dieneke Van de Sompel, University College Ghent, Belgium Neal Van Loock, University College Ghent, Belgium Iris Vermeir,Ghent University, Belgium

1

Introduction

As the overload of advertising in traditional media is causing annoyance with the audience (Fennis and Bakker, 2001) leading to advertising avoidance (Woltman-Elpers et al., 2003), marketers are exploring various new ways to convey brand-related messages to consumers. One of the proposed techniques is product or brand (name) placement, which is defined as the incorporation of a brand name into an entertainment vehicle (Ferraro and Avery, 2000). Past research concerning product placement mainly focused on product and brand name placement in movies (Babin and Carder, 1996; d’Astous and Chartier, 2000; Wiles and Danielova, 2009; Yang and Roskos-Ewoldsen, 2007), television programs (Gould and Gupta, 2006; La Ferle and Edwards, 2006; Matthes et al., 2007; Russell, 2002; Russell and Stern, 2006; Van Reijmersdal et al., 2007), music videos (Schemer et al., 2008), games (Lee and Faber, 2007; Mackay et al., 2009; Nelson, 2002; Schneider and Cornwell, 2005) or novels (Brennan, 2008; Friedman, 1985). Recent academic research suggests that brand placement can be implemented by including brand names in song lyrics (Delattre and Colovic, 2009), which often occurs in practice. Agenda Inc. created the American Brand Stand, which tracked all the mentions of brand names in the lyrics of the Billboard Top 20 singles chart in 2003 and 2004. In 2003, Mercedes was the most-mentioned brand name with 114 mentions in lyrics, and in 2004 Cadillac was mentioned 70 times. Other oftenmentioned brands were Hennessy, Rolls Royce, Gucci, to name but a few (Agenda Inc, 2004). These results indicate that the use of brand names in song lyrics is a common practice. For example, McDonald’s cooperated with Maven Strategies, a professional agency which sought rappers to incorporate “Big Mac” in their texts. McDonald’s was prepared to pay 1 to 5 dollar each time the brand name was mentioned on the radio in their song lyrics (Graser, 2005). On the other hand, the hip-hop band Outkast advised their audience to “Shake it Like a Polaroid Picture” and Busta Rhymes wrote a song called “Pass the CourS. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_2, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

22

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

voisier”, which led to a significant increase in sales for the Courvoisier brand. Important to note is that these artists were not paid at first for this action, but a settlement was arranged afterwards (Debruyne, 2007, Holloway, 2002). Consequently, organizations cannot always exert control over which artists include their brand name in song lyrics. If organizations select the artists themselves, they can pick artists who are positively perceived by the audience, whereas artists who include brand names freely might be negatively perceived by the audience. In this paper, we investigate the moderating role of the attitude toward the artist on the relationship between the amount of brand name repetitions in song lyrics and the attitude toward the brand. Our study contributes to the literature in the following ways. First, prior research only demonstrated a link between the amount of brand name repetitions in song lyrics on brand recall and recognition (Delattre and Colovic, 2009). Second, we extend prior research by examining the moderating role of attitude toward the artist. To date, research on brand name placement in song lyrics did not yet address this issue, although highly relevant given that organizations are faced with the potential danger that their brand name is included in a song by an artist which is negatively evaluated by the audience. Prior research demonstrated evaluative conditioning effects for product placement in music videos. Schemer et al. (2008) found that the pairing of a brand with positively evaluated artists produces positive attitudes toward the brand, while the pairing of a brand with negatively evaluated artists produces negative attitudes toward the brand. However, to date it remains unknown whether the same evaluative conditioning effects hold for brand name placement in song lyrics instead of music videos. Moreover, Schemer et al. (2008) did not investigate evaluative conditioning effects across a different number of pairings, although research on evaluative conditioning shows that evaluative conditioning effects become stronger across multiple pairings (De Houwer et al., 2001, 2005; Hoffman et al., 2010). In order to assess the impact of the amount of brand name placement in song lyrics on brand attitudes and the moderating role of attitude toward the artist, we rely on an experimental design making use of a song especially developed for the purpose of this study. In order to test whether brand placement effects become stronger with different numbers of brand exposures, three different versions of the song were created, with zero, two or five brand name placements.

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

2

23

Theoretical Background and Hypothesis Development

To the authors’ knowledge, only one study investigated the effectiveness of brand name placement in song lyrics. Delattre and Colovic (2009) selected two well-known songs (over 300,000 units sold) for the purpose of their research, namely ‘Tes Parents’ by Vincent Delerm and ‘Wonderbra’ by MC Solaar. In both songs, 17 brand names were included. Their results indicate that listeners recalled and recognized brands in both songs, especially in cases where the brand name placement was prominent (occurred at least twice) and pronounced slowly. However, their research did not address the impact of brand name placement in song lyrics on brand attitudes. Addressing this issue is important given that past literature often found an absence of correlations between memory (brand recall and recognition) and attitude measures (Cowley and Barron, 2008; Petty et al., 1983). Therefore, recall and recognition measures might be poor predictors of persuasion (Mackie and Asuncion, 1990). As a result, investigating brand attitudes might enhance our understanding of the effectiveness of brand name placement in song lyrics. In a broader context, research in the area of product and brand name placement found that a placement has the ability to positively influence brand attitudes. For example, Matthes et al. (2007) found that the frequency of placement has a positive impact on the evaluation of the brand in a TV show: the more frequent a brand is placed within a television show, the more favorable the attitudes toward that brand become. In a similar vein, Van Reijmersdal et al. (2007) showed that the image of brands placed in a TV show changed after two or more exposures. These results indicate that the brand attitudes should become more favorable with an increasing number of repetitions. This paper proposes that the effect of brand name repetition on brand attitude in song lyrics will vary depending on audiences’ attitude toward the artist. Theoretical anchoring for this expected effect can be found in conditioning theory. The most well-known experiment in classical conditioning is the one from Pavlov, who paired an unconditioned stimulus (dog food) with a conditioned stimulus (a ringing bell). After repeated pairings of unconditioned and conditioned stimulus, responses which only occur after the unconditioned stimulus also occur after the conditioned stimulus. That is, the dogs in Pavlov’s experiment already started salivating when hearing bell ringing (Schiffman and Kanuk, 2006). A variation on classical conditioning theory is evaluative conditioning, which is defined as the changes in the evaluation of a stimulus that result from pairing that stimulus with other positive or negative stimuli (De Houwer et al., 2001, 2005; Hoffman et al., 2010). The basic idea is that if a certain object is associated with a positive stimulus, a person might have a

24

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

positive evaluation of that object. In contrast, if a certain object is associated with a negative stimulus, a person’s evaluation of that object might become negative. Consumer research investigated how evaluative conditioning can be used to influence brand attitudes. For example, an experiment by Sweldens et al. (2010) paired a brand with positive and neutral images and found that a brand associated with positive images was evaluated more positively by the participants than the same brand which was associated with neutral images. In an advertising context, conditioning has been applied to explain the effectiveness of celebrity endorsement (Till et al., 2008). If a product is paired with a fameous celebrity, it is expected that the celebrity can transfer meaning to the product (McCracken, 1989). In contrast, if a product is paired with a negatively perceived celebrity, this might have a negative impact on the brand attitude (Till and Shimp, 1998). To date, only one study investigated the impact of evaluative conditioning in a product placement context. Schemer et al. (2008) investigated how the evaluation of a brand placed in a music video differed depending on the audience’s attitude toward the artist. Their findings indicated that the pairing of a brand with a negatively perceived artist led to lower brand attitudes than the pairing of a brand with a positively perceived artist. In this research, we aim to replicate the findings of Schemer et al. (2008) in a different product placement context (song lyrics versus music video) and extend them by investigating different amounts of brand exposures. Prior research on evaluative conditioning indicated that evaluative conditioning effects become stronger when multiple pairings of conditioned and unconditioned stimuli occur (Baeyens et al., 1992; McSweeney and Bierley, 1984). In our research context, we might expect that the more a brand name is mentioned in a song lyric by a positively evaluated artist, the more positive the brand attitude might become. In contrast, the more a brand name is mentioned in a song lyric by a negatively evaluated artist, the more negative the brand attitude might become. This leads to the following hypothesis: H1: The attitude toward the artist moderates the relationship between brand name repetition in song lyrics and the brand attitude.

3

Method

Two hundred and two undergraduate students (43.6% male, Mage=19.94) participated in a 3x2 between-subjects experiment, in which we manipulated the occurrence of a brand name (mentioned zero, two or five times) and the attitude toward the artist (positive or negative). As Martin and Collins (2002) found that brands or products were mostly depicted in hip-hop and rap music videos, we

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

25

used a hip-hop song as stimulus in our research. A study by Herd (2005) indicated that between 1979 and 1997, the number of hip-hop songs that include alcohol beverage brand name references increased from 0% to 71%, indicating that brand name placement is common in hip-hop songs. Given that youngsters are the privileged targets of the music industry (Delattre and Colovic, 2009; Schemer et al., 2008) and thus students represent a large part of the hip-hop audience, the use of a student sample in this study is justified. In this research, we applied the Theater Methodology as introduced by Russell (2002). This involved the creation of a stimulus (in this case, a song) that is completely under the experimenters’ control instead of using existing songs. This way, we rule out unwanted noise which might challenge the internal validity of results. Consequently, a hip-hop song was especially developed for the purpose of this study, for which we cooperated with an experienced but to the participants’ unknown hip-hopper. Three identical versions of the song were recorded which only differed on the number of brand name repetitions. All three versions of the song were sung in the native language of the hip-hopper (and respondents) and had duration of 2’13. The production of the stimulus material applied to professional standards, which ensured that the song did not raise the audiences’ suspicion of lacking authenticity (Yang and Roskos-Ewoldson, 2007). In the song, the artist sang about his life as hip-hopper and his devotion to hip-hop music. In case no brand name was added to the text, the artist mentioned listening to music on a “MP3 player”. In case a brand name was added to the text, “MP3 player” was replaced by “Jo-Box!”, which acts as a fictitious brand name for a MP3 player. By using an unknown brand name, potential confounding elements such as prior brand knowledge, attitude and usage could not influence our results. In a procedure similar to Schemer et al. (2008), participants’ attitude toward the artist was manipulated by using two different texts which provided background information about the artist. In the “negative attitude toward the artist”condition, the hip-hopper was depicted as greedy by exploiting his fans and overpricing his CDs. Moreover, he has been arrested for several crimes such as acts of violence and the use of drugs. In the “positive attitude toward the artist”condition, the hip-hopper was depicted as authentic, altruistic and loyal to his fans. Moreover, he gives donations for projects of public utility and gives guidelines to other young hip-hoppers who want to make a career. A pretest (n=71 undergraduate students) was conducted to assess whether the developed scenarios induced a positive or a negative attitude toward the artist. This variable was measured using a three-item seven-point Likert scale (α=0.94) adopted from Till and Shimp (1998). Results demonstrated that our manipulation was successful, as respondents who read a positive scenario had a significantly higher attitude

26

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

toward the artist (M=5.21, SD=0.75) in comparison to respondents who read the negative scenario (M=2.43, SD=1.39) (t(23.94)= 7.356, p<0.001).

3.1

Measures

Dependent variables Respondents rated their attitude toward the brand (α=0.86) on a seven-item seven-point semantic differential scale developed by Stuart et al. (1987). Manipulation checks Similar to the pretest, the attitude toward the artist (α=0.95) was measured by using a three-item 7-point Likert scale adopted from Till and Shimp (1998). Covariates As attitude toward the music genre might be a confounding factor, we decided to measure the respondent’s attitude toward hip-hop music using a single-item measure as in Schemer et al. (2008). Furthermore, as the song was sung in a local dialect, we found this might potentially confound our results. As such, we controlled for this potential bias by calculating the distance between the hip-hoppers’ and respondent’s hometown which was added as covariate to our analysis. Therefore, the postal code of the respondent’s hometown was asked.

3.2

Procedure

Questionnaires were administered in small groups (+/- 20 respondents per group). Participants were first randomly assigned to one of the two scenarios, in which a positive or negative description about the artist was given. After reading the scenario, respondents were questioned about their attitude toward the artist. Afterwards, the respondents listened twice to one of the three versions of the song and subsequently were asked to rate their attitude toward Jo-Box!, their attitude toward hip-hop and to provide some socio-demographic data. After completing the experiment, respondents were debriefed and thanked for their participation.

27

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

4 4.1

Main Study Manipulation Check

A manipulation check revealed that respondents who read the positive scenario had a positive attitude toward the artist (M=5.34, SD=0.95); respondents who read the negative scenario had a negative attitude toward the artist (M=2.08, SD=0.95) (t(198)=24.229, p<0.001). Hence, we conclude that the manipulations of the attitude toward the artist work as intended.

4.2

Results

Our hypothesis predicted that the attitude toward the artist would moderate the relationship between brand name repetition in song lyrics and the brand attitude. We expect that the evaluation of a brand which is paired with a positively perceived artist will become more and more positive as the number of brand name repetitions increases, whereas the evaluation of a brand which is paired with a negatively perceived artist will become more and more negative as the number of brand name repetitions increases. An ANCOVA was run in order to test our hypothesis, of which results are provided in Table 1. First, attitude toward hip-hop significantly influences the attitude toward the brand (p<0.001), while distance between the hip-hopper’s hometown and the respondent’s hometown was marginally significant as covariate (p=0.071). However, it is important to note that our results, which will be discussed in subsequent section, did not alter with the inclusion or exclusion of both covariates. In the remainder of this manuscript, we report results using both attitude toward hip-hop and distance as covariates. Table 1: ANCOVA results of brand attitude

Source Attitude toward hip-hop Distance Level of placement Attitude toward the artist Level of placement x Attitude toward the artist

F 16.532 3.296 3.567 10.648

df 1 1 2 1

p <0.001 0.071 0.030 0.001

η² 0.079 0.017 0.035 0.052

0.145

2

0.865

0.001

28

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

Table 1 reveals that the main effects of brand name placement and attitude toward the artist had a significant impact on brand attitude, but the interaction between the brand name placement and the attitude toward the artist was not significant (p=0.865). The interaction was plotted in Figure 1.

4,2

Brand attitude

4,0

Negative attitude toward artist

3,8 3,6

Positive attitude toward artist

3,4 3,2 3,0 0

4

10

Number of brand name repetitions Figure 1: Interaction between brand name placement and attitude toward the artist

Our results indicate an evaluative conditioning effect: if a brand name is associated with an artist who is negatively perceived by the audience, this will result in lower brand attitudes. Contrary to our hypothesis, an increasing number of brand repetitions or associations between the brand name and the artist has a positive impact on the brand attitude regardless of the attitude toward the artist. Figure 1 shows that when a brand is associated with an artist who is negatively perceived, the brand attitude of respondents was lowest with zero (M=3.46) or four repetitions (M=3.37) but increased with ten repetitions (M=3.73). A similar pattern was found for associations between a brand name and artists who are positively perceived. The brand attitude was higher after ten repetitions (M=4.02) when compared to zero (M=3.75) or four repetitions (M=3.83).

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

29

Interesting to note is that the amount of repetitions of the brand name has a significant impact on brand attitude. More specific, when the brand name was not mentioned, respondents had a mean brand attitude of 3.60. Respondents who heard the brand name four times did not have a significant higher brand attitude (M=3.62). If the brand name was mentioned ten times, the mean brand attitude increased significantly to 3.87. These findings suggest that overall, brand name placement in song lyrics can be an interesting advertising strategy, when the brand is sufficiently repeated.

5

Discussion

This study experimentally explores the impact of the amount of brand repetitions in song lyrics on the brand attitude and tests whether this relationship is moderated by the attitude toward the artist. Contrary to our expectations, our results indicate that for both artists who are either positively or negatively perceived by the audience, higher repetitions of the brand name have a higher positive influence on the brand attitude. These results contradict prior research on the moderating effects of attitudes toward the artist in a product placement context. Research by Schemer et al. (2008) on product placement in music videos found that pairing a brand with positively evaluated artists produced positive attitudes toward the brand, whereas pairing a brand with negatively evaluated artists produced negative attitudes toward the brand. A possible explanation for these contradictory results could be that our pairing with a negatively evaluated artist did not result in a negative brand attitude, although the brand attitude was less positive compared to the positive pairing condition. This could be explained by the context in which this research was conducted. Given the high prevalence of brand name placements in hip-hop and rap songs (Herd, 2005; Martin and Collins, 2002), we examined the effectiveness of brand name placement in song lyrics in a hip-hop context. However, many people in the hip-hop nation were condemned by the government or have been locked up (Butler, 2004). As a result, it might be the case that respondents took the negative attitude toward the artist into account to a lesser degree when evaluating the brand name placement, as a negatively perceived artist is an integral part of the hip-hop scene. Our results are in line with mere exposure research. Mere exposure research shows that repeated exposure to a stimulus enhances attitudes toward that stimulus (Zajonc, 1968) often explained by perceptual fluency or ease of processing (Lee and Labroo, 2004). Respondents had positive brand attitudes in both posi-

30

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

tive and negative ‘attitude toward the artist’ conditions resulting in enhanced attitudes after repeated exposure. When considering the main effect of brand name repetitions in song lyrics, our findings demonstrate that the attitude toward the brand significantly increases after ten brand name placements. This provides further support for the impact of repetition on brand attitudes, an issue which received only very limited attention in prior product placement research (Van Reijmersdal et al., 2009). Following our results, managers might consider brand name placement in song lyrics as a potential context for product or brand name placement, next to more researched areas as movies, television shows, games or novels. When considering brand name placement in song lyrics as an advertising strategy, managers should be aware of the differential effects of brand name placement on brand recall, recognition and attitudes. Delattre and Colovic (2009) argued that at least two repetitions of a brand name should be included in the song lyrics, as this has a positive impact on brand recall and recognition. Results of our research clearly demonstrate that the brand name should be repeated at least ten times in order to enhance brand attitudes. This finding also differs from prior research on the number of repetitions that is necessary in order to increase brand evaluations with product placements. Van Reijmersdal et al. (2007) found that in a TV show, at least two repetitions were necessary in order to increase brand attitudes. The reason for this difference might be due to the use of visual versus auditory stimuli. In Van Reijmersdal et al.’s (2007) study, the product placement was both auditory and visual, while in our study the brand name placement was only auditory. Research shows that visual images or pictures tend to be more memorable than their verbal counterparts, because pictures are coded with both pictorial and verbal codes (dual coding theory, Paivio, 1986). The extra code associated with pictures gives them an advantage during retrieval. This stronger memory trace could also explain why fewer exposures are necessary for visual versus auditory product placement in order to influence respondents’ attitudes. Another important managerial implication emerges from this study. We did not find a significant interaction between brand name placement in song lyrics and the attitude toward the artist, leading to the conclusion that the impact of brand name placement in song lyrics on brand attitudes is similar for artists who are positively evaluated by the audience and artists who are negatively evaluated by the audience. As a result, marketers benefit from brand name placement in song lyrics, regardless of the artist who mentions the brand name. Given our current findings, managers can try to stimulate artists to include brand names in their song lyrics, even if the artists are not always positively perceived.

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

5.1

31

Limitations and Future Research

Although our research provides evidence for the effectiveness of brand name placements in music songs, some limitations remain. First, our research is situated within the specific context of hip-hop- and rap songs which opens up the question whether our results would also apply in other music genres. Second, the number of brand name repetitions was limited to ten in our manipulations. Future research might explore whether an overload of brand name repetitions might cause irritation with respondents (see Cowley and Barron, 2008), resulting in negative effects of brand name placement in song lyrics on brand attitudes. Third, in our manipulation the hip-hopper sang about his life as hiphopper, while the brand we included involved a MP3 player. As such, congruence between song lyrics and the brand name might exist. Future research might also explore whether incongruent brand name placements might yield different results. Fourth, given that some artists voluntarily include brand names in their song lyrics whereas others include brand names based on a contract with an organization, question remains whether different effects might be obtained if the audience is aware of the fact that an artist might be paid to include brand names. Finally, brand name placement in song lyrics differs from product placement in movies or television programs because songs might be broadcasted multiple times a day. As such, future research might explore whether time intervals between the exposure to the brand name (for example, five times a day versus once a day for five sequential days) influence the effectiveness of brand placement in song lyrics.

6

References

Agenda Inc. (2004), “American Brandstand 2004”, http://www.brandchannel.com/images/Papers/245_brandstand04_final.pdf Babin, L.A. & Carder, S.T. (1996). “Viewers’ recognition of brands placed within a film”, in: International Journal of Advertising, 15(2), 140-151. Baeyens, F.; Eelen, P.; Van den Bergh, O. & Crombez, G. (1992), “The content of learning in human evaluative conditioning: Acquired valence is sensitive to US-revaluation”, in: Learning and Motivation, 23, 200-224. Brennan, I. (2008) “Brand placement in novels: A test of the generation effect”, in: International Journal of Advertising, 27(4), 495–509. Butler, P. (2004), “Much respect: Toward a hip-hop theory of punishment”, in: Stanford Law Review, 56, No. 983. Cowley, E. & Barron, C. (2008) “When product placement goes wrong – The effects of program liking and placement prominence”, in: Journal of Advertising, 37(1), 89-98. d’Astous, A. & Chartier, F. (2000) “A study of factors affecting consumer evaluations and memory of product placement in movies”, in: Journal of Current Issues and Research in Advertising, 22(2), 31-40.

32

Van Vaerenbergh, Van de Sompel, Van Loock and Vermeir

Debruyne, B. (2007), “Hiphopmarketing is bling bling”, in: Bizz, March 2007, 30-31. Delattre, E. & Colovic, A. (2009) “Memory and perception of brand mentions and placement of brands in songs”, in: International Journal of Advertising, 28(5), 807-842. De Houwer, J.; Baeyens, F. & Field, A.F. (2005), “Associative learning of likes and dislikes: Some current controversies and possible ways forward”, in: Cognition and Emotion, 19, 161-174. De Houwer, J.; Thomas, S. & Baeyens, F. (2001), “Associative learning of likes and dislikes: A review of 25 years of research on human evaluative conditioning”, in: Psychological Bulletin, 127(6), 853-869. Fennis, B.M. & Bakker, A.B. (2001) “Stay tuned - We will be back right after these messages: Need to evaluate moderates the transfer of irritation in advertising”, in: Journal of Advertising, 30(3), 15-25. Ferraro, R. & Avery, R.J. (2000) “Brand appearances on prime-time television”, in: Journal of Current Issues and Research in Advertising, 22(2), 1-15. Friedman, M. (1985) “The changing language of a consumer society: Brand name usage in popular American novels in the post-war era”, in: Journal of Consumer Research, 11(4), 927-938. Gould, S.J. & Gupta, P.B. (2006) “Come on down: How consumers view game shows and the products placed in them”, in: Journal of Advertising, 35(1), 65-81. Graser, M. (2005), “McDonald’s buying way into hip hop song lyrics”, in: Advertising Age, March 23, 2005. Herd, D. (2005), “Changes in the prevalence of alcohol use in rap song lyrics, 1979-1997”, in: Addiction, 100, 1258-1269. Hoffman, W., De Houwer, J., Perugini, M., Baeyens, F. & Crombez, G. (2010), “Evaluative conditioning in humans: A meta-analysis”, Psychological Bulletin, 136(3), 390-421. Holloway, L. (2002), “Hip hop sales pop: Pass the Courvoisier and count the cash”, in: New York Times, September 2, 2002. La Ferle, C. & Edwards, S.M. (2006) “Product placement: How brands appear on television”, in: Journal of Advertising, 35(4), 65-86. Lee, A. & Labroo, A.A. (2004), “The effect of conceptual and perceptual fluency on brand evaluation”, in: Journal of Consumer Research, 31, 151-165. Lee, M. & Faber, R.J. (2007) “Effects of product placement in on-line games on brand memory: A perspective of the limited-capacity model of attention”, in: Journal of Advertising, 36(4), 75-90. Mackay, T., Ewing, M., Newton, F. & Windisch, L. (2009) “The effect of product placement in computer games on brand attitude and recall”, in: International Journal of Advertising, 28(3), 423-438. Mackie, D.E. & Asuncion, A.G. (1990) “On-line and memory-based modification of attitudes: Determinants of message recall-attitude change correspondence”, in: Journal of Personality and Social Psychology, 59(1), 5-16. Martin, B.A.S. & Collins, B.A. (2002) “Violence and consumption imagery in music videos”, in: European Journal of Marketing, 36(7/8), 855-873. Matthes, J., Schemer, C. & Wirth, W. (2007) “More than meets the eye – Investigating the hidden impact of brand placements in television magazines”, in: International Journal of Advertising, 26(4), 477-503. McCracken, G. (1989), “Who is the celebrity endorser? Cultural foundation of the endorsement process”, in: Journal of Consumer Research, 16(3), 310-321. McSweeney, F.K. & Bierley, C. (1984), “Recent developments in classical conditioning”, in: Journal of Consumer Research, 11, 619-631. Nelson, M. R. (2002) “Recall of brand placements in computer/video games”, in: Journal of Advertising Research, 42(2), 80-92. Paivio, A. (1986) “Mental representations”. New York: Oxford University Press.

The Impact of Brand Name Placement in Song Lyrics on Brand Attitudes

33

Petty, R.E., Cacioppo, J.T. & Schumann, D. (1983) “Central and peripheral routes to advertising effectiveness: The moderating role of involvement”, in: Journal of Consumer Research, 10(2), 135-146. Russell, C.A. (2002) “Investigating the effectiveness of product placements in television shows: The role of modality and plot connection congruence on brand memory and attitude”, in: Journal of Consumer Research, 29(3), 306-318. Russell, C.A. & Stern, B.B. (2006) “Consumers, characters and products: A balance model of sitcom product placement effects”, in: Journal of Advertising, 35(1), 7-21. Schemer, C., Matthes, J., Wirth, W. & Textor, S. (2008) “Does ‘Passing the Courvoisier’ always pay off? Positive and negative evaluative conditioning effects of brand placements in music videos”, in: Psychology & Marketing, 25(10), 923-943. Schiffman, L.G. & Kanuk, L.L. (2006), “Consumer behavior”, ninth edition, Pearson Prentice Hall, Upper Saddle River, NJ. Schneider, L.-P. & Cornwell, T.B. (2005) “Cashing in on crashed via brand placement in computer games”, in: International Journal of Advertising, 24(3), 321-343. Stuart, E.; Shimp, T.A. & Engle, R.W. (1987), “Classical conditioning of consumer attitudes: Four experiments in an advertising context”, in: Journal of Consumer Research, 15, 334-349. Sweldens, S.; Van Osselaer, S.M.J. & Janiszewski, C. (2010), “Evaluative conditioning procedures and the resilience of conditioned brand attitudes”, in: Journal of Consumer Research, 37, 473489. Till, B.D. & Shimp, T. (1998) “Endorsers in advertising: The case of negative celebrity information”, in: Journal of Advertising, 27(1), 67-74. Till, B.D., Stanley, S.M. & Priluck, R. (2008), “Classical conditioning and celebrity endorsers: An examination of belongingness and resistance to extinction”, in: Psychology & Marketing, 25(2), 179-196. Van Reijmersdal, E.A., Neijens, P.C. & Smith, E.G. (2007) “Effects of television brand placement on brand image”, in: Psychology & Marketing, 24(5), 403-420. Van Reijmersdal, E.A., Neijens, P.C. & Smith, E.G. (2009) “A new branch of advertising: Reviewing factors that influence reactions to product placement”, in: Journal of Advertising Research, 49(4), 429-449. Wiles, M.A. & Danielova, A. (2009) “The worth of product placement in successful films: An event study analysis”, in: Journal of Marketing, 73(4), 44-63. Woltman-Elpers, J.L.C.M., Wedel, M. & Pieters, R.G.M. (2003) “Why do consumers stop viewing television commercials? Two experiments on the influence of moment-to-moment entertainment and information value”, in: Journal of Marketing Research, 40 (4), 437-453. Yang, M. & Roskos-Ewoldson, D.R. (2007) “The effectiveness of brand placements in the movies: Levels of placements, explicit and implicit memory, and brand-choice behavior”, in: Journal of Communication, 57(3), 469-489. Zajonc, R.B. (1968), “Attitudinal Effects of Mere Exposure”, in: Journal of Personality and Social Psychology Monographs, 9, 1-27.

Differences in Children’s Processing of Advergames and TV Commercials Martin K. J. Waiguny, University of Klagenfurt, Austria Ralf Terlutter, University of Klagenfurt, Austria

1

Introduction

Advergames are a new and fast-growing category of interactive advertising tools (Faber et al, 2004). Most of the common advergames can be described as casual browser games constructed around one particular brand, and in most cases they are available for free either on the manufacturer’s own site or on a high-traffic website (Lee and Youn, 2008; Svahn, 2005; Waiguny and Terlutter, 2009). Like TV advertising, advergames are part of marketing communications and usually promote only one brand. From the advertisers’ point of view, the main objectives of marketing communications are to foster brand awareness, to emotionalize brand attitude, and to foster brand relationships. However, advergames are different from TV advertising in several key ways (active vs. passive reception, non/low vs. high interactivity while exposed to the brand, exposure time etc.).The basic research question addressed by this study is: In terms of advertising effects, how do advergames differ from traditional TV advertising? A further research aim was to gain a better understanding of the ways in which children recognize the commercial and persuasive intent of marketing media (Eagle, 2007; Singer and Singer, 1998). Children from age 8 or so start to develop cognitive defense capabilities against commercial persuasion (Nairn and Fine, 2008). But little is known as to whether these defense mechanisms are effective when children are confronted with new and convergent forms of persuasive media (Wright, Friestad and Boush, 2005). The insights into this question gained from this research would have implications for media literacy education (Turnipseed and Rask, 2007; Wollslager, 2009). Therefore a further research focus of this study is: Do children understand the commercial intent of advergames as compared to TV adverts? Does children’s persuasion knowledge influence their perception of advergames?

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_3, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

36 2

Waiguny and Terlutter

Differences between Advergames and TV Commercials

Playing an advergame is very different from watching television. Nakatsu et al. (2005) characterize playing games as an integrated experience of physical and mental presence which leads to a higher degree of activation and arousal than is the case with regular TV advertising (Nakatsu et al., 2005). While playing an advergame, children are likely to be more active and their cognitive efforts will be greater than when watching a TV commercial (Grodal, 2000). Hence, we propose our first hypothesis: H1: Advergames will be perceived as more exciting than TV ads. With regard to cognitive efforts, the cognitive process is constrained by individuals’ limited information processing capacity (Eyseneck and Keane, 2000; Kahneman, 1973; Annie Lang, 2000; A. Lang, 2006). Gamers have to use this limited capacity to process the gameplay and anticipate the ongoing gaming situation (Grodal, 2000), which might distract players from the featured brand (Grigorovici and Constantin, 2004; Nelson, 2005). TV ads are constructed around a single strong brand message supported by audiovisual elements (Maher et al., 2006) which are probably not as distracting as the gameplay. A comparison between sponsorship (brand placement) in a gaming environment and in a TV setting of an NASCAR race found that brand recall in the TV setting was higher than brand recall in the game setting (Walsh et al., 2008). Hypothesis H2 reads: H2: Brand recall will be lower in the advergame setting than in the TV ad setting. However, encouraging brand recall is not the only aim of an advergame. Fostering associations with the brand or the product is a further important objective as the brand will become embedded in a network of mental associations (e.g., Sutherland and Sylvester, 2000). Games thus make it possible to convey messages through simulation and direct interaction (Ritterfeld and Weber, 2006) as well as through the rhetoric of the gameplay (Bogost, 2007). In consequence, an incidental and implicit learning effect occurs (Auty and Lewis, 2003; Ritterfeld and Weber, 2006) as the associative networks in children’s procedural memories are extended or strengthened by the showing of a brand or a product in a special scene (Auty and Lewis, 2003). Thus it is not only brand recall which is important, various episodes, characters, rhetorical scenes (like scenes of what you can do when consuming the product) assume a significance. Gamers spend much more time on and have more exposure to message elements than is the case when viewers watch a scripted TV commercial. Moreover, gamers pay a lot of attention to the game as learning the gameplay is the major

Differences in Children’s Processing of Advergames and TV Commercials

37

task in this situation (van Lankveld et al., 2010). As a consequence, if elements of the brand or product are placed within the gameplay area rather than outside the main gameplaying area, these elements will be recognized to a higher extent than from watching a TV commercial although in the case of TV no distracting elements feature in the commercial. H3: The recognition of product related elements will be higher in the advergame setting than in the TV ad setting. In most cases, entertainment leads to positive emotions and a feeling of control of and excitement regarding the situation (Früh, 2003). It can be expected that such feelings will tend to be stronger with advergames than in the TV viewing situation (Grodal, 2000). Gamers are active and are more likely to enter into a flow experience (Csikszentmihalyi, 1988; Sherry, 2004), whereas watching TV is more likely to be a passive task (Nakatsu et al., 2005). According to the excitation transfer theory (Zillmann, 1971; Zillmann et al., 1972), this excitement is likely to be transferred to the current situation and to other stimuli (i.e. the brand). Hence, we expect that gamers will like an advergame more than a TV ad (treatment liking, i.e. the attitude towards the ad). Given the fairly robust link between attitude towards the ad (in terms of cognitive and affective judgement) and attitude towards the brand (Edell and Burke, 1987; MacKenzie and Lutz, 1989; Shiffman and Kanuk, 2007, p. 247f), especially in the fast-moving consumer goods markets (FMCG), we also expect that gamers will thus have a more positive attitude towards the brand. Youn and Lee (2005) have already demonstrated this relationship in the context of advergames. Ahn (2008) showed that liking the game positively influenced perception of the brand (brand beliefs, brand personality). Hence, also a choicepreference for the brand directly after watching or playing should be influenced more strongly by an advergame than by a TV commercial. We further analyze WOMI (Word Of Mouth Intention) – the intention to recommend the advergame/TV ad - as well as the intention to reuse the advergame or the intention to view the TV ad again. Most children at elementary school level do not actually make buying decisions themselves, but rather influence their parents’ buying decisions through pestering (McNeal, 1992; Valkenburg and Cantor, 2001). Pestering intention on behalf of the brand is included in the study and we expect higher pestering intention in the advergame setting than in the TV ad setting due to more positive brand attitudes. Thus we derive the following hypotheses: H4: Children in the advergame setting as compared to children in the TV ad setting will show higher values in all of the following areas:

38

Waiguny and Terlutter

     

Liking of the treatment (TL) Intention to recommend the treatment to others (WOMI) Intention to reuse the treatment (RUI) Attitude towards the promoted brand (ATTB) Intention to pester for the brand (PI) Preference for the brand (PREF)

In this study, research is conducted not only into the affect-driven effects of the entertainment provided by the different media, but also into children’s more cognition-driven persuasion knowledge (PK). Children develop their PK of advertising as they grow older and receive more education (Boush et al., 1994; Friestad and Wright, 1994; John Roedder, 1999a; John Roedder and Cole, 1986; Roedder and Whitney, 1986). According to the Persuasion Knowledge Model (Friestad and Wright, 1994), individuals typically understand the persuasive purpose of a message if they recognize the agent, if they have a general awareness that a message is being delivered with the intention of persuading them, and if they know about the tactics and structures employed by the persuasive message. PK is also related to attitudes towards advertising in general: PK comes into play as skepticism and counterargument if an individual’s attitude towards advertising is more negative (Campell and Kirmani, 2008). In an advertising context, PK only then becomes activated and leads to coping behavior if the ad is recognized as such (Wei et al., 2008). However, very young children are not able to distinguish between the different forms and intentions of commercial, informational or entertaining TV programs. This type of cognitive ability develops between the ages of 4 and 7. Children aged from 8 to 10 are more likely to identify commercials and understand both the commercial content and the persuasive intent. At 10, they will probably be able to understand both the persuasive intent and various persuasion tactics which are being employed, whereas children older than 12 will generally understand the full range of advertising tactics (John Roedder, 1999b; John Roedder, 2002; John Roedder and Whitney, 1986; Robertson and Rossiter, 1974; John Roedder, 1981; Valkenburg and Cantor, 2001). However, PK is not only influenced by children’s physical and cognitive development but also by the education they have received regarding media and media impact (Friestad an Wright, 1994; Kunkel, 2001). Above all, training in media literacy, i.e. a special educational focus that encourages children to review TV programmes critically (Singer and Singer, 1998), especially commercials (Eagle, 2007), influences their PK. Hence, we expect that children’s PK will negatively influence their liking of the treatment and the outcome variables related to brand attitude. H5: Active persuasion knowledge will negatively influence the

Differences in Children’s Processing of Advergames and TV Commercials

   

39

Liking of the treatment (TL) Intention to recommend the treatment to others (WOMI) Intention to reuse the treatment (RUI). Attitude towards the promoted brand (ATTB)

However, most media literacy training is tailored to support understanding of the persuasive intent of TV commercials. New forms of persuasive techniques and subliminal communication tools such as advergames are frequently not included in children’s media literacy education. Hence, we expect that PK will have less of an impact on responses to advergames than is the case with TV advertising. An expected consequence is that psychological coping behaviors such as a more negative and skeptical evaluation of the message sender or the brand which is being promoted (Brown and Krishna, 2004; Campbell and Kirmani, 2000; Wood, Nelson, Atkinson, and Lane, 2008) will have a lower impact on advergames than on TV advertising. The expectation is that children are likely to have been educated to decode TV advertising, but not to decode advergames. In addition, as a result of the more engaging gaming situation the cognitive resources available to identify commercial content in a game are more limited (Annie Lang, 2009; Nairn and Fine, 2008). Hence we expect practically no PK influence, even when it is activated, on the variables analyzed in this study in the advergame setting. Recent studies focusing on advergames and PK indeed have found no PK influence on the evaluation of the advergame and the brand in question (Mallinckrodt and Mizerski, 2007), or only very weak influence (Waiguny et al., 2010). This means that children - even if they already have high PK - will evaluate the treatment (advergame vs. TV), the brand, and the behavioral measures more favorably in the advergame setting than in the TV ad setting. H6: Active persuasion knowledge will moderate the dependent variables. In the TV ad setting, children with high persuasion knowledge will evaluate the game and subsequently all game-related and brand-related attitudinal and intentional variables less favorably than children with low persuasion knowledge, whereas in the advergame setting, persuasion knowledge will not have a moderating effect.

3

The Role of Challenge in Influencing Children through Games

To gain a deeper insight into what actually happens when children are playing advergames and its relationship to persuasion, we finally investigate the role of challenge in the game. In general, we predict that advergames will be more entertaining than TV commercials; but it is also necessary to understand what

40

Waiguny and Terlutter

happens if an advergame itself is not complex enough to truly challenge the gamer. On the one hand, if the level of game challenge is low, gamers will not pay full attention to the gameplay, and will also need less time and feel bored by the game (Chen, 2007; van Lankveld et al., 2010). On the other hand, if a game is too challenging or hard to use or interact with, children might feel overchalleged and even feel negative emotions. However, as they engage in trying to understand the game, they will have a much longer exposure time. If the challenge of a game matches kids’ expectations and skills, they will feel positive emotions, and the time spent on playing the game will be perceived as passing very quickly (Gilleade et al., 2005). In terms of persuasion, the level of challenge is important as it, firstly, regulates the focused attention and the time spent on the advergame. As mentioned previously, learning effects via advergames are somewhat incidental, not conscious (Bogost, 2007), and the more a player engages with the game the stronger such effects are. Hence we predict that if the challenge level is low, children will need less time to finish the advergame and subsequently will remember fewer elements of the game since they do not focus on the gameplay to the same extent as optimallychallenged or over-challenged children. Drawing on the perceptual fluency theory (e.g., Reber et al., 2004; Reber et al., 1998) and the mere exposure effect (Zajonc, 1968, 2001; Zajonc et al., 1974), attitudes will increase by the repetition of exposure and an associative learning effect will occur. Secondly, attitude towards the brand is also regulated by the affect transfer, as only in the case of optimal challenge do positive affects occur. Over- and underchallenged children will thus show a negative affect transfer (Lazzaro, 2008). To sum up, optimally challenged children will show the highest level of influence on the attitude towards the brand. As over-challenged children need more time to complete the advergame, they will remember most of the game elements. However, the persuasive effect will be diminished by the negative affect. The last hypothesis therefore reads: H7: The level of challenge is related to time, recognition and attitude towards the brand. Under-challenged children will need less time, remember the fewest elements and will have the least favourable attitude towards the brand. Optimally-challenged children will display the most favourable level of attitude towards the brand, needing a moderate amount of time for the game and subsequently exhibiting a moderate degree of recognition of gaming elements. Over-challenged children will need much more time and recognize the most elements, but will only have a moderately favourable attitude towards the brand.

Differences in Children’s Processing of Advergames and TV Commercials

4

41

Study

In order to test the proposed hypotheses, a study involving 51 children was carried out in January 2010 in an elementary school. To control for developmental factors in the understanding of the intentions underlying advertising, only children aged between 8 and 10 were included in the study, all of whom, according to the literature, can be expected to be at roughly the same developmental stage of knowledge about advertising (John Roedder, 2002; Livingstone and Helsper, 2006).

4.1

Method and Measures

Design and Treatment: A 2 (TV vs. advergame) x 2 (high vs. low PK) quasiexperimental design was chosen. First demographic questions about gaming, TV and eating habits were asked by trained interviewers in one-to-one interviews. Afterwards the children were randomly divided into two groups: one group (29 children) was asked to play an advergame and the second group (22) children were shown a TV advertisement. The brand “Nesquik Duo” was chosen for the purposes of the study. Nesquik Duo broadcasts a TV spot in which the Nesquik mascot runs on a brown-and-white (brown and white chocolate) colored table, finds a package of the product, which it jumps into using a spoon made of chocolate. Then the cereal is poured into a bowl and the mascot also falls into the bowl, whereupon everything turns into a swirl and the mascot wakes up from a dream and starts a breakfast featuring the product and children. The corresponding advergame uses the same content elements. The player has to collect the product in the brown-and-white environment of a table with a breakfast bowl. This part of the game is offered at 4 levels with time limits (between 35 and 25 seconds); each level ends when the player locates a swirl, which is the exit to the next level. At the end, the player is allowed 3 chances to jump into a package with the aid of a brown spoon. The children were to play the game to the end in order to exclude the negative influence of interrupting a pleasurable experience. Afterwards the children were questioned by the interviewers about their perceptions of the game and then about their knowledge of advertising tactics. The time taken to complete the game was recorded for the individual gamers. During the questioning of the children, terms such as advergame or advertisement were deliberately not used to refer to the treatment. Measures of independent variables: Knowledge about the persuasive tactics employed in advertising and beliefs about advertising in general were measured using a simplified version, specially adapted for this age group, of the scale of beliefs about advertiser tactics used by Boush et al. (1994), This scale consists of beliefs about how credible advertising is, beliefs about the intention and

42

Waiguny and Terlutter

persuasive character of advertising, and the selling intent of advertising. Identification of commercial character was measured by asking: “Did any brand appear in the game/film?” If the child named any brand, this was coded as 1, and if not as 0. In addition, general attitude towards advertising was measured by asking: “Please indicate how much you like advertising in general!” In addition, in the case of the advergame group, the level of challenge was measured by four variables, namely “The time passed rather quickly”, “The game was difficult”, “I quickly understood what I was meant to do in the game”, “I quickly found out how to handle the game”. Measures of the dependent variables: Brand recall (BR) was measured by asking the children an open-ended question: “Did any brand appear in the game?” If the child named Nesquik, this was coded 1, any other or no brand was coded 0. Brand attitude (ATTB) was measured using the brand attitude scale for children (Pecheux and Derbaix, 1999). Excitement (EX) was measured by asking whether “The game/film was exciting.” Recognition was measured by 18 elements, some of which appeared in the game/ TV ad, whereas others did not. If the child recognized the element correctly (i.e. correctly identified whether or not it had been featured in the game or ad) it was coded with +1. If recognized incorrectly, it was coded -1. An index (RECOG) ranging from -18 to +18 was calculated, the higher the index score the higher the correct recognition of elements. The time taken was measured in seconds. All question-scales were measured on a 6-point Likert scale supported by a visual classification of the end-points of the scale continuum by using a hook and a cross (indicating right or wrong) (Tinson, 2009). Liking advertising (AL) in general and liking the treatment (TL) were measured by using a 6-point smiley scale, ranging from “like it very much” to “don’t like it at all”.

4.2

Results

26 male and 25 female children aged 8 (33.3%), 9 (45.1%), and 10 (21.5%) took part in the study. 98% of the children stated that they have access to a computer. Table 1 provides an overview of the children’s TV viewing and computergaming habits. First, after correcting the data for input errors and missing values, it was checked whether there are significant differences in the independent variables in the groups with regard to gender, age, TV viewing and videogame habits. These checks indicated that there was no significant difference on p<.1. Due to low loadings, three items of the attitude scale had to be removed. From the remaining items, a mean index was calculated.

43

Differences in Children’s Processing of Advergames and TV Commercials

4.2.1

Results of the Mean Comparisons

On average it took the children 4:50 minutes (min: 3:06, max: 13:50) to complete the advergame. The length of the TV ad was 0:31 minutes, hence the average exposure to brand-related stimuli is 8½ times longer in the advergame setting. To test the first set of hypotheses (H1-H4) one-way ANOVA and χ²analysis were conducted. Table 1 reports the findings that hypotheses H1, H2, H3 and H4 are all supported by the data at least at p<.1 level. Children perceive the advergame as more exciting (Madvergame: 4.72; MTV: 2.68) but in the advergame setting significantly fewer children (advergame: 44.8%, TV: 81,8%) were able to recall the brand unaided. Furthermore, as demonstrated in Table 1, both liking the treatment and brand-related measures are higher in the advergame setting than in the TV ad setting. This indicates that children’s perception of the advergame is more favourable than their perception of the TV ad and also the related behavioral intentions are higher. In terms of recognition (RECOG), children can remember far more elements correctly in the advergame setting (Madvergame: 7.03) than in the TV ad setting (MTV: 4.09). Table 1: Results of the One Way-ANOVA and Cross-Tabs Analysis Dependent Variable EX RECOG TL ATTB RI WOMI RC of brand PI PREF

4.2.2

M (overall)/ % (overall) 3.84 5.76 4.88 4.08 3.55 3.24 60.8% 26.5% 28.6%

M (TV)/ % (TV) 2.68 4.09 4.32 3.48 2.37 2.59 81.8% 13.6% 18.6%

M (Advergame)/ % (Advergame) 4.72 7.03 5.31 4.54 4.45 3.72 44.8% 37.0% 37.0%

F/ χ² 13.70 14.04 9.58 4.55 16.23 4.25 7.18 3.41 2.11

p .00** .00** .00** .04** .00** .05** .01** .06* .16n.s.

Results of the Interaction with Activated Persuasion Knowledge

To introduce PK in the context of the study, beliefs about the tactics employed in advertising (credibility of advertising, persuasive intent of advertising, selling intent of the advertiser), attitude towards advertising, and the identification of the commercial character of the game/ad were used to divide the group of subjects into an activated PK and a non-activated PK group by using two-step cluster analysis. The results indicate that for children with activated PK the knowledge about advertising tactics was higher and the attitude towards

44

Waiguny and Terlutter

advertising in general was less favourable. Hence ability to identify the commercial character is related to the level of persuasion knowledge. A twoway MANOVA for the dependent variables was computed, with the grouping variable and the calculated cluster variable as independent variables. Table 4 reports the results. Table 2: Means of the dependent variables of the 2x2 design Grouping Variables TL TV ad PK act PK inact Total adver- PK act game PK inact Total total PK act PK inact Total ATTB TV ad PK act PK inact Total adver- PK act game PK inact Total total PK act PK inact Total

m std.dev. 4.06 1.11 5.50 0.58 4.32 1.17 5.38 0.77 5.25 1.34 5.31 1.11 4.61 1.17 5.30 1.22 4.88 1.23 3.03 1.86 5.50 0.61 3.48 1.95 4.62 1.52 4.48 1.74 4.54 1.62 3.69 1.87 4.69 1.62 4.08 1.83

n Grouping Variables 18 RUI TV ad PK act 4 PK inact 22 total 13 adver- PK act game PK inact 16 29 total 31 total PK act 20 PK inact 51 total 18 WOMI TV ad PK act 4 PK inact 22 total 13 adver- PK act game PK inact 16 29 total 31 total PK act 20 PK inact 51 total

m std.dev. 1.83 1.34 4.75 1.5 2.36 1.76 5.23 1.01 3.81 2.2 4.45 1.88 3.26 2.08 4.00 2.08 3.55 2.09 2.22 1.63 5.25 0.96 2.77 1.93 5.23 1.24 5.50 1.37 5.38 1.29 3.48 2.1 5.45 1.28 4.25 2.05

n 18 4 22 13 16 29 31 20 51 18 4 22 13 16 29 31 20 51

On the inclusion of PK in the model, the main effect of the grouping variable becomes insignificant except for recommendation intention and re-usage intention, which indicates that all brand-related dependent variables are significantly influenced by PK except for the beliefs that the brand is healthy, and the re-usage intention regarding the treatment. Hence, H5 is partly supported. The interaction effects between the treatment groups and PK showed that all dependent variables interact as predicted, indicating that children with activated PK who watch the commercial will evaluate it less positively and will also be more skeptical regarding the brand, whereas no differences are observable between high and low PK in the gaming group. Table 5 reports the means and tests of the MANOVA analysis.

45

Differences in Children’s Processing of Advergames and TV Commercials

Table 3: Main and Interaction Effects Source Group

PK

Group X PK

4.2.3

Dependent Variable TL ATTB RUI WOMI TL ATTB RUI WOMI TL ATTB RUI WOMI

df

F 1 1 1 1 1 1 1 1 1 1 1 1

2.195 .260 5.239 11.910 3.235 4.353 1.944 12.192 4.701 5.382 16.270 8.535

p .15n.s. .61n.s. .03** .00** .08* .04** .17n.s. .00** .04** .03** .00** .01**

Results of the Influence of Challenge on Recognition and Persuasion

To analyze the level of challenge, a centroid cluster analysis was used to divide the children who played the advergame into three groups. Table 4 reports the distribution and the mean values. Using this grouping variable a MANOVA was calculated. As predicted under-challenged (M: 248 sec.), need significantly (F=5.867, p=.008) less time than optimally- (M: 411 sec.) and over-challenged (M: 607 sec.) children, to finish the game. In terms of recognition underchallenged subjects (M: 4.40) remember less than optimally- (M: 7.48) and over-challenged kids (M: 7.00), however this effect was only significant on p<.1 level (F=2.500, p=.100). Checking the influence of challenge on attitude towards the brand, the predicted pattern that the optimally-challenged (M: 4.58) will show higher values than under-challenged (M: 3.77) and over-challenged (M: 4.05) children was found, but the influence was not on a statistically significant level (F=.513, p=.605). Hence H7 is partially supported by our data.

46

Waiguny and Terlutter

Table 4: Results of the Challenge Grouping Item

The time passed rather quickly I quickly found out how to use the game. The game was difficult. I quickly understood what I was meant to do in the game.

4.3

underchallenged (n=5) 2.80

optimallychallenged (n=18) 5.94

overchallenged (n=6) 3.33

5.80

5.44

3.33

1.40

1.94

4.67

5.80

5.61

3.17

Selected Limitations

This study is limited by several factors. First only one game (with the same content elements as the corresponding TV ad) was tested. Moreover the sample was very small and only from one school, thus factors like social environment, and peer and family influence on PK could not be evaluated. Finally, the lowPK group in the TV advertising setting was very small because of the high recognition rate of persuasive intent in the case of the TV ad.

5

Discussion and Implications

This is the first study which directly compares the effects of an advergame with the effects of a comparable TV commercial in the field of advertising to children. The results clearly indicate that children perceive the game as more exciting. Moreover, evidence was found of the impact of limited information processing capacity on responses as the children who played the game were less likely than those in the TV group to recall the brand promoted in-game because they had been distracted by the gameplay. Furthermore, the children indicated that in the advergame setting they tended to rate the advergame and subsequently also all brand-related measures more positively. This could be the end of the story, but this more positive evaluation of the advergame is not only a direct result of the entertainment provided by the game. It is also the result to some extent of the fact that some of the children were not aware that an advergame is a form of advertising. Even in the case of those children who did realize that they had encountered a type of advertising, commercial PK is not activated to anything like the same extent as with TV advertising. In the case of the TV ad, the activated PK leads to a more skeptical

Differences in Children’s Processing of Advergames and TV Commercials

47

evaluation as a coping behavior. An obvious conclusion is that effective education is needed to help kids cope with these new forms of branded entertainment which target children. As Wollslager (2009) and also Stern and An (2009) have demonstrated, even brief training/induction to product placement in games leads to enhanced rates of identifying the commercial character of the advergame. But merely identifying the underlying character of the game is not enough. Like persuasive games in education (Bogost, 2007), advergames employ a lot of hidden persuasive content. The more the gameplay conveys the brand message (congruence of game goals and brand communication goals) and the higher the integration of the product, the stronger the persuasive attempts become (Smith and Just, 2009). We also showed that exposure time to the advergame is much longer than exposure to a TV ad. With the advergame experience thus much stronger than the experiencing of a TV ad, the persuasive power of advergames could well be rather strong. Also the level of challenge relates to persuasion, as optimally challenged children play longer and remember more message elements. Thus associative networks are built. However, no evidence for the relationship between challenge and persuasion in terms of brand attitude was found in this data. This may well be due to the small sample size. The general tendency for the relationships followed the predictions. The results of the study have clear implications for media literacy issues. Thus it would seem advisable to train children to be able to identify the persuasive elements of advergames. In addition to being able to recognize the bare fact that a brand or product is being placed in a game, media literacy training should also aim to help clarify more complex relationships such as typical messages delivered by the gameplay like: “If you eat the product, you’ll be able to jump higher.”

6

References

Ahn, D. (2008). The Interpretation of the Messages in an Advergame: The Effects on Brand Personality Perception. Paper presented at the Annual meeting of the Association for Education in Journalism and Mass Communication. Retrieved from http://www.allacademic.com/meta/p272095_index.htm Auty, S. & Lewis, C. (2003). The "Delicious Paradox": Preconscious Processing of Product Placements by Children. In L. J. Shrum (Ed.), The Psychology of Entertainment Media: Blurring the Lines Between Entertainment and Persuasion (pp. 117-133). Mahawah: Lawrence Erlbaum. Bogost, I. (2007). Persuasive Games: The Expressive Power of Video Games. Cambridge: MIT Press.

48

Waiguny and Terlutter

Boush, D. M., Friestad, M. & Rose, G. M. (1994). Adolescent Skepticism toward TV Advertising and Knowledge of Advertiser Tactics. Journal of Consumer Research, 21(1), 165-175. Brown, C. L. & Krishna, A. (2004). The skeptical shopper: A metacognitive account for the effects of default options on choice. Journal of Consumer Research, 31(3), 529-539. Campbell, M. C. & Kirmani, A. (2000). Consumers' Use of Persuasion Knowledge: The Effects of Accessibility and Cognitive Capacity on Perceptions of an Influence Agent. Journal of Consumer Research, 27(1), 69-83. Campell, M. C. & Kirmani, A. (2008). I Know What You're Doing, and Why You're Doing it: The Role of Persuasion Knowledge Model in Consumer Research. In C. P. Haugtvedt, F. Kardes & P. M. Herr (Eds.), Handbook of Consumer Psychology (pp. 549-572). New York: Lawrence & Erlbaum. Chen, J. (2007). Flow in Games (and everything Else). Commnications of the ACM, 50(4), 31-34. Csikszentmihalyi, M. (1988). The flow experience and its significance for human psychology. In M. Csikszentmihalyi & I. S. Csikszentmihalyi (Eds.), Optimal experience: Psychological studies of flow in consciousness (pp. 15-35). New York: Cambridge University Press. Eagle, L. (2007). Commercial Media Literacy What Does it Do, to Whom-and Does it Matter? Journal of Advertising, 36(2), 101-110. Edell, J. A. & Burke, M. C. (1987). The Power of Feelings in Understanding Advertising Effects. Journal of Consumer Research, 14(3), 421-433. Eyseneck, M. W., & Keane, M. T. (2000). Cognitive Psychology a Student's handbook. Hove: Psychology Press. Faber, R. J., Lee, M. & Nan, X. (2004). Advertising and the Consumer Information Environment Online. American Behavioral Scientist, 48(4), 447-466. Friestad, M. & Wright, P. (1994). The Persuasion Knowledge Model: How People Cope with Persuasion Attempts. The Journal of Consumer Research, 21(1), 1-31. Früh, W. (2003). Triadisch-dynamische Unterhaltungstheorie (TDU). In W. Früh & H.-J. Stiehler (Eds.), Theorie der Unterhaltung: Ein interdisziplinärer Diskurs (pp. 27-56). Köln: Herbert von Halem Verlag. Gilleade, K.M., Dix, A. & Allanson, J. (2005). Affective Videogames and Modes of Affective Gaming: Assist Me, Challenge Me, Emote Me. DiGRA, 1-7. Grigorovici, D. M. & Constantin, C. D. (2004). Experiencing Interactive Advertising beyond Rich Media: Impacts of Ad Type and Presence on Brand Effectiveness in 3D Gaming Immersive Virtual Environments. Journal of Interactive Advertising, 5(1), 1-18. Grodal, T. (2000). Video Games and the Pleasures of Control. In D. Zillmann & P. Vorderer (Eds.), Media Entertainment: The Psychology of its Appeal (pp. 197-213). Mahwah: Lawrence Erlbaum. John Roedder, D. (1999a). Consumer Socialization of Children: A Retrospective Look at Twenty Five Years of Research. Journal of Consumer Research, 26(3), 183-213. John Roedder, D. (1999b). Through the Eyes of a Child. Children’s Knowledge and Understanding of Advertising. In M. C. Macklin & L. Carlson (Eds.), Advertising to children. Concepts and controversies (pp. 3-26): Thousand Oaks. John Roedder, D. (2002). Consumer socialization of children: A retrospective look at twenty-five years of research. In F. Hansen (Ed.), Children - Consumption, Advertising and Media (pp. 2589). Copenhagen. John Roedder, D. & Cole, C. A. (1986). Age Differences in Information Processing: Understanding Deficits in Young and Elderly Consumers. Journal of Consumer Research, 13(3), 297-315. John Roedder, D. & Whitney, J. C. (1986). The Development of Consumer Knowledge in Children: A Cognitive Structure Approach. Journal of Consumer Research, 12(4), 406-417. Kahneman, D. (1973). Attention and Effort. Englewood Cliffs: Prentice Hall.

Differences in Children’s Processing of Advergames and TV Commercials

49

Kunkel, D. (2001). Children and Television Advertising. In D. G. Singer & J. R. Singer (Eds.), Handbook of Children and the Media (pp. 375-394). Thousand Oaks: Sage. Lang, A. (2000). The Limited Capacity Model of Mediated Message Processing. Journal of Communication, 50(1), 46-70. Lang, A. (2006). Using the limited capacity model of motivated mediated message processing to design effective cancer communication messages. Journal of Communication, 56. Lang, A. (2009). The Limited Capacity Model of Motivated Mediated Message Processing. In R. L. Nabi & M. B. Oliver (Eds.), The SAGE Handbook of Media Processes and Effects (pp. 193204). Los Angeles: SAGE. Lazzaro, N. (2008). Why we play: Affect and the Fun of Games: Designing Emotions for Games, Entertainment Interfaces and Interactive Products. In A. Sears & J. A. Jacko (Eds.), The HumanComputer Interaction Handbook: Fundamentals, Evolving Technologies, and Emerging Applications (pp. 679-700). New York: Lawrence Erlbaum. Lee, M. & Youn, S. (2008). Leading National Advertisers' Uses of Advergames. Journal of Current Issues & Research in Advertising, 30(2), 1-13. Livingstone, S. & Helsper, E. J. (2006). Does Advertising Literacy Mediate the Effects of Advertising on Children? A Critical Examination of Two Linked Research Literatures in Relation to Obesity and Food Choice. Journal of Communication, 56(3), 560-584. MacKenzie, S. B. & Lutz, R. J. (1989). An empirical examination of the structural antecedents of attitude toward the ad in an advertising pretesting context. (advertising). Journal of Marketing, v53(n2), p48(18). Maher, J. K., Hu, M. Y. & Kolbe, R. H. (2006). Children's recall of television ad elements: an examination of audiovisual effects. Journal of Advertising, 35(1), 23(11). Mallinckrodt, V. & Mizerski, D. (2007). The Effects of Playing an Advergame on Young Children's Perceptions, Preferences, and Requests. Journal of Advertising, 36(2), 87-100. McNeal, J. U. (1992). Kids as Customers: a Handbook of Marketing to Children. New York: Lexington Books. Nairn, A. & Fine, C. (2008). Who's Messing with My Mind? The Implications of Dual-process Models for the Ethics of Advertising to Children. International Journal of Advertising, 27(3), 447-470. Nakatsu, R., Rauterberg, M. & Vorderer, P. (2005). A New Framework for Entertainment Computing: From Passive to Active Experience Entertainment Computing - ICEC 2005 (pp. 112). Nelson, M. R. (2005). Exploring Consumer Response to "Advergaming". In C. P. Haugtvedt, K. A. Machleit & R. Yalch (Eds.), Online Consumer Psychology: Understanding and Influencing Consumer Behaviour in the Virtual World (pp. 167-194): Lawrence Erlbaum. Pecheux, C. & Derbaix, C. (1999). Children and Attitude toward the Brand: A New Measurement Scale. Journal of Advertising Research, 39(4), 19-27. Reber, R., Schwarz, N. & Winkielman, P. (2004). Processing Fluency and Aesthetic Pleasure: Is Beauty in the Perceiver's Processing Experience? Personality and Social Psychology Review, 8(4), 364-382. Reber, R., Winkielman, P. & Schwarz, N. (1998). Effects of Perceptual Fluency on Affective Judgments. Psychological Science, 9(1), 45-48. Ritterfeld, U. & Weber, R. (2006). Video Games for Entertainment and Education. In P. Vorderer & J. Bryant (Eds.), Playing Video Games: Motives, Responses and Consequences (pp. 399-412). Mahwah: Lawrence Erlbaum. Robertson, T. S. & Rossiter, J. R. (1974). Children and Commercial Persuasion: An Attribution Theory Analysis. The Journal of Consumer Research, 1(1), 13-20.

50

Waiguny and Terlutter

Roedder, D. L. (1981). Age Differences in Children's Responses to Television Advertising: An Information Processing Approach. Journal of Consumer Research, 8, 144-153. Sherry, J. L. (2004). Flow and Media Enjoyment. Communication Theory, 14(4), 328-347. Shiffman, L. G. & Kanuk, L. L. (2007). Consumer Behavior (9 ed.). New Jersey: Pearson. Singer, D. G. & Singer, J. L. (1998). Developing Critical Viewing Skills and Media Literacy in Children. Annals of the American Academy of Political and Social Science, 557, 164-179. Smith, J. H. & Just, S. N. (2009). Playful Persuasion: The Rhetorical Potential of Advergames. Nordicom Review, 30(2), 53-68. Stern, S. & An, S. (2009). Increasing Children’s Understanding of Advergames’ Commercial Nature: Does an Advertising Literacy Lesson or Ad Break Make a Difference? Paper presented at the Annual Meeting of the International Communication Association. Retrieved from http://www.allacademic.com/meta/p298021_index.html Sutherland, M. & Sylvester, A. K. (2000). Advertising and the Mind of the Consumer: What Works, What Doesn't, and why. St. Leonards: Allen & Unwin. Svahn, M. (2005). Future-proofing Advergaming: a Systematisation for the Media Buyer Proceedings of the second Australasian Conference on Interactive entertainment (Vol. 123, pp. 187-191). Tinson, J. (2009). Conducting Research with Children and Adolescents: Design, Methods and Empirical Cases. London: Goodfellow Publishing. Turnipseed, T. & Rask, A. (2007). Children’s Health and Advergaming: A Theoretical Study of Advertisement Driven Video Games, Product Placement, and Integrated Marketing Communication. Paper presented at the Annual Meeting of the NCA 93rd Annual Convention. Retrieved from http://www.allacademic.com/meta/p193018_index.html Valkenburg, P. M. & Cantor, J. (2001). The Development of a Child into a Consumer. Journal of Applied Developmental Psychology, 22(1), 61-72. van Lankveld, G., Spronck, P., van den Herik, H. & Rauterberg, M. (2010). Incongruity-Based Adaptive Game Balancing. In H. van den Herik & P. Spronck (Eds.), Advances in Computer Games (Vol. 6048, pp. 208-220): Springer Berlin / Heidelberg. Waiguny, M. K. J., Nelson, M. R. & Terlutter, R. (2010). Persuading Playfully? The Effects of Persuasion Knowledge and Positive Affect on Children's Attitudes, Brand Beliefs and Behaviors. In W.-N. Lee (Ed.), Proceedings of the 2010 Conference of the American Academy of Advertising (pp. 67-70). Minneapolis: AAA. Waiguny, M. K. J. & Terlutter, R. (2009). Advergames: The Role of Entertainment in BrandRelated Outcomes for Children. In R. Terlutter, S. Diehl, M. Karmasin & E. Smit (Eds.), Proceedings of the 8th ICORIA. Klagenfurt: Alpen-Adria-University Klagenfurt. Walsh, P., Kim, Y., & Ross, S. D. (2008). Brand Recall and Recognition: A Comparison of Television and Sport Video Games as Presentation Modes. Sport Marketing Quarterly, 17, 201208. Wei, M. L., Fischer, E. & Main, K. J. (2008). An Examination of the Effects of Activating Persuasion Knowledge on Consumer Response to Brands Engaging in Covert Marketing. Journal of Public Policy and Marketing, 27(1), 34-44. Wollslager, M. E. (2009). Children's Awareness of Online Advertising on Neopets: The Effect of Media Literacy Training on Recall. SIMILE, 9(2), 31-53. Wood, M. L. M., Nelson, M. R., Atkinson, L. & Lane, J. B. (2008). Social Utility Theory: Guiding Labeling of VNRs as Ethical and Effective Public Relations. Journal of Public Relations Research, 20(2), 231-249. Wright, P. Friestad, M., & Boush, D. M. (2005). The Development of Marketplace Persuasion Knowledge in Children, Adolescents, and Young Adults. Journal of Public Policy & Marketing, 24, 222-233.

Differences in Children’s Processing of Advergames and TV Commercials

51

Youn, S. & Lee, M. (2005). Advergame Playing Motivations and Effectiveness. In M. R. Stafford & R. J. Faber (Eds.), Advertising, Promotion and New Media (pp. 320-347). Armonk: M.E.Sharpe. Zajonc, R. B. (1968). Attitudinal Effects of Mere Exposure. Journal of Personality and Social Psychology, 9(2 PART 2), 1-27. Zajonc, R. B. (2001). Mere Exposure: A Gateway to the Subliminal. Current Directions in Psychological Science, 10(6), 224-228. Zajonc, R. B., Markus, H. & Wilson, W. R. (1974). Exposure Effects and Associative Learning. Journal of Experimental Social Psychology, 10(3), 248-263. Zillmann, D. (1971). Excitation Transfer in Communication-mediated Aggressive Behavior. Journal of Experimental Social Psychology, 7(4), 419-434. Zillmann, D., Katcher, A. H. & Milavsky, B. (1972). Excitation transfer from physical exercise to subsequent aggressive behavior. Journal of Experimental Social Psychology, 8(3), 247-259.

Part II

Social Issues and Advertising

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages Martine Lewi, University of Antwerp, Belgium Patrick De Pelsmacker, University of Antwerp, Belgium

1

Introduction

Alzheimer’s disease is a devastating, incurable disease. The disease pathway is typically characterised by a long preclinical phase in which patients experience mild cognitive impairments (Visser et al, 2002). Medical research states that through early diagnosis of symptoms the disease process can be slowed down (Seltzer, 2007; Bain et al., 2008). The main concern is then to make the population at risk aware of the importance of early diagnosis and stimulate them to take responsible actions at the earliest stage of the onset of the disease. Very little is known about the efficacy of health communication in the domain of a fatal disease, where one can merely intervene in order to lengthen the time period in which the patient can still enjoy a reasonable quality of life. However, results of previous research on the emotional and cognitive responses to various message formats of (health) communication can serve as frameworks to analyze different types of health messages (Williams-Piehota et al., 2006; Morris et al., 2002; Bagozzi et al., 1999; Brown and Stayman, 1992; Petty et al., 1993; Mitchell, 1988; Petty and Cacioppo, 1986; Chaiken, 1980). The present study extends prior work in the area of emotional and rational persuasive communication to the domain of health communication, especially in the domain of Alzheimer’s disease. More particularly, this study aims at entangling the role of affect in the information processing and at investigating how affective reactions of older people to Alzheimer awareness messages varying in argument strength and affective tone of visuals are processed through affective and cognitive responses. Further, we investigate differences between men and women and between lowly and highly educated individuals.

2

Conceptual Framework and Research Questions

An Alzheimer awareness message may be processed either peripherally or centrally, depending on the involvement with the issue. Standard expectations (Petty and Cacioppo, 1986) imply that in high-involvement situations people S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_4, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

56

Lewi and De Pelsmacker

would centrally process message arguments, while in low-involvement situations, peripheral processing would take place, and affective reactions to the message would drive responses. However, the question remains to what extent rational and affective message elements and affective and cognitive responses to these message formats play a role in central or peripheral message processing and judgment formation.

2.1

The Role of Affect in Developing Cognitive Responses about Alzheimer Awareness Messages

Forgas’ Affect Infusion Model (AIM) distinguishes four judgmental strategies, two low-affect infusion strategies (direct access and motivated processing) and two high-affect infusion strategies (heuristic and substantive processing). The AIM predicts that affect is unlikely to influence judgments during direct access or motivated processing. Affect should have an effect when heuristic or substantive processing is used (Schwarz and Clore, 2003; Fiedler, 1988, 1990; Fiedler et al., 1991; Forgas, 1991, 1992, 1995; Schwarz and Bless, 1991). Two underlying mechanisms of affect infusion are affect-as-information and affect priming. The affect-as-information theory suggests that rather than computing a judgment on the basis of recalled features of a target, individuals may ask themselves: 'How do I feel about it?' and in doing so, may be guided by their feelings to judge a message (Schwarz,1990; Schwarz and Clore, 2003). In the affect priming theory, affect can indirectly inform judgments by facilitating access to related cognitive categories (Bower, 1981; Isen, 1987). Hence, affect can prime the encoding, retrieval, and selective use of information in the constructive processing of judgments. Affect-priming is considered a broad and parsimonious principle that has implications for a variety of cognitive processes and suggests multiple avenues through which indirect affect infusion into judgments may occur (Forgas and Bower, 1988; Singer and Salovey, 1988). Within the AIM model, it is implied that it is in the course of substantive, constructive processing that affect is most likely to play a significant informational role in what is perceived, learned, and recalled and how stimulus information is interpreted. Thus stimuli that are perceived as positively affective should lead to paying closer attention to positive information, better learning such details, and making more positive interpretations of ambiguous information (Forgas, 1995). The AIM also implies that judgments about more complex stimuli (substantive processing context), requiring more elaborate processing and made without the benefit of objective evidence, should show greater affective effects. According to this research, substantive processing is more likely when the message is personally relevant or ambiguous. Previous research (e.g., Petty et al., 1993) reported that positive affect produces more positive judgments in both high- and

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

57

low-elaboration conditions. Several studies have found significant affect congruence in judgments and thoughts about persuasive messages, mostly under conditions of substantive processing (i.e., high elaboration likelihood; Petty et al., 1993). The fact that cognitive approaches to persuasive messages are mediated by affective message components is a prominent theme in advertising research (e.g., Ray and Batra, 1983; Batra and Ray, 1986; Edell and Burke, 1987; Cohen and Areni, 1991; Geuens and De Pelsmacker, 1999; Raghunathan and Pham, 1999). Their results indicate that affectively valenced messages can be important triggers of cognitive responses to messages. Applied to the domain of health, Salovey and Birnbaum (1989) found that affective reactions had a major influence on people's perceptions of symptoms of illness, their health efficacy judgments, and their expectations of future disease. Salovey also concluded that these effects are due to affect-priming mechanisms operating during substantive processing (Salovey et al., 1991). Affective responses to messages that are processed in a heuristic or a substantive way are expected to at least partly mediate the effect of message components on the cognitive reactions (thoughts) about the message. Since the role of the message components studied her (argument strength and affective tone of the visual) are unclear, we formulate a research question: RQ1: Does the affective reaction to Alzheimer messages mediate the effect of message components (the strength of the arguments and the affective valence of the visual in the message) on thoughts about the message?

2.2

The Role of Cognitive Responses about the Message in Developing Responses to the Alzheimer Issue

Message-evoked affective responses can influence subsequent evaluations of the (Alzheimer) issue in different ways. On the one hand, message-evoked affect may directly transfer to the evaluation of and the responses to the message and the issue, due to peripheral transfer or assimilation processes, such as the Affect Transfer Mechanism (e.g., Coulter, 1998). On the other hand, a message may be processed more centrally, in which case message-evoked thoughts may play a greater role, leading to a more systematic evaluation of the ad. This systematic evaluation is often assumed to be objective, but could be biased as well. The latter is the case, for example, when message-evoked feelings bias the viewers’ message thoughts in a feeling-congruent way and the viewers base their evaluation of the message and the issue on these thoughts (Cfr. ELM, Petty and Ca-

58

Lewi and De Pelsmacker

cioppo, 1981, 1986). The Alzheimer issue may for most people be a serious concern and therefore a simple transfer or assimilation effect seems less likely. People may be motivated to process the message more centrally because the message effectively transports them into a serious reality (Jeandrain 2001, 2004; Coyle and Thorson, 2001). Consequently, biased processing may be expected in which case message-evoked thoughts will substantially mediate the effect of message-evoked affect on the overall judgments about the message and the issue, and will bias these judgments in a communication-thoughts consistent way. Since it is unclear to what extent these mechanisms will prevail, we again formulate a research question: RQ2: Do thoughts about the Alzheimer awareness message mediate the effect of the affective response to the message on the attitude towards the message, perceived message substance and thoughts about Alzheimer disease? An overview of the mechanisms studied is given in Figure 1.

Figure 1. Conceptual framework of the study

Additionally, we explore to what extent these mechanisms differ across demographic groups. More particularly, we investigate whether men and women and low and high education segments of the population differ with respect to the mediating mechanism of affect and message thoughts.

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

3 3.1

59

Method Stimulus Development

Six Alzheimer awareness messages were developed in which two levels of argument strength (weak and strong) were combined with three pictures of different affective valence, i.e. negative, neutral and positive. These messages were aimed at convincing people to seek early treatment. Initially, nine Alzheimerrelated images were selected eliciting positive, neutral and negative affective reactions and six weak (but still realistic) and strong message texts were prepared. The message texts were based on the medical literature (Lopez et al., 2002). In a first pre-test, these images and texts were tested among a group of 326 individuals. The objective of the first pre-test was twofold: 1) test what images were perceived as affectively most negative, positive and neutral; 2) test which message wordings and combination of arguments received the lowest and highest score on argument strength. In a second pre-test, we tested one extra fine-tuned weak message text and three additional strong message texts. The three images used in the main study obtained the following significantly different scores on a seven-point affective valence scale: negative image (M=3.44), neutral image (M=4.65), positive image (M=5.88). After the second pre-test, two message texts were selected that obtained the following significantly different ratings on the seven-point message strength scale: weak message (M=4.23) and strong message (M=5.68). The resulting three images and two texts were then combined into six Alzheimer’s disease awareness building messages and used in the main study (see Appendix for examples).

3.2

Procedure and Measures

A 2 (message strength: weak – strong) X 3 (affective valence of image: negative - neutral – positive) between subjects design was used to collect the data. The survey was conducted in Belgium and Spain (n=495). Two hundred and fifty Belgian and 245 Spanish individuals of between 55 and 65 years old were recruited by InSites, an Internet research company (240 females and 255 males). They received an e-mail with a link to a web survey. After answering a number of demographic questions, they viewed one of the six Alzheimer’s disease messages, to which they were randomly assigned. Subsequently, they answered questions on scales evaluating their responses to the message. The two independent variables, namely valence of the image in the message and argument strength, were manipulated in the experiment. Perceived message affect was measured on a 5-item 7-point scale (happy-sad, hopeful-desperate,

60

Lewi and De Pelsmacker

assured-worried, warm hearted-cold hearted, secure-unsafe) (α=.805). Perceived message substance was measured with a 5-item 7- point scale (persuasive-not persuasive, informative-not informative, credible-not credible, strong-weak, important-unimportant) (α=.904). The attitude towards the ad (Aad) was measured through a 3-item 7-point scale (I find the health communication strong/weak; this communication makes a good/bad impression on me; I like/dislike this communication) (α=.915). After exposure to the message, participants were asked to freely elicit their thoughts about the message and Alzheimer’s disease. All thoughts were coded according to the valence (negative, neutral, positive) and the focus (message or Alzheimer’s disease) of the thoughts by two independent judges. They agreed on 94% of the ‘Message thoughts’ and 91% of the ‘Alzheimer thoughts’. Disagreement was solved by a third judge. The variables ‘Message thoughts’ and ‘Alzheimer thoughts’ used in the analyses were defined as the sum of the number of positive (+1), negative (1) and neutral (0) thoughts. Finally also gender and education level were measured. Education level was recoded into a dichotomous variable, with ‘low’ education defined as equal as or less than high school.

4 4.1

Results Analysis on the Complete Data Set

In the first part of the analysis (RQ1), the mediating role of the affective reaction to the message on the relation between argument strength (STR) and image valence (IMA) on message thoughts is investigated using the Baron and Kenny approach (1986).

61

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

Table 1: Mediating role of message affect on the effect of argument strength and image valence on message thoughts (ANOVA) Baron &Kenny

Relation

F

Partial Eta2

Step1

Argument strength Image valence Strength x valence -> Message thoughts

F(1, 481) = 16.835 F(2,481) = 2.467 F(2,481) = .284

.034 .010 .001

<.001 .086 .753

Step2

Argument strength Image valence Strength x valence -> Message affect

F(1,489) = 6.043 F(2,489) = 3.553 F(2,489) = .029

.012 .014 .000

.014 .029 .971

Step3

Argument strength Image valence Strength x valence Message affect -> Message thoughts

F(1, 480) = 12.117 F(2,480) = 2.445 F(2,480) = .252 F(1,480) = 49.026

.025 .010 .001 .093

.001 .088 .778 <.001

p

In the first step, the effects of the independent variables STR and IMA on the dependent variable ‘message thoughts’ are estimated by means of ANOVA. The results show a significant main effect of STR (F(1,481)= 16.835, p<.001) and a marginally significant main effect of IMA (F(2, 481)= 2.467, p=.086). The interaction effect is not significant (F(2,481)= .284, p=.753) (Table 1). Strong messages (M=1.208) lead to a significantly better score on the thoughts about the message than weak messages (M=.351). Messages with a positive image (M=1.012) score as well as messages with a neutral image (M=.869). Both score marginally significantly better than messages with a negative image (M=.465). In the second step of the procedure, the effect of the independent variables on the mediating variable (message affect) is investigated. The ANOVA results show two significant main effects of STR (F(1,489)= 6.043 p=.014) (strong arguments generate more positive affective reactions than weak arguments) and IMA (F(2,489)= 3.553, p=.029). Messages with a neutral image (M=4.524) score significantly better than messages with a negative (M=4.251) or a positive (M=4.346) image. The latter two do not generate significantly different affective reactions. The interaction of affect and strength is not significant (F(2,489)=.029, p=.971). In the third step the effect of the mediating and the independent variables on the dependent variable is simultaneously esti-

62

Lewi and De Pelsmacker

mated. Therefore, message affect was added as a covariate in the step 1 ANOVA. The results demonstrate that message affect has a strong significant positive effect on message thoughts (F(1,480)= 49.026, p<.001). The main effect of argument strength remains significant (F(1,480)= 12.117, p=.001). The main effect of affect remains marginally significant (F(2,480)= 2.445, p=.088). The interaction effect of STR and IMA is again not significant (F(2,480)= .252, p=.778). Partial eta squares show that the impact of message affect (.093) is substantially stronger than that of message strength (.025) and image affect (.010). Message affect thus partially but substantially mediates the effect of message strength and image valence on message thoughts. The mediating role of message thoughts on the relationship between message affect and the dependent variables Aad, perceived message substance and Alzheimer thoughts was tested by means of three regression analyses (RQ2) (Tables 2-4). Step 1 shows a significant positive effect of the independent variable (message affect) on the dependent variable Alzheimer thoughts (β=.153, p=.001). The second step shows a significant positive impact of message affect (independent variable) on message thoughts (mediator) (β=.318, p<.001). In the third step, the effect of message affect and message thoughts on Alzheimer thoughts are simultaneously assessed. The impact of message affect on Alzheimer thoughts is significant (β=.142, p=.003), while the effect of message thoughts is not significant (β=.060, p=.206). Message thoughts do not mediate the effect of message affect on Alzheimer thoughts. Table 2: Mediating effect of message thoughts on the effect of message affect on Alzheimer thoughts (regression analysis) Baron &Kenny

Relation

R2

E

P

Step1

Message affect -> Alzheimer thoughts

.023

.153

.001

Step2

Message affect -> Message thoughts

.101

.318

<.001

Step3

Message affect Message thoughts -> Alzheimer thoughts

.029

.142 .060

.003 .206

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

63

Table 3: Mediating effect of message thoughts on the effect of message affect on perceived message substance (regression analysis) Baron &Kenny

Relation

R2

E

P

Step1

Message affect -> Perceived message substance

.181

.425

<.001

Step2

Message affect -> Message thoughts

.101

.318

<.001

Step3

Message affect Message thoughts -> Perceived message substance

.318

.298 .393

<.001 <.001

The same mediating analysis is conducted for the dependent variable perceived message substance. The results of step 1 indicate a significant positive effect of message affect on perceived message substance (β=.425, p<.001). In the previous analysis, we observed in the second step a significant positive effect of message affect on message thoughts (β=.318, p<.001). The third step shows that the effect of message affect on perceived message substance is significant (β=.298, p<.001). Also the effect of message thoughts is significant (β=.393, p<.001). Message thoughts partially mediate the effect of message affect on perceived message substance. Table 4: Mediating effect of message thoughts on the effect of message affect on the attitude towards the ad (regression analysis) Baron &Kenny

Relation

R2

E

P

Step1

Message affect -> Attitude towards the ad (Aad)

.269

.519

<.001

Step2

Message affect -> Message thoughts

.101

.318

<.001

Step3

Message affect Message thoughts -> Attitude towards the ad (Aad)

.390

.401 .368

<.001 <.001

64

Lewi and De Pelsmacker

Finally, the test is carried out for the dependent variable Aad. The results of step 1 indicate a significant positive effect of message affect on Aad (β=.519, p<.001). The results of the third step show that the effect of message affect (β=.401, p<.001) and message thoughts (β=.368, p<.001) on Aad are significant. Message thoughts partially mediate the effect of message affect on Aad.

4.2

Gender Differences

We have tested for gender effects in these mediation mechanisms by conducting an additional analysis in which a gender dummy variable and interaction variables between gender and the original independent variables in the basic analysis are added. In the first step of analysis 1, gender does not have a significant main effect on message thoughts (F(1,477)=.085, p=.770). The interaction effects are not significant for gender with STR (F(1,477)=.190, p=.663) and for gender with IMA (F(2,477)=.285, p=.752). From step 2 we can conclude that gender does not have a significant main effect on the mediating variable message affect (F(1, 485)=.029, p=.864). In the third step, where we analyse the effect of the mediating and the independent variables on the dependent variable simultaneously, we observe that gender does not have a significant main effect (F(1,475)=.180, p=.672). In addition, the interaction effects between gender on the one hand and argument strength and image valence on the other, are not significant [STR*gender: (F(1,475)=.613, p=.434); IMA*gender: (F(2,475)= .529, p=.589)]. The overall conclusion is that there is no significant effect of gender on the mediation role of message affect between argument strength and image valence on message thoughts. To test the moderating role of gender in the second model, we also repeated the mediating analysis for the dependent variables Alzheimer thoughts, message substance and the attitude towards the ad, again by adding a gender dummy variable and interaction effects between this variable and the other independent variables in the equations. Gender has no significant main effect on Alzheimer thoughts (β=.065, p=.728) or message thoughts (β=.020, p=.913). There is no significant effect of gender on the relation between message affect and Alzheimer thoughts (β=.013, p=.947). The interaction effects of gender with message thoughts and message affect were also not significant [message thoughts*gender: (β=-.017, p=.800); message affect*gender: (β=.131, p=.521)]. The same mediating analysis is conducted for the dependent variable perceived message substance. Again, gender has no significant main effect on perceived message substance (β=-.060, p=.731) and on message thoughts (β=.020, p=.913). Gender has no significant effect on the relation between message affect and perceived message substance (β=-.116, p=.480). The interaction effects of gend-

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

65

er with message thoughts and message affect were also not significant [message thoughts*gender: (β=-.082, p=.152); message affect*gender: (β=.157, p=.360). Finally, we test whether there is a difference between the gender groups when the analysis is carried out for the dependent variable Aad. Gender has no significant main effect on Aad (β=.152, p=.351) and on message thoughts (β=.020, p=.913). Gender has no significant effect on the relation between message affect and Aad (β= .123, p=.424). The interaction effects of gender with message thoughts and message affect were not significant either [message thoughts*gender: (β=-.048, p=.376); message affect*gender: (β=-.007, p=.968). Summing up, the mediating role of message thoughts for the relationship between message affect on the one hand and Alzheimer thoughts, message substance perception and Aad is not any different between men and women. All in all, the two mediating mechanisms of message affect and message thoughts found are not different between gender groups. They play the same role for men and women.

4.3

Differences in Education Level

We have tested for the effect of level of education in a similar way as for gender. The two defined groups consist of a low education group (high school or less) and a high education group (higher education) (variable EDU). First, we investigate the moderating effect of the education level on the mediating role of message affect (model 1). Education does not have a significant main effect on message thoughts (F(1,477)=.736, p=.392). The interaction effects are not significant for education with STR and for education with IMA either [STR*EDU (F(1,477)=.363, p=.547); IMA*EDU (F(2,477)=.066, p=.936)]. From step 2 we can conclude that education level does not have a significant main effect on the mediating variable message affect (F(1,485)=.179, p=.673). In the third step of the analysis, where we analyse the effect of the mediating and the independent variables on the dependent variable simultaneously, we observe that education level does not have a significant main effect (F(1,475)=.321, p=.571). In addition, the interaction effects are not significant [STR*EDU (F(1,475)=.307, p=.580); IMA*EDU (F(2,475)=.096, p=.909)]. In conclusion, there is no significant moderating effect of education on the mediating role of message affect on message thoughts. We now proceed with the analysis of the mediating role of message thoughts on the relationship between message affect and the dependent variables Alzheimer thoughts, perceived message substance and Aad. Again the education

66

Lewi and De Pelsmacker

dummy variable and the interaction between this variable and the independent variables in the original analysis were added to the model. First, the analysis was carried out for the dependent variable ‘Alzheimer thoughts’. Education level has no significant main effect on Alzheimer thoughts (β=-.189, p=.313) nor on message thoughts (β=-.071, p=.629). In the third step, education level has no significant effect on Alzheimer thoughts either (β=-.131, p=.499), and none of the interaction effects with education is significant [message thoughts*EDU: (β=.037, p=.624); message affect*EDU: (β=.114, p=.573)]. Education does not significantly change the mediating role of message thoughts on the relation between message affect and Alzheimer thoughts. The same analysis is conducted for the dependent variable perceived message substance (Table 5). Education level has a marginally significant main effect on perceived message substance (β=-.325, p=.058). Also the interaction effect between education level and affect is significant (β=.391, p=.025). Education level has no significant main effect on message thoughts (β=-.071, p=.692), but in the third step has a significant main effect on perceived message substance (β=-.424, p=.008). The interaction effects of education level with message affect and of education level with message thoughts are also highly significant (message affect*EDU: β=.518, p=.002; message thoughts*EDU: β=.195, p=.002). Higher educated individuals generally perceive Alzheimer messages as having less substance than lower educated individuals. The mediating role of message thoughts is still there, but highly educated people appear to be less influenced by message thoughts and more by their affective reaction to the message than lowly educated ones in developing perceptions of message substance. Finally, we test whether there is a difference between the education level groups when the analysis is carried out for the dependent variable Aad (Table 6). Education level has a marginally significant main effect on Aad (β=-.267, p=.098). Also the interaction effect between education level and affect is marginally significant (β=.315, p=.057). Education level has no significant main effect on message thoughts (β=-.071, p=.692). In step 3, education level has a marginally significant main effect on Aad (β=-.277, p=.070).

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

Table 5: Mediating effect of message thoughts on the effect of message affect on perceived message substance (regression analysis) – Moderating role of education level Baron &Kenny Step1

Step2

Step3

Relation Message affect EDU EDU x message affect -> Perceived message substance Message affect EDU EDU x message affect -> Message thoughts Message affect Message thoughts EDU EDU x message affect Edu x message thoughts -> Perceived message substance

R2

E

p

.191

.336 -.325 .391

<.001 .058 .025

.105

.289 -.071 .131

<.001 .692 .476

.339

.177 .532 -.424 .518 -.195

.001 <.001 .008 .002 .002

Table 6: Mediating effect of message thoughts on the effect of message affect on Aad (regression analysis) – Moderating role of education level Baron &Kenny Step1

Step2

Step3

Relation Message affect EDU Edu x message affect -> Attitude towards the ad Message affect EDU EDU x message affect -> Message thoughts Message affect Message thoughts EDU EDU x message affect EDU x message thoughts -> Attitude towards the ad

R2

E

p

.275

.446 -.267 .315

<.001 .098 .057

.105

.289 -.071 .131

<.001 .692 .476

.395

.327 .410 -.277 .317 -.060

.<.001 <.001 .070 .048 .316

67

68

Lewi and De Pelsmacker

The interaction effect between education level and message thoughts is not significant (β= -.060, p=.316). Partially consistent with the results for message substance, higher educated individuals generally perceive Alzheimer messages as less positive. The partially mediating role of message thoughts is still there, but highly educated people are less influenced by message thoughts and more by their affective reaction to the message (message affect x EDU: β= .317, p=.048) than lowly educated ones in their attitude towards the message.

5

Conclusions, Implications and Further Research

The role of affective and cognitive responses in processing Alzheimer awareness messages is mixed. The first analysis reveals that message thoughts are strongly driven by people’s affective response to Alzheimer messages. The more positive the affective responses to the message, the more positive message thoughts are. This is consistent with the affect priming and the affect-asinformation principle (Isen, 1987; Schwarz, 2003; Forgas, 1995). However, argument strength also remains a strong determinant of message thoughts, regardless of the affective response it evokes. Stronger arguments evoke more positive message thoughts than weaker ones. All together, this suggests a process of substantive and central processing that is partly mediated by the affect infusion mechanism. Indeed, both image valence and argument strength appear to lead to affective reactions that shape message thoughts, above and beyond the direct impact of argument strength. A remarkable result is that messages with images of neutral valence lead to more positive affective reactions and message thoughts than negatively or positively valenced images. Negatively valenced images may cause fear and distress that subsequently lead to negative affect and thoughts. Apparently also a positively valenced image leads to more negative responses than a neutral one. An explanation could be that people feel that a positive image does not fit with a serious problem like Alzheimer disease and dismiss it as inappropriate. Affect transfers to evaluative judgments about the message and the Alzheimer issue in different ways, but again the process is largely affect-driven. One part of the second analysis suggests that the development of Alzheimer thoughts is a peripheral affect transfer process (Coulter, 1998), since communication thoughts do not mediate the impact of affective response on Alzheimer thoughts. However, the effect of message affect on Aad and perceived message substance is partially mediated by message thoughts which points at more centrally but biased processing (Petty and Cacioppo, 1986).

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

69

These processes do not seem to be moderated by gender differences. The mediating role of message affect and message thoughts is essentially the same for men and women. However, the level of education does seem to have a significant moderating effect, at least on the perception of message substance and the appreciation of the message (Aad). In general, more highly educated individuals like the messages less and are less convinced about their substance. Highly educated people may be more prone to scrutinizing the message, and as a result may be more critical and skeptical about them. In developing perceptions about messages, these highly educated people appear to be significantly more influenced by their affective reactions to the message than by their thoughts about the message than their more lowly educated counterparts. This is, in a way, surprising. One might expect highly educated individuals to be more thought-oriented and cognitive in developing message evaluations. The contrary appears to be the case. The results of the present study have implications for health communication professionals. Processing health messages and forming health issue judgments following these messages is largely an affect-driven process. Positive affective reactions are an important prerequisite for the development of positive attitudes and thoughts. Nevertheless, argument strength remains an important driver of positive message processing too. Second, various message components may evoke positive affective reactions. Argument strength is one of them, but also the type of visual is an important driver of affective responses. Remarkably, there is no straightforward relationship between image valence and affective message response. On the basis of the present study, neutrally valenced images seem to be the optimal choice in this type of message. On the basis of these results, the same communication strategies can be used for different demographic groups based on gender and education, albeit that in case of highly educated groups, there could be more focus on the affective quality to the message. However, this result needs further confirmation. Further research should confirm these findings for different types of health messages. Target groups other than seniors, such as caregivers or doctors, may process various types of messages differently. Individual differences have been shown to moderate the effects of persuasive communication (e.g., Haughtvedt et al., 1992; Geuens and De Pelsmacker, 1999). Next to socio-demographic differences, a particularly important factor may be the level of experience an individual has with the disease. Caregivers, or people who have experience with Alzheimer due to contacts with a member of the family or a close friend developing Alzheimer may process Alzheimer messages differently from people who only have a vague idea of what Alzheimer does to people. One might assume that

70

Lewi and De Pelsmacker

more experienced people would also be more affectively involved, thus increasing the role of affect in processing Alzheimer messages. On the other hand, they might know more about the disease, and pay more attention to the arguments and thoughts about it. This is an avenue for further research. Another next step is to study the impact of message components, and affective and cognitive responses to messages on various aspects of behavioural intention and behaviour, such as visit to a doctor, search for more information, vigilance about early warning signals, etc.

6

References

Batra, R. & Ray, M.L. (1986), “Affective responses mediating acceptance of advertising,” in: Journal of Consumer Research, 13 (2), 234-249. Bagozzi, R.P., Gopinath, M. & Nyer, P.U. (1999), “The role of emotions in marketing,” in: Journal of the Academy of Marketing Science, 27(2), 184-206. Bain, L.J., Barker, W., Loewenstein, D.A. & Douara, R. (2008), “Towards an Earlier Diagnosis of Alzheimer Disease,” in: Proceedings of the 5th MCI Symposium 2007, Alzheimer Disease & Associated Disorders, 22(2), 99-110. Baron, M. R. & David A. Kenny (1986), “The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations,” in: Journal of Personality and Social Psychology, 51 (6), 1173-1182. Bower, G.H. (1981), “Mood and memory,” in: American Psychologist, 36, 129-148. Brown, S.P. & Stayman, D.M. (1992), “Antecedents and consequences of attitude towards the ad: A meta-analysis,” in: Journal of Consumer Research, 19, 34-51. Chaiken, S. (1980), “Heuristic versus systematic information processing and the use of source versus message cues in persuasion,” in: Journal of Personality and Social Psychology, 39, 752766. Cohen, J. B. & Areni, C. S. (1991), “Affect and Consumer Behavior,” in: T.S. Robertson, H.H. Kassarjian (Eds.): Handbook of Consumer Behavior, Englewood Cliffs NJ: Prentice-Hall, 188240. Coulter, K.S. (1998), “The effects of affective responses to media context on advertising evaluations,” in: Journal of Advertising, 27(4), 41-51. Coyle, J.R. & Thorson, E. (2001), “The effects of progressive levels of interactivity and vividness in web marketing sites,” in: Journal of Advertising, 30(3), 65-77. Edell, J. A. & Burke, C. M. (1987), “The power of feelings in understanding advertising effects,” in: Journal of Consumer Research, 14(3), 421-433. Fiedler, K. (1988), “Emotional mood, cognitive style, and behaviour regulation,” in: K. Fiedler, J. Forgas (Eds.): Affect, cognition and social behavior, Gottingen/Germany: Hogrefe, 100-119. Fiedler, K. (1990), “Mood-dependent selectivity in social cognition,” in: W. Stroebe, M. Hewstone (Eds.): European review of social psychology, New York: Wiley, Vol. 1, 1-32. Fiedler, K., Asbeck, J., Nickel, S. (1991), “Mood and constructive memory effects on social judgment,” in: Cognition and Emotion, 5, 363-378. Forgas, J. P. (Ed.). (1991), “Emotion and social judgments,” Elmsford, NY: Pergamon Press. Forgas, J. P. (1992), “Affect in social judgments and decisions: A multiprocess model,” in: M.

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

71

Zanna (Ed.): Advances in experimental social psychology, San Diego, CA: Academic Press, Vol. 25, 227-275. Forgas, J.P. (1995), “Mood and Judgment: The Affect Infusion Model (AIM),” in: Psychological Bulletin, Vol. 117, 1, 39-66. Forgas, J. P. & Bower, G. H. (1988), “Affect in social judgments,” in: Australian Journal of Psychology, 40, 125-145. Geuens M. & De Pelsmacker, P. (1999), “Individual differences and the communication effects of different types of emotional stimuli: "affect intensity" revisited,” in: Psychology and Marketing, 16(3), 195-209. Haugtvedt, C. P., Petty, R. E. & Cacioppo, J.T. (1992), “Need for cognition & advertising: Understanding the role of personality variables in consumer behavior,” in: Journal of Consumer Behavior, 3(1), 239-260. Isen, A. (1987), “Positive affect, cognitive processes and social behaviour,” in L. Berkowitz (Ed.): Advances in experimental social psychology, San Diego, CA: Academic Press, Vol. 20, 203253. Jeandrain, A.C. (2004), “Why and how do the telepresence dimensions influence persuasive outcome?,” in: Proceedings of the 7th International Workshop on Presence, Valencia, 73-79. Lombard, M. & Ditton, T. (1997), “At the heart of it all: The concept of presence,” in: Journal of Computer Mediated Communication, 3 (2), 1-42. Lopez, O.L., Becker, J.T., Wisniewski, S., Saxton, J., Kaufer D.I. & DeKosky, S.T (2002), “Cholinesterase inhibitor treatment alters the natural history of Alzheimer’s disease,” in: Journal of Neurol Neurosurg Psychiatry, 72: 310–314. Mitchell, A.A. (1988), “Current perspectives and issues concerning the explanation of feelings advertising effects,” in: Hecker, S., Stewart, D.S. (eds.): Non-verbal communication in advertising, Lexington, MA: Lexington Books, 122-144. Morris, J.D., Woo, C., Gaeson, J.A. & Kim, J. (2002), “The power of affect: predicting intention,” in: Journal of Advertising Research, 42(3), 7-17. Petty, R.E. & Cacioppo J.T. (1981), “Attitudes and persuasion: Classic and contemporary approaches,” Dubuque, IA: William C. Brown. Petty, R.E. & Cacioppo J.T. (1986), “Communication and Persuasion: Central and Peripheral Routes to Attitude Change,” New York: Springer-Verlag. Petty, R.E.; Schumann, D.W., Richman, S.A. & Strathman, A.J. (1993), “Positive mood and persuasion: Different roles for affect under high and low elaboration conditions,” in: Journal of Personality and Social Psychology, 64(1), 5-20. Raghunathan, R. & Pham, M.T. (1999), “All negative moods are not equal: Motivational influences of anxiety and sadness on decision making,” Organizational Behaviour and Human Decision Processes, 79(1), 56-77. Ray, M. L. & Batra, R. (1983), “Emotion and persuasion in advertising: What we do and don't know about affect,” in: Advances In Consumer Research, 10. Salovey, P. & Birnbaum, D. (1989), “Influence of mood on health-related cognitions,” in: Journal of Personality and Social Psychology, 57, 539-551. Salovey, P., O'Leary, A., Stretton, M., Fishkin, S., & Drake, C.A. (1991), “Influence of mood on judgments about health and illness,” in: J.P. Forgas (Ed.): Emotion and social judgments, Elmsford, NY: Pergamon Press, 241-262. Schwarz, N. (1990), “Feelings as information: Informational and motivational functions of affective states,” in: Higgins, E.T., Sorrentino R. (Eds.): Handbook of motivation and cognition: Foundations of social behavior, New York: Guilford Press, Vol. 2, 527-561.

72

Lewi and De Pelsmacker

Schwarz, N. & Bless, H. (1991), “Happy and mindless, but sad and smart? The impact of affective states on analytic reasoning,” in: J. P. Forgas (Ed.): Emotion and social judgments, Elmsford, NY: Pergamon Press, 55-71. Schwarz, N. & Clore G.L. (2003), “Mood as information: 20 years later,” in: Psychological Inquiry, 14 (3&4), 296-303. Seltzer, Ben (2007), “Early detection of Alzheimer’s disease,” in: Aging Health, 3(4), 449-456. Singer, J.A. & Salovey, P. (1988), “Mood and memory: Evaluating the network theory of affect,” in: Clinical Psychology Review, 8, 211-251. Visser, P.J., Verhey, F.R.J., Scheltens, Ph., Cruts, M., Ponds, R.W., Van Broeckhoven, C.L. & Jolles, J. (2002), “Diagnostic accuracy of the Preclinical AD Scale (PAS) in cognitively mildly impaired subjects,” in: Journal of Neurology, 249, 312-319. Williams-Piehota, P., Pizarro, J., Navarro Silvera, S.A., Mowad, L. & Salovey, P. (2006), “Need for cognition and message complexity in motivating fruit and vegetable intake among callers to the cancer information service,” in: Health Communication, 19(1), 75-84.

Affect is an Important Factor in Processing Alzheimer Disease Awareness Messages

Appendix

Translation Tr rannsllat ation ion fr ffrom rom om Dutch Dut utch ch to to English: Engl Engl En glish: Alzheimer Alz zhei eiimer m disease d sea di eaasee – Treeatm ment ent im mpro p ves pr veess daily d il da ily ly y fu funct n ion nc nct ioning oning ing and an and alleviates alleviatess behavioural beehav av viou io oural raal symptoms. syympt mptoms mp om oms ms. ms. Treatment improves functioning Eldderlly Elderly

73

Environmental Information Influences on Pro-environmental Behaviour Aimed at Fighting Climate Change: Main and Moderating Effects Nieves García de Frutos, University of Almería, Spain José Manuel Ortega Egea, University of Almería, Spain

1

Introduction

Climate change arguably represents the most challenging environmental threat of our time, owing to its magnitude, complexity, and irreversible effects (European Commission, 2008; Moser, 2010). This issue has garnered increased attention since the late eighties (Nerlich et al., 2010). At first, a great deal of media and public debate centered on the existence of climate change; later, the discussion shifted to the origins of climate change and global warming, be it anthropogenic (ie, human-induced) or part of the natural cycle of the Earth’s climate (Moser, 2010). Nowadays, there is general agreement in the scientific community regarding the existence and anthropogenic causes of climate change (Nerlich et al., 2010). In the last decade, intensified media coverage of climate change and global warming issues has substantially raised public awareness and concern for global environmental degradation (Mobley et al., 2010). For the most part, perhaps best exemplified by the film ‘An Inconvenient Truth’, communication efforts have been directed towards the dissemination of scientific knowledge on the existence, wide-ranging consequences, and human causes of climate change to the general public. Recent data from the European Union (EU) suggest that, despite continued attempts to promote more environmentally-friendly citizen and consumer behavior, there is still much room for improvement in most EU countries (European Environment Agency, 2010). This situation warrants further work (especially cross-national studies) into the precursors of environmentally responsible behaviors (Maibach et al., 2008). Individual ecological behavior is a function of a wide variety of internal and external influences. Psychographics consistently stand out as (direct or indirect) determinants of pro-environmental behavior (Bamberg and Möser, 2007). However, the predictive value of psychological factors can vary significantly across situations and population groups (Thøgersen, 2007). The links in proenvironmental behavior models may be contingent upon various sorts of “nonvolitional, socio-cultural” factors (Kaiser et al., 1999), including demographics (Diamantopoulos et al., 2003), risk perceptions (O’Connor et al., 1999), nationality (Franzen and Meyer, 2010), or the type of behavior assessed (Barr, 2007). Relatively few studies have examined individual actions specifically aimed at fighting climate change. S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_5, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

76

García de Frutos and Ortega Egea

The present study provides a better understanding of the factors that lead EU citizens to fight climate change. For this purpose, citizens’ level of information, motivations, and attitudes towards climate change are evaluated as potential determinants of pro-environmental behavior. Based on expectancy-value theory (Ajzen 1991; Fishbein and Ajzen, 1975), the research model proposes a direct link between environmental attitudes and pro-environmental behavior, and examines the role of (altruistic and egoistic) ecological motivations and level of information about climate change in developing favorable attitudes towards the environment. Four additional influences (threat perceptions, gender, age, and country value orientation) are examined as possible moderators of the hypothesized effects of (externally controllable) environmental information on ecological motivations and attitudes towards climate change. Hence, this work extends the available evidence in environmental research by assessing both the main and moderating effects of internal and exogenous factors on an ‘aggregate’ measure of pro-environmental behaviors. 2

Theory and Hypotheses

‘Green’ behavior is usually conceptualized as a mixture of self-interest and prosocial concerns (Bamberg and Möser, 2007). Researchers seeing individual ecological behavior as motivated by the desire to ‘do good’ for society often use Schwartz’s norm-activation model (NAM) (Schwartz, 1977) as a theoretical framework. In contrast, researchers regarding self-interest as the most important factor driving pro-environmental behavior tend to rely on Ajzen’s theory of planned behavior (TPB) (Ajzen, 1991; Bamberg and Möser, 2007). In addition, the TPB and NAM frameworks have been successfully combined to provide a more thorough analysis of pro-environmental decisions (Bamberg and Möser, 2007). 2.1

‘Core’ Research Model

A considerable amount of research has been devoted to assessing the impact of psychological variables on ecological behavior (eg, Leiserowitz, 2006). The predictive importance of psychographics has remained fairly stable over time (Bamberg and Möser, 2007), suggesting their potential role as determinants of individual pro-environmental behaviors. In the ‘core’ research model, three psychographics (i.e., level of information / knowledge, motivations, and attitudes towards climate change) are modeled as interrelated precursors of proenvironmental behavior aimed at fighting climate change.

Environmental Information Influences on Pro-environmental Behaviour

77

Figure 1. Research model: main and moderator effects

Attitude represents an individual’s positive o negative feelings about a psychological object (Kollmus and Aygeman, 2002). Such psychological objects comprise a wide array of discriminable aspects of the physical world, such as a physical entity, a person or group of people, an abstract concept, or a behavior (Ajzen and Gilbert, 2008). Environmental studies often include attitude as a contributing factor to ecological intention and behavior (Bamberg and Möser, 2007; Kaiser et al., 1999). Consistent with expectancy-value models (eg, Ajzen, 1991) and previous analyses of the ‘attitude–environmental behavior’ linkage (Bamberg and Möser, 2007; Kaiser et al., 1999), individuals with positive attitudes towards the environment will be more likely to engage in environmentally responsible behavior (Kaiser, et al., 1999; Kollmus and Aygeman, 2002). Nonetheless, the association between environmental attitudes and proenvironmental behavior is rather inconsistent (at best, moderate) across studies (Bamberg and Möser, 2007; Kaiser et al., 1999). Two factors suggest the importance of direct attitude effects on ecological behavior: the corresponding level of measurement generality in attitudes and behaviors, both being linked to the climate change issue; and the moderate to high degree of volitional control over performance of the behaviors, which are modeled as the ‘aggregate’ of eleven self-reported pro-environmental activities (Ajzen, 1991). H1: Attitudes towards climate change will have a positive effect on proenvironmental behavior aimed at fighting climate change. Motivation is understood as the inner driving force behind behavior (Kollmus and Aygeman, 2002). Environmental studies have explored a number of values underlying people’s motivation for environmentally responsible behavior (Schultz and Zelezny, 1999). Following Schwartz’s value model

78

García de Frutos and Ortega Egea

(Schwartz, 1977), many authors have distinguished between altruistic and egoistic values. Stern et al., (1993) further separated altruistic values into social and biospheric values. Gagnon-Thompson and Barton (1994) made a distinction between ecocentric and anthropocentric values. Ecocentric individuals are concerned about the environment for its own sake, and tend to undertake proenvironmental actions, even if it involves some sacrifice on their part (Gagnon Thompson and Barton, 1994); to a large extent, this behavioral pattern is anchored in biospheric values (Stern et al., 1993). Anthropocentric people’s behavior is grounded more in egoistic and social-altruistic values; accordingly, these individuals would engage in pro-environmental activity, to the extent it has positive consequences for mankind and does not decrease their quality of life or wealth (Gagnon-Thompson and Barton, 1994). Different motivational mechanisms (eg, altruistic and egoistic environmental motivations) may be operative in a person’s environmentally responsible decision-making (Maibach et al., 2008; Schultz and Zelezny, 1999). The available empirical evidence suggests that both ecocentric (biospheric) and anthropocentric (social or egoistic) values can result in pro-environmental behavior (Stern et al., 1993), particularly when different types of ecological values come into play in a coherent way (De Young et al., 1993). However, the motivational complexity of environmentally-friendly behavior may cause conflicts between an individual’s selfish and altruistic motives. Based on expectancy-value theory (Ajzen 1991; Fishbein and Ajzen, 1975), which postulate that attitude develops from salient beliefs about the outcomes of the behavior (eg, altruistic vs. selfish beliefs), and considering the ‘value/belief’ nature of ecological motivations (Grunert and Juhl, 1995), attitude will mediate the effect of (altruistic and egoistic) ecological motivations on proenvironmental behavior aimed at combating climate change. H2. Motivations to combat climate change will have a positive effect on attitudes towards climate change. The level of information or knowledge about environmental issues is regarded as a necessary (but not sufficient) antecedent of ecological awareness and behavior (Kollmus and Aygeman, 2002). Attitudes are not innate, but are acquired and altered in response to external stimuli, such as people’s access to various sources and types of information (Ajzen and Gilbert, 2008). The level and type of environmental information (and knowledge) can affect the attitudinal and behavioral responses to environmental information (Kaiser and Fuhrer, 2003; Kollmus and Aygeman, 2002). Basic knowledge is necessary to correctly interpret environmental threats and motivate pro-environmental actions (Kollmus and Aygeman, 2002). However, excessive information or very detailed technical data, concerning a complex and global-scale environmental issue such as climate change, can bring about frustration or denial of the issue (Kollmus and Aygeman, 2002). With respect to the type of environmental information or knowledge, information about the nature and causes of environmental problems, and information on ecological action strategies, appear

Environmental Information Influences on Pro-environmental Behaviour

79

to be especially useful in promoting pro-environmental behavior (Kaiser et al., 1999). Consequently, it seems reasonable to expect that better-informed citizens concerning the causes, ways of fighting, and consequences of climate change will behave in more environmentally responsible ways, compared to badlyinformed individuals. In line with social psychology literature (Ajzen and Gilbert, 2008) and most environmental studies (Bamberg and Möser, 2007; Kaiser et al., 1999), the ‘core’ model posits that respondents’ level of information about climate change has indirect effects on pro-environmental behavior, through environmental attitudes. H3a: Environmental information level will have a positive effect on attitudes towards climate change. Environmental information is also hypothesized to exert a significant effect on individual motivations for engaging in ecological behaviors (Thøgersen, 1994). Previous studies suggest that the ‘information–motivation’ linkage will be stronger for people with lower environmental concern, lacking the internal motivation to act ecologically (Schultz et al., 1995). Thus, providing highquality information about relevant climate change issues should be a powerful instrument for raising people’s levels of environmental motivation (Schultz et al., 1995; Young et al., 2010). Environmental information and messages appealing to both egoistic and altruistic ecological motivations should be more effective in encouraging pro-environmental behaviors (De Young et al., 1993). H3b: Environmental information level will have a positive effect on motivations to combat climate change. 2.2

Moderating Effects on ‘Environmental Information’ Linkages

The importance of analyzing moderation effects has been acknowledged in environmental research (Kaiser et al., 1999). Moderation analyses are consubstantial to social science; their abundance reflects the degree of maturity and sophistication of a research field (Frazier et al., 2004). Moderating effects are often examined when there are unexpectedly weak or inconsistent relationships between predictor and outcome variables across settings and population groups. In particular, moderation analyses help to clarify the questions of ‘when’ and ‘for whom’ a given variable is significantly or more strongly predictive of an outcome variable (Frazier et al., 2004). As depicted in Figure 1, the present study examines moderating effects on ‘environmental information’ linkages; that is, the moderation analyses will clarify some conditions and socio-demographic segments for which the hypothesized effects of environmental information on ecological motivations and attitudes towards climate change vary in magnitude. A deeper understanding of the predictive conditions of environmental information (about climate change) is important for interventions aiming to promote environmentally-friendly behavior patterns, owing to the ‘externally

80

García de Frutos and Ortega Egea

controllable’ character of environmental information and messages. The consistency of ‘environmental information–attitudes’ and ‘environmental information–motivation’ links is likely to be contingent on a variety influences (Diamantopoulos et al., 2003; Kaiser et al., 1999; O’Connor et al., 1999). Potential moderators include not only information traits (e.g., type of environmental content, or tone of messages), but also psychological and sociodemographic characteristics of the target audience (von Borgstede and Anderson, 2010). In this work, three external traits (gender, age, and country value orientation) and one internal constraint (perceived threat of climate change) are examined as moderating variables. Perceived threat of climate change. Many studies on pro-environmental behavior have included some measure of perceived risk or threat of environmental problems (eg, climate change or global warming) to explain differences between more and less environmentally proactive people (Leiserowitz, 2006; O’Connor et al., 1999). Theoretical and empirical evidence exists to suggest moderating influences of perceived threat of climate change: environmental knowledge and value orientations may relate to increased and lowered risk perceptions (O’Connor et al., 1999; Withmarsh, 2008); research also suggests that the relative importance of internal reasons (eg, ecological motivations) for pro-environmental behaviors, such as those aimed at fighting climate change, will be higher for people high in perceived environmental threat (Schultz et al., 1995). This latter argument suggests, somewhat exploratively, that individuals rating climate change as a more serious problem (ie, high threat perceptions) will be less influenced by external stimuli in their proenvironmental decision-making. H4a: For respondents with ‘high’ perceived threat of climate change, the effect of environmental information level on attitudes towards climate change will be ‘attenuated’. H4b: For respondents with ‘high’ perceived threat of climate change, the effect of environmental information level on motivations to combat climate change will be ‘attenuated’. Gender. In most analyses of gender differences in pro-environmental behavior, women have displayed lower levels of environmental knowledge than men (Diamantopoulos et al., 2003; Kollmus and Aygeman, 2002). This pattern has been attributed to women’s traditional roles and socialization process (Straugham and Roberts, 1999); for instance, the traditional ‘wife and housekeeper’ role has kept women away from scientific and technological fields (Olli et al., , 2001). However, women have been found to be more responsive to ‘green’ advertising (Haytko and Matulich, 2010). Hence, female respondents will more likely be influenced by environmental information than men (von Borgstede and Anderson, 2010). Consistent with this pattern, gender should be an important moderator of the environmental information links in the ‘core’ research model.

Environmental Information Influences on Pro-environmental Behaviour

81

H5a: For female respondents, the effect of environmental information level on attitudes towards climate change will be ‘intensified’. H5b: For female respondents, the effect of environmental information level on motivations to combat climate change will be ‘intensified’. Age. There is much controversy concerning the effects of age on environmental knowledge, attitudes, and pro-environmental behavior. There is some agreement that younger people tend to be more aware and concerned about environmental issues (Diamantopoulos et al., 2003; Olli et al., 2001; Straugham and Roberts, 1999), mostly due to overlapping lifecycle and cohort effects (Franzen and Meyer, 2010). Regarding environmental knowledge, it is not possible to find clear age patterns, with age having either non-significant (Cottrell, 2003), or weak negative relations to knowledge (Diamantopoulos et al., 2003). In line with these findings, younger individuals were found to be more knowledgeable about an environmental problem such as acid rain, compared to their older counterparts (Arcury et al., 1987). Based on the available evidence, two preliminary hypotheses are developed for the moderating effects of age on ‘environmental information/knowledge’ linkages. H6a: For younger respondents, the effect of environmental information level on attitudes towards climate change will be ‘intensified’. H6b: For younger respondents, the effect of environmental information level on motivations to combat climate change will be ‘intensified’. Country value orientation. It is widely believed that differences in environmental knowledge, concern and behavior may be a consequence of shared values within a culture (Schultz and Zelezny, 1999). Thereby, most studies have relied on ‘postmaterialism’ to explain country differences (see Franzen and Meyer, 2010). Postmaterialist theory posits that environmental concern emerges only once basic individual needs are fulfilled (Dunlap and Mertig, 1995). Consistent with this, people from countries with a predominant postmaterialist value orientation would be more concerned and engage more often in pro-environmental behaviors (Franzen and Meyer, 2010). ‘Postmaterialist’ countries are also expected to show higher levels of environmental knowledge; these societies put greater emphasis on immaterial resources, such as providing citizens with improved education and freedom of information (Dunlap and Mertig, 1995). On this basis, the following hypotheses are proposed suggesting a greater predictive importance of respondents’ environmental information level in countries with predominant postmaterialist values. H7a: For ‘postmaterialist’ countries, the effect of environmental information level on attitudes towards climate change will be ‘intensified’. H7b: For ‘postmaterialist’ countries, the effect of environmental information level on motivations to combat climate change will be ‘intensified’.

82 3 3.1

García de Frutos and Ortega Egea

Methods Sample

The empirical analyses are conducted on the survey database ‘Eurobarometer 69.2 – Europeans’ attitudes towards climate change’. This cross-national dataset covers a representative sample of the population (aged 15 and over) of the EU27 member states, plus three candidate countries (Croatia, Turkey, and the Former Yugoslav Republic of Macedonia) and the Turkish Cypriot Community. The fieldwork was carried out between March 25th and May 4th 2008 through face-to-face interviews in people’s homes, and in the appropriate national language. In all countries, a stratified, multistage probability design was used for selection of sampling units. A total of 30,170 responses were collected (with few exceptions, approximately 1000 per country). The questionnaire addressed European citizens’ attitudes, motivations, level of knowledge, and engagement in proenvironmental behaviors aimed at fighting climate change; other relevant measures were available, including materialist/post-materialist values, perceived seriousness of climate change, and socio-demographic indicators, such as gender, age, education level, political ideology, and country. Access to the Eurobarometer data was provided by the GESIS Data Archive for the Social Sciences (Cologne, Germany). 3.2

Statistical Procedures

3.2.1

‘Core’ Model Tests

First, exploratory factor analysis (EFA) is used to check the dimensional structure of the ‘core’ research model. Next, confirmatory factor analysis (CFA) and structural equation modeling (SEM), using the EQS v6.1 statistical software, are respectively applied to examine the psychometric properties of the scales and test the hypothesized relations in the ‘core’ model. In addition, several statistical tests and criteria (such as Cronbach’s alpha values, composite reliabilities, the average variance extracted (AVE), and chi-square difference tests) are used to analyze measurement reliability and convergent and discriminant validity. 3.2.2

Moderation Analyses

Two common methods for testing moderating effects have been described: multiple regression and multi-group analysis (multi-group SEM). Owing to the characteristics of the ‘core’ research model and moderator variables (i.e., perceived threat of climate change, gender, age, and country value orientation), the multi-group SEM approach is used in this work. Hence, differences in the hypothesized ‘environmental information / knowledge’ effects were assessed

Environmental Information Influences on Pro-environmental Behaviour

83

between the following groups of respondents: ‘high’ vs. ‘low’ perceived threat of climate change (median split defined groups); ‘male’ vs. ‘female’ gender categories; three age groups split into terciles: ‘15–38 years’, ‘39–56 years’, and ‘more than 57 years old’; and three country groups based on materialist vs. postmaterialist values: ‘materialist countries’, ‘countries with mixed values’, and ‘post-materialist countries’. 4

Results

From the initial sample of 30,170 cases, the authors selected those respondents who ‘agreed’ or ‘tended to agree’ with the statement ‘You personally have taken actions aimed at helping to fight climate change’, with the aim of testing the hypotheses of the study. In the year 2008, EU citizens undertaking some form pro-environmental activity accounted for 51.7% of the sample (n=17,233); only these individuals were included in subsequent analyses. 4.1

‘Core’ Model Results

4.1.1

Psychometric Properties of the Scales

The dependent factor (pro-environmental behavior) captures respondents’ level of individual environmental proactivity, through the aggregation of eleven specific pro-environmental actions into a single, unidimensional construct. Both exploratory and confirmatory factor analyses suggested the distinction between positive and negative attitudes toward climate change. Consistent with previous literature (eg, Thøgersen, 2007), the analyses supported a second-order factor of motivations to combat climate change, formed by altruistic and egoistic ecological motivations. The obtained Cronbach’s alpha and Raykov’s rho coefficients were well above 0.8 for the construct of ‘environmental information level’ (Straub, 1989). For both positive and negative attitudes towards climate change, and ecological motivations, Cronbach’s alphas were around 0.5, but all Raykov’s rho coefficients exceeded the more lenient cut-off of 0.6 (Hair et al., 1998). These results were expected, owing to the use of Eurobarometer data with ‘predetermined indicators’. Thus, the internal consistency of the multi-item scales may be acceptable in this study (Hair et al., 1998). Two separate tests were sequentially performed to assess the discriminant validity of each pair of constructs. First, following Anderson and Gerbing’s (1988) recommendations, a confidence interval test was used to assess the discriminant validity of each pair of factors. Next, chi-square difference tests were applied only to those pairs of factors showing problems of discriminant validity, according to the confidence interval test. All tests indicated that discriminant validity exists between the analyzed pairs of factors.

84 4.1.2

García de Frutos and Ortega Egea

Explaining Pro-environmental Behavior

The results provided support for all the hypothesized relationships in the ‘core’ model (see Figure 2). Despite accounting for only 15.3% of the variance, positive attitude towards climate change was the single direct predictor of proenvironmental behavior (std. β=0.389; p<0.01), thus providing support for H1. Conversely, negative attitude did not significantly detract from ecological behavior aimed at fighting climate change. The results also supported the positive, direct effects of ecological motivations and environmental information level on respondents’ attitudes towards climate change (support for H2 and H3a); these effects were larger on positive than on negative environmental attitude. Motivation had a greater impact than information level on positive attitude (std. β=0.758; p<0.01 vs. std. β=0.438; p<0.01), both explaining a large percent of the variance. As expected, ecological motivations exerted a negative influence on negative environmental attitude (std. β=-0.265; p<0.01). Surprisingly though, the effect of information level (about climate change) on negative attitude was positive, but weak (std. β=0.093; p<0.01). In line with H3b, the level of information about climate change had a positive, though modest, impact on (altruistic and egoistic) motivations to combat climate change (std. β=0.189; p<0.01). The final version of the ‘core’ model showed a very good fit to the data. The final version of the ‘core’ model showed a very good fit to the data. The assessment of absolute fit revealed that the scaled chi-square (S-B χ2) was significant at p<0.01. Owing to the sensitivity of the chi-square statistic to large sample sizes, alternative indices were assessed (Hair et al., 1998). RMSEA (0.033) fell below the recommended level of 0.08. Incremental fit measures also indicated a good model fit, with all values (NNFI, NFI, CFI, IFI, and MFI) exceeding the rigorous threshold of 0.95. 4.2

Moderation Results

The four moderating variables examined here (threat perceptions, gender, age, and country value orientation) significantly intensified or attenuated at least one causal effect flowing from ‘environmental information level’, according to chisquare difference tests between constrained and unconstrained pairs of models – ie, paths constrained to be equal vs. paths freely estimated across subgroups. Moreover, two variables (age and country value orientation) moderated all three structural paths from environmental information level to positive and negative attitudes, and to ecological motivations, as included in the final ‘core’ model (see Figure 2).

85

Environmental Information Influences on Pro-environmental Behaviour

Table 1: Multi-group results Information level → Moderating Positive attitude variables / subgroups Std. β t-value Perceived threat of climate change High threat 0.301 18.792*** Low threat 0.321 18.792*** Gender Male 0.305 12.288*** Female 0.336 16.600*** Age 15–38 years 0.318 15.007*** 39–56 years 0.292 15.007*** 57+ years old 0.352 13.521*** Country value orientation Post-materialist 0.380 12.090*** Materialist -0.004 -0.062 n.s. Mixed values 0.448 12.090***

Information level → Negative attitude Std. β t-value

Information level → Ecological motivations Std. β t-value

-0.001 0.156

-0.002 n.s. 8.795***

0.163 0.198

10.956*** 10.956***

0.082 0.085

6.772*** 6.772***

0.190 0.194

11.865*** 11.865***

0.038 0.113 0.116

1.787 n.s. 7.555*** 7.555***

0.142 0.231 0.212

5.251*** 11.636*** 11.636***

-0.020 -0.042 0.103

-0.807 n.s. -1.477 n.s. 5.109***

0.193 0.371 0.200

9.462*** 5.825*** 9.462***

*p<0.05, **p<0.01, ***p<0.001.

As shown in Table 1, perceived threat of climate change moderated only the non-hypothesized and, unexpectedly, positive relationship between environmental information level and negative attitude towards climate change. This path became non-significant for respondents with ‘high’ threat perceptions; thus, neither H4a nor H4b were supported. Concerning gender differences, the ‘information level–positive attitude’ effect was intensified for female respondents (partial support for H5a), but no significant differences were found between men and women in the paths from environmental information level to negative environmental attitude and to ecological motivations (lack of support for H5b). The environmental attitudes (particularly, negative ones) and ecological motivations of younger EU citizens appear to be shaped, to a lesser extent, by their level of environmental information or knowledge; the opposite effects were preliminarily proposed in H6a and H6b. As for respondents’ country, the path from environmental information level to positive environmental attitude was significantly attenuated in the ‘materialist’ country group, compared to the ‘post-materialist’ and ‘mixed’ countries; this finding is partly in line with H7a. Non-significant influences of information level on negative environmental attitude were observed both in ‘materialist’ and ‘post-materialist’ countries. By contrast, negative attitude was most strongly (positively) influenced by environmental information in countries with a ‘mixed’ value orientation. Finally, the ‘information–motivation’ linkage was attenuated in the ‘postmaterialist’ country group; this pattern is contrary to that hypothesized in H7b.

86

García de Frutos and Ortega Egea

Figure 2: Assessment of the final ‘core’ model and moderation results. Note: The figure shows the standardized path coefficients and t-values in the final ‘core’ model and depicts the significant moderation effects.

5

Discussion

The findings of this study shed light on individual pro-environmental behavior aimed at fighting climate change. The analyses examined the role of EU citizens’ level of information, motivations, and attitudes towards climate change in their pro-environmental decision-making. Moreover, three external traits (gender, age, and country value orientation) and one internal constraint (perceived threat of climate change) were examined as possible moderators of ‘environmental information level’ influences. The ‘core’ research model was consistent with expectancy-value theory (Ajzen, 1991; Fishbein and Ajzen, 1975), and supported the idea that environmental attitudes are important precursors of individual behavior aimed at fighting climate change. Unlike most other studies, two attitudinal factors were differentiated –ie, positive and negative attitudes; on the positive side, only positive environmental attitude was significantly (and positively) predictive of pro-environmental behavior. The modest explained variance (approximately 15%) points to unmodeled antecedents of pro-environmental activity (most notably, behavioral intentions). Respondents’ (altruistic and egoistic) ecological motivations and level of information about climate change were respectively identified as internal and external factors influencing both positive and negative attitudes towards climate change. Environmental information level was also found to have a positive direct effect on ecological motivations to combat climate change. Based on the findings, informing about environmental issues, such as climate change, may undesirably contribute to negative environmental

Environmental Information Influences on Pro-environmental Behaviour

87

attitudes. Although this effect was rather weak, it suggests that not all types of environmental messages and information (eg, information about the causes, ways of fighting, and consequences of climate change) can be assumed to be equally effective in promoting environmentally responsible behaviors. Further understanding of the predictive conditions and socio-demographic segments for which environmental information and messages (about climate change) are more or less effective is important for interventions and campaigns aimed at promoting more environmentally-friendly citizen and consumer behavior. The analyses provided preliminary evidence of significant moderating influences of various internal and exogenous variables on ‘environmental information’ linkages (ie, ‘information level–attitudes’ and ‘information level– motivations’). Variations in environmental information effects between respondent groups, defined by threat perceptions, gender, age, and country value orientation, may be partly attributed heterogeneous interest and concern about environmental issues –eg, women typically show higher levels of environmental concern (Haytko and Matulich, 2010). Moderation results also highlight the importance of carefully designing environmental information and messages to account for socio-demographic and psychological influences (Leiserowitz, 2006). In this regard, there may be special challenges in environmental awareness campaigns targeted at ‘younger’ population segments and ‘materialistic’ countries. 5.1

Implications for Environmental Awareness Campaigns

The assessment of main and moderating effects on ‘environmental information level’ linkages provides important insights for promoting pro-environmental activity at the individual level. Overall, environmental information and knowledge appear to be necessary, but not sufficient, preconditions for environmentally responsible behavior. To a lesser extent, the findings warn of unintended effects of environmental information and messages, eliciting negative environmental attitudes. The effectiveness of environmental messages (and information) is dependent on its alignment with people’s attitudes, motivations and current goals (Maibach et al., 2008; von Borgstede and Anderson, 2010). However, a great deal of environmental communication appeals to ‘global well-being’ and the ‘preserving the Earth for the young and future generations’. Such environmental messages will be adequate for population segments with predominant altruistic and self-transcendent values, but not for individuals with an egoistic value orientation. Targeting environmental messages to egoistic ecological motivations, such as climate change dangers to personal health and safety or the monetary benefits of saving resources, may have greater effect on certain population groups, such as countries with a predominant ‘materialistic’ value orientation.

88

García de Frutos and Ortega Egea

In climate change communication, ‘what’ is said is as important as ‘how’ is it said (Moser, 2010). Thus, visual imagery, colors, music, and other non-verbal cues can be very effective in eliciting motivational and attitudinal responses towards climate change and global warming issues. The choice of communication channel and spokesperson can be used to target tailored environmental information and messages to the characteristics of specific sociodemographic and psychographic segments, in accordance with the identified moderating effects on ‘environmental information level’ relationships. Widely available reports on international media use will greatly aid in this process. Finally, environmental awareness campaigns on climate change and global warming issues should avoid vague, contradictory, and complex ecological messages. On the contrary, the arguments used in environmental information and messages need to be clear and simple; in addition, different kinds of environmental information (eg, about the causes, ways of solving, and consequences of climate change) “must work together in a convergent manner” (Kaiser and Fuhrer, 2003). Research on climate change communication has found that environmental messages tend to be more effective when focused on the serious environmental consequences from climate change, while at the same time clarifying that each person can do something about it (Maibach et al., 2008). 5.2

Limitations and Future Research

This study is not free of limitations. Maybe the most important one relates to the use of Eurobarometer data. Despite its benefits, such as providing a rich, crossnational source of information, using this type of data had restrictive effects on research design. Thus, the authors could not participate in the development of questionnaire items. In addition, the use of Eurobaromenter data precluded the inclusion of other relevant factors, such as behavioral intentions, subjective norm, or perceived behavioral control (Ajzen, 1991). A social desirability effect may also bias the results of survey research focusing on ecological behavior –ie, respondents might have exaggerated their responses to project a socially desirable image of themselves. The proposed research model aims to explain complex and dynamic human behaviors with ethical and ecological implications. Thus, future studies should consider including ‘unmodeled’ factors such as citizens’ perceptions of selfrelevance or involvement with environmental issues. In addition, alternative theoretical frameworks, such as Fazio’s MODE model (Fazio, 1990), could be used to assess the consistency of the links between motivations, attitudes, and pro-environmental behavior. Further research should also considerate additional socio-demographic variables as moderators; among other traits, political ideology and level of education tend to be related with higher commitment to

Environmental Information Influences on Pro-environmental Behaviour

89

pro-environmental behaviors. For instance, political ideology may be useful in selecting the most appropriate spokesperson and media (Straugham and Roberts, 1999).

6

References

Arcury, T.A.; Scollay, S.J. & Johnson, T.P. (1987), “Sex Differences in Environmental Concern and Knowledge: the Case of Acid Rain,” in: Sex Roles, Vol. 16 (9/10), 463-472. Ajzen, I. (1991), “The Theory of Planned Behavior,” in: Organizational behavior and human decision processes, Vol. 50 (2), 179-211 Ajzen, I. & Gilbert, N. (2008), “Attitudes and the Prediction of Behavior,” in: W. D. Crano and R. Prislin (eds.): Attitudes and Attitude Change, Psychology Press, New York, NY, 289-311. Anderson, J.C. & Gerbing, D.W. (1988), “Structural Equation Modelling in Practice: A Review and Recommended Two-step Approach,” in: Psychological Bulletin, Vol. 103 (3), 411-423. Bamberg, S. & Möser, G. (2007), “Twenty Years after Hines, Hungerford and Tomera: a New Meta-analysis of Psychosocial Determinants of Pro-environmental Behavior,” in: Journal of Environmental Psychology, Vol. 27 (1), 14-25. Barr, S. (2007), “Factors Influencing Environmental Attitudes and Behaviors: a U.K. Case Study of Household Waste Management,” in: Environment and Behavior, Vol. 39 (4), 435-473. Cottrell, S. P. (2003), “Influence of Sociodemographics and Environmental Attitudes on General Responsible Environmental Behaviour Among Recreational Boaters,” in: Environment and Behavior, Vol. 35 (3), 347-375. De Young, R.; Duncan, A.; Frank, J.; Gill, N.; Rothman, S.; Shenot, J.; Shotkin, A. & Zweizig, M. (1993), “Promoting Source Reduction Behavior. The Role of Motivational Information,” in: Environment and Behavior, Vol. 25 (1), 70-85. Diamantopoulos, A.; Schlegelmilch, B. B.; Sinkovics, R. R. & Bohlen, G. M. (2003), “Can sociodemographics Still Play a Role in Profiling Green Consumers? A Review of the Evidence and an Empirical Investigation,” in: Journal of Business Research, Vol. 56, 465-480. Dunlap, R. E. & Mertig, A. G. (1995), “Global Concern for the Environment: Is Affluence a Prerequisite?” in: Journal of Social Issues, Vol. 51, 121-137. European Commission (2008). Europeans’ Attitudes Towards Climate Change. Available as 30112009 from http://ec.europa.eu/public_opinion/archives/ebs/ebs_300_full_en.pdf European Environment Agency (2010). Tracking Progress Towards Kyoto and 2020 targets in Europe. EEA Report No 7/2010. EEA, Copenhagen, 2010. Available as 16102010 from http://www.eea.europa.eu/publications/progress-towards-kyoto/at_download/file Fazio, R. H. (1990), “Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework,” in: M. P. Zanna (ed.), Advances in experimental social psychology, Vol. 23, Academic Press, New York, NY, 75-109. Fishbein, M. & Ajzen, I. (1975), “Belief, attitude, intention, and behavior: an introduction to theory and research,” Addison-Wesley Reading, MA. Franzen, A. & Meyer, R. (2010), “Environmental Attitudes in Cross-national Perspective: A Multilevel Analysis of the ISSP 1993 and 2000,” in: European Sociological Review, Vol. 26 (2), 219-234 Frazier, P.; Tix, A. P. & Barron, K. E. (2004), “Testing Moderator and Mediator Effects in Counseling Psychology Research,” in: Journal of Counseling Psychology, Vol. 51 (1), 115-134 Gagnon Thompson, S. & Barton, M. (1994), “Ecocentric and Anthropocentric Attitudes Toward the Environment,” in: Journal of Environmental Psychology, Vol. 14 (2), 149-157. Grunert, S.C. & Juhl, H.J. (1995), “Values, Environmental Attitudes and Buying of Organic Foods,” in: Journal of Economic Psychology, Vol. 16 (1), 39-62.

90

García de Frutos and Ortega Egea

Hair, J.F.Jr., Anderson, R.E., Tatham, R.L., Black, W.C. (1998). “Multivariate data analysis,” 5th edition, Prentice Hall, Englewood Cliffs. Haytko, D. L. & Matulich, E. (2010), “Green Advertising and Environmentally Responsible Consumer Behaviors: Linkages Examined ,” in: Journal of Management and Marketing Research, Vol. 1, 1-11. Kaiser, F.G. & Fuhrer, U. (2003), “Ecological behavior’s dependency on different forms of knowledge,” Applied Psychology: An International Review, Vol. 52 (4), 598-613. Kaiser, F. G.; Wölfing, S. & Fuhrer, U. (1999), “Environmental Attitude and Ecological Behavior,” in: Journal of Environmental Psychology, Vol. 19 (1), 1-19. Kollmuss A. & Agyeman, J. (2002), “Mind the Gap: Why do People act Environmentally and What are the Barriers to Pro-environmental Behavior?,” in: Environmental Education Research, Vol. 8 (3), 239-260. Leiserowitz, A. (2006), “Climate Change Risk Perception and Policy Preferences: The Role of Affect, Imagery, and Values,” in: Climate Change, Vol. 77, 45-72. Maibach, E., Roser-Renouf, C. & Leiserowitz, A. (2008), “Communication and Marketing as Climate Change Intervention Assets: A Public Health Perspective,” in: American Journal of Preventive Medicine, Vol. 35, 488-500. Mobley, C., Vagias, W.M., & DeWard, S.L. (2010), “Exploring Additional Determinants of Environmentally Responsible Behavior: The Influence of Environmental Literature and Environmental Attitudes,” in: Environment & Behavior, Vol. 42 (4), 420-447. Moser, S.C., (2010), “Communicating Climate Change: History, Challenges, Process and Future Directions”, in: Wiley Interdisciplinary Reviews: Climate Change, Vol. 1 (1), 31-53. Nerlich, B.; Koteyko, N. & Brown, B. (2009), “Theory and Language of Climate Change Communication,” in: Wiley Interdisciplinary Reviews: Climate Change, Vol. 1 (1), 97-110. Olli, E.; Grendstand, G. & Wollebaek, D. (2001), “Correlates of Environmental Behaviors: Bringing Back Social Context,” in: Environment and Behavior, Vol. 33 (2), 181-208. O’Connor, R. E.; Bord, R. J. & Fisher, A. (1999), “Risk Perceptions, General Environmental Beliefs, and Willigness to Address Climate Change,” in: Risk Analysis, Vol. 19, 461-471. Schultz, P. W.; Oskamp, S. & Mainieri, T. (1995), “Who Recycles and When: a Review of Personal and Situational Factors,” in: Journal of Environmental Psychology, Vol., 15 (2), 105-121 Schultz, P.W. & Zelezny, L. (1999), “Values as Predictors of Environmental Attitudes: Evidence for Consistency Across 14 Countries,” in: Journal of Environmental Psychology, Vol. 19 (3), 255265. Schwartz, S. H. (1977), “Normative Influences on Altruism,” in: L. Berkowitz (ed.), Advances in Experimental Social Psychology, Vol. 10, Academic Press, New York, 221-279. Stern, P. C.; Dietz, T. & Kalof, L. (1993), “Value Orientations, Gender and Environmental Concern,” in: Environment and Behavior, Vol. 25 (5), 322-348. Straub, D.W. (1989), “Validating Instruments in MIS Research” in: MIS Quarterly, Vol. 13(2), 147168. Straughan, R.D. & Roberts, J.A. (1999), “Environmental Segmentation Alternatives: a Look at Green Consumer Behavior in the New Millennium,” in: Journal of Consumer Marketing, Vol. 16, 558-575. Thøgersen, J. (1994), “A model of recycling behaviour, with evidence from Danish source separation programmes,” International Journal of Marketing Research, Vol. 11 (2), 145-163. Thøgersen, J. (2007), “The Motivational Roots of Norms for Environmentally Responsible Behavior,” in: Proceedings of the Nordic Consumer Policy Research Conference. von Borgstede, C. & Andersson, K. (2010), “Environmental Information. Explanatory Factors for Information Behavior,” in: Sustainability, 2 (9), 2785-2798. Whitmarsh, L. (2008), “Are flood victims more concerned about climate change than other people? The role of direct experience in risk perception and behavioural response,” in: Journal of Risk Research, Vol. 11 (3), 351-374. Young, W., Hwang, K., McDonald, S., & Oates, C. J. (2010), “Sustainable consumption: Green consumer behaviour when purchasing products,” Sustainable Development, Vol. 18 (1), 20-31.

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption: Affective Responses and Advertising Effectiveness Mª José Miquel-Romero, University of Valencia, Spain Franco Manuel Sancho-Esper, University of Alicante, Spain

1

Introduction

It is widely accepted that a framework is necessary to regulate advertising activities. Advertising has a responsibility within our economic system, to provide information on new and existing products and services. Economic health of both buyers and competitors is affected if the advertising system is injected with false or misleading claims. Thus, Government and self-regulation associations should create a set of rules which define and control the behaviour of the advertising companies. Taking into account the power attributed to advertising for influencing consumer behaviour, some of those mentioned rules are established to protect the audience. In this sense, broad social issues have been raised for decades, concerning the role of advertising. Questions such as whether it is ethical to advertise cigarettes or to engage in alcohol advertising, which might persuade young people to start smoking or drinking, have become topics of controversial debate for years. For these products in several countries, severe measures have been applied, like tax increase, prohibition of broadcasting tobacco/alcohol advertising in specific mass media or specific public places, prohibition of its consumption and/or sale in specific public places, or even a clear identification in the product packaging of the health problems caused by the consumption of the product (tobacco), or the recommendation of a responsible consumption (alcohol). Alcohol consumption is nowadays a generalised behaviour, and although its consumption has occurred for thousands of years, many of the varied health effects have been discovered fairly recently (WHO, 2010). It has health and social consequences via intoxication (drunkenness), dependence (habitual, compulsive and long-term drinking), and other biochemical effects, and its harmful use causes an estimated 2.5 million deaths every year, of which a significant proportion occur in the young.

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_6, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

92

Miquel-Romero and Sancho-Esper

According to the World Health Organization (2010), globally alcohol consumption has increased in recent decades, especially in developing countries. It is the third leading risk factor for poor health globally. The consideration of “risky” factor is even more notorious among youth: automobile crash injury and death, suicide and depression, missed classes and decreased academic performance, loss of memory, blackouts, fighting, property damage, and so forth, are some of the consequences directly related to drinking in adolescents (Bonomo et al., 2001). Adolescents are frequent users of alcohol and they increasingly consume it in a risky fashion. In the Spanish context, and according to the Spanish Ministry of Health and Consumption (2007), alcohol consumption has been trivialized and the perceived risk associated to its consumption has diminished. Two main reasons are identified for this situation: easy access to alcoholic drinks by adolescent and under eighteen young people, and a higher presence of stimulus promoting its consumption, such as advertising. A lot of studies have demonstrated the relationship between persuasive messages through mass media (alcoholic drinks ads) and young people consumption decision (e.g. Atkin et al., 1984; Wallack et al., 1990; Collins et al., 2003). Anderson et al. (2009) concluded that alcohol advertising and promotion increases the likelihood that adolescent will start to use alcohol, and to drink more if they are already drinking alcohol. However, there is not enough research about how these messages affect their decisions. Although in different countries law protects adolescents forbidding advertising targeting this group, we cannot obviate that youths are exposed to ads, and in some way, ads influence them. In fact, several studies have shown that alcohol advertising contains elements that appeal to underage youth (Austin et al., 2006; Chen et al., 2005; Waiters et al., 2001). This reality makes necessary to analyse how this advertising influences on both, adolescents and young adults, in order to work on their effect to minimize the likelihood of having potential future problems. From these evidence, the purpose of this study is to analyse how alcoholic beverages advertising can influence young people consumption, and whether that influence is the same between people over and under eighteen years old, the legal age in Spain to buy and drink alcoholic drinks, as well as to target alcohol advertising. The specific advertising measures into consideration are ad liking, ad recognition, brand recall and slogan recall.

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

2

93

Literature Review

Although the study of factors determining alcohol consumption among young people comes from different fields, it could be argued that it is the psycho-social perspective the one that has shed light towards understanding how the social context affects the young person decision. There are three main key agents, parents, peers and mass media (Moore et al., 2002), that influence the social development of an individual (Hurrelman, 1988). In this paper, we focus our analysis on the influence of alcohol advertising, as part of the mass media influence. If we take into account that adolescents spend less time with their parents in favor of media - what makes them vulnerable to any commercial message -, that they have greater vulnerability to alcohol than adults - as they have no context or reference point for assessing or regulating their drinking -, and, furthermore, that they have built up no tolerance to alcohol (Anderson et al., 2009; Dotson and Hyatt, 1994), it is imperative that law protects this segment of population.

2.1

Advertising Influence: Affective Responses and Ad Effectiveness

Consumer’s affective responses to ads have increased its interest to both academics and practitioners in advertising. Ad liking experienced by the individual is an affective response which contributes to the efficient delivery of the message, leveraging the advertiser’s media investment (Walker and Dubitsky, 1994). Likeability of an advertisement may be attributed to its creativity (Stone et al., 2000), its perception as meaningful and relevant (Biel and Bridgewater, 1990), or by using special elements such as the characters, action, storyline, music, and visual elements (Walker and Dubitsky, 1994). In the specific case of alcohol advertising, Chen et al. (2005) found that youth’s perceived likeability of beer ads was a function of the positive affective responses evoked by elements more closely related to humour and story than to music, animal characters, and people characters, cues also identified in other researches (eg. Waiters et al., 2001). There is a solid and positive relationship between ad liking and persuasion. Research evidence has offered strong support to the belief that better liked ads lead to more positive evaluations of what it is engaged in them, and even a higher purchase intention of the advertised brand. In fact, literature suggests that ad liking is one of the best indicator of advertising effectiveness (Biel, 1990; Haley and Baldinger, 1991). According to this and considering that if people

94

Miquel-Romero and Sancho-Esper

like an ad become involved with it (Brown, 1991), we propose the following hypotheses: H1: There is a positive relationship between the ad liking of alcoholic beverages and the brand recall of alcoholic beverages. H2: There is a positive relationship between the ad liking of alcoholic beverages and the advertising recognition of alcoholic beverages. H3: There is a positive relationship between the ad liking of alcoholic beverages and the slogan recall of alcoholic beverages. Alcohol advertising is one of the many factors that can potentially encourage youth drinking, as recent studies have evidenced (e.g., Anderson et al., 2009; Collins et al., 2007; Henriksen et al., 2008; Smith and Foxcroft, 2009; Snyder et al., 2006). Much of this existing literature derives from a pioneer research developed by Grube and Wallack in 1994. In that study, the authors showed that alcohol advertising does have effect on the brand and slogan knowledge, the expectancies (beliefs), the consumption intention, and consequently on the actual alcohol consumption of young people, but only if it is attended and recalled. Grube and Wallack’s ideas about the relationship between alcohol advertising and brand knowledge are supported by the increasing role that brands play on young people daily life. Brand knowledge and preferences are becoming fundamental, even for very young children (Dotson and Hyatt, 2000). In fact, brand awareness is usually one of the main goals that advertisers look for, and that awareness is reinforced by the use of a slogan, which tries to tie the name of the brand in the memory of the consumer, associating the brand, in many cases, with positive feelings. Based on these ideas, the following two hypotheses are proposed: H4: There is a positive relationship between the advertising recognition of alcoholic beverages and the brand recall of alcoholic beverages. H5: There is a positive relationship between the advertising recognition of alcoholic beverages and the slogan recall of alcoholic beverages. If alcoholic beverage advertising is analyzed, it is easy to identify in their messages that social acceptance achieved by means of consuming alcohol is related with several successful roles, values and traits (sexual success, sex appeal, romance, adventure, rebelliousness, elegance, sociability, and so forth) Those roles, values and traits (many of them explicitly forbidden by law) are admired by young adults but also by adolescents. As mentioned before, brands

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

95

would try to associate their products with these values or ideas by means of the brand names or brand slogans, in order to increase the positive expectancies and reduce the negative expectancies of alcohol consumption. So, we propose that: H6: There is a positive relationship between alcoholic beverage brand recall and the positive expectancies towards alcohol consumption. H7: There is a negative relationship between alcoholic beverage brand recall and the negative expectancies towards alcohol consumption. H8: There is a positive relationship between alcoholic beverage slogan recall and the positive expectancies towards alcohol consumption. H9: There is a negative relationship between alcoholic beverage slogan recall and the negative expectancies towards alcohol consumption. In addition, in most alcohol beverages advertising, human roles are young attractive people. Hence, it could be stated that these alcohol advertisements try to communicate young people the social value of drinking (Henrinksen et al., 2008); values that the young individual will try to adopt in order to “behave properly” in the marketplace (social legitimacy). Departing from the process explained by Grube and Wallack (1994), that has been largely contrasted in several contexts (e.g. Smith et al., 1995; Williams and Clark, 1998; Dunn and Yniguez, 1999; Fleming et al., 2004), the positive expectancies towards alcohol consumption have an influence, both direct and positive, over the final alcohol consumption intention. In spite of the fact that there is less empirical support for stating the direct relationship between the negative expectancies and the consumption intention (Austin et al., 2006; Armeli et al., 2005), it is finally considered since this relationship plays a key role at the classical low-involvement information processing model (Krugman, 1965; Ray, 1982). In those cases, cognitive changes, for instance over expectancies or beliefs, have direct effect on behavioral aspects, such as purchase or intention to do it. For these reasons, the following hypotheses are stated: H10: There is a positive relationship between the positive expectancies towards alcohol consumption and the actual alcohol consumption (over 18 years) / consumption intention (under 18 years) of the young man/woman. H11: There is a negative relationship between the negative expectancies towards alcohol consumption and the actual alcohol consumption (over 18 years) / consumption intention (under 18 years) of the young man/woman.

96

Miquel-Romero and Sancho-Esper

Evidence shows that children who are more aware of alcohol advertising have greater intention to drink (Grube, 1999). In fact, one of the main goals of alcoholic beverage companies is to create valuable brand equity, and one of the main tools for achieving this objective is by means of advertising, where they try to create positive associations between their brands (and slogans) and the final product consumption. This idea supports the last two hypotheses: H12: There is a direct and positive relationship between alcoholic beverage brand recall and the actual alcohol consumption/consumption intention of the young man/woman. H13: There is a direct and positive relationship between alcoholic beverage slogan recall and the actual alcohol consumption/consumption intention of the young man/woman. The stated framework previously described is depicted in figure 1. Moreover, a further research question is proposed: The analysis of the possible difference in terms of influence, in the specific variables considered (ad liking, advertising recognition, brand and slogan recall, positive and negative expectancies, and alcohol consumption/consumption intention) according to the age of the audience: underage versus overage individuals.

3

Methodology

To obtain data, a self-administered questionnaire was used. The two populations were defined, on one hand as underage students (12-17 years old) and, on the other hand, overage students (18-40 years old). Convenience quota sampling methods were used1, which intended to reflect the socio-economic diversity of the young people of the city under study. The questionnaires were administered, for the underage sample at two private highschools and one public high-school and, for the overage sample at the Main Public University of the Province. The total sample size was 667 questionnaires, 304 among under aged and 363 among over aged students.

1

The relevant control variables which defined the quotas were: sex and age. The selected sample did not statistically differ from the population proportions in any of them (Source: INE, Spanish National Statistics Institute).

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

97

Figure 1.- Suggested model

The procedure was as follows: in a classroom, the individuals were shown a set of images, of five alcoholic beverages ads, which were representative of each alcoholic beverage ad or brand (without showing either the brand or the logo); for each ad, the subjects had around ten seconds to recall if they had seen the ad, and if the answer was positive, to write the corresponding brand and slogan. After that, two 5-likert scales were used to measure both the positive and the negative expectancies towards alcohol consumption. Consumption intention was measured by means of a dichotomous variable (yes/no) one for each alcoholic beverage studied (beer, liquor). In order to measure ad liking, five ads of alcoholic beverages were broadcasted to the students. Afterwards, they had to rate them into a 5-item likert scale. The final consumption intention indicator was an ordinal variable which derived from the addition of two dichotomous variables. Given that it is not legal and ethically possible to ask for actual alcohol consumption to people under eighteen, for the youngest sub-sample, the alcohol consumption intention was considered, and for the oldest, the actual alcohol consumption. All the scales used (table 1) had been widely used within the literature.

98

Miquel-Romero and Sancho-Esper

Table 1: Variables and scales used in the questionnaire: underage and overage Variables: Indicators and Constructs

Authors

Scale description

Brand Recall (indicator)

Grube & Wallack, 1994

Aggregation of two items of suggested brand recall: beer and spirits (range 010)

Slogan Recall (indicator)

Grube & Wallack, 1994

Aggregation of two items of suggested slogan recall: beer and spirits (range 0-10)

Ad Recognition (scale)

Grube & Wallack, 1994

Three items related to ad recognitions: ad, product and brand. 5-likert scales.

Ad Liking (scale)

Chen, Grube, Bersamin, Waiters & Keefe, 2005

Five items to assess ad liking: music, actors, humour, and so on. 5-likert scales.

Positive Expectanc. Austin, Chen, & Grube, (scale) 2006

Ten items related to positive expectancies towards consuming alcohol. 5likert scales.

Negative Expectanc. (scale)

Austin, Chen, & Grube, 2006

Seven items related to negative expectancies towards consuming alcohol. 5likert scales.

Consumption Intention* (scale)

Flemming, Thorson & Atkin, 2004

Two-items of consumption intention: beer and spirits. 5-likert scales.

Actual Consumption** (scale)

Flemming, Thorson & Atkin, 2004

Two-items of consumption intention: beer and spirits. 5-likert scales.

Source: Own elaboration. * only at underage model, ** only at overage model.

4 4.1

Results Validation of the Measurement Instruments

For assessing reliability and validity of measurement instruments, several steps were followed. Firstly, the maximum-likelihood estimation of the Confirmatory Factor Analysis over the four factors previously defined was done. Those items with factorial loadings lower than 0.6 (Bagozzi and Baumgartner, 1994;

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

99

Bagozzi and Yi, 1988), or those suggested by the Lagrange multiplier test (Hatcher, 1994), were deleted in order to assure the convergent validity of the instrument. Almost all the factorial loads were statistically significant and greater than 0.6 (Bagozzi and Yi, 1988), which confirms the convergent validity of both instruments. Related to the reliability, Cronbach Alphas (Cronbach, 1951), the Compound Reliability Index and the Average Extracted Variance (Fornell and Larcker, 1981), were calculated for each factor, supporting that used instruments presented acceptable levels of reliability and convergent validity. The discriminant validity was also analyzed and supported2.

4.2

Hypotheses Testing

After identifying the models (underage and overaged), they were estimated by the maximum likelihood procedure using EQS 6.1. The results of the initial model estimation for the two groups are presented in table 2, and show a reasonable adjustment level in each sub-sample. In table 2 is shown that, for the under aged audience, all the hypotheses are supported except those connecting slogan recall with expectancies (positives and negatives), and with alcohol consumption intention. When considering the over aged audience, positive as well as negative expectancies are influenced only by brand recall; there is no relationship between slogan recall and positive or negative expectancies on the one hand, and actual alcohol consumption, on the other hand. The main differences compared to under aged audience arise when analyzing the effect of advertisement liking on recognition and recall. While all the relationships hold for under aged students, there is only a direct and positive relationship between ad liking and ad recognition, but neither there is relationship between ad liking and brand recall nor slogan recall for over aged students. According to these results and the analysis of the standardized loadings, we can affirm that the way advertising influences on young audience varies depending on the age of the individual. Advertisement liking (affective response) has a higher influence on recall and recognition at a younger age. Moreover, advertising recognition influences more on brand recall and less on slogan recall for under aged students than over aged; in addition, the influence of positive expectancies on alcohol consumption intention/actual consumption is higher for under 2

Contact the correspondence author for further details on validation process.

100

Miquel-Romero and Sancho-Esper

aged than over aged individuals, but the influence of the negative ones on alcohol consumption intention/actual consumption is higher for over aged individuals With these comments the research question proposed is also answered.

Table 2: Hypotheses testing Under aged samplea

Over aged sampleb

Standard loading

t-value

Standard loading

t-value

H1: Ad likingÆBrand recall

0.242**

4.11

0.041ns

0.76

H2: Ad likingÆAd recognition

0.147*

2.35

0.245**

4.45

H3: Ad likingÆSlogan recall

0.161*

2.86

0.073ns

1.40

H4: Ad recognitionÆBrand recall

0.292**

5.20

0.208*

3.92

H5: Ad recognitionÆSlogan recall

0.397**

7.07

0.349**

6.94

H6: Brand recallÆPositive expectancies

0.165*

2.65

0.137*

2.533

H7: Brand recallÆNegative expectanc.

-0.333**

-5.15

-0.248**

-4.54

H12: Brand recallÆConsump. intention

0.161*

2.69

0.231*

3.90

H8: Slogan recallÆPositive expectanc.

-0.065ns

-1.05

-0.089ns

-1.65

H9: Slogan recallÆNegative expectanc.

0.001ns

0.01

0.049ns

0.90

H13: Slogan recallÆConsump. intention

-0.050ns

-0.91

-0.037ns

-0.72

H10: Positive expect.ÆConsum. intent.

0.445**

7.04

H11: Negative expect.ÆConsum. intent.

-0.189*

-2.95

H10: Positive expect.ÆActual consump.

0.168*

2.95

H11: Negative expect.ÆActual consum.

-0.309**

-4.69

Hypotheses

2

(a) Adjustment under age model: Chi : 609.3 (288 dof, p<0.01); BBNFI: 0.839; BBNNFI: 0.895; CFI: 0.907; AGFI: 0.840; RMSEA: 0.061 (b) Adjustment over age model: Chi2: 746.1 (241 dof, p<0.01); BBNFI: 0.882; BBNNFI: 0.904; CFI: 0.916; AGFI: 0.820; RMSEA: 0.076 ** p<0.1; *p<0.05; ns non-significant

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

5

101

Discussion and Conclusions

The present research has shown the role of advertising as a key factor which determines youth alcohol consumption and consumption intention. According to our results, it is possible to affirm that this influence is different depending on the age of the audience. Although alcohol beverages advertisers are not legally able to target their products and commercials to under aged individuals, several studies have previously shown that young children are aware of alcohol advertising and this awareness increases with age (Aitken. 1989; Aitken et al.,1988). These results become more relevant if we consider one of the realities identified in the Global Strategy To Reduce The Harmful Use Of Alcohol report (WHO, 2010), which exhibits that the degree of risk for harmful use of alcohol varies with age. One of the main conclusions to be mentioned is that alcohol advertising has a clear positive influence in alcohol consumption/consumption intention, through expectancies, an those expectancies play an important role in alcohol consumption. However, it is really important to point out that the influence of advertising is more powerful for under aged audience (which is not the legal target of the product) than for over aged. As a starting point and taking into consideration the affective responses to advertisements, the role of the ad execution is very important. It could be affirmed that the influence of ad liking has, in general, much more power on underage individuals; in this sense, ad liking for younger people influences the three advertising effectiveness measures analyzed (brand recall, ad recognition and slogan recall), whereas for older people only influences in one of them (ad recognition). Although the influence of ad liking on ad recognition is higher for older ones, the direct influence on expectancies comes from brand recall, which is only influenced by ad liking in underage people. Moreover, the influence of advertising recognition on brand recall is higher for underage students, and that brand recall influences in both types of expectancies, positives and negatives, as well as on alcohol consumption intention in both sub-samples. It has to be taken into consideration that brand recall, for both types of audiences, has a higher influence on negative expectancies than on the positive ones; but when comparing the sub-samples, the influence of brand recall on both types of expectancies is higher for younger people. It is also necessary to add that the direct relationship between brand recall and consumption is higher for older audience. Considering the task played by expectancies, it is especially relevant to mention that the influence of those expectancies (determined by brand recall) in

102

Miquel-Romero and Sancho-Esper

alcohol consumption/consumption intention is different depending on the audience; for underage people, positive expectancies have a higher positive influence in alcohol consumption intention than for over age students. However, negative expectancies have a higher negative influence in alcohol consumption than that for underage individuals. Moreover, positive expectancies have a higher influence than negative expectancies for underage people, and the opposite happens to over age individual: negative expectancies have a higher influence in alcohol consumption than positive expectancies. This can be due to the fact that under age audience likes to emulate the roles and lifestyles presented at alcoholic beverages advertising (Austin and Knaus, 2000), what makes them to assign higher positive expectancies to alcohol drinking; but they have less experience with alcohol what decreases their negative expectancies. On the other hand, older people are used to have more experience with alcohol, what makes them judge what advertisements show (imageries, roles, and so forth); for this reason, their positive expectancies influences alcohol consumption in a less extent than the negative ones. Although brand recall becomes an essential mediating variable between advertising recognition and actual consumption/consumption intention, given that it has a direct influence on them, and also an indirect influence through expectancies, this cannot be said with respect to slogan recall; although there is a clear influence from ad recognition to slogan recall for both audiences considered, no other hypotheses related to slogan recall stand statistically significant for none of both groups, what lead us to say that slogan plays a poor role in alcohol consumption. On the light of this evidence, there are many ideas and proposals to be made. Firstly, it is shown that advertising restrictions undertaken by public regulators on media are useful for reducing alcohol consumption intention among young people. In fact, in many countries there should be much more restrictive regulation, as it is extended the time and ways that underage individuals are exposed to mass media. The significant relationship between advertising and brand recall increases the likelihood of youth alcohol consumption, both enhancing motivations which increase positive expectancies of younger people, and reducing the constraints that generate the negative expectancies for both, younger and older individuals. Therefore, unless it is obvious that advertising is not the only socialization agent, a responsible self-control among professional and creative executives should be implemented in order to restrict the different roles shown at advertisements, and also a wider restriction on the media and contexts where alcohol ads can be displayed. Secondly, for advertisers, creative executives and product managers whose target group is individuals over eighteen years old, the marginal role of slogan recall on the attitude modification process does not have

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

103

a direct effect on alcohol consumption but is a useful tool for positioning a brand at the mind of the customer. Since alcohol advertising processing, in this product category, seems to be peripheral (Petty and Cacioppo, 1981; 1986), slogans should emphasize tactical aspects and associate the brand with positive aspects of execution, such as roles, locations, colors, special effects, and so on. Nevertheless, and considering that for underage people the level of ad liking influences in all advertising effectiveness measures analyzed in this paper, creative executives should separate from cues specially relevant for adolescents and towards which they respond positively, such as animals, humor or musical elements as Waiters et al (2001) have proved. To conclude, it is worthy to remark the fact that although law throughout different countries gives special attention to protecting young consumers from alcoholic beverages advertising, parents should explain, as much as possible, what are the goals of advertising and how they have to be interpreted. Moreover, information programs should be implemented in both primary and secondary schools, in order to reduce positive expectancies and increment the negatives ones related to alcohol consumption. The results obtained in this study, as well as the proposed actions, are in line of the suggestions made by the World Health Organization (WHO, 2010) about the policy options and interventions wished for reducing the harmful use of alcohol, in the marketing context (Area 6 of the policy options and interventions, pp.15). The WHO considers the necessity to reduce the impact of marketing, particularly on young people and adolescents, and recognizes the difficulty to target young adult consumers without exposing cohorts of adolescents under the legal age to the same marketing, so the content of alcohol marketing as well as the amount of exposure of young people to that marketing are crucial issues. The World Health Organization (2010, pp.15) proposes setting up regulatory or co-regulatory frameworks, preferably with a legislative basis, and supported when appropriate by self-regulatory measures for alcohol marketing by: (i) regulating the content and the volume of marketing, as well as (ii) regulating direct or indirect marketing in certain or all media, between other measures.

6

References

Aitken, P.P. (1989), “Television alcohol commercials and under-age drinking”, in: International Journal of Advertising, Vol. 8, 133-150. Aitken, P.P., Leathar, D.S. & Scott, A.C. (1988), “Ten- to sixteen-year-olds’ perceptions of advertisements for alcoholic drinks”, in: Alcohol and Alcoholism, Vol. 23, 491-500. Anderson, P., Bruijn, A., Angus, K., Gordon, R. & Hasting. G. (2009), “Impact of Alcohol Advertising and Media Exposure on Adolescent Alcohol Use: A Systematic Review of Longitudinal Studies”, in: Alcohol & Alcoholism, Jan., 1-15.

104

Miquel-Romero and Sancho-Esper

Armeli, S., Mohr, C., Todd, M. & Maltby, N. (2005), “Daily evaluation of anticipated outcomes from alcohol use among college students”, in: Journal of Social and Clinical Psychology, Vol. 24 (6), 767-792. Atkin. C.K., Hocking, J. & Bolck, M. (1984), “Teenage drinking: does advertising make a difference?”, in: Journal of Communication, Vol. 34 (2), 157-167. Austin, E.W., Chen, M.J. & Grube, J.W. (2006), “How does alcohol advertising influence underage drinking? The role of desirability, identification and skepticism”, in: Journal of Adolescent Help, Vol. 38 (4), 376-384. Austin, E.W. & Kanaus, C. (2000), “Predicting the potential for risky behavior among those “too young” to drink as the result of appealing advertising”, in: Journal of Health Communication, Vol. 5 (1), 13-27. Bagozzi, R.P. & H. Baumgartner (1994),”The evaluation of structural equation models and hypothesis testing”, at Bagozzi. R. (eds). Principles of Marketing Research. Blackwell Publishers. Cambridge. MA, 386-422. Bagozzi, R.P. & Y. Yi (1988), “On the evaluation of structural equation models”, in: Journal of the Academy of Marketing Science, Vol. 16 (1), 74-94. Biel, A.L. (1990), “Love the ad. Buy the product? Why liking the advertising and preferring the brand aren’t such strange bedfellows after all”, in: Admap, Vol. 26 (9), N. 299, September, 21-25. Biel, A.L. & C.A. Bridgewater (1990), “Attributes of likeable television commercials”, in: Journal of Advertising Research, 30, 38-44. Bonomo, Y., Coffey, C, Wolfe R., et al. (2001), “Adverse outcomes of alcohol use in adolescents”, in: Addiction, 96, 1485–96. Brown, G. (1991), “How advertising affects the sales of packaged goods brands: A working Hypothesis for the 1990s”, Warwick, Millward Brown International plc. Cited by McDonald (1992) op. cit. Chen, M., Grube, J.W., Bersamin, M., Waiters & E. Keefe (2005), “Alcohol advertising: what makes it attractive to youth?”, in: Journal of Health Communication, Vol. 10 (6), 553-565. Collins, R.L., Ellickson, P.L., McCaffrey, D. & K. Hambarsoomians (2007), “Early adolescent exposure to alcohol advertising and its relationship to underage drinking”, in: Journal of Adolescent Health, Vol. 40 (6), 527-534. Collins, R.L., Schell, T., Ellickson, P. & D. McCaffrey (2003), “Predictors of beer advertising awareness among eighth graders”, in: Addiction, 93, 1297-1306. Cronbach, L.J. (1951), “Coefficient alpha and the internal structure of tests”, in: Psychometrika, Vol. 16, (3), 297-334. Currie, C., Roberts, C., Morgan, A., et al. (2004), “Health Behaviour in School-Aged Children (HBSC) Study”, in: International Report from the 2001/2002 Survey. Copenhagen: WHO. Dotson, M.J. & E.M. Hyatt (1994), “The Impact of changes in the household on the consumer socialization process”, in: Proceedings of the Southern Marketing Association, New Orleans. LA. November, 156-60. Dotson, M.J. & E.M. Hyatt (2000), “A comparison of parents' and children's knowledge of brands and advertising slogans in the United States: implications for consumer socialization”, in: Journal of Marketing Communications, Vol. 6 (4), 219-30. Dunn, M.E. & R.M. Yniguez (1999), “Experimental demonstration of the influence of alcohol advertising on the activation alcohol expectancies in memory among fourth- and fifth-grade children”, in: Experimental and Clinical Pharmacology, Vol. 7 (4), 473-483. Fleming, K., Thorson, E. & C.K. Atkin (2004), “Alcohol adverting exposure and perceptions: Link with alcohol expectancies and intentions to drink or drinking in underage youth and young adults”, in: Journal of Health Communication, Vol. 9 (1), 3-29. Fornell, C. & D.F. Larcker (1981), “Evaluating structural equations models with unobservable variables and measurement error”, in: Journal of Marketing Research, 18 (February), 39-50. Grube, J.W. (1999), “Alcohol Advertising and Alcohol Consumption: A review and Recent Research”, in: National Institute on Alcohol Abuse and Alcoholism 10th Special Report to Congress on Alcohol and Health. Bethesda (MD): National Institute on Alcohol Abuse and Alcoholism.

The Influence of Alcohol Advertising in Underage versus Overage Alcohol Consumption

105

Grube, J.W. & L.Wallack (1994), “Television beer advertising and drinking knowledge, beliefs and intentions among schoolchildren”, in: American Journal of Public Health, Vol. 84 (2), 254-259. Haley, R.L. & A.L. Baldinger (1991), “The ARF Copy Research Validity Project”, Journal of Advertising Research, 31 (2), 11-32. Hatcher, L. (1994), “A Step by Step Approach to Using the SAS System for Factor Analysis and Structural Equation Modeling”, in: Cary. NC: SAS Institute Inc. Henriksen, L., Feighery, E.C., Scheleicher, N.C. & S.P. Fortmann (2008), ”Receptivity to Alcohol Marketing Predicts Initiation of Alcohol Use”, in: Journal of Adolescent Health, 42, 28-35. Hurrelman, K. (1988), “Social Structure and Personality Development: the individual as productive processor”, New York: Cambridge University Press. Krugman, H.E. (1965), “The impact of television advertising: learning without involvement”, in: The Public Opinion Quarterly, Vol. 29 (3), 349-356. McDonald, C. (1992), “How Advertising Works. A review of current thinking”, NTC PublicationsLimited. The Advertising Association, Biddles Ltd, Guilford and Kings Lynn, Great Britain. Moore, J.N., Raymond, M.A., Mittelstaedt, J.D. & J.F. Jr. Tanner (2002), “Age and consumer socialization agent influences on adolescents' sexual knowledge, attitudes and behavior: Implications for social marketing initiatives and public policy”, in: Journal of Public Policy & Marketing, Vol. 21 (1),37-52. Petty, R. E. & J. T. Cacioppo (1981), Attitudes and Persuasion: Classic and Contemporary Approaches. Dubuque, IA: Wm. C. Brown. Petty, R. E. & J. T. Cacioppo (1986), “Communication and Persuasion: Central and Peripheral Routes to Attitude Change”, New York: Springer-Verlag. Ray, M.L. (1982), “Advertising and communication management”, Englewood Cliffs. NJ: Prentice Hall. Smith, G.T., Goldman, M.S., Greenbaum, P.E. & B.A. Christiansen (1995), “Expectancy for facilitation from drinking: The divergent paths of high-expectancy and low-expectancy adolescents”, in: Journal of Abnormal Psychology, Vol. 104 (1), 32-40. Smith, L.A. & D.R. Foxcroft (2009), “The Effect of Alcohol Advertising, Marketing and Portrayal on Drinking Behaviour in Young People: Systematic review of Prospective Cohort Studies”, in: BMC Public Health, Vol. 9 (51). Available from: http://www.biomedcentral.com/14712458/9/51. Snyder, L.B., Milici, F.F., Slater, M., Sun, H. & Y. Strizhakova (2006), “Effects of Alcohol Advertising Exposure on Drinking Among Youth”, in: Archives of Pediatrics & Adolescent Medicine, 160 (January), 18-24. Spanish Ministery of Health and Consumption (2007), “Ministerio de Sanidad y Consumo. Anteproyecto de Ley de medidas sanitarias para la protección de la salud y la prevención del consumo de bebidas alcohólicas por menores”. Available at http://www.msc.es/novedades/docs/Anteproyecto_de_ley_de_medidas_sanitarias.pdf Stone, G., Besser, D. & L.E. Lewis (2000 ), “Recall, liking and creativity in TV commercials: A new approach”, in: Journal of Advertising Research. Vol. 40, 7-18. Strasburger, V.C. (2004), “Children, Adolescent, and the Media”, in: Current Problems in Pediatric and Adolescent Health Care, 35, 54-113. Unger, J.B., Schuster, D., Zogg, J., Dent, C.W. & A.W.Stacy (2003), “Alcohol advertising exposure and adolescent alcohol use: A comparison of exposure measures”, in: Addiction Research and Theory, Vol. 11 (3), 177-193. Waiters, E., Treno, A. & J. W. Grube (2001), “Alcohol advertising and youth: A focus-group analysis of what young people find appealing in alcohol advertising”, in: Contemporary Drug Problems, Vol. 28 (4), 695. Walker, D. & T.M Dubitsky (1994), “Why liking matters”, in: Journal of Advertising Research, 34, 9-18. Wallack, L.; Cassady, D. & J.W. Grube (1990), “Tv beer commercials and children: Exposure, attention, beliefs, and expectations about drinking as an adult”, Falls Church, VA: AAA Foundation for Traffic Safety. Mentioned by Chen, Grube, Bersamin, Waiters and Keefe (2005) op.cit..

106

Miquel-Romero and Sancho-Esper

Williams, A. & D. Clark (1998), “Alcohol consumption in university students: The role of reasons for drinking, copying strategies, Expectancies and personality traits”, in: Addictive Behavior, Vol. 23 (3), 371-378. WHO (2010), World Health Organization. “Global Strategy to reduce the harmful use of alcohol”. WHO Library Cataloguing-in-Publication Data. Available at : http://www.who.int/substance_abuse/publications/en/ in 20th Decemb. 2010.

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages for a Fair Trade Campaign Erlinde Cornelis, Ghent University, Belgium Leen Adams, Hogeschool-Universiteit Brussel, Belgium Verolien Cauberghe, Ghent University, Belgium

1

Introduction

According to research on message framing, messages that are factually equivalent, but framed differently, often lead to differences in persuasion (Yi and Baumgartner, 2009). Most previous research on message framing has compared message frames that differ in overall valence, i.e., positive frames versus negative frames (Rothman and Salovey, 1997; Levin et al., 1998). Message framing in terms of a prevention versus promotion outcome focus lately received an increased academic interest. This way of framing is based on the self-regulatory focus theory of Higgins (1997), stating that people can be classified as either relatively more promotion-focused or relatively more prevention-focused. Furthermore, this theory proposes that message (frame) effectiveness depends on consumers’ underlying motivational orientations (e.g., Aaker and Lee, 2001). That is, when the message frame is congruent to the viewer’s self-regulatory focus, the message is assumed to be more effective (Higgins, 2002). Although ample evidence supporting this regulatory congruency principle has been found (e.g., Latimer, et al., 2007; Yi and Baumgartner, 2008), the boundaries of this principle are still insufficiently studied. So far, studies have mainly focused on differently framed rational messages. To our knowledge, no studies were conducted yet to scrutinize the effectiveness of differently framed emotional messages despite the importance of emotions in consumer decision-making and their use in realistic ads (Pham, 1998). Therefore, the current study wants to complement prior academic research by examining the moderating role of message type (i.e., rational vs. emotional) on the effectiveness of regulatory congruent versus incongruent messages.

2

Hypotheses

Higgins’ self-regulatory focus theory (1997) describes a promotion focus as a tendency to focus on the presence or absence of positive outcomes and a prevenS. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_7, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

108

Cornelis, Adams and Cauberghe

tion focus as a tendency to focus on the presence or absence of negative outcomes. The self-regulatory focus can be either measured as a personality trait or can be temporarily induced (Higgins, 1997). An important finding in the framework of the self-regulatory focus theory is the congruency principle (Higgins, 2000). That is, people experience regulatory congruency when exposed to a message that matches with their inner selfregulatory focus (Higgins, 2000). Regulatory congruency has been found to boost message effectiveness (e.g., Zhao and Pechmann, 2007; Aaker and Lee, 2001). However, it is questionable whether a congruency effect always holds. Previous studies did found mixed results concerning the congruency effect: in certain contexts, an incongruency effect or no effect at all was found. For example, schema congruity theory argues that sometimes, incongruency between the message and its receiver can be more effective than congruency (Mandler, 1982). Also, building on the self-regulatory focus, Wang and Lee (2006) found a regulatory congruency effect in a low involvement context, but not in a high involvement context. Moreover, previous studies have mainly addressed the regulatory congruency principle in a merely rational context (e.g., Yi and Baumgartner, 2009), while, to our knowledge, an emotional context has been neglected so far. Therefore, the current study will assess the moderating role of the message type (i.e., rational versus emotional messages) on the effectiveness of regulatory congruent versus incongruent messages. According to the ‘affect-as-information’ heuristic, when a message is emotional, respondents can use emotions as a valuable input for their subsequent evaluations (Pham, 1998; Schwarz, 1990). However, not all people are naturally inclined to rely on their emotions. It has been shown that reliance on affect (versus cognition) depends on one’s self-regulatory focus (Pham and Avnet, 2004). More precisely, prevention-focused individuals are less likely to rely on affect than promotion-focused individuals. They tend to be more vigilant and externally rather than internally focused (Pham and Avnet, 2009). As a result, it is expected that affect is less salient to prevention-focused than to promotion-focused individuals (Kramer and Yoon, 2007). However, the salience of emotional arguments could also be determined by the characteristics of the message itself. More precisely, affect could be more salient by making the message frame incongruent to the self-regulatory focus of the receiver (Kramer and Yoon, 2007). This is in line with schema congruity theory, postulating that people make judgments by evaluating new stimuli against existing expectations (Mandler, 1982). While congruency results in little arousal, incongruency results in higher

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages

109

levels of arousal (Mandler, 1975, 1982) and, as a result, attracts more attention and is more salient than congruent information (Mandler, 1982). Based on the above, we expect that salience, created by regulatory incongruency, will be especially important for prevention-focused people when processing an emotional message, as they are not used to rely on affect (Pham and Avnet, 2004, 2009; Kramer and Yoon, 2007). Therefore, we hypothesize the following: H1: When the message is emotional, prevention-focused individuals will indicate a more positive Aad, Ab and PI when the message is incongruent with their self-regulatory focus than when the message is incongruent with their self-regulatory focus. This effect will be less pronounced for promotionfocused individuals. When a message is rational (instead of emotional), comprehensibility, rather than salience, will be of crucial importance for message effectiveness (Eagly, 1974). Prior research has found that messages being congruent (versus incongruent) with one’s self-regulatory focus could enhance comprehensibility, processing fluency and subsequently result in more favourable judgments (Lee and Aaker, 2004). Following this reasoning, we can expect that, when exposed to a rational message, people will prefer a message to be congruent (as opposed to incongruent) to their self-regulatory focus. In line with this, Avnet and Higgins (2006) suggested that the regulatory congruency effect might be especially important in a rational condition. Again, we expect a difference between promotion- and prevention-focused individuals, as Pham and Avnet (2004, 2009) found that prevention-focused individuals are naturally more inclined to rely on message substance than on affect, whereas the reverse is true for promotion-focused individuals. So, we could assume that promotion-focused individuals rely less on message substance than prevention-focused individuals. Therefore, when exposed to a rational message, there will be stronger congruency effects for promotion-focused people (as opposed to prevention-focused people), as promotion-focused people are not used to rely on message substance (Pham and Avnet, 2004, 2009; Lee and Aaker, 2004). Hence, we hypothesize the following: H2: When the message is rational, promotion-focused individuals will indicate a more positive Aad, Ab, and PI when the message is congruent with their regulatory focus than when the message is incongruent with their regulatory focus. This effect will be less pronounced for prevention-focused individuals.

110 3 3.1

Cornelis, Adams and Cauberghe

Method Design and Stimuli

A 2 (chronic self-regulatory focus: prevention vs. promotion) x 2 (message’s regulatory focus: prevention vs. promotion) x 2 (message type: rational vs. emotional) between-subject factorial experimental design was used. Message frame and message type were manipulated into four different messages. In addition, each respondent’s chronic self-regulatory focus was measured using a selfreported scale. The stimuli were pretested among 53 respondents. The results show that the emotional message (M = 5.21) scored significantly higher on the emotionality (versus rationality) scale than the rational message (M = 2.84), (t (51) = 5.83, p < .001). Similarly, respondents rated the promotion focused message (M = 5.96) significantly higher on promotion focus (versus prevention focus) scale than the prevention focused message (M = 2.75), (t (51) = 7.26, p < .001). For this study, ads for fair trade bananas were chosen, because fair trade is a suitable topic for developing rational as well as emotional campaigns. In reality, both rational and emotional fair trade campaigns have been launched with success, indicating that fair trade is a topic fitting both in a rational as well as an emotional persuasion context.

3.2

Participants and Procedures

Data were collected from 284 participants by an online questionnaire. The sample consisted of 54.6 % women, with an average age of 33.12, ranging between 16 and 74 years. The participants were randomly assigned to the four experimental conditions.

3.3

Measures

Chronic self-regulatory focus was measured using the 18 items seven-point Likert scale of Lockwood et al. (2002), containing a promotion dimension (e.g., “I typically focus on the success I hope to achieve in the future.”) (α = .73) and a prevention dimension focus (e.g., “I often imagine myself experiencing bad things that I fear might happen to me.”) (α = .64). Message’s regulatory focus was measured by means of one single item on a seven-point semantic differential scale (i.e., “This advertisement mainly focuses on the avoidance of negative results for the farmers in the South.” versus “This advertisement mainly focuses

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages

111

on the achievement of positive results for the farmers in the South.”). Message type (rational versus emotional) was measured by a four-item seven-point semantic differential scale by Liu and Stout (1987) (e.g., “I think this advertisement is logical versus emotional.”) (α = .801). Aad was measured by the semantic scale of Mitchell and Olson (1981) (e.g., “I like versus dislike this message”) (α = .92). Ab was assessed by seven bipolar items of Bower and Landreth (2001) (e.g., “I consider the brand to be good versus bad”) (α = .94). The scale of Petrevu and Lord (1994) was used to asses purchase intention (e.g., “It is very likely that I will purchase Oxfam Fairtrade”) (α = .89).

4 4.1

Results Manipulation Check

The manipulation check revealed that the emotional message (M = 4.37) was considered significantly more emotional (versus rational) than the rational message (M = 2.91), (t (282) = 12.51, p < .001). In addition, the respondents rated the promotion focused message (M = 5.58) significantly higher on promotion focus (versus prevention focus) scale than the prevention focused message (M = 4.27), (t (282) = 7.49, p < .001).

4.2

Test of the Hypotheses

To assess the hypotheses, three one-way ANOVA’s were performed. A significant three-way interaction effect appeared among respondents’ regulatory focus, the message’s regulatory focus, and message type on Aad (F (1, 262) = 6.29, p < .05), on Ab (F (1, 262) = 3.92, p < .05), and on PI (F (1, 262) = 9.68, p < .005).

4.2.1

Emotional Messages

Prevention-focused respondents reported a significantly higher Aad towards the incongruent (i.e., promotion-focused) emotional message (M = 5.47) than towards the congruent (i.e., prevention-focused) emotional message (M = 4.53), (t (69) = 4.29, p < .001), while for promotion-focused respondents, no distinct preference for the incongruent (M = 5.32) over the congruent emotional message (M = 5.38) was found (t (62) = 0.27, p = 0.78) (see figure 1). Preventionfocused people also indicated a significantly higher Ab after exposure to the incongruent (i.e., promotion-focused) emotional message (M = 5.74) than to the congruent (i.e., prevention-focused) emotional message (M = 4.89), (t (69) =

112

Cornelis, Adams and Cauberghe

3.84, p < .001), while for promotion-focused respondents, no different response was found after exposure to the incongruent emotional message (M = 5.49) versus after exposure to the congruent emotional message (5.75), (t (62) = 1.03, p = 0.31) (see figure 2). Finally, prevention-focused people also indicated a significantly higher PI after being exposed to the incongruent (i.e., promotionfocused) emotional message (M = 4.69) than after being exposed to the congruent (i.e., prevention-focused) emotional message (M = 3.56), (t (69) = 3.96, p < .001). While again, for promotion-focused respondents, no difference in reaction was found after seeing the incongruent emotional message (M = 4.75) versus after seeing the congruent emotional message (M = 4.41), (t(62) = -0.94, p = 0.35) (see figure 3). Hence, hypothesis 1 was fully supported.

Figure 1: Attitude toward the ad (Aad) for an emotional message

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages

Figure 2: Attitude toward the brand (Ab) for an emotional message

Figure 3: Purchase intention (PI) for an emotional message

113

114 4.2.2

Cornelis, Adams and Cauberghe

Rational Messages

Promotion-focused respondents reported a significantly higher Aad towards the congruent (i.e., promotion-focused) rational message (M = 5.10) than towards the incongruent (i.e., prevention-focused) rational message (M = 4.51), (t (66) = 2.10, p < .05). While for prevention-focused respondents, no preference between the congruent rational message (M = 4.40) or the incongruent rational message (M = 4.59) was found (t (58) = 0.69, p = 0.49) (see figure 4). Promotion-focused people also indicated a significantly higher Ab for the congruent (i.e., promotion-focused) rational message (M = 5.55) than for the incongruent (i.e., prevention-focused) rational message (M = 4.96), (t (66) = 2.18, p < .05), while for prevention-focused respondents, no preference between the congruent rational message (M = 4.85) or the incongruent rational message (M = 5.00) was found (t(58) = 0.59, , p = 0.56) (see figure 5). Furthermore, promotion-focused people indicated a significantly higher PI towards the congruent (i.e., promotionfocused) rational message (M = 4.50) than towards the incongruent (i.e., prevention-focused) rational message (M = 3.75), (t (66) = 2.58, p < .05), while for prevention-focused respondents, no preference between the congruent rational message (M = 4.23) or the incongruent rational message (M = 4.46), (t(58) = 0.69, p = 0.49) was found (see figure 6). Hence, hypothesis 2 was fully supported.

Figure 4: Attitude toward the ad (Ad) for a rational message

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages

Figure 5: Attitude toward the brand (Ab) for a rational message

Figure 6: Purchase intention (PI) for a rational message

115

116 5

Cornelis, Adams and Cauberghe

Discussions and Conclusions

The regulatory congruency principle (Higgins, 2000) emphasizes the importance of regulatory congruency in order to obtain more effective messages. However, this study found that the effectiveness of regulatory congruency depends on the message type (i.e., rational or emotional messages). For emotional messages, the results show that, an incongruency effect appeared for prevention focused individuals, who do not generally monitor their internal states. For them, in order for emotions to be relied on by and to be effective for prevention-focused individuals, these emotions should be made salient by making the message’s regulatory frame incongruent to the viewers’ chronic self-regulatory focus (Kramer and Yoon, 2007). In line with schema congruity theory (Mandler, 1982), this result appears to be driven by the salience of incongruent versus congruent information. Furthermore, our finding that promotion-focused people have no significant preference in terms of Aad, Ab, or PI for congruent or incongruent emotional messages, supports the idea that promotion-focused individuals tend to rely on emotions in general, regardless of the message’s regulatory frame (Pham and Avnet, 2004; Kramer and Yoon, 2007). Looking at rational messages, we rather found the opposite. Promotionfocused people reacted more favourably to a rational message that is congruent (versus incongruent) with their self-regulatory focus, whereas congruency or incongruency was of no importance for prevention-focused people. We believe that, as promotion-focused people are less used to rely on the substance of a message, it is more difficult for them to process rational messages (Pham and Avnet, 2004). Matching the rational message’s regulatory focus with viewers’ self-regulatory focus enhances comprehensibility, processing fluency and could, therefore, result in more positive evaluations. Hence, the finding that promotionfocused individuals indicate a more positive Aad, Ab, and PI when a rational message is congruent (as opposed to incongruent) to their self-regulatory focus, supports earlier found congruency effects (Latimer, et al., 2007; Zhao and Pechmann, 2007). Our finding that prevention-focused people have no distinct preference in terms of Aad, Ab, or PI for congruent or incongruent rational messages, supports the idea that prevention-focused people have less difficulties with processing and relying on substance, and are, therefore, also less influenced by the message’s regulatory frame (Pham and Avnet, 2004; Kramer and Yoon, 2007).

The Effectiveness of Emotional and Rational Regulatory (In)congruent Messages

6

117

Limitations and Further Research

In this study, only one type of emotions, namely positively valenced emotions, are examined. This limitation provides possibilities for further research to assess the effect of different types of emotions, including negatively framed emotions. Also, the use of mixed-emotions within prevention- or promotion framed messages could be an interesting direction for further research. Secondly, only one message topic is examined in the current study, namely “fair trade”. Further research can look at other topics as well. For instance, an advertisement for products that are considered more utilitarian versus hedonic could evoke different preferences in terms of regulatory orientation of the ad (Pham and Avnet, 2006). Next, based on Feldman and Lynch’s (1988) accessibility-diagnosticity model, further research should also measure the ‘accessibility’ and the ‘perceived diagnosticity’, to examine the underlying mechanism of the congruency effect. In addition, it would also be highly relevant for further research to unravel the underlying mechanism supporting the incongruency effect. Lastly, the sample in this study was based on the snowball method. Further studies should enlarge the sample size and use different sampling methods.

7

References

Aaker, J.L. & Lee, A.Y (2001). “I” Seek Pleasures and “we” Avoid Pains: The Role of SelfRegulatory Goals in Information Processing and Persuasion. Journal of Consumer Research, 28, 33-49. Avnet, T. & Higgins, E.T. (2006). How Regulatory Fit Affects Value in Consumer Choices and Opinions. Journal of Marketing Research, 43, 1-10. Bower, A. B. & Landreth, S. (2001). Is Beauty Best? Highly versus Normally Attractive Models in Advertising. Journal of Advertising, 30, 1-12. Cesario, J. & Higgins, E.T. (2008). Making Message Recipients “Feel Right”. How Nonverbal Cues Can Increase Persuasion. Psychological Science, 19 (5), 415-420. Chandy, R.K., Gerard, J.T., Macinnis, D.J. &. Thaivanich, R. (2001). What to Say When: Advertising Appeals in Evolving Markets. Journal of Marketing Research, 38, 399-414. Eagly, A.H. (1974). Comprehensibility of Persuasive Arguments as a Determinant of Opinion Change. Journal of Personality and Social Psychology, 29 (6), 758-773. Feldman, J.M. & Lynch, J.G.Jr. (1988). Self-Generated Validity and Other Effects of Measurement on Belief, Attitude, Intention, and Behavior. Journal of Applied Psychology, 73, 421-735. Higgins, E.T. (1997). Beyond Pleasure and Pain. American Psychologist, 52 (12), 1280-1300. Higgins, E.T. (2000). Making a Good Decision: Value From Fit. American Psychologist, 55, 121-1230. Kramer, T. &. Yoon, S. (2007). Approach-Avoidance Motivation and the Use of Affect as Information. Journal of Consumer Psychology, 17 (2), 128-138. Latimer, A.E. Rivers, S.E., Rench, T.A., Katulak, N.A., Hicks, A., Hodorowski, J.K.; Higgins, E.T. & Salovey, P. (2007). A Field Experiment Testing the Utility of Regulatory Fit Messages for Promoting Physical Activity. Journal of Experimental Social Psychology, 44, 826-832.

118

Cornelis, Adams and Cauberghe

Lee, A. & Aaker, J. (2004). Bringing the Frame into Focus: The Influence of Regulatory Fit on Processing Fluency and Persuasion. Journal of Personality & Social Psychology, 86, 205-218. Liu, S. S. & Stout, P.A. (1987). Effects of Message Modality and Appeal on Advertising Acceptance. Psychology & Marketing, 4 (3), 167-187. Lockwood, P.; Jordan, C. H. & Kunda, Z. (2002). Motivation by Positive or Negative Role Models: Regulatory Focus Determines Who Will Best Inspire Us. Journal of Personality & Social Psychology, 83, 854-864. Mandler, G. (1975). Mind & Emotions, New York: Wiley. Mandler, G. (1982). The Structure of Value: Accounting For Taste, in Margaret, H.; Clarke, S. and S. T. Fiske (eds). Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, 336. Hillsdale, NJ: Lawrence Erlbaum. Petrevu, S. & Lord, K.R. (1994). Comparative and Noncomparative Advertising: Attitudinal Effects Under Cognitive and Affective Involvement Conditions. Journal of Advertising, 23 (2), 77-90. Pham, M. T. & Avnet, T. (2004). Ideals and Oughts and the Reliance on Affect versus Substance in Persuasion. Journal of Consumer Research, 30, 503-517. Pham, M. T. & Avnet, T. (2009). Contingent Reliance on the Affect Heuristic as a Function of Regulatory Focus. Organizational Behavior and Human Decision Processes, 108 (2), 267-278. Pham, M. T. & Avnet, T. (2009). Rethinking Regulatory Engagement Theory. Journal of Consumer Psychology, 29 (2), 115-123. Schwarz, N. (1990). Feelings as Information: Informational and Motivational Functions of Affective States. In R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of Motivation and Cognition (521- 561). New York: Guilford. Wang, J. & Lee, Y.A. (2006). The Role of Regulatory Focus in Preference Construction. Journal of Marketing Research, 43 (1), 28-38. Yi, S. & Baumgartner, H. (2008). Motivational Compatibility and the Role of Anticipated Feelings in Positively Valenced Persuasive Message Framing. Psychology & Marketing, 25 (11), 10071026. Yi, S. & Baumgartner, H. (2009). Regulatory Focus and Message Framing: A Test of Three Accounts. Motivation & Emotion, 33 (4), 435-443. Zhao, G. & Pechmann, C. (2007). The Impact of Regulatory Focus on Adolescents’ Response to Anti Smoking Advertising Campaigns. Journal of Marketing Research, 44, 671-687.

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention in HIV/AIDS Fear Appeal Advertising Marlize Terblanche-Smit, University of Stellenbosch, South Africa Nic S. Terblanche, University of Stellenbosch, South Africa

1

Introduction

Social advertising campaigns address anti-social behavior and pandemics, like HIV/AIDS, with the intention to bring about social change. The effectiveness of advertising is strongly influenced by its ability to target a specific, preferably homogeneous, target market. Advertising in South Africa is often targeted at a heterogeneous audience with a standardized message. The HIV/AIDS pandemic in South Africa is a major concern, especially amongst adolescents, and the main advertising campaign LoveLife does not seem to be producing the desired results. The use of fear as an advertising appeal raises the question on the appropriate severity of the threat. Furthermore theory points out that external influences should be included when evaluating beliefs. These influences include cultural background, perceived vulnerability to HIV/AIDS infection and personality traits, which may have a mediating influence on attitudes, norms and selfefficacy beliefs (Fishbein, 2000). This study investigated whether the use of fear in HIV/AIDS communication increases the likelihood of adopting appropriate behavior amongst adolescents. The influence of personality traits on efficacy beliefs, attitude and behavioral intention were examined.

2

Contextualisation and Literature Overview

Fear is one of the most important secondary drivers of human action (Hergenhahn and Olson, 2003). Fear appeal literature indicate that fear can be described by mood adjectives, including feeling frightened or anxious, and also via ratings of concern or worry (Henthorne et al., 1993; LaTour and Tanner, 2003; LaTour and Rotfeld, 1997; Rogers, 1983). Fear is caused by a threat that is perceived to S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_8, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

120

Terblanche-Smit and Terblanche

be substantial and personally relevant to individuals (Easterling and Leventhal, 1989; Ortony and Turner, 1990). The question often raised in fear appeal advertising is the appropriate severity of the threat. (Rotfeld, 2000). Different models to improve the effectiveness of fear appeal have been proposed. Tay, Ozanne and Santiono (2000) recommend the utilization of fear appeals should be segment specific as fear appeals have been found to influence various population segments differently (Burnett and Wilkes, 1980; Burnett and Oliver, 1979; Quinnet al., 1992). According to the EPPM (Extended Parallel Process Model), based on previous fear appeal theoretical approaches (Janis, 1967; Leventhal, 1971; Rogers, 1983), threat motivates action, and perceived efficacy determines whether the action taken controls the danger (protective behavior) or controls the fear (inhibits protective behavior), as indicated in Figure 1. Perceived response efficacy refers to an individual's beliefs that a response effectively prevents the threat (i.e. “I believe condoms prevent HIV contraction”), and perceived self-efficacy refers to an individual's belief in his or her ability to perform a recommended response (i.e. “I think that I can use condoms to prevent HIV contraction”) (Rogers, 1983). If the results of a survey indicated high threat and high efficacy, then it could be stated that the intervention was producing the desired actions (Witte and Allen, 2000). Individual differences however influence the assessment of threat and efficacy. Individuals evaluate the components of a message relative to their prior experiences, culture, and personality characteristics (Witte, 1992). Fishbein (2000) pointed out that self-efficacy is one variable that can be seen as a function of underlying determinants that include beliefs about the outcome of behaviour, social and normative prescriptions within a population, and specific barriers to such actions. Therefore, external influences should be included when evaluating the following beliefs: cultural background, perceived vulnerability to HIV infection and personality traits that may have a mediating influence on attitudes, norms and self-efficacy beliefs.

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention

121

Action: Control danger Control fear

Threat

Perceived efficacy

Figure1: The Extended Parallel Process Model (EPPM) (adapted from Witte and Allen, 2000)

HIV/AIDS research shows that about 50 percent of HIV infections in South Africa are transmitted to people before the age of 20, with more than 5 million HIV positive people in a country with a population of 48 million (Avert, 2008). LoveLife is South Africa’s foremost multi-million dollar HIV/AIDS prevention campaign (US$12 million/annum) that does not seem to be producing the desired results in changing social behaviour. Subsequently, the campaign became the world's first organisation to have its funding discontinued by the Global Fund to Fight AIDS, Tuberculosis and Malaria, during December 2005. The discontinuance of monetary support reflects debate about the effectiveness of LoveLife's HIV-prevention programme and the viability of behaviour-changing AIDS education (Peng, 2006).

3

Purpose of the Study

The focus of this research study is on the use of fear appeal advertising pertaining to HIV/AIDS and whether the use of fear increases the likelihood of adopting appropriate safe sexual behavior among adolescents. It examines the influence of high, medium, and low/no fear appeal based advertising on fear for HIV/AIDS infection. The influence of personality traits as an external influence on efficacy beliefs, attitude and behavioral intention are also examined. Based on the research objectives, the following hypotheses were proposed when using low-, medium- and high-fear appeal advertising stimuli respectively. H01 Fear arousal is positively influenced by the level of fear appeal; H02 Attitude towards the advertisement is positively influenced by the level of fear appeal; H03 Intention to adapt sexual behaviour is positively influenced by the

122

Terblanche-Smit and Terblanche

level of fear appeal; H04 Efficacy level is negatively influenced by the level of fear appeal. H05 Personality characteristics positively influence efficacy, fear, attitude and intention.

4

Research Method

Exploratory qualitative research was conducted to select appropriate advertisements to be used in the study. This was followed by an experimental research study to measure individual constructs.

4.1

Exploratory Qualitative Research

Advertisements selection was made from various sources including fifty different websites using a key word search for ‘HIV/AIDS advertising’. The age of the actors in the advertisement (18-30 years) were used as principle selection criteria. Nine print advertisements and ten television commercials were selected for the qualitative research, and used as experimental stimuli. The different advertisements were ranked on a continuum of low/no fear to high fear by the respondents, and shown from low to high fear to prevent desensitizing respondents.

4.2

Experimental Research

Based on the results from the qualitative research three advertisements including low, medium and high fear appeals for print and television were selected for use in the experimental research. The advertisements were tested for face validity by senior academics, as well as advertising and marketing research practitioners. A pre-test post-test, 3 X 2 between subjects, experimental design was used to collect data from 360 respondents. A 17 item self-completion questionnaire (5 point Likert type scale) with statements based on risk behavior diagnosis (Witte, 1998) was completed prior to intervention and post-intervention. Three experimental groups were exposed to print, and three to television advertisements, as an intervention, with each cell exposed to only one type of fear appeal. Fear arousal was measured postintervention by ratings of six mood adjectives to indicate the fear experienced. Adapted instruments of LaTour and Tanner (2003) and LaTour and Rotfeld (1997) were used to measure attitude towards the various fear appeal advertisements (5 statements on a 5 point Likert type scale), and behavioural intent (4

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention

123

statements of a 5 point Likert type scale), after the intervention. Efficacy measurements included response efficacy (effectiveness of suggested response, i.e. “Using condoms is effective in preventing HIV infection”) and self efficacy (ability to carry out the suggested response, i.e. “I am able to use condoms to prevent getting HIV infection”). The Single-Item Measures of Personality (SIMP) scale, were used to measure five personality traits (extraversion, agreeableness, emotional stability, conscientiousness and openness/ intellect) of respondents. The SIMP demonstrated acceptable psychometric properties in previous research (Woods and Hampson, 2005) and its use is specifically recommended for research studies where time is limited, for pilot research, and for other research that require respondents to provide multiple personality ratings of themselves (Saucier and Goldberg, 1996; Smith and Snell, 1996; Van der Zee et al., 2002; Woods and Hampson, 2005).

4.3

Data Collection and Sample

Adolescents between the ages of 18 and 24 years, from the three main racial groups in South Africa were identified as the research population, and were invited to participate in the research via an independent fieldwork research company. Quota sampling was used to draw respondents from the target population. Respondents represented educated, middle to upper income groups who had access to television and print media and advertising. The sample size for the four focus groups included 10 respondents per group. The experimental research (n=360) included 60 respondents per experimental cell with an equal male and female split, and 20 respondents per racial group, namely Black African, White and Coloured.

5

Major Findings

The following major findings have been derived at based on the qualitative and experimental research.

5.1

Fear, Attitude, Intention and Efficacy

Reliability of summed scales used to measure fear, attitude and behavioral intention was confirmed with Cronbach alphas of 0.81, 0.72 and 0.71 respectively.

124

Terblanche-Smit and Terblanche

Analysis of variance (ANOVA) comparisons indicated significant mean differences for type of fear appeal (p<0.0001), namely low, medium and high fear appeals. This showed that respondents were more afraid of contracting HIV after exposure to medium fear appeals than low fear appeals, as well as for high fear appeals than low fear appeals, confirming H01. Attitude and behavioral intention constructs’ ANOVA comparisons indicated significant mean differences for fear appeal (p < 0.01). This revealed that respondents’ attitude toward the advertisements, as well as their intention to adapt behavior was more positive after exposure to medium fear appeals than low fear appeals, as well as for high fear appeals than low fear appeals, confirming H02 and H03. A smaller change in attitude was evident from the medium to high groups than from the low to medium groups, as depicted in Figure 2. Interestingly the Black respondents’ attitude was more negative after exposure to high than medium fear appeal advertisements. This could indicate that their perceived efficacy made them feel less in control at the highest fear level, and they could possibly react by controlling the fear, which inhibits protective behavior.

Current effect: F(4, 326)=2.7033, p=.03053 Vertical bars denote 0.95 confidence intervals 5

attitude

4

3

2

1 low fear

medium fear appeal

high fear

Race black Race coloured Race white

Figure 2: Means plot for attitude and fear appeal type per respondent group

Pre- and post-test response- and self-efficacy measures were conducted for both condom usage and safe sexual behavior. Cronbach’s alpha scores for individual items ranged between 0.6 and 0.7 for both pre- and post-tests. ANOVA

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention

125

comparisons indicated that pre-and post-tests for all groups pertaining to efficacy for condom usage resulted in a significant interaction for time with fear (p=0.007). Respondents’ efficacy showed a downward movement after exposure to medium and high fear appeals, and an upward movement when exposed to low fear appeals. Medium and high fear appeal exposure thus caused respondents to be more afraid of HIV/AIDS and less in control (lower efficacy) in terms of condom usage in general and believing they can easily use condoms to prevent HIV infection, confirming H04. ANOVA comparisons denoted no significant time effect for efficacy in terms of safe sexual behavior, which indicated that respondents’ response efficacy in terms of safe sexual behavior didn’t change after exposure to the various fear appeal advertisements.

5.2

Personality

Respondents rated themselves as average to more extraverted ( x = 4.3) on the reversed extraversion scale (1=highest; 9=lowest). The agreeableness score ( x = 6.2), indicated that respondents viewed themselves as generally trusting and forgiving, interested in people, but can sometimes be taken for granted, as per the description of this trait. The emotional stability score ( x = 4.6) illustrated that respondents saw themselves as average in terms of emotional stability, thus somewhat sensitive and relaxed, somewhat excitable and unemotional, somewhat tense and rarely irritated. Respondents’ conscientiousness score ( x = 4.7), was average for the reversed scale for conscientiousness (1=highest; 9=lowest), revealing somewhat diligent and precise, and somewhat inflexible personality traits. The openness/intellect score ( x = 5.9) was slightly above average for openness/intellect, and respondents were therefore somewhat more reflective, with an active imagination (Smith and Snell, 1996; Van der Zee et al., 2002; Woods and Hampson, 2005). Since this research study was based on the fear emotion, it was decided to investigate the emotional stability of respondents to ascertain whether there were any differences among the different respondent groups. A one-way analysis of variance for emotional stability indicated significant mean differences for race (p < 0.02). The means plot, Figure 3, illustrates a downward trend for white respondents. White respondents thus felt less emotionally stable than black and coloured respondents, and saw themselves as below average in terms of emotional stability. This finding was confirmed with Bonferonni tests indicating significant differences between black and white respondents, as well as between coloured and white respondents. Both black and coloured respondents rated themselves as average in terms of emotional stability.

126

Terblanche-Smit and Terblanche

In this instance, white respondents felt more sensitive and excitable and indicated that they can be tense. Black and coloured respondents rated themselves as somewhere in between, thus somewhat sensitive and relaxed, somewhat excitable and unemotional, somewhat tense and rarely irritated (Woods and Hampson, 2005:388).

Current effect: F(2, 335)=4.0297, p=0.02 Kruskal-Wallis p=0.03 Vertical bars denote 0.95 confidence intervals 9 8

emotional stability

7 6 5 4 3 2 1 black

coloured

white

Race

Figure 3: Means plot for emotional stability and race

5.2.1

Personality Correlations

The correlation coefficients between the personality variables and fear, attitude, intention, self-efficacy (use of condoms) and response efficacy (safe sexual behavior), were calculated. Spearman’s ps were used as a measure of nonmetric correlation to examine the correlations between the said variables. The measures relied on rankings and varied between -1.0 and +1.0 (Malhotra and Birks, 2007). It is important to note that the correlations were not strong, however they were still deemed useful to indicate the relationships between variables. The following relationships between variables were evident, thus confirming a slightly positive relationship as indicated in H05.

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention

127

A positive correlation (r = 0.10) was found between agreeableness and selfefficacy (belief that I can use condoms to prevent HIV/AIDS). This pointed out that respondents’ agreeableness was associated with their efficacy levels for condom usage; consequently, the higher respondents measured on agreeableness, the higher their efficacy levels for condom usage. Emotional stability and attitude returned a negative correlation (r = -0.09). This finding indicated that the higher respondents measured on emotional stability, the lower their attitude towards the various fear appeal advertisements. A negative correlation (r = -0.10) was also found for emotional stability and intention, which points to a decrease in intention to adapt sexual behavior when emotional stability increased. Conscientiousness and attitude towards the advertisements presented a negative correlation (r = -0.14). A low score for conscientiousness, on the reversed scale, indicated that respondents were highly conscientious. This indicated that respondents’ conscientiousness influenced their attitudes towards the various advertisements positively. A negative correlation (r = -0.10) was found for openness/intellect and fear. Consequently, the higher respondents measured on openness/intellect, the lower their fear experienced.

6

Implications and Conclusions

Fear as an approach to increase vulnerability is an important element in HIV/AIDS communication that should be considered for future advertising development. The results of this study support the notion that respondents will change their attitude and behavior (safe sex and abstinence) after exposure to medium and high fear advertisements. Self-efficacy levels for condom usage was high even after exposure to medium and high fear appeal messages, and suggests that adolescents believe they can use condoms to prevent HIV/AIDS infection. High response-efficacy levels pertaining to safe sexual behavior highlights that respondents would weigh their risk of actually experiencing the threat (high fear appeal) against actions they can take that would minimize or prevent the threat of HIV/AIDS infection and in this instance revert to safe sexual behavior. Results of the personality correlations support the notion that HIV/AIDS advertising can be tailored to specific combined personality profiles for adolescents. Adolescents’ agreeableness added to their self-efficacy belief that they can use condoms to prevent HIV/AIDS infection. Conscientiousness also influences their attitudes towards advertisements positively. The openness/intellect of this group is also high and this aspect causes them to experience slightly less fear.

128

Terblanche-Smit and Terblanche

Furthermore, when adolescent’s emotional stability is higher their attitude and intention to adapt their sexual behavior is less. White respondents, however, felt less emotionally stable than Black and Coloured respondents. If adolescents feel above average in terms of emotional stability, this could drive them to be over confident and less scared of contracting HIV/AIDS. A key issue of this research study concerns the overall implications of the outcomes for future advertising. Future HIV/AIDS advertising campaigns targeted at adolescents should be developed to communicate the target audiences’ susceptibility to the disease and should ensure that adolescents experience relevant fear levels. This will drive them to change the way they think about HIV/AIDS, which will ultimately influence them to modify their sexual behaviour to safe sexual behaviour. The understanding of behaviour remains the key for advertising specialists to take up the challenge to curb the HIV/AIDS pandemic in a scientifically responsible manner. From this study it is evident that racial characteristics also play a role in reaction to fear appeals. It is however important to note that different approaches can be used to develop HIV/AIDS communication that is effective in changing behaviour. The problem is not easily solved and when this approach is placed in a larger societal context other approaches apart from targeted advertising campaigns aimed at specific racial groups can be used. Other approaches could include investigating adolescents’ susceptibility to HIV/AIDS or investigating message content to improve effectiveness. Emotional development in adolescence could play a role in young people’s responses to fear appeals. Therefore a replication of this study among younger adolescents, and comparison of different age groups, could prove to assist further development of fear appeal advertising. Furthermore, to see how the different dependent variables interrelate a model could be designed and tested using structural equation modeling.

7

References

Avert (2008), “The South African national HIV survey.” Available via ww.avert.org/safricastats.html.Accessed 10 Mar 2008. Burnett, J.J. & Wilkes, .E. (1980), “Fear appeals to segments only,” in: Journal of Advertising Research, 20(5), 21-24. Burnett, J.J. & Oliver, R.L. (1979), “Fear appeal effects in the field: a segmentation approach,” in: Journal of Marketing Research, 16, 181-190. Easterling, D.V. & Leventhal, H. (1989), “Contribution of concrete cognition to emotion: Neutral symptoms as elicitors of worry about cancer,” in: Journal of Applied Psychology, 74, 787-796.

The Impact of Personality Differences on Efficacy, Attitude and Behavioural Intention

129

Fishbein, M. (2000), “The role of theory in HIV prevention,” in: AIDS Care, 12(3), 77-278. Henthorne, T.L., LaTour, M.S. & Nataraajan,R. (1993), “Fear appeals in print advertising: an analysis of arousal and ad response,” Journal of Advertising, 22(2), 59-68. Hergenhahn, B.R. & Olson, M.H. (2003), “An introduction to theories of personality,” (6th ed.) Pearson Education, New Jersey. Janis, I.L. (1967), “Effects of fear arousal on attitude change: recent developments in theory and experimental research,” in Berkowitz, L. (ed.), Advances in experimental social psychology (3rd ed.), Academic Press, New York, 166-225. LaTour, M.S. & Rotfeld, H.J. (1997), “There are threats and (maybe) fear-caused arousal: Theory and confusions of appeals to fear and fear arousal itself,” Journal of Advertising, 26(3), 45-59. LaTour, M.S. & Tanner,J.F. (2003), “Radon: Appealing to our fears,” Psycholology and Marketing, 20(5), 377-394. Leventhal, H. (1971), “Fear appeals and persuasion: the differentiation of a motivational construct,” American Journal of Public Health, 61(6), 1208-1224. Malhotra, N.K. & Birks, D.F. (2007), “Marketing research: An applied approach,” (3rd ed.). Prentice Hall, London. Ortony, A. & Turner, T.J. (1990), “What's basic about basic emotions!,” Psychological Review, 97, 315-331. Peng, T. (2006), “South Africa: LoveLife Faces Up to Funding Cuts And Critics,” In: Business Day. Available via. http://allafrica.com/stories/html, Accessed 12 Jun 2008. Quinn, V., Meenaghan, T. and Brannick, T. (1992), “Fear appeals: segmentation is the way to go,” International Journal of Advertising, 11, 355-366. Rogers, R.W. (1983), “Cognitive and physiological processes in fear appeals and attitude change: A revised theory of protection motivation,” in: L. Cacioppo and R. Petty (eds.), Social Psychophysiology, Guilford, New York, 153-176. Rotfeld, H.J. (2000), “The textbook effect: Conventional wisdom, myth, and error in marketing,” Journal of Marketing, 64(2), 122-127. Saucier, G. and Goldberg, L.R. (1996), “Evidence for the Big Five in analyses of familiar English personality adjectives,” European Journal of Personality, 10(1), 61-77. Smith, D.R. & Snell, W.E. (Jr.) (1996), “Goldberg's bipolar measure of the Big-Five personality dimensions: reliability and validity,” European Journal of Personality, 10(4), 283-299. Tay, R.; Ozanne, L. and Santiono, J. (2000), “Advertising and road safety: a segmentation approach,” Proceedings of ANZMAC 2000 Conference Visionary Marketing for the 21st Century: Facing the Challenges, 1248-1251, Australia. Van der Zee, K., Thijs, M. & Schakel, L. (2002), “The relationship of emotional intelligence with academic intelligence and the Big Five,” European Journal of Personality, 16(2), 103-125. Witte, K. (1992), “Putting the Fear Back into Fear Appeals: The Extended Parallel Process Model,” Communication Monograms, 59, 329-349. Witte, K. (1998), “Theory-based interventions and evaluation of outreach efforts,” Seattle, WA. in: National Network of Libraries of Medicine, Pacific Northwest Region. Available via http://www.nnlm.nlm.nih.gov/pnr/eval/witte.html, Accessed 17 May 2008. Witte, K. & Allen, M. (2000), “A Meta-Analysis of Fear Appeals: Implications for Effective Public Health Campaigns,” Health Education and Behavior, 27 (5), 591-615. Woods, S.A. & Hampson, S.E. (2005), “Measuring the Big Five with single items using a bipolar response scale,” European Journal of Personality, 19, 373-390.

Part III

Generation

The (B)old and the Beautiful: Investigating the Preference of Senior Consumers Concerning (the Age of) Models Used in Advertising Katarina Panic, Ghent University, Belgium Verolien Cauberghe, Ghent University, Belgium Delphine Verhoye, Ghent University, Belgium

1

Introduction

Today, it is estimated that 7.6 per cent of the world population is aged over 65 (CIA, 2009), making it the fastest-growing age segment in the world. Considering that older citizens do not only grow in number, but also in life expectancy and economic power, one would expect the elderly to receive a great amount of attention from advertisers. But even though marketing executives are aware of the importance of senior consumers, older models are rarely used in mainstream advertising. When they appear in advertisements, elderly are primarily depicted in a negative, stereotypical way (Gunter, 1998; Sikkel, 2004). Older models are mainly found in ads for ‘senior products’ like health products, often suggesting some form of physical impairment. However, these cliché images are far from the current reality. Today’s seniors are socially active, living an active and dynamic life. They are also a powerful consumer group with the means and the willingness to try out new things (Hanson, 1998). Nevertheless, almost no models older than 50 can be found in advertisements for general, age-universal products. This underrepresentation and stereotyping evokes feelings of dissatisfaction, frustration and even insult (Dutta-Bergman, 2006). Many older consumers become alienated from and disinterested in advertisements because they do not identify with the models used in advertisements. These emotions can negatively influence seniors’ attitudes towards advertising (and brands) and affect their purchase intention. While a series of studies revealed that older consumers would like to see more models of their own age in ads (e.g. Long, 1998), other research suggests that seniors do not wish to be confronted more models of their age group in advertisements (Greco, 1989; Carrigan and Szmigin, 1999). Therefore, the main question addressed in this study is: do senior consumers prefer seeing models of their own age, models who match their subjective age, or even younger models in advertising?

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_9, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

134

Panic, Cauberge and Verhoye

Past research has found that the age of the model used in advertisements also has an influence on the models' credibility (Gunter, 1998). In this study, the relationship between the age of the models used in advertising and the perceived credibility is investigated for older consumers. In addition, the moderating influence of gender is studied to determine whether preferences regarding model likeability and credibility differ for male and female seniors. To conclude, the impact of model likeability and credibility on the advertising effectiveness (measured through attitude towards the ad, attitude towards the brand and purchase intentions) is explored.

2 2.1

Conceptual Framework and Hypothesis Development Understanding the Older Consumer

After recognizing the potential of the growing group of senior consumers, the next step is to develop advertising campaigns that actually appeal to the elderly. This appears to be more complex than many marketers expect. Older consumers have different motivations and needs than the younger, mainstream audience that marketers are used to target. As consumers grow older, their preferences, taste and priorities change. Therefore, communication needs to be adjusted accordingly. Some authors (e.g., Chingching, 2008) state that it is best to use older models when targeting the more mature market. Others advise the use of models who are ten to fifteen years younger than the target audience (Loro, 1989; Sikkel, 2004), or argue that one should portray the target audience in the same way as they see themselves (e.g., Tréguer, 1998). So the question arises: how ‘old’ do older people see themselves? Who do they identify themselves with? In order to answer these questions, we investigate the role of models used in advertisements since they set an image of the product user and his or her lifestyle. After seeing a certain product being advertised by a certain model, consumers will categorize it as ‘something for me’ or ‘not something for me’. This way, the age of the model has an important influence on the consumers’ interest in the product and eventually on the purchase decision (Chingching, 2008).

2.2

The Role of Subjective Age

To disentangle the impact of the model’s age on advertising effectiveness, it is necessary to look at the concept subjective age or “the age one perceives one’s self to be” (Stephens, 1991, p.37). Subjective age considers how old people feel and how they perceive themselves, as opposed to how old they actually are (the

The (B)old and the Beautiful

135

objective or chronological age). Research shows that 60 to 75% of the people aged over 60 tend to see themselves ten to fifteen years younger than they actually are (Sudbury and Simcock, 2009). Accordingly, a study by the NCOA (2002) revealed that nearly half of the people aged over 65 actually consider themselves middle-aged or young. As consumers tend to buy brands and products that reflect their aspirational identity rather than their objective identity (Wolfe and Snyder, 2003), subjective age helps to achieve better targeting and more effective advertising content than objective age. According to Chingching (2008), people will evaluate products as less appropriate for them to use if the estimated age of the model in the advertisement differs too much from their own subjective age, making subjective age a much better criterion for predicting consumer responses to advertisements and purchase intention than the consumer’s objective age. The age of the model used in advertisements can also influence the perceived credibility of the message. The use of older characters can, for example, increase credibility of an advertisement (Gubernick, 1996). Since seniors identify themselves most with models who’s age matches their own subjective age (i.e. models who are ten to fifteen years younger than they are), we expect that seniors will like these models most and also perceive them as the most credible spokesperson. This leads to the following hypothesis: H1: “An advertisement showing a model from the subjective age category will lead to a higher model likeability and a higher perceived credibility than an advertisement showing a younger model, or a model from the objective age category” According to Seccombe and Ishii-Kuntz (1991), both men and women consider women as ‘old’ at an earlier age than men. Also, negative social stereotypes define women as unattractive and sexually inactive after a certain age (Gannon, 1994). As they try to reject this stigma of old age, maintaining a young identity becomes much more important to women than to men. For men, maturity is associated with power and social prestige. Their sexual attractiveness is less likely to be linked to appearance and they are more likely to change their subjective age identity to ‘being old’ after retirement (Pliner et al., 1990). In order to preserve a positive self-image, older women are more eager to look younger than older men. Therefore, they are more attracted to the use of younger models in advertisements than to models of their subjective, or objective age. Thus, we expect the respondents’ gender to have an effect on his or her model likeability and the perceived model credibility: H2a: “For male respondents, the model likeability and credibility will be higher for the model of the subjective age category than for the model of the younger, or the objective age category.”

136

Panic, Cauberge and Verhoye

H2b: “For female respondents, the model likeability and credibility will be higher for the youngest model than for the model of the subjective or objective age category” Past research has demonstrated that model likeability and credibility have a positive impact on advertising effectiveness in terms of attitude toward the ad, attitude toward the brand and purchase intentions (Lafferty and Goldsmith, 1999). There are no reasons to expect differences for older consumers. This leads to the following hypothesis: H3: “Both model likeability and credibility have a positive impact on the attitude-toward-the-ad, attitude-toward-the-brand and purchase intention”

3

Materials and Methods

A between subjects 2 (model’s gender: male – female) x 3 (age category: young – subjective age – objective age) factorial design was used. Six different advertisements were created using photographs of models (3 males and 3 females). The ads were constructed using either a picture of a model who was 25 to 30 years younger than the respondents (young model), a picture of a model who was about 10 years younger than the respondents (subjective age category) or a picture of a model who was the same age as the respondents (objective age category). Since we wanted to measure self-perception, male respondents were asked to evaluate an advertisement showing a male model and female respondents were shown an advertisement with a female model. To make the picture look like an advertisement, a package shot of a bottle of water was added across all six conditions. We used bottled water because this is a neutral product which is frequently used by senior consumers, but not exclusively by them. Further, a nonexistent brand name (Givet) and a self-invented slogan (‘purifies body and soul’) were added to the advertisement. The advertisements were all of the same size, had the same background color and showed a close up of a model looking into the lens of the camera. None of the models wore glasses and they were all shown in a stagnant pose to keep the advertisements as neutral and similar as possible (cf. figure 1-6).

137

The (B)old and the Beautiful

Figure 1-3: Male models (shown to male respondents)

Figure 1: Young model

Figure 2: Subjective age model

Figure 3: Objective age model

Figure 4-6: Female models (shown to female respondents)

Figure 4: Young model

Figure 5: Subjective age model

Figure 6: Objective age model

The respondents for the main study were recruited from different senior organizations and events. After they received information about the general purpose of the study, each participant was asked to take a close look at the advertisement he or she was given and fill in a standardized questionnaire. The seniors were randomly exposed to one of the six conditions (see table 1 for an overview). In total, 185 respondents participated in the study (48.6% male, average age of 67.38).

138

Panic, Cauberge and Verhoye

Table 1

Gender

N

Mean objective age (respondents)

Self-reported subjective age

Estimated age of the model

Respondents who saw the model from the first age category ( young model): Males 30 67.03 55.18 42.07 Females 31 67.06 55.49 39.45 Respondents who saw the model from the second age category (subjective age): Males 29 67.93 57.93 58.11 Females 28 67.25 58.14 57.79 Respondents who saw the model from the third age category (objective age): Males 30 67.60 58.80 70.67 Females 29 67.45 57.11 67.11 Attitude towards the spokesperson likeability was measured using the scale of Whittler and Dimeo (1991) (α=.910) and the credibility of the model using the scale of Ohanian (1991) (α=.915). To measure the level of attitude toward the ad, the scale of Simons and Carey (1998) (α=.886) was used. Brand attitude was measured using the scale of Mahnnis and Park (1991) (α=.867) and purchase intention by the scale of Baker and Churchill (1977) (α=.876). To exclude confounding effects, model attractiveness for controlled for. The attractiveness of the model was measured by the scale of Ohanian (1990) (α=.932). Although various previous studies (e.g. Markides and Boldt, 1983; NCOA, 2002; Sudbury and Simcock, 2009) show that the subjective age of an 65+ year old individual indeed differs approximately 10 years from his or her chronological age, this assumption was tested in this study. To measure the respondents’ subjective age, Barak and Schiffman’s (1981) scale was used (α=.814).

4 4.1

Results Pretests

In the first pretest, the estimated age of 56 models was measured to define the age manipulation. Ten randomly selected respondents (50% male, 65 to 70 years old) were shown 28 pictures of models of their own gender and were asked to

The (B)old and the Beautiful

139

estimate the age of these models. The pictures were all coloured and equal in size, composition and background colour. Based on the mean estimated age of each model, three male and three female models were selected to represent each of the three age categories. Also, involvement with the advertised product (bottled water) was pretested. Results show no significant difference in involvement between male and female seniors for bottled water (Mmale= 3.06 vs. Mfemale= 3.16, t(183)=.699; p=.485). To assess the success of our manipulations, a second pretest was conducted. 20 randomly selected respondents (50% male, 65 to 70 years old) were asked to estimate the age and rate the attractiveness of the 18 models (9 male and 9 female) selected in pretest 1 on a 5-point Likert scale. Male respondents rated the pictures of the 9 male models, female respondents were asked to do the same for the 9 female models. Results confirm the age estimations made in the first pretest. Finally, one picture of a male and one picture of a female model was selected for each of the three age categories, making sure that the perceived attractiveness did not differ significantly between the models.

4.2

Manipulation Check

The manipulation checks confirmed that the perceived age of the models matched the age categories they were assigned to (male models: M young= 42.07 years; Msubj= 58.11 years; Mobj= 70.67 years. Female models: Myoung= 39.45 years; Msubj= 57.79 years; Mobj= 67.11 years). Also, the respondent’s mean subjective age was measured. Based on past studies, we estimated that the subjective age of the respondents should be approximately ten years younger that the objective age (which means 57 years for our respondents). This was confirmed by the manipulation check (Msubj.age= 57.09). There were no significant differences in subjective age between male and female respondents (Mmale= 57.3 vs. Mfemale= 56.9; t(182)=.426, p=.671).

4.3

Analysis

The effect of two independent variables (the respondents’ gender and age of the models) on model likeability and credibility was analyzed using two separate ANOVA’s. There appeared a main effect of the model’s age on the model likeability (F(5, 184)= 7.699, p=.001). The results show that the model of the subjective age category was liked significantly better than the model of the objective age category (Msubj=3.623 vs. Mobj=3.276; t(120)=1.976, p=.050).

140

Panic, Cauberge and Verhoye

However, model likeability did not differ significantly between the younger model and the model of the subjective age category (Msubj= 3.623 vs. Myoung= 3.922; t(121)=1.887, p=.062). The results show a trend: respondents tend to like the younger model more than the model of the subjective age. The same results were found for credibility. Results show a significant main effect of the models' age on the perceived credibility of the model (F(5, 184)= 5.547, p=.005). The model of the subjective age category is perceived as more credible than the model of the objective age category (Msubj= 3.348 vs. Mobj=3.006; t(120)=2.323, p=.022). However, the model of the subjective age category is not perceived as more credible than the youngest model (Msubj= 3.348 vs. Myoung= 3.483; t(121)=.883, p=.379). H1 is partly supported. Gender appeared to have a moderating influence on the relation between the models' age and model likeability F(5, 184)= 3.226, p=.042) as well as the relation between age and model credibility (F(5, 184)= 3.373, p=.036). The results of the interaction effect (cf. Figure 7) show that male respondents like the model of the subjective age category (M= 3.75) more than the objective age category (M= 3.07, t(58)=2.964, p=.004). However, they did not like model of the subjective age category (M= 3.75) more than the youngest model (M= 3.65, t(58)=.413, p=.681). The same trend occurred for model credibility (cf. Figure 8). H2a is partly supported.

Figure 7: Perceived model likeability

The (B)old and the Beautiful

141

As expected, female respondents liked the youngest model (M= 4.17) more than the model of the subjective age category (M= 3.49) and the objective age category (M= 3.47; t(46.401)=3.539, p=.001). There was no significant difference in liking between the subjective and the objective age category (t(60)=.090, p=.928) (cf. figure 7). Again, results show the same trend for model credibility (cf. figure 8). H2b is supported.

Figure 8: Perceived Model Credibility

In the last analysis, the impact of model likeability and credibility is measured on advertising effectiveness. Correlation analyses shows a strong, positive relation between model likeability and attitude towards the ad (r=.608; p<.001), attitude-towards-the-brand (r=.603; p<.001) and purchase intention (r=.467; p<.001). There also appears a strong, positive relation between model credibility and attitude towards the ad (r=.565; p<.001), attitude towards the brand (r=.550; p<.001) and purchase intention (r=.535; p<.001). H3 is thus supported.

5

Discussion, Conclusion and Further Research

In this study, we investigated the impact of the age of models used in advertisements on advertising effectiveness for older consumers. We found that seniors evaluate the youngest model and the model who match with their subjective age

142

Panic, Cauberge and Verhoye

significantly more likeable and also more credible than models of their own age. In addition, the results show that older men liked the model from the subjective age category and the youngest model significantly more - and perceive these models as more credible – than the model of their own age. Elderly women like the youngest model significantly more - and evaluate this model as more credible – than the older models. These results confirm that maintaining a young identity is much more important to women than to men. In addition, both model likeability and credibility appear to have a positive impact on consumer’s attitude towards the ad, attitude towards the brand and purchase intention. These results confirm previous findings that senior consumers prefer models who match their aspirational (or subjective) identity, rather than their objective identity. Also, results confirm that seniors are more likely to buy products when these are shown in advertisements that reflect this aspirational identity (e.g., Wolfe and Snyder, 2003). Further, Chingching’s (2008) statement that subjective age is a much better criterion for predicting consumer responses to advertisements and purchase intention than objective age is confirmed, although using 25-years younger models seems to be even more effective when targeting elderly women. From this overall preference for younger models, we can conclude that older people do not look for a representative view of the world in advertising. An explanation for these results may be the trend in today’s society, characterized by a fear of aging and a strong need to stay and look youthful. Not only do seniors like models who match their subjective age more, they also find them more credible. The limitations of the current study suggest directions for further research. First, this study used only one model per age category. Although perceived attractiveness was taken into account during the selection, further research should use more models per age category. Second, for the young age category, we used models who were 25 to 30 years younger than the respondents (resp. 45 years old). These models were perceived as highly likeable and credible, especially with the female seniors. However, it is possible that this preference for younger models does not hold for extremely young models. Gunter (1998) already found that senior consumers perceive models younger than 35 as less credible than older models. Therefore, it would be interesting to include pictures of models who are, for example, 40 to 50 years younger than the respondents. Finally, we only tested one product category. It is possible that the use of older models is more lucrative when designing advertisements for certain product categories, while younger models may be better for promoting products associated with youth and rejuvenation.

The (B)old and the Beautiful

6

143

References

Baker, M. & Churchill, G. (1977), “The impact of physically attractive models on advertising evaluations” in: Journal of Marketing research, 14, 538-555. Barak, B. & Schiffman, L.G. (1981), “Cognitive age: a nonchronological age variable” in: K.B. Monroe (ed.). Advances in consumer research, 8. Ann Arbor: Association for consumer research. Carrigan, M. & Szmigin, I. (1999), “The portrayal of older characters in magazine advertising” in: Journal of Marketing Practice, 5 (6/7/8), 248-261. Central Intelligence Agency (CIA) (2009, October 5). https://www.cia.gov/library/publications/theworld-factbook/ geos/xx.html Chingching, C. (2008), “Chronological age versus cognitive age for younger consumers” in: Journal of Advertising, 37 (3), 19-32. Dutta-Bergman, M. (2006), “The demographic and psychographic antecedents of attitude toward advertising” in: Journal of Advertising Research, 46 (1), 102-112. Gannon, L. (1994), “Sexuality and menopause” in: Choi, P., Nicolson, P. (Eds). Female sexuality: psychology, biology and social context. London: Harvester Wheatsheaf. Greco, A.J. (1989), “Representation of the elderly in advertising: crisis or inconsequence?” in: Journal of Consumer Marketing, 6 (1), 37-44. Gubernick, L. (1996), “Gray Hair Is Cool” in: Forbes, 157 (9), 116-118. Gunter, B. (1998), “Understanding the older consumer: the grey market”, London: Routledge Lafferty, B. & Goldsmith, R. (1999), “Corporate Credibility’s Role in Consumers’ Attitudes and Purchase Intentions When a High versus a Low Credibility Endorser Is Used in the Ad“ in: Journal of Business Research, 44, 109–116. Long, N. (1998), “Broken down by age and sex: exploring the ways we approach the elderly consumer,” in: Journal of the Market Research Society, 40 (2), 73-91. Loro, L. (1989), “Reaching 50-plus market without alienating younger consumers takes finesse,” in: Advertising Age, May 22, S7. Markides, K. & Boldt, J. (1983), “Change in subjective age among the elderly: a longitudinal analysis” in: The Gerontologist, 23 (August), 422-427. NCOA, (2002), “American perceptions of aging in the 21st century. The NCOA’s continuing study of the myths and realities of aging”. Washington DC. Ohanian, R. (1990), “Construction and validation of a scale to measure celebrity endorsers’ perceived expertise, trustworthiness and attractiveness” in: Journal of Advertising, 19, 39-52. Ohanian, R. (1991), "The Impact of Celebrity Spokespersons' Perceived Image on Consumers' Intention to Purchase," in: Journal of Advertising Research, 31,1, 46-54. Pliner, P., Chaiken, S. & Flett, G.L. (1990), “Gender Differences in Concern with Body Weight and Physical Appearance over the Life Span” in: Personality and Social Psychology Bulletin, 16(2), 263-273. Seccombe, K. & Ishii-Kuntz, M. (1991), “Perceptions of problems associated with aging: Comparisons among four older age cohorts” in: The Gerontologist, 31, 527–533. Sikkel, D. & Keehnen, E. (2004), “Ervaren maar veranderlijk: Het consumentengedrag van de vijftigplusser.“ Amsterdam: Kluwer. Simons, J. & Carey, K. (1998), “A Structural Analysis of Attitudes Toward Alcohol and Marijuana Use” in: Personality and Social Psychology Bulletin, 24 (7), 727-35. Stephens, N. (1991), “Cognitive age: a useful concept for advertising?” in: Journal of Advertising, 20 (4), 37-48. Sudbury, L. & Simcock, P. (2009), “Understanding older consumers through cognitive age and the list of values: A U.K.-based perspective” in: Psychology and marketing, 26(1), 22-38. Tréguer, J.P. (1998), “Le senior marketing,” Paris: Dunod.

144

Panic, Cauberge and Verhoye

Whittler, T. & DiMeo, J. (1991), “Viewers’ reactions to racial cues in advertising stimuli” in: Journal of Advertising Research, 31 (December) 37-46. Wolfe, D., & Snyder, R. (2003), “Ageless Marketing,” USA: Dearborn Trade Publishing.

Children’s Advertising Literacy: Do BMI, Body Shape Perception, Self-Esteem and TV Exposure Matter? Julia Spielvogel, University of Klagenfurt, Austria Ralf Terlutter, University of Klagenfurt, Austria

1

Introduction

The purpose of this study is to analyze whether the body mass index (BMI), body shape perception (BSP) and self-esteem influence children’s development of TV advertising literacy. Given the present rise in overweight among children and the dominance of advertising for highly sugared, salted or fatty foods, the relationship between the three above-mentioned factors and advertising literacy deserves attention (e.g. Desrochers and Holt, 2007; Harker et al., 2007). Furthermore, the influence of TV exposure on advertising literacy as well as the relationship between advertising literacy and the effects of advertising on children are researched. Concerning the nature of these relationships a consensus has not yet been reached in the literature and further empirical research is required. Advertising literacy is seen as important in children’s maturation into responsible and competent consumers. For decades, the age of a child has been regarded as the most critical factor influencing the development of advertising literacy. Nevertheless, many empirical studies show significant differences between same-aged children in this respect (Oates et al., 2003; Roedder John, 1999). Obviously, there are other factors that might also make an impact, but their effects are far from clear today. While research has focused on the influence of the family, school or peers, little attention has been devoted to children’s personal characteristics and appraisal such as BMI, BSP and self-esteem. However, this study shows that these factors may drive or impede advertising literacy. Additionally, there are important inconsistencies in the results of existing studies concerning the influence of TV exposure on advertising literacy as well as a lack of empirical results concerning the relationship of advertising literacy and the effects of advertising. Therefore, this study wants to contribute to the state of research by methodically collecting and analyzing empirical data in this field. The focus of this study is on television advertising since television represents the medium that is most frequently and most preferably used by children aged 8 to 11 (Feierabend and Klingler, 2009). Due to this reason and S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_10, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

146

Spielvogel and Terlutter

due to the low requirements related to TV consumption (e.g. no reading or writing skills necessary), television advertising also represents the kind of advertising that is most strongly perceived by children (Baacke et al., 1999).

2 2.1

Theoretical Background and Research Gaps Advertising Literacy and its Factors of Influence

To describe advertising literacy, different labels have been used in the literature ranging from attitude towards advertising (e.g. Rossiter, 1977) over knowledge and understanding of advertising (e.g. Roedder John, 1999) to advertising literacy (e.g. Livingstone and Helsper, 2006). However, when comparing the exact definitions, it appears that the authors are referring to similar constructs. The development of TV advertising literacy can be described as a progress in four steps that goes along with children’s maturation on a cognitive, social and experiential level (Roedder John, 1999). First, children gain the ability to distinguish TV commercials from the regular program. At the beginning, prominent visual and sound effects of advertisements help to differentiate since they lead to more attention and therefore a higher probability of recognizing a commercial as such (Gunter and Furnham, 1998; Wartella and Ettema, 1974). Second, children start to understand the persuasive intent of advertising. At this point commercials are no longer interpreted as a form of objective product information or entertainment (Roedder John, 1999). Often a very general, negative view of advertising emerges at the same time (Bever et al., 1975). Third, children acquire the ability to recognize specific instances of bias and deception in advertising. They start to evaluate individual advertisements and put the respective truthfulness into question, which usually leads to less trust and less liking (Robertson and Rossiter, 1974; Roedder John, 1999). Fourth, usually during early adolescence knowledge of advertising tactics and appeals manifests (Roedder John, 1999). While skepticism towards advertising is more or less formed and stable at the end of childhood, knowledge of tactics and appeals is developing over the following years (Boush et al., 1994). From a theoretical point of view, different concepts are necessary to explain the effects and processes of the development of advertising literacy (Robertson and Rossiter, 1974; Roedder John, 1999; Zigler and Child, 1969). First, Piaget’s (e.g. 1932, 1952) theory of cognitive development and its four stages are em-

Children’s Advertising Literacy

147

ployed to explain the gradual acquirement of abilities such as an advertising literate behavior. Cognitive development is understood as a psychological process, which leads to a logical conception and adjustment to the environment (Ginsburg and Opper, 1969; Roedder John, 1999). Second, Selman’s (e.g. 1980; 1975) stage model of social perspective taking is important. The ability to recognize certain characteristics and the perspective of others is the basis of interpersonal understanding and awareness, which plays an essential role for example in gaining the ability to understand advertisers’ motives and intent (Cooney and Selman, 1980; Roedder John, 1999). Third, social learning theories have to be considered. They focus on socialization agents (e.g. parents, teachers, friends, television), which supposedly transfer norms, attitudes, motivations and behaviors to children during interactions (Bandura, 1962, 1977; Moschis and Churchill, 1978). Over the last 40 years several factors have been researched that might influence the development of advertising literacy. In this regard, age is the factor that has been examined most extensively. In accordance with the stages of cognitive development, researchers have identified age groups indicating when the majority of children reach the different levels of advertising literacy. Beyond that, the following influencing factors can be found in the literature: family environment (interaction with parents and siblings, social class, parents’ education, parenting style, number of siblings), TV and advertising exposure, school and media education, format of advertisements, sex, peers, intelligence (for a more detailed analysis of the different factors see Terlutter and Spielvogel, 2010). The state of research varies widely between the factors and there has not yet been an attempt to survey them all at once. Therefore, it is necessary to conduct further empirical research to contribute to an understanding of the complex development of advertising literacy. In addition, the question arises whether there are other factors that have not yet received attention in the literature. This study focuses on overweight and a possible relation to advertising literacy. These days, this issue is of crucial importance both to society and academic research (e.g. Harker et al., 2007; Robinson, 1999; Young, 2003). Although several studies have been conducted on the effects of TV/advertising exposure on food choice and obesity, little is known about the influence of BMI, BSP and self-esteem on advertising literacy especially concerning food advertising.

2.2

New Factors of Influence and Hypotheses

Rosenberg et al. (1995, p. 141) define global self-esteem as: “… the individual’s positive or negative attitude towards the self as a totality”. From a theoretical

148

Spielvogel and Terlutter

point of view, the self-enhancement theory serves as a basis and describes selfesteem as a fundamental human need and motive. To a certain degree, the selfesteem is responsible for the psychological wellbeing of a person. Therefore, every person strives for the protection and enhancement of the feelings of selfworth and suffers from psychological distress when this desire becomes frustrating (Rosenberg et al., 1995). As studies have shown, children feel this need the same way as anybody else, while at the same time different processes and influences shape their self-esteem (Piers and Harris, 1964; Rosenberg and Pearlin, 1978). The physical appearance is the aspect of the self that is available and recognizable at first sight. In our society, overweight imposes a social identity on the person that is often devaluated by the environment (Miller and Downey, 1999). Therefore, affected people often feel stigmatized (Crocker et al., 1993; Puhl and Brownell, 2002) and have a negative perception of their body (Collins, 1991; Lewis and Hill, 1998). In this context, Miller and Downey (1999) conducted a meta-analysis of more than seventy studies that examined the relationship of body weight or body dissatisfaction and self-esteem. They showed that lower self-esteem associates with higher weight whereby the correlation is stronger for studies that focus on the self-perceived weight or body shape perception than for those focusing on the actual body weight. Although the effect is stronger for adolescents than for children, the latter are also affected by it (Miller and Downey, 1999). This leads to the following hypotheses: H1: Children with a higher BMI have a more negative BSP (higher BSP score) than children with a lower BMI. H2: Children with a higher BMI have a lower self-esteem than children with a lower BMI. H3: BSP and self-esteem of children are negatively related. Individuals with low self-esteem tend to have less trust in their own abilities as well as to react more sensitively on failure and rejection (Epstein, 1979). Furthermore, studies have shown a negative relationship between persuasibility and self-esteem (Hovland et al., 1953; Leventhal and Perloe, 1962; Maile, 1977). Lesser and Abelson (1959) reported that this relationship is also applicable to children. The lower the self-esteem, the more likely a person will adopt someone else’s (more appreciated) opinion. In contrast to that, a person with high self-esteem and confidence is in general not as easily swayed (Neubauer, 1976). Therefore, we believe that on the one hand children with a negative BSP and on the other hand children with low self-esteem are at risk of being less

Children’s Advertising Literacy

149

skeptical and more easily persuaded by advertisements. If children do not dare or are too doubtful about themselves to take a skeptical position, the development of advertising literacy might be impeded. We do not only test the influence of self-esteem but also of BSP on advertising literacy since the latter might be a good indicator of children’s self-esteem in association with body weight. Therefore the following hypotheses are expected: H4: Children with a more negative BSP (higher BSP score) have a lower advertising literacy than children with a more positive BSP (lower BSP score). H5: Children with a lower self-esteem have a lower advertising literacy than children with a higher self-esteem. In addition to the new factors of influence that are described above, TV exposure of children is a much-discussed topic in the academic literature on advertising literacy. However, existing studies often resulted in contradictory findings. For instance, Mangleburg and Bristol (1999) as well as Moschis and Churchill (1978) found a positive influence of TV exposure on advertising literacy or consumer socialization, while Chan and McNeal (2006), Kapoor and Verma (2005) as well as D’Alessio et al. (2009) showed a negative influence. However, empirical studies resulting in the identification of a negative influence seem to prevail. In this regard, D’Alessio et al. (2009) reported that children, who more often watch television, believe more in advertising claims and exhibit a lower understanding of the nature and purpose of advertising. They use the cultivation theory (Gerbner and Gross, 1976) as a possible explanation for this relationship. Accordingly, television program and advertising may ‘cultivate’ a distorted view of one’s social reality on the basis of the reality shown on TV. With a more frequent exposure to television the credence increases that TV presents reality in the actual way. Additionally, the mere exposure effect (Zajonc, 1968) may explain the formation of a more positive attitude towards advertisements through repeated exposure (Eagle, 2007; Gorn and Goldberg, 1980; Gunter and Furnham, 1998). Following this perspective, a negative influence of TV exposure on advertising literacy is expected in hypothesis 6: H6: Children with higher TV exposure have a lower advertising literacy than children with lower TV exposure. According to Livingstone and Helsper (2006), the relationship between advertising literacy and the effects of advertising lacks integration in the research. On the one hand, a widespread assumption exists among researchers that this relationship is clear and well established. On the other hand, when examining

150

Spielvogel and Terlutter

the literature available, little empirical evidence can be found to substantiate this claim (Roedder John, 1999). In accordance to the theoretical background presented above, the development of cognitive abilities to understand advertising and its intent (often described as cognitive defenses or persuasion knowledge) is supposed to lead to lower susceptibility to advertising effects. Correspondingly, lower advertising literacy is expected to result in higher susceptibility to advertising effects. Therefore, hypothesis 7 is expected: H7: Advertising literacy and susceptibility to advertising effects are negatively related. Figure 1 shows the research model applied in this study.

Figure 1: Research model

3 3.1

Empirical Study Method, Sample and Measures

Two hundred face-to-face interviews with children aged 8 to 11 were conducted in three elementary schools in Austria. All children were in the third and fourth grade. The interviews were conducted outside the classroom in a separate setting with one interviewer asking one child at a time. The interviewers were trained and provided with a standardized questionnaire that is appropriate for children at this age. The interviewers read out the questions, presented colorful cards with the answer choices, explained the questions if necessary and wrote down the answers. The questions focused on food advertising and in case of comprehension problems interviewers asked the children to think of advertising for sweets. Additionally, they measured the children’s height and weight using

151

Children’s Advertising Literacy

portable scales and tape measures. Parents were informed and had agreed to the child’s participation in advance. The characteristics of the sample concerning age, grade and sex are presented in Table 1.

Table 1: Characteristics of the sample Variable Age

Grade

Sex

Categories

Frequency

Percentage

8 years

53

26.5%

9 years

102

51.0%

10 years

42

21.0%

11 years

3

1.5%

3rd grade

101

50.5%

4th grade

99

49.5%

112

56.0%

88

44.0%

Male Female

Variables were measured using scales from the literature. The respective measurement is reported in Table 2 in the Appendix. With regard to the measurement of advertising literacy, Rossiter’s (1977) scale was employed since it is one of the few quantitative instruments in this field and has been frequently used in the literature for its brevity, objectivity and reliability. This monodimensional construct comprises different aspects such as the credibility and likeability of commercials as well as their persuasive power (D’Alessio et al., 2009). A test of internal-consistency reliability led to the elimination of the item “GOODONLY” for further analyses. The remaining six items resulted in a Cronbach’s alpha of .65 (Rossiter, 1977: .69) and a mono-dimensional structure.

3.2

Results of the Empirical Study

Children’s BMI was calculated from their height and weight. Following the recommendations of Kromeyer-Hauschild et al. (2001) for German and Austrian children, the sample was divided into groups by means of BMI percentiles considering variances due to age and gender. To define the weight groups the following percentiles were applied, which serve as a standard in medicine in Ger-

152

Spielvogel and Terlutter

man-speaking countries (Kiefer et al., 2006; Kromeyer-Hauschild et al., 2001): underweight: <10th percentile (24 children), normal weight: 10th – 90th percentile (140), overweight: 90th – 97th percentile (21), obesity: >97th percentile (14). In this study 10.6% of the children fall in the overweight and 7.0% in the obesity category. This result is in accordance with data from other Austrian studies (Kiefer et al., 2006). H1 expected that children with a higher BMI have a more negative BSP (higher BSP score) than children with a lower BMI. Means of BSP are presented in Table 3. Table 3: Means of BSP and self-esteem compared across BMI groups Underweight N

Normal weight

Overweight

Obesity

24

140

21

14

BSP

-.13

.44

.52

1.38

Self-esteem

6.92

6.71

6.71

6.07

Following Collins (1991), ANOVA was applied and revealed a significant influence of BMI on BSP (F(3, 194) = 7.88; p = .000). Post-hoc Scheffé tests showed significant differences between the obesity group and the underweight group (p = .000) and the normal weight group (p = .005). Differences between the obesity and the overweight group were not significant on a 5%, but on a 10%-level (p = .068). Thus, H1 is supported. H2 hypothesized that the BMI influences children’s self-esteem. Means of self-esteem are presented in Table 3. ANOVA was not significant on a 5%-level (F(3, 195) = 2.34; p = .075). However, pair wise tests using contrasts showed that the obesity group has significantly lower self-esteem than the underweight (p = .011) and the normal weight group (p = .020). Comparison of the obesity and overweight group was not significant on a 5%, but on a 10%-level (p = .058). Therefore, H2 receives at least partial support by the data. As the study focuses on children’s overweight, children with a negative score on BSP, meaning that they want to gain weight, are excluded from further analysis. This leads to a remaining sample of 184 children. H3 stated that a negative relationship is supposed between BSP and self-esteem. A Pearson correlation was used to test this prediction and resulted in a significant, negative correlation (r = −.24; p (1-tailed) = .000). This implies that more positive BSP

153

Children’s Advertising Literacy

(BSP score a 0) is associated with higher self-esteem as well as more negative BSP (BSP score > 0) with lower self-esteem. Hence, H3 is supported by the data. To test H4, the sample was divided into three groups by means of BSP (positive, negative, strongly negative BSP). The mean of advertising literacy of each BSP group can be found in Table 4.

Table 4: Means of advertising literacy compared across BSP groups

BSP score N Advertising literacy

Positive BSP

Negative BSP

Strongly negative BSP

0

1 or 2

3 or 4

111

67

11

18.45

18.36

14.50

ANOVA showed a significant influence of BSP on advertising literacy (F(2, 181) = 4.20; p = .017). Post-hoc Scheffé tests indicated significant differences between the strongly negative BSP group and the positive BSP group (p = .017) as well as the negative BSP group (p = .023). Additionally, the correlation between BSP and advertising literacy was calculated (r = −.13; p (1-tailed) = .044) in order to control for self-esteem. By comparing the original correlation to the partial correlation using self-esteem as control variable (r = −.10; p (1-tailed) = .085), it showed that the controlled correlation was not significant on a 5%, but on a 10%-level and the coefficients were almost the same. Therefore, we conclude that there is a direct, negative link between the BSP score and advertising literacy and H4 is supported. To examine the influence of self-esteem on advertising literacy as expected in H5, the sample was divided into a low and a high self-esteem group whose characteristics are presented in Table 5.

Table 5: Means of advertising literacy compared across self-esteem groups

N Advertising literacy

Low self-esteem

High self-esteem

23

161

16.96

18.48

154

Spielvogel and Terlutter

ANOVA revealed a significant influence of self-esteem on advertising literacy (F(1, 182) = 4.31; p = .039). Therefore, H5 is supported by the data. In H6 a negative influence of TV exposure an advertising literacy was expected. The means of advertising literacy of the four TV exposure groups are presented in Table 6.

Table 6: Means of advertising literacy compared across TV exposure groups

N Advertising literacy

Every day

Several times a week

Once a week

Less often

86

64

17

17

17.53

18.89

17.94

20.18

To test the hypothesis, ANOVA was used and resulted in a significant influence on advertising literacy (F(3, 180) = 4.30; p = .006). Post-hoc Scheffé tests showed a significant difference between the ‘every day’ and the ‘less often’ group (p = .026). The difference between the ‘every day’ and the ‘several times a week’ group was not significant on a 5%, but on a 10%-level (p = .095). Thus, H6 is supported by the data. H7 stated that a negative relationship exists between advertising literacy and susceptibility to the advertising effects. A Pearson correlation revealed a significant relationship between the variables (r = −.39; p (1-tailed) = .000), which supports H7. This means that children with higher advertising literacy are less vulnerable to the effects of advertising, i.e. are less susceptible or seduced by the advertisements in their desire and choice of (food) products.

4

Discussion and Implications

This is the first study that based on a theoretical deduction analyzes the interplay of BMI, BSP and self-esteem in the field of advertising literacy with a special focus on food advertising. Additionally, the influence of TV exposure on advertising literacy and the relationship between the latter and the effects of advertising were researched. The data from personal interviews with primary school children aged 8 to 11 lend support to the six hypotheses addressing the influence of BMI on BSP, the link between BSP and self-esteem and their respective influence on advertising literacy, the influence of TV exposure on avertising literacy as well as the relationship between advertising literacy and advertising

Children’s Advertising Literacy

155

effects. The hypothesized influence of BMI on self-esteem was at least partially supported. The results indicate that children with high BMI, negative BSP and low selfesteem, characteristics that often occur together, are at risk to lag behind concerning advertising literacy. Furthermore, advertising literacy might be impeded by heavy exposure to television. At the same time lower advertising literacy is associated with higher susceptibility to advertising effects. This might lead to a vicious circle since television advertisements for food are mainly for highly sugared, salted or fatty products. The findings of this study are of interest to companies as well as policymakers who decide about the regulation of advertising to children and the sponsorship of programs for media and advertising education. The at-risk group might need special enhancements and informing should be implemented. Thus, the existing initiatives for media and advertising education in primary schools might incorporate this knowledge and for example target possible traps of food advertisements in their programs. At the same time, centers for overweight and obese children might take up the point of the influence of advertisements and consider it in their counseling and treatments. The result concerning TV exposure emphasizes the importance of control and restriction of children’s time spent watching television through the parents. In addition, the family’s encouragement is essential to foster children’s self-esteem and BSP.

5

References

Aufenanger, S. (1997), "Verlockungen und Gefahren heutiger Werbewelten für Kinder," in: Meister, D. M.; Sander, U. (eds.): Kinderalltag und Werbung. Zwischen Manipulation und Faszination, Luchterhand, Neuwied et al., 28-44. Baacke, D., Sander, U., Vollbrecht, R. & Kommer, S. (1999), "Zielgruppe Kind: Kindliche Lebenswelt und Werbeinszenierungen," Leske + Budrich, Opladen. Bandura, A. (1962), "Social learning through imitation," in: Jones, M. R. (eds.): Nebraska symposium on motivation, University of Nebraska Press, Lincoln, 211-269. Bandura, A. (1977), "Social learning theory," Prentice-Hall, Englewood Cliffs. Bever, T.G., Smith, M.L., Bengen, B. & Johnson, T.G. (1975), "Young viewers’ troubling response to TV ads," in: Harvard Business Review, Vol. 53, Iss. 6, 109-120. Boush, D.M., Friestad, M. & Rose, G.M. (1994), "Adolescent skepticism toward TV advertising and knowledge of advertiser tactics," in: Journal of Consumer Research, Vol. 21, Iss. 1, 165-175. Chan, K. & McNeal, J.U (2006), "Chinese children’s understanding of commercial communications: A comparison of cognitive development and social learning models," in: Journal of Economic Psychology, Vol. 27, Iss. 1, 36-56. Collins, M.E. (1991), "Body figure perceptions and preferences among preadolescent children," in: International Journal of Eating Disorders, Vol. 10, Iss. 2, 199-208.

156

Spielvogel and Terlutter

Cooney, E.W. & Selman, R.L. (1980), "Children's use of social conceptions: Toward a dynamic model of social cognition," in: Personnel and Guidance Journal, Vol. 58, Iss. 5, 344-352. Crocker, J., Cornwell, B. & Major, B. (1993), "The stigma of overweight: Affective consequences of attributional ambiguity," in: Journal of Personality and Social Psychology, Vol. 64, Iss. 1, 6070. D’Alessio, M., Laghi, F. & Baiocco, R. (2009), "Attitudes toward TV advertising: A measure for children," in: Journal of Applied Developmental Psychology, Vol. 30, Iss. 4, 409-418. Desrochers, D. M. & Holt, D.J. (2007), "Children’s exposure to television advertising: Implications for childhood obesity," in: Journal of Public Policy and Marketing, Vol. 26, Iss. 2, 182-201. Eagle, L. (2007), "Commercial media literacy. What does it do, to whom—and does it matter?," in: Journal of Advertising, Vol. 36, Iss. 2, 101-110. Epstein, S. (1979), "Entwurf einer Integrativen Persönlichkeitstheorie," in: Filipp, S.-H. (eds.): Selbstkonzept-Forschung. Probleme, Befunde, Perspektiven, Klett-Cotta, Stuttgart, 15-45. Feierabend, S. & Klingler, W. (2009), "Was Kinder sehen. Eine Analyse der Fernsehnutzung Dreibis 13-Jähriger 2008," in: Media Perspektiven, Vol. 3, Iss. -, 113-128. Gerbner, G. & Gross, L. (1976), "Living with television: The violence profile," in: Journal of Communication, Vol. 26, Iss. 2, 172-199. Ginsburg, H.P. & Opper, S. (1969), "Piaget’s theory of intellectual development," Prentice-Hall, Englewood Cliffs. Gorn, G.J. & Goldberg, M.E. (1980), "Children's responses to repetitive television commercials," in: Journal of Consumer Research, Vol. 6, Iss. 4, 421-424. Gunter, B. & Furnham, A. (1998), "Children as consumers. A psychological analysis of the young people’s market," Routledge, London. Harker, D., Harker, M. & Burns, R. (2007), "Tackling obesity: Developing a research agenda for advertising researchers," in: Journal of Current Issues and Research in Advertising, Vol. 29, Iss. 2, 39-51. Hovland, C.I., Janis, I.L. & Kelley, H.H. (eds.) (1953), "Communication and persuasion. Psychological studies of opinion change," Yale University Press, New Haven. Kapoor, N. & Verma, D.P.S. (2005), "Children’s understanding of TV advertisements: Influence of age, sex and parents," in: Vision – The Journal of Business Perspective, Vol. 9, Iss. 1, 21-36. Kiefer, I., Rieder, A., Rathmanner, T., Meidlinger, B., Baritisch, C. & Lawrence, K., et al. (2006), "Erster österreichischer Adipositasbericht 2006. Grundlage für zukünftige Handlungsfelder: Kinder, Jugendliche, Erwachsene," Altern mit Zukunft, Wien. Kromeyer-Hauschild, K., Wabitsch, M., Kunze, D., Geller, F., Geiß, H.C., Hesse, V., et al. (2001), "Perzentile für den Body-mass-Index für das Kindes- und Jugendalter unter Heranziehung verschiedener deutscher Stichproben," in: Monatsschrift Kinderheilkunde, Vol. 149, Iss. 8, 807-818. Kutteroff, A., Behrens, P., König, T., Schmid, T., Feierabend, S. & Rathgeb, T. (2009), "KIMStudie 2008: Kinder+Medien, Computer+Internet: Basisuntersuchung zum Medienumgang 6bis 13-Jähriger in Deutschland," Medienpädagogischer Forschungsverbund Südwest, Stuttgart. Lesser, G. S. & R. P. Abelson (1959), "Personality correlates of persuasibility in children," in: Janis, I. L.; Hovland, C. I. (eds.): Personality and persuasibility, Yale University Press, New Haven, 187-206. Leventhal, H. & Perloe, S.I. (1962), "A relationship between self-esteem and persuasibility," in: Journal of Abnormal and Social Psychology, Vol. 64, Iss. 5, 385-388. Lewis, M. K. & Hill, A.J. (1998), "Food advertising on British children’s television: a content analysis and experimental study with nine-year-olds," in: International Journal of Obesity, Vol. 22, Iss. 3, 206-214.

Children’s Advertising Literacy

157

Livingstone, S. & Helsper, E.J. (2006), "Does advertising literacy mediate the effects of advertising on children? A critical examination of two linked research literatures in relation to obesity and food choice," in: Journal of Communication, Vol. 56, Iss. 3, 560-584. Maile, C.A. (1977), "The apparent lack of self-esteem and persuasibility relationships," in: Journal of Psychology, Vol. 96, Iss. 1, 123-130. Mangleburg, T.F. & Bristol, T. (1999), "Socialisation and adolescents’ scepticism toward advertising," in: Macklin, M. C.; Carlson, L. (eds.): Advertising to children. Concepts and controversies, Sage, Thousand Oaks et al., 27-48. Martensen, A. & Hansen, F. (2002), "Children’s knowledge and interpretation of commercial advertising – Intentions, truthfulness and viewing habits," in: Hansen, F. e. a. (eds.): Children – Consumption, advertising and media, Copenhagen Business School Press, Copenhagen, 149180. Miller, C.T. & Downey, K.T. (1999), "A meta-analysis of heavyweight and self-esteem," in: Personality and Social Psychology Review, Vol. 3, Iss. 1, 68-84. Moschis, G.P. & Churchill, G.A. (1978), "Consumer socialization: A theoretical and empirical analysis," in: Journal of Marketing Research, Vol. 15, Iss. 4, 599-609. Neubauer, W.F. (1976), "Selbstkonzept und Identität im Kindes- und Jugendalter," München, Basel. Oates, B., Blades, M., Gunter, B. & Don, J. (2003), "Children’s understanding of television advertising: A qualitative approach," in: Journal of Marketing Communications, Vol. 9, Iss. 2, 59-71. Piaget, J. (1932), "The moral judgement of the child," Harcourt, New York. Piaget, J. (1952), "The origins of intelligence in children," International University Press, New York. Piers, E.V. & Harris, D.B. (1964), "Age and other correlates of self-concept in children," in: Journal of Educational Psychology, Vol. 55, Iss. 2, 91-95. Puhl, R. & Brownell, K. (2002), "Stigma, discrimination, and obesity," in: Fairburn, C.; Brownell, K. (eds.): Eating disorders and obesity: A comprehensive handbook, Guilford Press, New York, 108-112. Robertson, T.S. & Rossiter, J.R. (1974), "Children and commercial persuasion: An attribution theory analysis," in: Journal of Consumer Research, Vol. 1, Iss. 1, 13-20. Robinson, T.N. (1999), "Reducing children’s television viewing to prevent obesity: A randomized controlled trial," in: Journal of the American Medical Association, Vol. 282, Iss. 16, 1561-1567. Roedder John, D. (1999), "Through the eyes of a child. Children’s knowledge and Understanding of Advertising," in: Macklin, M. C.; Carlson, L. (eds.): Advertising to children. Concepts and controversies, Sage, Thousand Oaks et al., 3-26. Rosenberg, M. (1965), "Society and the adolescent self-image," Princeton University Press, Princeton. Rosenberg, M. & Pearlin, L.I. (1978), "Social class and self-esteem among children and adults," in: American Journal of Sociology, Vol. 84, Iss. 1, 53-77. Rosenberg, M., Schooler, C., Schoenbach, C. & Rosenberg, F. (1995), "Global self-esteem and specific self-esteem: Different concepts, different outcomes," in: American Sociological Review, Vol. 60, Iss. 1, 141-156. Rossiter, J.R. (1977), "Reliability of a short test measuring children's attitudes toward TV commercials," in: Journal of Consumer Research, Vol. 3, Iss. 4, 179-184. Selman, R. (1980), "The growth of interpersonal understanding," Academic Press, New York. Selman, R.L. (1975), "Level of social perspective taking and the development of empathy in children: Speculations from a social-cognitive viewpoint " in: Journal of Moral Education, Vol. 5, Iss. 1, 35-43. Terlutter, R. & Spielvogel, J. (2010), "Einflussfaktoren auf die Entwicklung der Werbekompetenz bei Kindern," in: der markt - Journal für Marketing, Vol. 49, Iss. 1, 17-41.

158

Spielvogel and Terlutter

Von Ploetz, A. (1999), "Werbekompetenz von Kindern im Kindergartenalter: ein Experiment zum Erkennen von Werbung," KoPäd, München. Wartella, E. & Ettema, J.S. (1974), "A cognitive development study of children’s attention to television commercials," in: Communication Research, Vol. 1, Iss. 1, 69-88. Young, B. (2003), "Does food advertising influence children's food choices? A critical review of some of the recent literature," in: International Journal of Advertising, Vol. 22, Iss. 4, 441-459. Zajonc, R.B. (1968), "Attitudinal effects of mere exposure," in: Journal of Personality and Social Psychology, Vol. 9, Iss. 2, 1-27. Zigler, E. & Child, I.L. (1969), "Socialization," in: Lindzey, G.; Aronson, E. (eds.): The handbook of social psychology, Vol. 3, The individual in a social context, Addison-Weslay, Reading et al., 450-589.

159

Children’s Advertising Literacy

6

Appendix

Table 2: Measures Measures Body mass index (BMI)

Scale

Items

Percentiles:  Height Kromeyer-  Weight Hauschild et al. (2001)

Evaluation

Scores

 BMI = kg/m2  BMI percentile rankings (age, gender)  Grouping by BMI percentile

 Underweight: <10th p.  Normal weight: 10th – 90th p.  Overweight: 90th – 97th p.  Obesity: > 97th p.

 Identification  Neutral: positive of sameBSP, no wish to gender Self change and Ideal  Negative: negaSelf tive BSP, wish to  Discrepancy gain weight score (Self –  Positive: negative Ideal Self) BSP, wish to lose weight

Body shape Collins perception (1991) (BSP)

7 child figures (both genders) illustrating body weight from very thin to obese

Self-esteem Rosenberg (1965) adapted by Boush et al. (1994)

 4 answer  Self-confidence choices in one’s own abilities  Sum score  Self-contentment  (Optimism: removed due to age groups in the sample)

TV exposure

Kutteroff et  Times usually al. (2009); spent watching Moschis & television Churchill (1978); Von Ploetz (1999)

 4 answer choices

 Possible scores: 2 –8  Higher score = higher self-esteem

 Possible values: 1 –4  Higher value = higher TV exposure

160 Measures

Spielvogel and Terlutter

Scale

Advertising Rossiter literacy (1977)

Items

Evaluation

Scores

 Possible scores: 6  Perceived truth-  4 answer fulness choices – 24  Perceived persu-  Sum score  Higher score = asive power higher advertising  Factor analyliteracy  Believability of sis: monocharacters dimensional structure  Potential annoying qualities  Overall likability  Trustworthiness as guides to product purchase  (Objectivity in describing advertised products)

 Desire to try Advertising Adapted  4 answer  Possible scores: 4 effects from Maradvertised prodchoices – 16 tensen and ucts (e.g. sweets)  Sum score  Higher score = Hansen  Propensity to higher susceptibil Factor analy(2002); feel hungry due ity to advertising sis: monoAufenanger to food advereffects dimensional (1997) tisement structure  Desire for advertised (food) product in store  Propensity for choice of advertised (food) product in store

The Relevance of Life Changing Situations for Media Usage and their Relevance as a Segmentation Strategy for Media Companies and Advertisers Andrea Leopold, University of Klagenfurt, Austria Sandra Diehl, University of Klagenfurt, Austria

1

Introduction

Nowadays it is impossible to go through daily life without the use of media; the patterns of our daily lives are determined by and intertwined with media use. But what happens to daily routines and media usage when an event occurs that drastically changes the circumstances of our lives? Peoples´ lives can be transformed by major changes: some might be intentional, such as occupational changes or relocations, e.g. leaving the parental home for the first time; other changes may happen unintentionally, e.g. illnesses or the death of relatives or loved ones. How can such a situation of change be understood or constructed? What happens to the daily routines of the people affected? What do they require in order to cope with the changed situation? Which challenges are they confronted with? How can life changing situations be operationalized by media companies and advertisers? How can they support individuals in life changing situations? These questions will be addressed by the authors in an ongoing research project with the underlying assumption that new needs arise in these life changing situations and that most of these needs can be met by the use of media respectively the media usage supports satisfying the arising needs. The underlying assumption states that media use and the appropriate equipment for it can help individuals deal with new situations. In this article, basic considerations in relation to the underlying assumption and first pre-studies are presented, and the relevance of life changing situations for people in those situations, for media companies and advertisers - as a possible segmentation strategy – are discussed; and further steps in this research project are presented. In the second section, the construct of “life changing situation” is theoretically discussed and a pre-study dealing with the understanding of life changing situations is presented. A diagram explains the life changing phase.

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_11, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

162

Leopold and Diehl

The next section addresses existing segmentation strategies and the identified gap which could be closed by a more situational approach, i.e. considering life changing situations as a new segmentation strategy; and the resulting theses are presented. In section four, a possible win-win-win-situation is introduced: People in life changing situations, media companies and advertisers could all benefit from the situational approach of considering life changing situations as a segmentation strategy. In section five, the results of a first pre-study concerning the life changing situation of first-term university students is discussed. Further research steps are described in the last section.

2

The Construct of Life Changing Situations

As far as the authors know, the term “life changing situations” has not been defined yet. Before we define life changing situations, we discuss related terms and present a qualitative pre-study to further conceptualize the construct of life changing situations.

2.1

Life Events

The quite similar term life event was understood first in an epistemological way by Holmes and Rahe in 1967 as occurrences that were likely to bring about readjustment-requiring changes in people’s usual activities to determine how stressful life events could affect mental and physical health. This definition is still used in this way in later publications (see for example Dohrenwend, 2006). “Social readjustment includes the amount and duration of change in one’s accustomed pattern of life resulting from various life events. As defined, social readjustment measures the intensity and length of time necessary to accommodate to a life event regardless of the desirability of this event.” (Holmes and Rahe, 1967: 34) For Holmes and Rahe, the historical starting point was the Life Chart invented by Meyer (Lief, 1948), who was interested in the relationship of three open-ended disciplines – biology, psychology, and sociology – to the processes of health and disease in human beings, which means the relation between life situations, emotions and diseases. (Holmes, 1989) “The entries on the life chart

The Relevance of Life Changing Situations for Media Usage

163

describe life situations – experiences having to do with growth, development, maturation, and senescence – as well as the patient’s emotional responses to those situations.” (Holmes, 1989) In 1967, Holmes and Rahe developed the Social Readjustment Rating Scale (SRRS) to quantify human experience, which enabled them to measure “the amount of change that certain events require of people: the readjustment necessary in their coping style and life-style. Both ‘desirable’ and ‘undesirable’ events require readjustment.” (Holmes and David, 1989: 31) The SRRS combined with the later developed Schedule of Recent Experience (SRE) “[…] reflected the magnitude of change required in ongoing life adjustment by each event, and the values were defined in terms of life change units (LCU). The scale ranged from 11 LCU for a minor violation of the law to 100 LCU for death of spouse. The Social Readjustment Scale has since been validated by cross-cultural and by comparison with other scaling methods.” (Rahe and Holmes, 1989: 193) In this way, a “[…] life crisis was therefore stipulated to mean any clustering of life change events whose individual values summed to 150 LCU or more in one year. The duration of a life crisis was found to be generally 1 – 2 years; rarely, it was longer.1” (Rahe and Holmes, 1989 and see also Rahe, 1969) The term “life event” is mainly operationalized from a medical point of view (stress caused by life events and having impact on health) of changing situations. The authors, however, try to see the occurring events from a more psychic and psychological perspective – without disregarding diseases, which could cause a life changing situation. Yet, whereas Holmes and Rahe put their focus on diseases as a consequence of a life event, we consider diseases in themselves to be life changing situations.

2.2

Understanding of Life Changing Situations (Pre-Study)

In order to identify the most important life changing situations and to develop a concept of which events people regard or experience as a life changing situations, a pre-study focusing on the definition was conducted. The term “life changing situation” was described as mentioned above and easily understood by the respondents. 50 people (average age 36.3 years, 33 female, 17 male, 17 students and 33 non-students) were surveyed in an open questionnaire about the 1

For example, Rahe and Holmes (1989) detected that people with scores between 150 and 199 LCU, 37 % had an associated health change. This association rose up to 51 % for people with scores between 200 and 299 LCU and to 79 % for those with scores greater than 300 LCU. However, the authors concede that not all persons (21 % with a LCU greater than 300) experiencing a life crisis are facing a health change in the following two years. In these cases they assume a high resistance to health change.

164

Leopold and Diehl

circumstances which had made their daily lives change profoundly. They were asked to name the three most important events that had triggered this change, which resulted in 137 answers. According to this survey, the changes which had the greatest impact on the respondents’ lives were the birth of a child (26), followed by the beginning of studies including being independent for the first time (14)2, the entrance into the first employment (13), the death of a loved person or relative (13) and marriage (13). Of course, it has to be taken into consideration that the respondents were aged 36.2 years on average. Life changing situations vary according to the age. For persons aged 50 plus (14), the most frequently mentioned life changing situations were childbirth (12) and wedding (7). An interesting finding was that women ranked the birth of a baby highest (9 out of 12) more often than men did (3 out of 12). Men tended to consider the birth of a baby as the second or third important event in their lives (7 out of 14 namings in second or third place).

Namings 30

26 21

25 20 15

14

13

13

13

10

10

7

5

5

5

5

5 0

Figure 1: Understanding of Life Changing Situations in a restricted region of Austria, n = 137 (own figure)

The named life changing situations are quite similar to the events in the SRRS, invented by Holmes and Rahe (1967). 2

Due to the fact that 17 students were asked.

The Relevance of Life Changing Situations for Media Usage

165

Based on these remarks and their own considerations, the authors understand a life changing situation as a special time in life, after a dramatic incident has occurred, which drastically changes the daily life of the person concerned. The occurrences can be intended or unintended, desirable or not desirable, can occur suddenly or slowly over time. In all instances, people must cope with the new situation as dictated by their new needs. This phase lasts until the person is able to re-establish a new daily routine and readjust daily patterns (see following figure).

Figure 2: Construct of a Life Changing Situation (own figure)

3

Theoretical Background and Development of the Thesis

People who are confronted with life changing situations typically adjust or reorientate their daily lives and routines. Old routines, which worked in the past and provided them with security, do not work anymore. They face new challenges, new problems and insecurities. They also experience strong negative or positive emotions and are forced to create new ways of life, leading to intensive cognitive processes as they are focused on creating their new life. In order to deal with their changed situation, people attempt to maintain their existing social relationships while developing new ones at the same time. In order to do so, the use of media plays an important role. The new situation can also lead to a change in media and equipment usage habits. The people concerned still use their old media sources when available, while at the same

166

Leopold and Diehl

time they have to look for new services to satisfy their new needs. Other factors are the financial aspect and the changes in time management – both of which can also trigger modified media usage. Therefore, it can be assumed that people in life changing situations are characterized by a high degree of involvement in and attention for both adequate editorial information (media content) and information on products and services (advertisements) (see also Kroeber-Riel and Weinberg, 2003, Petty and Cacioppo, 1981). According to the Elaboration Likelihood Model of Petty and Cacioppo (1986), it could be assumed that people in life changing situations have a high involvement in relevant issues, so that they will elaborate the information by the central route. Involvement, according to Zaichowsky (1985), means that the relevance of the information on advertisement to a person is based on inherent needs, values and interests. High involvement is characterized by a high activation of the persons, combined with strong emotions and intensive cognitive processes, which lead to extensive buying behavior (Kroeber-Riel and Weinberg, 2003). Rahe and Holmes (1989: 197) describe them from a medical point of view, saying that “[…] the process of human reaction and readjustment to a variety of significant life situations, many body systems (for example, cardiovascular, neuroendocrine, respiratory, musculoskeletal, and autonomic nervous) have been shown to become physiologically activated, some with sustained and eventually injurious effects. The findings […] suggest the possibility that, the greater the magnitude of the crises, the greater the probability that the involved subject’s psychophysiologic activation will result in tissue, organ and body system dysfunction,” which indicates the seriousness of the impact of life changing situations on the concerned persons and their activation. People in life changing situations are thus supposed to have an extensive buying behavior and they are furthermore expected to make extended problemsolving consumer decisions, meaning – according to Solomon et al. (2006) – that they have a high consumer involvement, are unfamiliar with product classes and brands, take more time to think, search and purchase a product, and that they buy more expensive products. High risk and high involvement are typical of this kind of buying decision (see also Peter and Olson 2008). For example, a student leaving the parental home actively searches for information on apartments, furniture, a new computer etc. He or she also looks for information on the city he or she is about to move to in order to study and on services available there, such as medical services etc. In order to find the desired information, the use of media is indispensable. The characteristics of people affected (high involvement, high interest for information) make them a very

The Relevance of Life Changing Situations for Media Usage

167

interesting target group for both media companies and advertising companies. But how do they fit into the typical variables for market segmentation?

3.1

Market Segmentation Strategy

“A market segment is simply a group of present or potential customers with some common characteristics which is relevant in explaining (and predicting) their response to a supplier’s marketing stimuli.“ (Wind and Cardozo, 1984: 209, see also Blackwell et al., 2006) Through segmentation „[…] a company can identify and group customers or countries according to common needs and wants.“ (Keegan and Green, 2005: 258; see also Blackwell et al., 2006) Therefore, they can use basic variables, like demographic segmentation (country, income, population, age, ethnic heritage), psychographic segmentation (attitudes, interests, opinions and lifestyles), behavioral segmentation (user status, usage rate) and benefits users seek (Keegan and Green, 2005; Kotler and Lee, 2008). Some segmentation variables have similarities to and can be combined with life changing situations. For example, Gilly and Enis (1982) suggest a recycled family circle, in which they describe stages in the life circle (e.g. bachelor stage, full nest, empty nest etc.). But Gilly and Enis do not consider what exactly happens within a family when a new situation occurs. The individuals´ feelings, their thoughts and needs in between these stages are neglected. The concept of life changing situations focuses on these exact points: The transitions, the breaks for the individuals, their needs, the changes of their media usage patterns are of interest to the authors. As defined before, every life changing situation turns into a daily routine after some time (threshold or immediate beginning of a new phase), but in a phase, e.g. in a family life circle stage, a situation which is part of the family life circle concept. Other commonly used segmentation criteria are lifestyles, defined “as patterns in which people live and spend time and money” (Blackwell et al., 2006: 277), as well as milieu, personality (Blackwell et al., 2006, Diehl and Terlutter, 2003, Sinus Sociovision, 2010) or affinity for advertising (Smit and Neijens, 2000). All these segmentation variables have in common that they only describe stable and enduring phases in life. The situational factor of breaks in life, life changing situations, is missing here as well. The following graphic shows the gap between the existing segmentation strategies:

168

Leopold and Diehl

Figure 3: Life Changing Situations as new Segmentation Strategy (own figure)

Even though people in life changing situations can certainly be found in existing market segments, the characteristics of their special situation are not sufficiently covered by the discussed segments. A more situational approach is needed – as you can see in the diagram – because the specific situation should be taken into consideration. To summarize these theoretical remarks without making claims of being complete, there is no comparable research on people in life changing situations regarding to market segmentation. Their meaning for media companies, advertisers or marketers has not yet been researched. What makes people in life changing situations especially attractive for market segmentation is that they are characterized by high involvement and special needs; they actively and selectively search for information. Highly involved information seeking persons are very important to media companies and advertisers. Nowadays, recipients are usually suffering from information overload. Information overload adds up to 99 %, with only 1 % of the information from mass media being perceived by the recipients (Meffert, 2000). As early as in 1984 Ferber argues that “[...] we are presented with so much

The Relevance of Life Changing Situations for Media Usage

169

information in our daily lives that we cannot possibly receive it all effectively. What information does get perceived depends both on our attention level and on what we selectively notice of the stimuli that are received by our senses. […] This selectivity may be based on the consumer’s own characteristics, such as needs and values, expectations, and past experiences; or the selectivity may be determined by product characteristics, such as distinctiveness, intensity, and novelty.” (Ferber, 1984: 127).

3.2

Relationship between Advertisers and Media Companies and Individuals

For the relationship between advertisers and media companies and individuals the reader’s engagement with the media vehicle can be brought up. Calder and Malthouse (2008) defined it as “the collective experiences that a reader has with the editorial content.” They proved that “engagement of the audience as a whole with the media vehicle is reflected in the effectiveness of an ad for the entire audience.” (Malthouse and Calder, 2009: 5). They propose a reflection of the price of an ad according to the level of engagement with the editorial part of magazines. Also Bronner and Neijens (2006) examined the relationship between advertising experience and the experience in which these advertisements were embedded. They found a strong correlation between media and advertising experience for print media, which suggests that this “could be that print media titles are rather specific and target their audiences more accurately, making it possible to tailor the advertisements to profiles of these media.” (Bronner and Neijens, 2006: 97) Therefore, taking the target group “people in life changing situations” into consideration can extend and supplement the existing market segmentation in a useful way. The additional variable would be a situational one, depending on the specific life changing situation. Another aspect is that people in life changing situations try to handle their new daily lives with the help of a combination of various media. Consequently, an intermedia survey has to be carried out, as recommended by Smit and Neijens (2000) and Bronner and Neijens (2006) in their media context research. A study by Folkman Curasi, Hogg and Maclaran (2001) focuses on the changing role and identity of mothers entering the empty nest stage as a kind of transition. Folkman Curasi, Hogg and Maclaran (2001) showed that mothers were altering their media usage and started participating in a bulletin board to find ‘communitas’ and advice from other women who had already successfully coped with this transitional period or life changing situation. This indicates that

170

Leopold and Diehl

media usage plays an important role in times of change and can fulfill new emotional and informational needs. Based on the theoretical remarks, especially the findings of Holmes and Rahe (1989) that life events mainly have an influence on the health of the people concerned, and the study by Folkman Curasi, Hogg and Maclaran (2001), the underlying assumption of the research project, which has also to be determined in further research, is: Media usage and media equipment help individuals to deal with new situations. In this way they, as well as the needs of people in these life changing situations, are subject to changes influencing them.

4

Win-Win-Win Situation?

The assumption is that media usage and media equipment play an important role in life changing situations as they help people deal with change. It provides them with orientation, information, entertainment and a kind of social life (Biagi, 2007). It can be argued that life changing situations lead to new needs resulting in a change in media usage patterns. Thus, it can be assumed that people affected by life changing situations are actively searching for useful information on products and services which help them. If a media company manages to satisfy the new needs of thus affected recipients, they in turn are likely to become permanent customers who remain loyal to the company’s products. Therefore, not only the recipients themselves benefit from appropriate offers by the media company, but also the media company benefits if it identifies them as an interesting group. By providing valuable information to people in life changing situations, media companies can also cooperate with advertising agencies which could advertise suitable products and services in their editorial section. Moreover, companies can offer an attractive setting for advertisers in which their ads are noticed, which in turn could lead advertisers to provide a higher budget. By addressing people in life changing situations and even define them as a new segment on the market, advertisers and media companies alike have a chance to target and win new consumers. The following diagram (figure 4) illustrates the possible win-win-win situation, with stronger relationships between the three partners being an additional beneficial outcome.

The Relevance of Life Changing Situations for Media Usage

171

Figure 4: Possible win-win-win situation for people in life changing situations, media companies and advertisers (own figure)

This theoretical framework as well as the underlying assumption will be determined in the ongoing research project. The starting point for these considerations are the results of a first pre-study, which are described and presented in the next chapter.

5

Methodology and First Results

At the beginning of this more comprehensive research project, the authors analyze this basic assumption for one specific life changing situation.

5.1

Methodology

This further pre-study analyzes a specific life changing situation: first year students at a middle-sized Austrian university are interviewed after having left their parental homes. The study consists of three methodical components: (1) a guided interview, (2) an observation of the living environment as well as the media equipment and (3) a questionnaire which pays attention to media equipment both at their former and their present homes. The reason for the

172

Leopold and Diehl

trisection of the study is the perspective that media usage cannot be considered separately from people’s daily lives. (1) In guided interviews (between 45 minutes and an hour each) 14 students are asked about their past media usage – at their parental homes – and their media usage in their new homes as students (all students were living alone in the first term, in an apartment-sharing community or a dormitory). The guided interviews are held at the students’ homes, which makes (2) an observation of their living environments and their media equipment and placement possible. In addition to the guided interviews and the monitoring, the students are asked to fill out (3) a questionnaire about their media infrastructure, that is to say their media equipment at their parental home and at their student’s home. For the interpretation of the gained data, the grounded theory is used, which allows for a combination of qualitative and quantitative findings to generate a new theory concerning the people’s behavior in life changing situations (Glaser and Strauss, 2005).

5.2

Results: The daily patterns are changing

In this second pre-study, we find that students in the first semester are likely to change their media usage pattern and that they are searching for new media content. For example, they search for doctors, leisure facilities and events nearby, they stay in touch with their families and friends over the internet – something they did not have to do before the change. Also, these students do not need a conventional telephone landline at their place of study, but they tend to use the same cell phone network provider as their family or friends. Ten interviewees have also switched to another main medium, i.e. the most used medium with regard to time and content. This was due to new challenges in the student workplace and modified leisure time activities. Thus, the computer/laptop with internet access becomes the new main medium. In addition to that, the observation reveals that the main medium is placed centrally within the living arrangements. Television, radio and to some extent also the internet act as a substitute for the family and the old friends. They help combat feelings of loneliness. Students tend to use more than one medium at the same time. A special form of media usage – concerning its function as a substitute for the family and friends – occurs at the dinner table: Either the television or the radio is turned on during meals. In addition to that, the internet is used or newspapers or books are read in order to masquerade the absence of a vis-á-vis. The computer/laptop with internet access is increasingly used for reading the newspaper, as first-year students do not have a permanent subscription to a print newspaper.

The Relevance of Life Changing Situations for Media Usage

173

It is worth mentioning that students are not restricted in their media usage anymore, for they are neither bound by time nor content. This means that they no longer have to be considerate of other family members as opposed to their school time, when e.g. watching television or facing the problem of several people surfing the web with only one internet access point or limited access. At the beginning of their studies, students learn that university or college life places a certain demand. In order to work properly and to succeed as a student they are required to have access to both the internet and the university’s intranet. As far as time is concerned, the study demonstrates that the lives of people affected by change were getting back to normal at a certain point. They are able to establish new patterns of daily routine. Services and products which are used and prove to be helpful during their life changing situations are still being used, such as newspapers or internet services. As the study shows, the students become permanent users of the media they consider helpful during the transitional period. This is the case because they appreciate the familiarity as well as the support these products render.

6

Discussion and Further Research

The first pre-study concerning the understanding of life changing situations and the second pre-study about the specific life changing situation of students in their first term are the basic steps for further research. In additional studies, it will be necessary to determine media usage and the importance of advertisements. The first outcomes could show that people are changing their behaviour and their media usage pattern due to a specific life changing situation and that new needs are evolving as a consequence. The analysis of people affected by these changes seems to be a promising field of research. Media companies and their products can function as “companions” in life changing situations, provided they are able to supply people with suitable, helpful products in these situations. In order to offer appropriate and appealing products, market research has to be conducted in order to identify and describe the challenges that have to be met in the near future. This strategic planning could contribute not only to the company’s benefit, but could also benefit advertising agencies, because it ensures that demands are met and desires are fulfilled. It would be important to address people in life changing situations, and to acquire them as loyal customers before they re-establish their daily routines. After the reestablishment of the daily routines, people could be assigned to already existing segments (parallel segmentation).

174

Leopold and Diehl

Further steps of the research project will involve the deeper analysis of other life changing situations (e.g. first employment, childbirth), with focus on the underlying psychological, emotional and cognitive processes of people in such situations. This includes people’s involvement, changes in their media usage, their decision making processes and their buying behavior. How products and services by media companies and advertisers should be tailored in order to meet the consumers’ needs will be analyzed. Additionally, it is worth investigating how involved people, who have become permanent and loyal users, are when they re-establish their daily routine. It might also be interesting to find out how male and female recipients respond to life changing situations and if there are gender-specific differences.

7

References

Blackwell, R.D., Miniard, P.W. & Engel, J. F. (2006), “Consumer Behavior”, International Student Edition, 10th ed, Mason, Thomson South-Western. Bronner, F. & Neijens, P. (2006), “Audience Experiences of Media Context and Embedded Advertising, A comparison of eight media,” in: International Journal of Market Research, 48 (1) 81 – 100. Biagi, S. (2007), “Media/Impact: An Introduction to Mass Media”, 8th ed., Belmont, Thomson Wadsworth. Calder, B. J. & Malthouse, E., C. (2008), “Media Engagement and Advertising Effectiveness,” in Calder (ed.): Kellogg on Advertising and Media, New Jersey, John Wiley & Sons, Inc . 1 – 36. Diehl, S. and Terlutter, R. (2003), “The Role of Lifestyle and Personality in Explaining the Attitude toward Ads and the Purchase Intentions,” in: Hansen, F.; Christensen, L.B. (eds.): Branding and Advertising, Copenhagen Business School Press, 306 – 331. Dohrenwend, B. P. (2006), “Inventorying stressful life events as risk factors for psychopathology: Toward resolution of the problem of intracategory variability,” in: Psychological Bulletin, Vol 132(3), May 2006, 477-495 (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1584216/ [25.01.2011]) Ferber, R. (1984), “What do We Know About Consumer Behavior?”, in: Kotler, P., Cox, K. (ed.): Marketing management and Strategy, A Reader, 3rd Edition, New Jersey, Prentice-Hall, Inc, 125 – 136. Folkman Curasi, C., Hogg, M.K. & Maclaran, P. (2001), “Entering the empty nest stage: A multimethod Exploration of Women’s Life Experiences and Coping Strategies in Periods of Life Stage Transition,” in European Advances in Consumer Research, 5, 260 – 267. Gilly, M.C. & Enis, B.M. (1982), “Recycling the Family Life Cycle: A Proposal for Redefinition,” in Advances in Consumer Research, 9, 271 – 276. Glaser, B.G. & Strauss, A.L. (2005), “Grounded Theory,” 2nd Edition, Bern, Hans Huber. Holmes, T. H. & David, E.M. (ed.) (1989). “Life Change, Life Events, and Illness”, Selected Papers, New York, Westport, Connecticut, London, Praeger Publishers. Holmes, T.H. (1989), “It Was in This Setting …,” in: Holmes, T. H.; David, E. M. (eds.) (1989), Life Change, Life Events, and Illness, Selected Papers, New York, Westport, Connecticut, London, Praeger Publishers.

The Relevance of Life Changing Situations for Media Usage

175

Holmes, T.H. & Rahe, R.H. (1967), “The Social Readjustment Rating Scale,” in Journal of Psychosomatic Research 11, 213 – 218, printed In: Holmes, T. H.; David, E. M. (eds.) (1989), Life Change, Life Events, and Illness, Selected Papers, New York, Westport, Connecticut, London, Praeger Publishers. Keegan, W.J. & Green, M.C. (2005) “Global Marketing,” International Edition, 4th Edition. Kotler, P. & Cox, K. (eds.) (1984), “Marketing management and Strategy”, A Reader, 3rd Edition, New Jersey, Prentice-Hall, Inc. Kotler, P. & Lee, N.R. (2008), “Social Marketing, Influencing Behaviors for Good,” 3rd Edition, London, Sage Publications. Kroeber-Riel, W. & Weinberg, P. (2003). Konsumentenverhalten, 8th Edition, München, Vahlen Verlag. Lief, A. (ed) (1948), “The commonsense Psychaitry of Dr. Adolf Meyer”, New York, McGraw-Hill. Malthouse, E.C. and Calder, B.J. (2009), “Media Engagement is as important as advertising execution,” in Terlutter, R.; Diehl, S., Karmasin, M. and Smit, E. (eds.): Proceedings of the 8th ICORIA International Conference on Research, on CD-Rom. May, P.R.A. & Wittenborn, J.R. (eds.), “Psychotropic Drug Response: Advances in Prediction”, Springfield, Illinois, Thomas S.S. Meffert, H. (2000), „Marketing: Grundlagen marktorientierter Unternehmensführung, Konzepte – Instrumente – Praxisbeispiele,“ 9th Edition, Wiesbaden, Gabler. Peter, J.P. & Olson, J. C. (2008), “Consumer Behavior and Marketing,” 8th International Edition, Boston, McGraw-Hill. Petty, R.E. & Cacioppo, J.T (1981), “Issue Involvement as a Moderator of the Effects on Attitude of Advertising Content and Context,” in Advances in Consumer Research, 8, 20 – 24. Petty, R.E. & Cacioppo, J.T. (1986), “Communication and Persuasion, Central and Peripheral Routes to Attitude Change,” New York, Springer. Rahe, R.H. & Holmes, T.H. (1989), “Life Crisis and Disease Onset: Qualitative and Quantitative Definition of the Life Crisis and Its Association with Health Change,” in Holmes, T. H., David, E. M. (eds.) (1989), in Life Change, Life Events, and Illness, Selected Papers, New York, Westport, Connecticut, London, Praeger Publishers, 193 – 201. Rahe, R.H. (1969), “Life Crisis and health Change,” in May, P. R. A.; Wittenborn, J. R. (eds.), Psychotropic Drug Response: Advances in Prediction, Thomas, S. S., Publisher: Springfield Illinois, 92 – 125. Sinus Sociovision (2010). Sinus-Mileus, http://www.sociovision.de/loesungen/sinus-milieus.html [30.01.11]. Smit, E.G. & Neijens, P.C. (2000), “Segmentation Based on Affinity for Advertising,” Journal of Advertising Research, (July/August), 35 – 43. Solomon, M., Bamossy, G.; Askegaard, S. and Hogg, M. K. (2006), “Consumer Behaviour,” A European Perspective, 3rd Edition, New Jersey, Prentice Hall, Inc. Wind, Y. & Cardozo, R. (1984), “Industrial Market Segmentation,” in: Kotler, P.; Cox, K. (eds.): Marketing management and Strategy, A Reader, 3rd Edition, New Jersey, Prentice-Hall, Inc, 208 – 219. Zaichkowsdy, J.L. (1985), “Measuring the Involvement Construct,” in Journal of Consumer Research, 12 (December), 341 – 352.

Part IV

Social Interaction and Advertising

An Investigation of Social Media Usage, Brand Consciousness, and Purchase Intention towards Luxury Products among Millennials Shu-Chuan Chu, DePaul University, USA Sara Kamal, American University in Dubai, United Arab Emirates*

1

Introduction

According to a recent report on the values, attitudes, behaviors and demographic characteristics of the Millennial generation (Pew Research, 2010a), the Millennial generation refers “those born after 1980 –the first generation to come of age in the new Millennium (p.4).” Compared to previous American generations, Milliennials (ages 18-33) are more likely to use the Internet wirelessly. They remain more likely to actively engage in a wide range of communication and entertainment-related activities online, such as using social networking sites and instant messaging, listening to music, reading blogs, and playing games online (Pew Research, 2010b). Among the many online activities, using social media (SM hereafter) such as Facebook, LinkedIn, and Twitter has appeared to be most popular among Millennials. A recent report by eMarketer (2008) forecasts that global advertising expenditures on SM will reach $ 2.6 billion by 2012. Given their potential impact on branding, marketers and advertisers have employed SM as a tool to develop positive brand relationships with young consumers by promoting advertising messages on these sites. In particular, the explosive growth of SM outlets represents huge opportunities for luxury brands in terms of demand (Okonkwo, 2005) and sales (Okonkwo, 2009). Recently, luxury brands are engaging with Millennials through increasingly sophisticated technology. For example, Giorgio Armani utilizes SM to target young consumers who want to be entertained and informed (Silverstein, 2010). Moreover, the increase of affluent Internet users, (defined as those with an annual income above $100,000) (eMarketer, 2007), highlights the importance of e-Retail of luxury goods (Okonkwo, 2009). This is particularly important as to the newly emerged online advertising formats, social media advertising (SMADV hereafter). SMADV may play a role in promoting luxury brands, serving as a source of brand aspiration for potential users (Balwani, 2009) such as Millennials. Of note is that the majority of Millennials are unmarried (Pew Research, 2010a), thus are more likely to have fewer finical responsi*All authors contributed equally, names are listed in alphabetical order. S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_12, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

180

Chu and Kamal

bilities and more discretionary income to spend on themselves than previous generations. Millennials represents an emerging consumer segment for luxury products and are active users of SM (Lenhart et al., 2010; Silverstein, 2010). Thus SMADV may serve as a useful tool in developing attitudes and beliefs towards the luxury brand and will eventually influence behavior (Lavidge and Stiener, 1961; Palda, 1966). Given the rampancy of SM, luxury brands will have to adopt SMADV in order to maintain salience among the new and younger generation, namely Millennials. Despite many luxury brands have opened up to online advertising, particularly SMADV, research to date has not examined if SM usage might be related to consumption of luxury products among young generations. With the mounting growth of affluent Internet users, the online promotion of luxury products has become increasingly imperative and necessary. At the same time, SM usage and advertising spending on SM platforms are also on the rise. Along this line, this study empirically investigated millennial’s SM usage and two luxury related constructs: brand consciousness and purchase intention towards luxury goods. Demographic factors such as gender differences are also examined. As more and more luxury brands rely on SM to engage Millennials (New York Times, 2010; Silverstein, 2010), this study examines if SM usage, brand consciousness and intention to purchase luxury products relates to beliefs, attitudes and behaviors towards SMADV. Further, this study explores whether SM usage relates to brand consciousness and purchase intention. Findings of this study will provide theoretical insight by applying traditional advertising model (Pollay and Mittal, 1993) to the new SM context. Managerial implications are provided as well as directions for further research.

2 2.1

Theorectical Framework and Hypotheses Beliefs, Attitudes, and Behaviors toward SMADV

Pollay and Mittal’s (1993) model of beliefs and attitudes about advertising has been widely used to assess consumer evaluation of advertising (Wang, et al., 2009; Wolin and Korgaonkar, 2003; Wolin et al., 2002). Within this model consumer’s attitudes towards advertising is determined by their beliefs toward advertising. For example, Wolin et al. (2002) applied the model to examine consumer evaluations of general advertising on the Web. This study posits that Pollay and Mittal’s model can be applied to the emerging Web for mats, namely SMADV. With respect to the online social sphere, the social nature of SM pro-

An Investigation of Social Media Usage, Brand Consciousness and Purchase Intention

181

vides advertisers with an ideal channel to develop messages that are relevant, personal, and engaging. The later examination has important implications for theory and practice—given the rapid adoption of SM by consumers and marketers. As a fundamental step towards understanding the impact of SMADV, this study first discusses beliefs, attitudes, and behaviors towards advertising in the SM context. Drawing from previous research, this study argued that an individual’s beliefs about SMADV are an important indicator of their overall attitude toward SMADV (Lavidge and Stiener, 1961; Palda, 1966; Ducoffe, 1996). Furthermore, researchers such as Lavidge and Steiner (1961), Wolin and Korgaonkar (2002), Stevenson et al. (2000) and Mehta (2000) noted that individual’s attitude towards advertising is an important antecedent of their behavioral response towards advertising, including their purchase intention. Thus, applying the same logic it is posited that SM users with favorable attitudes towards SMADV will tend to actively engage in advertising messages and seek information about the promoted products or brands.

2.2

Socail Media Usage

To explore the relationships among SM usage, brand consciousness, and purchase intention towards luxury products among the Millennial generation, SM usage was examined. Consumer beliefs towards advertising have often been associated with levels of media exposure. Specifically, previous studies (Churchill and Moschis, 1979; Richins, 1987) have noted that increased media usage is positively associated with high scores in advertising beliefs. Consumer beliefs and attitudes toward advertising involve the cognitive and affective evaluations of promotional messages. On the other hand, the behavioral responses toward advertising consist of online consumer actions in response to advertising stimulus (ad interaction, clicking through an advertisement and e-commerce activities). Millenials invest a significant amount of time on SM (Pew Research, 2010a), results from past research suggested that increased media usage is positively associated with high scores in advertising beliefs (Churchill and Moschis, 1979; Richins, 1987). Thus the time Millennials spend on SM may affect their beliefs, attitudes, and behaviors toward advertising within these online social platforms. Along this line, this study hypothesizes that high levels of SM usage may lead to increased beliefs, attitudes, and behaviors toward SMADV. Thus, the first hypothesis posited: H1: Millennial SM users with a higher level of SM usage will indicate more favorable (a) beliefs, (b) attitudes, and (c) behaviors toward SMADV than will Millennial SM users with a lower level of usage.

182 2.3

Chu and Kamal

Brand Consciousness and Purchase Intention toward Luxury Products

Two important luxury related constructs were also examined, namely: brand consciousness and purchase intention toward luxury products. Brand consciousness, the degree to which a consumer is oriented toward buying well-known branded products (Shim and Gehrt, 1996; Sproles and Kendall, 1986), is an individual personality trait likely to affect consumer shopping orientation and product evaluation (Jamal and Goode, 2001). LaChance et al. (2003) noted that brands play an essential role in the psychological processes that influence buying decisions for brand conscious consumers. Further, past research suggested that brand-conscious individuals perceive well-known brands as a reflection of his/her own personalities (Jamal and Goode, 2001) and use the symbolic meanings of brands as a vehicle for self-expression (Belk, 1988; Sirgy et al., 1997). As brand consciousness indicates how oriented an individual is towards buying well-known branded products (Nelson and McLeod, 2005), it is argued that this construct may impact how users attend to and perceive commercial information in new media vehicles such as SM. Thus, the level of consumer brand consciousness may indicate individual differences in online social sphere and can be viewed as a discriminator of beliefs, attitudes, and behaviors toward SMADV. As brand-conscious consumers tend to look to and appreciate the media for consumption-related information such as the latest fashions, trends and brands (Keum et al., 2004), it is argued that the perception of SMADV for highly brand conscious consumers will be will be favorable as compared to users with low levels of brand-consciousness. Thus, the following hypothesis was formulated: H2: Millennial SM users with a higher level of brand-consciousness will indicate more favorable (a) beliefs, (b) attitudes, and (c) behaviors toward SMADV than will Millennial SM users with a lower level of brandconsciousness. Past studies show that attitudes towards advertising had a positive and significant relationship with purchase intention. A few studies have looked into the relationship between online advertising attitudes and behaviors online. For example, Stevenson et al. (2000) found that attitude towards advertising on the Web and purchase intention was positively related. Similarly, Wolin and Korgaonkar (2002) reported that positive attitudes towards online advertising were significantly and positively related to higher levels of online spending and online purchasing frequency. Thus, the present study predicted that: H3: Millennial SM users with a higher level of purchase intention toward luxury products will indicate more favorable (a) beliefs, (b) attitudes,

An Investigation of Social Media Usage, Brand Consciousness and Purchase Intention

183

and (c) behaviors toward SMADV than will Millennial SM users with a lower level of purchase intention toward luxury products.

2.4

Social Media Usage, Brand Consciousness and Purchase Intention

Research on consumer socialization has noted the pivotal role of media consumption and commercial attitudes and behaviors (cf. Churchill and Moschis, 1979; Nelson and McLeod, 2005; Richins, 1987). Past studies examining the effect of commercial messages on consumption-oriented attitudes of young audiences, particularly teens and adolescents, found that increased media exposure lead to more positive consumption-oriented attitudes and behaviors (Brand and Greenberg, 1994). In two seminal studies Churchill and Moschis (1979) and Richins (1987) found a positive relationship between television consumption and consumption attitudes such as materialism. Similarly, in a later study, Nelson and McLeod (2005) noted that increased media consumption had a positive and significant relationship with consumer brand consciousness. As SM have become a popular destination for luxury campaigns (Balwani, 2009) and is widely used among users within the 18-28 age group (Pew Research, 2010a), there is a need for research examining the relationship between new media channels and brand-related values, particularly brand consciousness and luxury brand consumption among Millennials. Therefore the fourth hypothesis, predicted that: H4: Millennial’s SM usage is positively related to (a) brand consciousness and (b) purchase intention toward luxury products.

2.5

Gender, Beliefs, Attitudes, and Behaviors toward SMADV

Several studies have examined the relationship between demographic characteristics and attitudes toward advertising. Past research (Shavitt et al. 1998; Wolin and Korgaonkar, 2003) suggested that with respect to gender and beliefs, attitudes, and behaviors toward advertising, males tend to have more positive beliefs and attitudes toward advertising than females. Additionally, studies have noted that for Web advertising, male users had a higher level of behavioral response (e.g., Web purchases, click through rates, ad interaction) than compared to their female counterparts (Wolin and Korgaonkar, 2003). Accordingly, the following research question was put forth: RQ: How does gender of Millennial SM users relate to (a) SM usage, (b) beliefs, (c) attitudes, and (d) behaviors toward SMADV?

184 3

Chu and Kamal

Method

This study used an online survey to explore the relationships among the variables. Three hundred and six participants from a major university of the Midwest area in the U.S. completed the survey. The sample consisted of 131 males (42.8%) and 175 females (57.2%). More than two out of three respondents (65%) were between 21 and 30, and 34.3 percent were under 20 year old. As college students represent an important segment of millennial SM users, the use of student sample was deemed appropriate. First, SMADV was defined as “an online ad that incorporates user interactions that the consumer has agreed to display and be shared. The resulting ad displays these interactions along with the user’s persona (picture and/or name) within the ad content” (Interactive Advertising Bureau, p. 4). Next, Pollay and Mittal’s (1993) framework was applied to examine respondents’ beliefs toward SMADV (see Wolin et al. 2002). The measure included twenty-three items that tapped into the seven underlying belief factors: Good for Economy, Falsity/No Sense, Materialism, Value Corruption, Product Information, Social Role and Image, and Hedonic/Pleasure. Attitudes towards SMADV were examined by using the four items (Wolin, et al. 2002) developed by Mittal (1994) and Ducoffe (1996). In terms of behaviors towards SMADV, statements regarding participants’ behavioral interests, responses, and offence level (Wolin, et al. 2002) were used. Thus, four items were used to measure behavioral aspects of SMADV. A 5-point Likert scale (1= strongly disagree, 5= strongly agree) was used to measure beliefs and behaviors towards SMADV respectively. Attitudinal items were gauged on a 5-point semantic-differential scale. To measure brand consciousness, Sproles and Kendall’s (1986) 7-item scale was used. Another 5-item scale was utilized to examine purchase intention toward luxury products (Yoo and Lee, 2009). More specifically, purchase intention was measured by asking respondents how much they would like to purchase a luxury brand in the future. A list of luxury brands including Louis Vuitton, Gucci, Chanel, Rolex, and Hermes was identified through The Leading Luxury Brands 2008 (Interbrand, 2008). Items concerning brand consciousness and purchase intention were measured on a 5-point Likert scale respectively (1= strongly disagree, 5= strongly agree). Finally, SM usage was measured on a 6item scale by asking participants to indicate their media use from the previous day (LaFerle et al., 2000). As a result, a 10-point scale (1= 0, 10= more than 8 hours) was used to examine participants’ SM usage.

An Investigation of Social Media Usage, Brand Consciousness and Purchase Intention

4

185

Results

To examine the impacts of usage pattern on beliefs, attitudes, and behaviors toward SMADV, a median split was used to divide the participants into those with high and low SM usage. A series of independent samples t-tests compared beliefs, attitudes, and behaviors between the two groups of SM users. The results showed that high usage respondents reported having a more positive attitude (Mhigh usage = 3.23; Mlow usage = 2.94, t(1,304) = -3.04, p < .005) and behavior (Mhigh usage = 2.53; Mlow usage = 2.25, t(1,304) = -3.67, p < .001) toward SMADV than did low usage respondents. Thus t-tests results lent support of H1b and H1c. However, no significant difference was found in beliefs toward SMADV (Mhigh = 3.13; Mlow = 3.04, t(1,304) = -1.42, p = .16). Thus, H1a was not supported. When participants spent more time on SM, they remained more favorable attitudes and behaviors toward advertising messages on these social sites. Another median split was used to categorize the respondents into those with high and low levels of brand consciousness and purchase intention toward luxury goods. The results indicated that highly brand-conscious SM users had a more favorable belief (Mhigh = 3.18; Mlow = 2.97, t(1,304) = -3.10, p < .01) and attitude (Mhigh = 3.25; Mlow = 2.87, t(1,304) = -4.03, p < .001) toward SMADV than did low brand-conscious individuals. Thus, H2a and H2b were confirmed. A marginally significant difference was observed in behaviors toward SMADV (Mhigh = 2.44; Mlow = 2.30, t(1,304) = -1.84, p = .07), supporting H2c. In terms of purchase intention toward luxury products, users with a higher level of purchase intention (M = 3.16) were found to have more favorable beliefs toward SMADV than did low purchase intention users (M = 3.01) (t (1,304) = -2.05, p<.05). Thus, H3a was supported. As for behaviors toward SMADV, a marginally significant difference was found (Mhigh = 2.45; Mlow = 2.31, t(1,304) = -1.91, p = .06), and thus H3c was marginally supported. On the other hand, the results showed no significant difference in attitudes toward SMADV (Mhigh = 3.14; Mlow = 3.02, t(1,304) = -1.23, p = .22), disconfirming H3b. Regression analysis was conducted to examine the relationships among SM usage, brand consciousness, and purchase intention. Specifically, brand consciousness and purchase intention were regressed respectively on SM usage. The regression models were found to be significant (Brand consciousness: ( 2 2 Radj = .14), F (1, 304) = 49.74, p < .001; Purchase intention ( Radj = .10), F (1,

304) = 34.75, p < .001). As predicted, usage of SM was a significant predictor of respondents’ brand consciousness and purchase intention toward luxury products. Hence, H4a and H4b were supported.

186

Chu and Kamal

The last research question focused on how gender of SM users affects their SM usage patterns and related beliefs, attitudes, and behaviors toward SMADV. To examine whether there are gender differences with regard to respondents’ usage and evaluation of SMADV, independent samples t-tests were computed. A significant difference was found in SM usage and behaviors toward SMADV. Male respondents (M = 2.62) were more likely to use SM than did their female counterparts (M = 2.20) (t (1,304) = 2.62, p<.01). In addition, male participants (M = 2.47) reported having more positive behaviors toward SMADV than did female respondents (M = 2.31) (t (1,304) = 2.17, p<.05). On the other hand, the results showed no substantial gender differences in terms of beliefs (Mmale = 3.09; Mfemale = 3.07, t(1,304) = .24, p = .81) and attitudes (Mmale = 3.14; Mfemale = 3.03, t(1,304) = 1.10, p = .28). Overall, these results suggested that gender was associated with SM usage and its impact on behavioral responses to SMADV. That is, male participants were more likely to actively engage in SM activities and they tended to maintain more positive behaviors toward SMADV through looking for brand information on SM or clicking social advertisements.

5

Discussion and Conclusions

Millennials represent the future generation of luxury consumers, these users have a penchant for digital content and social interaction—nearly 75 percent of Millennials engage in SM on a daily basis (Pew Research, 2010a). Yet, it can be argued that the main appeal of luxury goods is the sense of exclusivity which is absent in Internet based communication. As luxury brands target this new consumer market to sustain future growth, an important question arises: What are the implications of SMADV for luxury brands with respect to young consumers? The current study explored how SM usage, brand consciousness and purchase intention for luxury products relates to beliefs, attitudes and behaviors towards SMADV among Millennials. Results from this study indicated that users with high levels of SM usage displayed more positive attitudes and behaviors towards SMADV than those with lower levels of SM usage. From a theoretical perspective, these findings are consistent with past research which showed positive linkages between consumer attitudes and behaviors (Ducoffe, 1996; Lavidge and Stiener, 1961; Palda, 1966; Wolin and Korgaonkar, 2002). Further, these results suggest that the consumer’s level of usage may impact the effectiveness of SMADV. One of the possible reasons why heavy users of SM showed more positive SMADV attitudes and behaviors than their peers with lower levels of usage, may be due to the level of familiarity or expertise with these users have with this new medium.

An Investigation of Social Media Usage, Brand Consciousness and Purchase Intention

187

These results imply that heavy users of SM will respond more favorably than those who use the medium less frequently. Thus more traditional avenues of advertising will continue to play an important role in developing favorable attitudes and behaviors for novice and low level SM users. From the managerial perspective a ‘one size fits all’ approach is not likely to work in SMADV-- to be successful, advertisers must identify power users and target them with SMADV messages. Another finding from this study was that users with high levels of brand consciousness reported more positive beliefs and attitudes towards SMADV and are more likely to attend to SMADV messages compared to their peers with lower levels of brand consciousness. This finding is supported by previous research (Keum et al., 2004) that is, people with a keen interest in brands are responsive to gaining more knowledge about those brands and product through alternative media outlets. This means that the effectiveness of SMADV for Millenials is largely driven by user characteristics i.e. interest in brands. Advertisers may be well suited to use SM as a means of providing enhanced information about their products for brand conscious consumers who ‘seek out’ shopping and consumption oriented messages. Conversley, users with low brand conscious levels showed less favorable beliefs and attitudes towards SMADV compared to users with high brand consciousness—this suggests that SMADV is unlikely to have an effect on targets who do not care about brands, these consumers will require attention from other elements of the marketing communication mix such as promotions, public relations and personal selling. Additionally, results showed that users with high levels of purchase intention for luxury goods exhibited more positive beliefs about SMADV than those users with low levels of purchase intention. This finding suggests that for luxury brands, SMADV may serve as a useful tool within the promotional mix. These findings are also explained by past research on selective perception which notes that as consumers are intending or likely to buy a product they are more interested in receiving messages relevant to that purchase (Aaker et al., 1992). Furthermore, results from the regression analysis showed that SM usage was a significant predictor of consumer brand consciousness and purchase intention of luxury goods. Currently many leading luxury brands have allocated a growing proportion of their marketing budgets towards Internet platforms (Chadel, 2009; Okonkwo, 2009). Most notably, luxury brands including Burberry, Gucci, Chanel, Armani, and Van Cleef and Arpel have embraced SM technologies such as iPhone applications, Twitter and Facebook pages as part of their promotional efforts to reel in youthful audiences (New York Times, 2010). Results from this study provide empirical evidence that suggests the importance of SMADV for

188

Chu and Kamal

brand conscious consumers, who are likely targets for luxury brands. This finding is important, given that many luxury brands are expanding their distribution strategies by integrating e-Retail into their websites (Okonkwo, 2010). Additionally, industry reports approximate that 73 percent of luxury brands found that user traffic generated from Facebook to their brand sites had doubled in 2010 (L2 Digital IQ® Index: Luxury, 2010). In light of the positive linkages found in this study between consumer brand consciousness, purchase intention of luxury goods and SM usage, such initiatives are likely to benefit luxury brands in building their market within the Millennial generation. Lastly, while beliefs and attitudes towards SMADV were consistent across male and female respondents, gender differences were noted with respect to behaviors toward SMADV and SM usage. Specifically, male users were more likely to provide a behavioral response to SMADV compared to their female counterparts. Also, male users indicated a higher level of usage over female users. This implies that advertisers should consider male users as a prime target for SMADV. As such, ad-vertisers should consider products, messages and tactics (games, applications, and etc) that are appropriate for this target group. In conclusion, this research contributes to the advertising literature by linking consumption values with the emerging advertising formats, SMADV. More research needs to be conducted to further uncover the impact of SM usage on young users’ beliefs, attitudes, and behaviors toward promotional messages in SM, as these insights will provide valuable implications for advertisers engaging youthful segments with new technologies.

6

References

Aaker, D.A.; Batra, R. & Myers, J.G. (1992), “Advertising Management,” 4th ed. Englewood Cliffs, NJ: Prentice-Hall, Inc. Balwani, S. (2009). Understanding luxury brands and social media. Retrieved Online at: http://mashable.com/2009/07/02/luxury-brands/ Belk, R.W. (1988), “Possession and the Extended Self,” in: Journal of Consumer Research, 15, 139168. Brand, J.E., & Greenberg, B.S. (1994), “Commercials in the Classroom: The Impact of Channel One Advertising,” in: Journal of Advertising Research, 34, 18-28. Chadel, R. (2009). Does the Internet really matter for the luxury goods industry? Retrieved Online at:http://www.chadel.ch/PDFs/Article_Does%20Internet%20really%20matter%20for%20the%2 0luxury%20goods%20industry_EHLITE%2023_final.pdf Churchill, G. A. & Moschis, G.P. (1979), “Television and Interpersonal Influences on Adolescent Consumer Learning,” in: Journal of Consumer Research, 6(1), 23-35. Ducoffe, R.H. (1996), “Advertising Value and Advertising on the Web,” in: Journal of Advertising Research, 36(5), 21-35.

An Investigation of Social Media Usage, Brand Consciousness and Purchase Intention

189

eMarketer (2007). Affluent Internet users: how the rich live online. Retrieved Online at: http://www.emarketer.com/Report.aspx?code=emarketer_2000430 eMarketer (2008). Social network marketing: ad spending update. Retrieved from http://www.emarketer.com Interactive Advertising Bureau (2009). IAB Social Advertising Best Practices, Retrieved Online at: http://www.iab.net/media/file/Social-Advertising-Best-Practices-0509.pdf. Interbrand (2008). The leading luxury brands 2008. Retrieved online at: http://www.interbrand.com/en/news-room/press-releases/2008-12-12.aspx. Jamal, A. & Goode, M. (2001), “Consumers' Product Evaluation: A Study of the Primary Evaluative Criteria in the Precious Jewellery Market in the UK,” in Journal of Consumer Behaviour, 1(2) (November), 140-155. Keum, H., Devanathan, N., Deshpande, S., Nelson, M.R. & Shah, D.V. (2004), “The CitizenConsumer: Media Effects at the Intersection of Consumer and Civic Culture,” in: Political Communication, 21, 369-392. L2ThinkTank.com (2010). L2 Digital IQ® Index: Luxury. Retreieved Online at: http://l2thinktank.com/?page_id=7644 LaChance, J.J., Beaudoin, P. & Robitaille, J. (2003), “Adolescents’ Brand Sensitivity in Apparel: Influence of Three Socialization Agents,” in: International Journal of Consumer Studies, 27, 4757. La Ferle, C.; Edwards, S. M. & W-N Lee (2000), “Teens' Use of Traditional Media and the Internet,” in: Journal of Advertising Research, 4(3), 55-64. Lavidge, R. J. & G.A. Steiner (1961), “A Model for Predictive Measurements of Advertising Effectiveness,” in: Journal of Marketing, 25, 59-62. Lenhart, A., Purcell, K., Smith, A. & Zickuhr, K. (2010). Social media & mobile Internet use among teens and young adults. Retrieved Online at: http://www.pewinternet.org/~/media//Files/Reports/2010/PIP_Social_Media_and_Young_Adults _Report.pdf Mehta, A. (2000), “Advertising Attitudes and Advertising Effectiveness,” in: Journal of Advertising Research, 40(3), 67-72. Mittal, B. (1994), “Public Assessment of TV Advertising: Faint Praise and Harsh Criticism,” in: Journal of Advertising Research, 34(1), 35-53. Nelson, M. & McLeod, L. (2005), “Adolescent Brand Consciousness and Product Placements: Awareness, Liking and Perceived Effects on Self and Others,” in: International Journal of Consumer Studies, 29(6), 515-528. New York Times (2010). Marketing to the Millennials. Retrieved Online at: http://www.nytimes.com/2010/03/03/fashion/03iht-rmil.html?_r=1 Okonkwo, U. (2005). Can the Luxury fashion brand store atmosphere be transferred to the Internet? Retrieved online at: http://www.brandchannel.com/images/papers/269_Lux_Goods_Online.pdf Okonkwo, U. (2009), “Sustaining the Luxury Brand on the Internet,” in: Journal of Brand Management, 16, 5(6), 302-310. Okonkwo, U. (2010). Luxury online: styles, systems, strategies. Publisher: Palgrave Macmilla, Hampshire, United Kingdom. Palda, K.S. (1966), “The Hypothesis of a Hierarchy of Effects: A Partial Evaluation,” in: Journal of Marketing Research, 3, 13-24. Pew Research Center (2010a). The Millennials: Confident. Connected. Open to Change. Retrieved Online at: http://pewresearch.org/pubs/1501/millennials-new-survey-generational-personalityupbeat-open-new-ideas-technology-bound Pew Research Center (2010b). Generations 2010. Retrieved Online at: http://www.pewinternet.org/Reports/2010/Generations-2010.aspx Pollay, R. & Mittal, B. (1993). “Here's the Beef: Factors, Determinants, and Segments in Consumer Criticism of Advertising,” in: Journal of Marketing, 57(3), 99-114. Richins, M.L. (1987), “Media, Materialism, and Human Happiness,” in: Advances in Consumer Research, 14(1), 352-356.

190

Chu and Kamal

Shavitt, S., Lowrey, P.M. & Haefner, J.E. (1998), “Public Attitudes Toward Advertising: More Favorable Than You Might Think,” in: Journal of Advertising Research, 7-22. Shim, S. & Gehrt, K.C. (1996), “Hispanic and Native American Adolescents: An Exploratory Study of Their Approach to Shopping,” in: Journal of Retailing, 72(3), 307-324. Silverstein, B. (2010). Luxury Brands Utilize New Technology To Engage Millennials. Retrieved Online at: http://www.brandchannel.com/home/post/2010/03/03/Luxury-Brands-Utilize-NewTechnology-To-Engage-Millennials.aspx Sirgy, M.J., Grewal, D., Mangleburg, T.F., Park, J.O., Chon, K., Claiborne, C.B., Johar, J. S. & Berkman, H. (1997), “Assessing the Predictive Validity of Two Methods of Measuring SelfImage Congruence,” in: Journal of the Academy of Marketing Science, 25(3), 229-241. Sproles, G.B. & Kendall, E.L. (1986), “A Methodology for Profiling Consumers’ Decision-Making Styles,” in: Journal of Consumer Affairs, 20(2), 267-279. Stevenson, J. S., Bruner II, G.C. & Kumar, A. (2000), “Web Page Background and Viewer Attitudes,” in: Journal of Advertising Research, 20(1 &2), 29-34. Wang, Y., Sun, S., Lei, W. & Toncar, M. (2009), “Examining Beliefs and Attitudes toward Online Advertising among Chinese Consumers,” in: Direct Marketing: An International Journal, 3(1), 52-66. Wolin, L., Korgaonkar, P. & Lund, D. (2002),“Beliefs, Attitudes and Behavior towards Web Advertising,” in: International Journal of Advertising, 21(1), 87-113. Wolin, L.D. & Korgaonkar, P. (2003), “Web Advertising: Gender Differences in Beliefs, Attitudes and Behaviour,” in: Internet Research, 13(5), 375-385. Yoo, B &Lee, S.H. (2009), “Buy Genuine Luxury Fashion Products or Counterfeits?” in: Advances in Consumer Research, 36, 280-286.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour Rafael Currás-Pérez, University of Valencia, Spain María José Miquel-Romero, University of Valencia, Spain Carla Ruiz-Mafé, University of Valencia, Spain Silvia Sanz-Blas, University of Valencia, Spain

1

Introduction

Nowadays, television media has evolved into an important channel of direct selling that brands and advertisers are exploiting. Many technological changes (e.g., digital television appearance or the integration of Internet in the TV medium and mobile technology) have increasingly led to consumers see television as a channel for purchasing products and services and as a source of information (Beyer et al., 2007). However, researchers have put little interest on teleshopping, one of the consumer behaviour phenomena. Traditionally, media exposure, television media dependency or teleshopping genre dependency have been key antecedents of the teleshopping behaviour. Thus, it has been shown that the intensity of the relationship between consumer and media predicts the likelihood of a media message having some sort of effects on the individual’s behaviour (Ballantine and Martin, 2005; Ball-Rokeach et al., 1993; Grant et al., 1991). Media Dependency Theory (Ball-Rokeach, 1985, 1989; Ball-Rokeach et al., 1984; Defleur and Ball-Rokeach, 1989; Loges, 1994) is one of the most useful theories to explain individual-media relations, and has been used as the conceptual framework for our study. The Media Dependency Theory makes possible to identify not only the needs and gratifications derived from using media but also the relationships between the goals pursued by the consumer and the available resources. The micro-level of this theory named as Individual Media Dependency (IMD) provides a concrete basis for measuring individual dependency relations with regard to a specific medium or genre (Grant et al., 1991; Loges, 1994). However, research has also shown that television viewers can develop a sense of social relationship with their favourite television personalities such as newscasters or soap opera characters (Perse and Rubin, 1989; Rubin et al., 1985). Later et al. (1998) also included the viewer-audience relationship in S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_13, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

192

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

genre personae relationships, which is different from the show-host parasocial interaction, on television shopping experience. On the basis of the Media Dependency Theory, Grant et al. (1991) and Skumanich and Kintsfather, (1998) presented and evaluated a teleshopping dependency model in which the ‘media dependency genre’ is a key variable in the purchase, integrating the genre personae relationships to capture the psychosocial dynamics of the dependency phenomenon. Indeed, interactivity comes as one of the deepest and revolutionary changes in the television media. Television is no longer seen as the device that only emits unidirectional messages given that television content managers are concerned with the fact that viewers should take part in the program by actively participating as well as empathizing with characters and hosts (Beyer et al., 2007). The search for a major interaction between individual-personae answers the fact that television consumerism is, every time to a larger extent, a vehicle for construction and expression of the personal identity of the subjects. It is throughout its token of link with programs and television characters that individuals can satisfy their self-definitional needs. This phenomenon encourages television content designers to create contexts and use tools to make consumer social relationship and integration with characters (host, guests) or other viewers that take part in the program easier. Some examples of such contexts are: choosing the host according to the audiences’ favourite profile, direct calls, SMS messaging, development of a social networks profile, starting up chats or web pages, etc. To sum up, both relationships individual-media and individual-personaeother viewers would determine the strong tendency towards teleshopping (Skumanich and Kintsfather 1998; Stephens et al., 1996). This study, looking at models proposed by Grant et al. (1991) and Skumanich and Kintsfather (1998), has the main objective to analyze the role of genre personae relationships (parasocial interaction and teleparticipation relationships) as well as media relationships (exposure and dependency) as antecedents of teleshopping behaviour. In order to achieve the objective this work is organized as follows: First, a conceptual review of both teleshopping key antecedents, the relationship with the people that get involved in the genre (parasocial interaction and teleparticipation) and media-genre dependency are presented. Next, hypothetical relationships between variables and the proposed theoretical model are explained in this research, allowing explaining integrally the phenomenon of teleshopping in terms of relations of the individual with media, the teleshopping genre and television genre personae. That is followed by a description of the methodology

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

193

used and the results. Finally, we present the conclusion, managerial implications and study limitations.

2 2.1

Conceptual Framework: Relations with Television Media Individual Media Dependency

An important area of study within the sphere of communication is to explore the functions of mass media from the consumer perspective, including individual media relations and any derived effects. Previous research focussed on the study of television shopper (Bigné et al., 2006; Grant et al., 1991; Ruiz and Sanz, 2007, 2008; Skumanich and Kintsfather, 1998), have identified the key variables of the shopping decision. These variables are exposure to the media-genre, dependency on the teleshopping media-genre, parasocial interaction and effects of teleparticipation. The Uses and Gratifications Theory (Anderson and Meyer, 1975; Blumler, 1979; Katz et al., 1974; Krotz and Eastman, 1999; Mcquail, 1995; Perse, 1986) have been considered one of the strongest influential theories in the study of mass media. Different to other approaches and theories, it has reached a better understanding of the needs that communication contents meets and of the gratifications provided to mass media audiences. Thus, it allows to examine the functions of media from the consumer’s perspective, trying to answer the question what is media used for (McQuail, 1995) and then, what are the needs and wishes it is able to fulfil (Anderson and Meyer, 1975). The previous theory has been improved and extended with the Media Dependency Theory (Ball-Rokeach, 1985, 1989; Ball-Rokeach and DeFleur, 1976; Ball-Rokeach, et al.,1990; Ball-Rokeach et al., 1984; Colman, 1990; DeFleur and Ball-Rokeach, 1989; Loges, 1994), allowing to identify not only the needs and gratifications derived from the use of media but also the relations between the pursued goals and the resources available. Similarly, unlike the Uses and Gratifications Theory, the Dependency Theory takes into account the context that surrounds communication, examining those factors and variables that can influence the access to media. The Media Dependency Theory studies the relations between the systems and its components (Ball-Rokeach, 1989). Mass media is considered, from this theory’s point of view, both a system and an important part of the actual social fabric, and thus, it can have relations with individuals, groups and organizations, as well as with other social systems (Ball-Rokeach, 1989). Therefore, media is

194

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

made up of three components: audience-media-society (Ball-Rokeach, 1985; Ball-Rokeach and DeFleur, 1976; DeFleur and Ball-Rokeach, 1989). According to the Dependency Theory, individuals, groups, organizations and systems fulfil their personal and collective goals having to depend or to rely on resources of other people, groups, organizations or systems, and vice versa (Ball-Rokeach, 1989). The macro-level of the MSD theory (Media System Dependency) (BallRockeach and DeFleur, 1976; DeFleur and Ball-Rokeach, 1989) has been a useful framework to examine, within a more abstract approach, the interdependencies of media with economic and institutional organizations. The micro-level of this theory known as individual media dependency (IMD) provides a concrete basis for measuring individual dependency relations with regard to a specific medium (Colman, 1990; Loges, 1994). Like other systems, the individual has also its own goals and some of them require access to computer resources being controlled by mass media (BallRokeach, 1985, 1989). IMD has been defined as “a relation where the individual’s capacity to reach his or her goals, depends to a certain extent on the medium’s information resources” (Ball-Rokeach et al., 1984, 3). The IMD has three dimensions or categories (see table 1): understanding, orientation and play (Ball-Rokeach, 1985, 1989; Grant, 1996; Grant et al., 1991; Loges, 1994; Loges and Ball-Rokeach, 1993; Skumanich and Kintsfather, 1998).

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

195

Table 1. Types of dependency relations between individuals and media

PLAY

ORIENTATION

UNDERSTANDING

SELF Relationship that maintains or expands the individual capacity to interpret its own beliefs, behaviour, self-esteem or personality.

Creation of own behavioural decisions out of different guides for specific behaviour provided by media.

SOCIAL Understanding the nature of other individuals, cultures and world-events (from the past, present and future). Including the perception of the role that someone plays in that society. Information regarding types of behaviour appropriate or effective in the personal relationship of the individual with other people.

SIGNIFICANT LITERATURE

Ball-Rokeach, 1985, 1989; Ball-Rokeach et al.,1984; Ball-Rokeach et al.,1993; Bigné et al., 2006; Colman, 1990; Defleur and Ball-Rokeach, 1989; Grant, 1989, 1996; Grant et al.,, 1991; Loges, 1994; Loges and Ball-Rokeach, 1993; Skumanich and Kintsfather, 1998

Capacity of the medium to provide conEntertainment, tents that stimulate stimulation or relaxation that the content provides intervention and social relationship among to the individual. people.

Source: Adapted from Ball-Rokeach (1989)

These dimensions have been identified in different media such as radio, press o magazines, television and Internet. Studies focused on the analysis of the individual dependency on television (Ball-Rokeach et al., 1984; Ball-Rokeach et al., 1993; Grant, 1989, 1996; Grant et al.,1991; Loges, 1994; Skumanich and Kintsfather, 1998), have allowed to identify the six levels of dependency previously described and thus, to show that television provides with enough news resources as to reach any of the set goals for understanding, orientation and play. This classification also allows measuring the dependency relationship within a specific product of a medium such as a genre of television programs, a specific television program or a film. Regarding the teleshopping genre, studies of dependency on teleshopping genre or on a specific shopping program have proved that the individual can reach certain goals such as self or social entertainment,

196

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

action or interaction orientation or social understanding (Bigné et al., 2006; Grant et al., 1991; Guthrie, et al., 1988; Skumanich and Kintsfather, 1998). Literature review has allowed identifying as key precursors of the television media-genre dependency the following: social systems, social environment (development), broadcasted contents, media exposure, interpersonal networks and social demographic variables (Bigné et al., 2006; Ball-Rokeach, 1985; Rosenstein and Grant, 1997; Ruiz and Sanz, 2008). Apart from that, as a result of the dependency it is to mention the intensification of the level of media and genre exposure, the identification and achievement of new goals or individual's needs (Bigné et al., 2006; Rosenstein and Grant, 1997), as well as parasocial interaction with television program characters or hosts (Auter and Moore, 1993; Grant et al,, 1991; Horton and Wohl, 1956; Levy, 1979; Nordlund, 1978; Perse, 1990; Perse and Rubin, 1989; Rubin et al,, 1985; Skumanich and Kintsfather, 1998).

2.2

Parasocial Interaction

The concept of parasocial interaction (PSI), a new form of social interaction created by the mass media, made its first appearance in the communication literature with Horton and Wohl’s (1956) study. These authors defined parasocial interaction as the apparent face-to-face interaction that can occur between media characters and their audience (Horton and Wohl 1956). PSI is one-sided communication and interaction between media users and media personae during media exposure (Schiappa et al., 2005; Schramm and Wirth, 2010). While audience members consist of mass media users, media characters can include several types of media figures (or personae) such as hosts, actors, or celebrities. Research on parasocial interaction has been conducted in contexts such as interactive reality television programs (Godlewski, 2005; Hartmann et al., 2008; Schiappa et al., 2006; Schramm and Hartmann, 2008; Schramm and Wirth, 2010), cable home shopping channel (Stephens et al., 1996) and Internet (Ballantine and Martin, 2005; Sanderson, 2008a; 2008b; Sanderson, 2009). PSI has been referred to as an illusion of a face-to-face relationship (Horton and Wohl 1956), as an imaginary social relationship (Alperstein 1991; Cohen, 2004), as an imaginary friendship (Perse 1990), and as an interpersonal interaction between the media user and the media being consumed (Auter, 1992; Giles, 2002; Rosengren, 1974; Rubin et al., 1985; Rubin and Step, 2000; Sanderson, 2009; Schramm and Wirth, 2010). McQuail et al. (1972) identified two essential functions of parasocial interaction: companionship and personal identity. Soap characters frequently reminded viewers of people they knew, and viewers used characters’ situations and behaviour as ways of understanding their own lives.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

197

Overall, parasocial relationships are similar to social relationships, and for many television viewers, relationships with television characters and other celebrities are a constant, large, and in many instances important part of their social world (Skumanich and Kintsfather 1998; Schramm and Wirth, 2010). Perse and Rubin (1989) state that parasocial interactions relationships resemble interpersonal friendships in three ways. First, parasocial relationships (like friendships) are voluntary, and often contain a personal focus. Second, both types of relationships provide companionship, and third, social attraction is a precursor to either type of relationship. Such parasocial relationships can be affected when another person reveals information about the media host and when they make comments about the media figure during viewing (Sanderson, 2009). In most studies, parasocial relations are regarded as the individual’s feelings of friendship and familiarity with other persons outside his/her immediate environment which can lead to regular exposure and make viewing behaviour habitual and routine (Grant et al., 1991; Horton and Wohl, 1956; Perse and Rubin, 1989, 1990; Rosenstein and Grant, 1997; Rubin et al., 1985; Stephens et al., 1996; Schramm and Wirth, 2010). When audience members possess strong feelings toward the character, host and celebrities, it seems plausible that they would turn to communication channels that would allow them to actively convey these sentiments to the media figure. Accordingly, it is not surprising that media users are increasingly turning to computer-mediated communicative channels that enable them to create parasocial interaction experiences (Sanderson, 2008b). By intensifying parasocial relations, an individual has the opportunity to satisfy certain personal definition, integration and group identification needs (McQuail et al., 1972). By identifying with a certain host or with a group of viewers with certain characteristics, teleshoppers can build self-concept by expressing something which they are or aspire to be (Mittal, 2006). To this end, they develop high degrees of commitment to the program which should lead to a greater need for exposure to it and greater purchase intentions (Grant et al., 1991; Skumanich and Kintsfather, 1998; Sanderson, 2009). More recently, parasocial interaction has been restricted to viewers’ relationships with program hosts (Skumanich and Kintsfather, 1998), thus distinguishing it from other types of teleparticipation relationships. Teleparticipation refers to the relationship between viewers and other members of the audience who intervene in the program, either in the studio, through chat rooms, telephone calls or SMS messaging from home. Viewers, by being able to share the comments and suggestions with other members of the audience, identify with them and may also experience feelings of liking or even friendship towards them.

198 3

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

Tested Hypothesis and Proposal for an Integrating Model

People admitting television as a useful medium that fulfils certain goals and aims will certainly develop a dependency with a specific genre of that medium (Ball-Rokeach, 1985). Thus, an individual should have a preexistent dependency with a medium as to be able to develop dependency with a certain genre (Ball-Rokeach, 1985; Defleur and Ball-Rokeach, 1989). Taking into account the aforesaid and the literature review of the conceptual framework, the first hypothesis of this research suggests that, first of all, individuals develop dependency relationships with a certain medium (herein the television medium), and subsequently, they are able to develop a dependency with a genre of the same medium (herein the teleshopping genre). H1: The greater the television dependency, the greater the dependency with the teleshopping genre. Dependency activation can be produced by the longest teleshopping exposure after perceiving that some needs and goals can be fulfilled through media content. In such cases, a dependency with the media genre is created and the content is considered of great importance for the fulfilment of those goals (BallRokeach, 1985; Ball-Rokeach et al.,1984; Defleur and Ball-Rokeach, 1989; Grant et al., 1991; Skumanich and Kintsfather, 1998; Ruiz and Sanz, 2008). Therefore, it is understandable that people who see television more frequently have a bigger possibility of watching adverts and/or shopping programs, and, therefore, they could develop a dependency relation with the teleshopping genre. H2: The greater the television exposure, the greater the dependency with the teleshopping genre. Literature points out that the greater the dependency with a genre or a program, the greater the level of attention during content exposure (Ball-Rokeach, 1989; Ball-Rokeach et al., 1984; Colman, 1990). Therefore, and in accordance with the previous hypothesis, parasocial links will come out only when an individual will have a greater or weaker dependency relationship with a certain genre (Grant et al., 1991). Teleshopping research has found that teleshopping program dependency can lead viewers to experience positive feelings towards the host. Thus, the viewer can develop feelings of “friendship” towards the host which may lead to an increased exposure/viewing of these programs to maintain that “friendship”

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

199

(Giles, 2002; Grant et al., 1991; Skumanich and Kinsfather, 1998). Consequently, teleshopping programs can be an ideal medium in the search for parasocial relations as these programs have little content for retaining viewers interest and it is precisely this parasocial relation which can achieve greater attention to the program (Grant et al., 1991; Skumanich and Kinsfather, 1998). Therefore: H3: The greater the genre dependency, the greater the parasocial interaction effects. Activation of dependency to intensify not only the parasocial relationship with the host, may also encourage teleparticipation (Giles, 2002; Skumanich and Kintsfather, 1998). This adds, therefore, the effects of teleparticipation, including the viewer’s relationship with other members of the audience which differs from the viewer-host relationship (Giles, 2002; Skumanich and Kintsfather, 1998), although both variables can lead to higher levels of exposure and purchase (Skumanich and Kintsfather, 1998). H4: Increased teleshopping genre dependency, increased teleparticipation effects. Moreover, previous research show that teleparticipation increases parasocial interaction. Therefore, a greater teleparticipation, that is, the implication of viewers with members of the program – comments from people of the public or of the streets, guests of the program, other people’s telephone calls- is said to increase the teleparticiparion interaction and, indirectly, the motivation of the teleshopping genre exposure. If the viewer identifies with the person speaking/calling the program and the host-speaker/caller interaction reinforces the viewer’s attitude towards the host, this will intensify the effects of parasocial interaction (Giles, 2002; Skumanich and Kinsfather, 1998). Thus: H5: Teleparticipation positively influences parasocial interaction effects. Parasocial interaction with a television character will increase the exposure to that genre or program (Ball-Rokeach, 1989; Ball-Rokeach, et al., 1984; Colman, 1990; Grant et al., 1991; Skumanich and Kintsfather, 1998). Previous researches have shown that parasocial interaction is a key element for intention, attention, exposure, viewing continuity, and media empathy (Auter and Moore, 1993; Levy, 1979; Palmgreen et al., 1980a, 1980b; Perse, 1990; Rosengren, 1974; Rosengren and Windahl, 1972; Rubin and McHugh, 1987; Rubin and Perse, 1987; Rubin, et al., 1985).

200

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

Given that the development of these parasocial relationships is not immediate, it is only possible to talk about a direct relationship between dependency and exposure to the genre when the viewer initially experiences low levels of parasocial interaction as a starting point for the development of a continuous relationship with the hosts of that television genre (Skumanich and Kintsfather, 1998). H6: The greater the parasocial interaction with teleshopping programs, the greater the levels of exposure to those programs. H7: The greater the dependency to the teleshopping genre, the greater the exposure to teleshopping programs. Exposure levels to the teleshopping genre can be influenced not only by the relations that an individual develop with the genre or television characters but also with the time expended watching television (Grant et al. ,1991; Skumanich and Kintsfather, 1998). Apart from that, there are also evidences stating that the use of television as a pastime or when there is nothing better to do, leads to important and considerable levels of viewing that in most cases do not address specific and favourite programs, therefore, being a no selective viewing (Abelman et al.,1997; Gerbner, 1982). In such cases, viewers can initially choose the exposure to the genre, for the purpose of being in company and/or as ground music, its attention being little focussed on the program, the dependency not being activated, and so specifying a direct relationship from television exposure to genre exposure (Skumanich and Kintsfather, 1998). Consequently: H8: The time used to watching television is a positive indicator of the exposure to the teleshopping genre. On the one hand, there are empirical evidences stating that the greater the dependency relationship with a media-genre the greater the probability of buying watched contents (Ball-Rokeach, 1985; Defleur and Ball-Rokeach, 1989). As for teleshopping, it is understandable that the buying behaviour is linked to the exposure and that the most frequent viewers have more possibilities to buying than the less frequent viewers (Bigné et al., 2006; Skumanich and Kintsfather, 1998). H9: The greater the teleshopping genre exposure, the greater the product purchase through these programs.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

201

It is when someone realizes that a medium helps achieving or fulfilling its own personal goals that its dependency to that medium increases (Grant et al., 1991). If someone’s goals are fulfilled as a result of the exposure to the content of a genre, the perceived value of the content increases and as a result there is a greater implication and exposure to that genre (Ball-Rokeach, 1985; Defleur and Ball-Rokeach, 1989). Many researches (Grant et al., 1991; Skumanich and Kintsfather, 1998) have emphasized that the product purchase through the television medium could be considered by someone as a goal (action orientation). Someone that has bought a product through television as to fulfil a goal will probably buy products again to fulfil other goals. There are evidences that show the existence of impulsive shoppers (Donthu and Gilliland, 1996) and even compulsive shoppers categorized by the daily watching and the high frequency of shopping (Faber, 1992; Skumanich and Kintsfather, 1998; Stephens et al., 1996). The analysis of literature suggests a possible direct relation between the buying behaviour and the teleshopping dependency, being showed in the following hypothesis: H10: Buying products through advertisement and/or teleshopping programs leads to increasing the dependency relationship with this genre of programs.

4 4.1

Methodology Data Collection and Sampling Procedure

To obtain the information necessary for the research, an initial questionnaire was produced as the result of the literature review and was pre-tested before settling on the definitive form. This initial questionnaire was reviewed by a group of 39 teleshoppers and non-teleshoppers to avoid possible interpretation errors. The target population were teleshoppers resident in Spain, 18 years and over who had purchased a product from telesales advertisements or programs over the last year, with a sample of 432 teleshoppers. Sample selection procedure was by gender and age quotas according to 2008 data on teleshopping genre audiences provided by Sofres A.M1.

1 According to Sofres A.M., 48.2% of people with teleshopping genre exposure are men and 51.8% women. With regard to age, 15.2% are 24 and under, 23.4% are between 25 and 34 28.6% are between 35 and 44, 17% are between 45 and 54, 10.2% are between 55 and 64 and, finally, 5.7% are 65 and over.

202

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

In relation to the viewing behaviour, almost 85% of all interviewees watch television for a maximum of 4 hours a day, although virtually half (40.1%) watch for less than 3 hours and only 15.3% watch more than 4 hours a day. Of all the time individuals spend watching television during the week, most of the sampled spend between 1 and 3 hours watching teleshopping advertisements and/or programs (76.9%). Most of those interviewed watch teleshopping advertisements and/or programs during the week and at the weekend (42%) and especially, at night (60.2%). Early morning and afternoon are the most popular times of the day for watching this television genre (28.7% and 22.2% respectively).

4.2

Measurement of the Variables

Television dependency, teleshopping genre dependency, teleparticipation and parasocial interaction variables were measured with the most commonly-used scales in the literature. In all cases, multi-item 5 point Likert scales were used. Television dependency was measured on an 18 item scale (with three representative terms for each of the six dependency dimensions), in a 7-point Likert format, scoring 1 for ‘totally disagree’ and 7 for ‘totally agree’. This scale was the result of the works of several researchers (Ball-Rokeach et al., 1993; Loges, 1994; Grant, 1996) who tried to create a standard measure of dependency that could be applied not only to television but also to the other media.Teleshopping genre dependency was measured using Grant et al. (1991) scale, which in turn was adapted on the basis of previous media dependency studies (Ball-Rokeach et al., 1984; Rubin, 1984). This is a 12-item scale, which has been subsequently used as the basis for further research in this field, such as the studies by Grant (1996) and by Skumanich and Kintsfather (1993, 1998). The effects of teleparticipation were measured using a six item scale developed by Skumanich and Kintsfather (1993). This scale was also used in a more recent study on teleshopping programs by the same authors (1998). Finally, parasocial interaction was analyzed with the 10-item scale proposed by Rubin and Perse (1987), with sufficient internal consistency. Two new items extracted from the study by Skumanich and Kintsfather (1998) were added to this measurement and attempt to measure interviewees’ feelings towards the program host. The literature is unanimous over the measurement of the view behaviour (Grant 1996; Grant et al., 1991; Perse 1986; Rubin 1981, 1983, 1984; Rubin et al., 1985; Skumanich and Kintsfather, 1998), using questions aimed at finding out the time spent viewing teleshopping genre. “How many hours, on average, do you watch teleshopping advertisements and/or programs?”

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

203

Finally, shopping behavior was determined by asking interviewees about the frequency of buying products offered in teleshopping advertisements and/or programs, a question which again, is one of the most commonly used in these studies in the literature (Grant et al., 1991; Skunmanich and Kintsfather, 1998).

4.3

Assessment of Measurement Scale Psychometric Properties

The measurement instrument was validated in two stages following Ulaga and Eggert (2005) recommendations. Confirmatory factor analysis with EQS 6.1 (Bentler, 1995) was done using maximum robust likelihood (Satorra and Bentler, 1988) on the indicators of the three-dimensions of television media dependency and teleshopping genre dependency (understanding, orientation and play), and after analyzing reliability and convergent and discriminant validity these dimensions became clear variables of the respective second-order construct through a summative scale of their initial indicators. Thus, media dependency and genre dependency were included in the structural model as a first order construct. In order to ensure the convergent validity of the television media dependency scale, those items with non-significant or fewer than .60 factor loads were removed. (Bagozzi and Baumgartner, 1994; Bagozzi and Yi, 1988). Particularly, items DTV4, DTV5 and DTV6 were removed. After removing the aforesaid items, all factor regression coefficients were significant (t>2.56), standardized loads higher than 0.6, besides, achieving a good fit model. These results confirmed the existence of convergent validity in the television media dependency scale. Convergent validity of the genre dependency scale was guaranteed when all the items were found to have significant factor loads over .60 (Bagozzi and Baumgartner, 1994; Bagozzi and Yi, 1988) and the Lagrange multipliers test did not suggest significant relations for a factor other than those they were indicators for (Hatcher, 1994). The measurement model also showed good fit. In terms of reliability, both media dependency and genre dependency scales, all Cronbach’s D (Cronbach, 1951) had to be above the recommended value of .70 (Nunnally and Bernstein, 1994). The composed reliability index was also calculated and for all values was over the recommended value of .70 (Fornell and Larcker, 1981). Average variance extracted (AVE) values were over .50 (Fornell and Larcker, 1981). Discriminant validity was evaluated following two criteria. Firstly, it was examined that none of the 95% confidence intervals for correlation estimations

204

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

between each pair of factors contained the value 1 (Anderson and Gerbing, 1988). Secondly, the variance shared between each pair of constructs (squared correlation), was below the corresponding variance extracted indexes (Fornell and Larcker, 1981). After verifying the psychometric properties of the transformation of the television dependency and genre dependency construct into first order factors, the measurement model associated to the structural model was assessed to ensure it also had the appropriate properties. The criteria for analyzing measurement model reliability and validity were the same as those applied in the previous validation process. The results showed no problems of convergent validity, as fit was reasonable (NFI=.905; NNFI=.947; CFI=.955; IFI=.955; RMSEA=.062) and all factor loads were significant and over 0.6 (Bagozzi and Baumgartner, 1994; Bagozzi and Yi, 1988). Nor were there any problems of reliability, in terms of Cronbach’s D, composed reliability (always above the cut off point of .70) and the average variance extracted (always above the cut-off point of .50). There were no problems of discriminant validity in the measurement model, in accordance with either of the two criteria used2.

5

Results

Once psychometric properties of the measurement tool are evaluated, an estimation of the structural model which summarizes the formulated hypothesis was carried out. It was estimated by the method of greatest robust credibility, using again EQS 6.1, due to the assurance of estimation consistency of this method even though some of the dependency variables are not common. Estimation of the first model showed good indicators of goodness of fit S-B χ2 =337.511; df=201; p<0.00; BBNFI=0.90; BBNNFI=0.94; CFI=0.95; IFI=0.95; RMSEA= 0.056). Table 2 shows the information of the estimated parameters that leads to acceptance and/or refusal of the formulated hypothesis.

2

Contact the authors for further details on validation process

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

205

Table 2. Tested hypothesis global sample β (t) Contrast Relation H1 Media dependency → Genre dependency .936 (8.805**) Accepted H2 Media exposure → Genre dependency .111 (2.248*) Accepted H3 Genre dependency → Parasocial Interaction .095 (2.040*) Accepted H4 Genre dependency → Teleparticipation .746 (8.475**) Accepted H5 Teleparticipation → Parasocial Interaction .873 (7.068**) Accepted H6 Parasocial Interaction → Genre exposure .135 (2.322*) Accepted H7 Genre dependency → Genre exposure .127 (2.292*) Accepted H8 Media exposure → Genre exposure .328 (4.943**) Accepted H9 Genre exposure → Purchase .294 (4.865**) Accepted H10 Purchase → Genre dependency .099 (2.059*) Accepted S-B χ2 (df = 201) =337.511 (p<0.00); NFI=.90; NNFI=.94; CFI=.95; IFI=.95; RMSEA=.056 ** p≤0.01; *p≤0.05

From table 2 we find that television media dependency is direct and positively related to teleshopping genre dependency (H1). This confirms media dependency as the main precursor of the genre dependency, and we support observations carried out by Ball-Rokeach (1985) and Defleur and Ball-Rokeach (1989). According to them, an individual should have a preexistent dependency with a medium as to develop a dependency with a genre. In addition, our findings suggest that there is a significant relationship between television exposure level and genre dependency (H2). That relationship is quite weak if we compare it with the one between medium dependency and shopping genre dependency. The greater exposure to the television medium the greater teleshopping genre dependency. The shopping genre dependency has a significant and positive influence either on parasocial interaction (H3 accepted, as its influence is very weak), or on audience’s teleparticipation (H4), which, at the same time, positively influences the parasocial interaction (H5). These findings suggest this common relationship of “distant intimacy” among mass media can emerge from viewers who watch selling programs/advertisement, and it is directly favoured by the dependency with the genre and teleparticipation, a variable that it is also influenced by the viewers’ shopping genre dependency. Moreover, according to data, teleshopping genre exposure is the result of the three variables initially considered: parasocial interaction (H6), with a low influence, shopping genre dependency (H7), also with a low influence, and the television media exposure level (H8), with a higher influence. These findings support data we looked up used in previous research (eg. Ball-Rokeach, 1985, 1989; Ball-Rokeach et al. 1984; Defleur and Ball-Rokeach, 1989; Grant et al., 1991; Skumanich and Kintsfather, 1993, 1998) according to which the development of a relationship both with a television genre and such

206

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

genre personae, is shown by the further exposure to that genre, in our case, programs and teleshopping advertisement. As to the relationship television exposure-genre exposure (H8) the time spent watching television has the same significant effect as the time spent watching advertisement and/or selling programs, for the reason that the more television exposure the more probability to find teleshopping genre. Besides, the greater the exposure to that genre, the greater the probability of the viewer buying (H9). As for the teleshopping, the shopping behaviour is expected to be linked with the exposure, and then more frequent viewers have bigger opportunities to purchase (Skumanich and Kintsfather, 1998), content they have watched (Ball-Rokeach, 1985; Defleur and BallRokeach, 1989). Focussing now on the causal relationship between the shopping and the genre dependency (H10), it is evident that buying more in teleshopping programs or advertisements will slightly influence although significantly on a bigger shopping genre dependency; in this respect, several are the authors who have proved in their researches the significance of that relationship (Grant et al., 1991; Skumanich and Kintsfather, 1998): if the shopping product allows to satisfy the viewers’ needs or goals, the content perceived value is boosted resulting in a bigger genre implication and dependency (Ball-Rokeach, 1985; Defleur and Ball-Rokeach, 1989). All in all, we can state that in the case being studied, the viewer dependency on the teleshopping genre is a key and essential variable for the understanding of that viewer buying behaviour. In this respect, of the ten hypotheses raised, all of them accepted, six directly link the variable shopping genre dependency with other variables linked to the teleshopping behaviour. This shows the need to benefit and increase such genre dependency if increasing the shopping through this way is searched for. Figure 1 shows graphically the results.

6

Conclusions and Implications

Many countries have experienced in the last years a great growing in the television supply, and consequently, a growing in some television genre that, until short time ago, had great existence. As for Spain, the new channel supply has brought more programs and formats, being one of them telesales, either through certain sales programs or adverts. It would be then important to know and analyze what determines the viewers’ purchase decision, as this could be a new commercial alternative for businesses.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

207

Results of this study show that buying through telesales programs and advertisements is favoured by the exposure to these sales programs: there is evidence that the greater the exposure, the greater the probability of purchase based on the content observed (Ball-Rokeach, 1985). In this respect, exposure to these programs is directly determined by the own television media exposure, by the shopping genre dependency level and by the parasocial interaction the program arouses. Thus, the greater the exposure to television media, the greater the probability of the individual being exposed to programs or sales advertisements, and also, a greater shopping genre dependency will led the individual to look for such programs exposure. If he/she is exposed, feelings of friendship and familiarity with different people who are also linked to the sales program or advertisement (parasocial interaction) can take him/her to a regular exposure and get the person to a regular and monotous watch (Grant et al., 1991; Horton and Wohl, 1956; Perse and Rubin, 1989; Perse and Rubin, 1990; Rosenstein and Grant, 1997; Rubin et al, 1985; Stephens et al., 1996).

Media dependency

.10*

Teleparticipation

.94** .75**

.87**

Genre dependency .10* .11*

Media exposure

Genre exposure

.13*

.29**

Purchase

.14* Parasocial interaction

.33**

Figure 1. Estimated model

Moreover, exposure to sales programs is also indirectly influenced by the viewer dependency on television media, and by teleparticipation in itself. On this note, the greater television media dependency, the greater the genre at issue dependency, a variable that directly influenced the exposure level of such programs. Besides, teleshopping genre dependency has an indirect influence, through genre personae relationships on teleshopping exposure, which in turn has a direct, positive influence on teleshopping behavior. In line with previous

208

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

studies (Ball-Rokeach, 1985, 1989; Ball-Rokeach et al., 1984; Grant et al., 1991; Skumanich and Kintsfather, 1998), this study confirms that the development of a television genre relationship manifests itself by greater exposure to the genre. Thus, parasocial interaction and teleparticipation have been found to be relevant in the research context, and are key mediator variables between dependency and exposure, intensifying genre exposure levels. Viewer’s dependency on the shopping genre positive and directly influences such sales programs exposure, but also indirectly, it has a great influence on teleparticipation and to a lesser extent, on parasocial interaction. It is precisely teleparticipation what would also positively and to a large extent determine parasocial interaction, ultimately influencing the exposure level to sales programs. The results suggest that genre personae relationships are not exclusively limited to the host but to the whole atmosphere surrounding the program, including audience interventions which help to generate those relationships. Lastly, it is stated that buying through sales programs develops a greater teleshopping genre dependency, despite having a quite weak influence. According to literature and as we have stated, individuals focussing on action, that is, buying through television as to fulfil a given goal, it is probable that they will buy again as to fulfil different goals (Grant et al., 1991; Skumanich and Kintsfather, 1998), thus, generating a bigger shopping genre dependency. A series of management implications emerge from the above results which may help companies, advertising, sales professionals and mass media, to obtain better results. Firstly, we believe telesales programs and/or advertisements can be boosted through other channels that broadcast and promote these programs. Telesales programs do not have the exclusivity but the general scheduling is used. On the one hand, this is how television media dependency is boosted. Should the viewer know about the television offer, it can sound interesting or atractive whether the need is that of personal or social understanding, orientation and/or play. This will boost a greater media dependency, a variable of influence on such genre dependency level. On the other hand, knowing about the television scheduling, should it be of interest among the audience, will also boost a greater television exposure, what could boost a greater shopping genre dependency and then, a greater probability of exposure to such telesales programs or advertisements. Therefore, as far as such telesales program or advertisement is concern, managers should design attractive, interesting sales programs to retain the audience’s attention. It is important, for instance, to persuade viewers to phone in and make the purchase, that products are presented as being the best in their category, emphasizing their unique characteristics, the opportunity to participate

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

209

in a competition or game (only if the product is acquired) or interesting discounts on product prices. In addition, television channels must allow viewers to communicate with the hosts and so they should include spaces in the programs for the audience to call in and ask questions or have their doubts resolved. We also suggest to allow SMS messaging to the programs and for the host to comment on some of them. Programs should also be supplemented with a web site where the audience can intervene at any time, even when the program is being broadcast live with their comments, doubts, questions, etc. All these different ways of participating and involving with the program should be easy for the viewer, and easily communicated. Thus, with such actions, teleparticipation is boosted and consequently parasocial interaction, both variables of considerable influence on teleshopping behaviour (Grant et al., 1991; Skumanich and Kinstfather, 1998), given that such program exposure is boosted. Interpretation of the results of this analysis is subject to certain limits due to the fact that not all consumers who buy different types of products and/or services are represented. Furthermore, part of the sample chosen may be viewers of a specific sales program and so the results are not necessarily generalizable to all viewers of teleshopping advertisements and/or programs. Future research, focussed on the study of media relationships, could be aimed at analysing the concept of dependency, either in a specific program within the teleshopping genre (for instance A3D) or applying other genre or programs of the media (such as news or entertainment programs). Moreover, the study of the dependency could be extended to other mass media. However, the study could be repeated with another sample of viewers that normally watch advertisement and/or sales programs but having never bought a product. We want to know whether it’s possible relationships with the media and/or genre differ from the one of the viewers who have bought products through television. The study focussed on the no-purchaser would not only have to aim at knowing whether it is dependent on medium or genre but also at knowing their opinions about the system, and therefore, finding out all the deciding aspects that will show the purchase option of the viewer.

7

References

Abelman, R.; Atkin, D. & Rand, M. (1997), “What viewers watch when they watch TV: Affiliation change as case study”, in: Journal of Broadcasting & Electronic Media, Vol. 41, No. 3, 360-379. Alperstein, N.M. (1991), “Imaginary social relationship with celebrities appearing in television commercials”, in: Journal of Broadcasting & Electronic Media, Vol. 35, No. 1, 43-58.

210

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

Anderson, J.A. & Meyer, T.P. (1975), “Functionalism and the mass media”, in: Journal of Broadcasting, Vol. 19 No. 1, 11-22. Anderson, J.C. & Gerbing, D.W. (1988), “Structural equation modelling in practice: A review and recommended two-step approach,” in: Psychological Bulletin, 103, 411-423. Auter, P.J. (1992), “TV that talks back: An experimental validation of a parasocial interaction scale”, in: Journal of Broadcasting & Electronic Media, Vol. 36, No. 2, 173-181. Auter, P.J. & Moore, R.L. (1993), “Buying from a friend: A content analysis of two teleshopping programs”, in: Journalism Quarterly, Vol. 70, No. 2, 425-436. Bagozzi, R.P. & Baumgartner, H. (1994), “The evaluation of structural equation models and hypothesis testing,” in: Bagozzi, R. (eds.): Principles of Marketing Research, Blackwell Publishers, Cambridge, MA, 386-422. Bagozzi, R.P. & Yi, Y. (1988), “On the evaluation of structural equation models,” in: Journal of the Academy of Marketing Science, Vol. 16, No. 1, 74-94. Ballantine, P., & Martin, B. (2005), “Forming parasocial relationships in online communities”, in: Advances in Consumer Research, Vol. 32, 197-201. Ball-Rokeach, S.J. (1985), “The origins of Individual Media-System Dependency: A sociological framework,” in: Communication Research, Vol. 12, No. 4, 485-510. Ball-Rokeach, S.J. (1989), “Media System Dependency Theory,” in: Defleur, M.L.; Ball-Rokeach, S. (eds.): Theories of Mass Communication, New York, Longman, 297-327. Ball-Rokeach, S.J. and Defleur, M.L. (1976), “A dependency model of mass-media effect”, in: Communication Research, Vol. 3, No. 1, 3-21. Ball-Rokeach, S.J.; Grant, A. & Horvath, A. (1993). On measuring individual´s relations with the media. Unpublished paper, Annenberg School for Communication, University of Southern California, Los Angeles. Cited by Loges (1994). Ball-Rokeach, S.J.; Rokeach, M. & Grube, J.B. (1984), “The great American values test: Influencing behaviour and belief through television,” Free Press. Bentler, P.M. (1995), “EQS structural equations program manual,” Multivariate Software Inc, Encino, CA. Beyer, Y., Enli, G.S., Maasø, A.J., & Ytreberg, M.E (2007), “Small talk makes a big difference. Recent developments in interactive, SMS-based television”, in: Television and New Media, Vol. 8, No.3, 213-234. Bigné, E.; Sanz, S., & Torán, F. (2006), “Dependency in consumer media relations. An application to the case of teleshopping”, in: Journal of Consumer Behaviour, No.5, 397-410. Cohen, J. (2004), “Parasocial break-up from favourite television characters: The role of attachment styles and relationship intensity”, in: Journal of Social and Personal Relationships, Vol. 21, 187202. Colman, W. (1990), “Health moves to prime time: Evaluating the impact of a prime-time television movie of the week on viewer´s content relevant health beliefs,” Unpublished Doctoral Dissertation, University of California. Cited by Grant et al., (1991) and Skumanich and Kintsfather (1998). Cronbach, L.J. (1951), “Coefficient alpha and the internal structure of tests,” in: Psychometrika, 16, 297-334. Defleur, M.L. & Ball-Rokeach, S.J. (1989), “Theories of mass communication,” Logman. Donthu, N. & D. Gilliland (1996), “Observations: The infomercial shopper”, in: Journal of Advertising Research, Vol. 36, No. 2, 69-76. Faber, R.J. (1992): “Money change everything: Compulsive buying from a biopsychosocial perspective”, in: American Behavioral Scientist, Vol. 35, No. 6, 809-819. Fornell, C. & Larcker, D.F. (1981), “Evaluating structural equations models with unobservable variables and measurement error,” in: Journal of Marketing Research, 18, 39-50.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

211

Gerbner, G. (1982), “Introductory comments”, in: Pearl D.; Bouthilet, L.; Lazar, J. (eds.): Television and Behaviour: Ten Years of Scientific progress and Implications for the Eighties, DHHS Publication No. ADM 82-1196, Vol.2, U.S. Government Printing Office, Washington, DC, 332-333. Giles, D. (2002), “Parasocial interaction: A review of the literature and a model for future research”, in: Media Psychology, Vol. 4, 279-305. Godlewski, L. (2005), “Interactive reality television: An analysis of variables influencing postexposure activity”, Unpublished Doctoral Thesis, Department of Communication, University of Delaware, USA. Grant, A.E. (1989), “Seven audiences: An exploration of Television viewing behaviour”, Paper presented at the Annual Meeting of the Broadcast Education Association, Las Vegas. Cited by Grant 1996. Grant, A.E. (1996), “Media dependency and multiple media sources,” in: Crigler, A.N. (eds.): The psychology of political communication, United States of America, University of Michigan Press, 199-210. Grant, A.E., Guthrie, K.K. & Ball-Rokeach, S.J. (1991), “Television shopping: A Media System Dependency perspective,” in: Communication Research, Vol. 18, No. 6, 773-798. Guthrie, K.K.; Grant, A.E. & Ball-Rokeach, S.J. (1988), “Media System Dependency Theory, television shopping, and television shoppers: Theoretical significance and empirical test”, Paper presented at the Annual Meeting of the International Communication Association (ICA), New Orleans. Cited by Ball-Rokeach (1989). Hartmann, T.; Stuke, D. & Daschmann, G. (2008), “Positive parasocial relationships with drivers affect suspense in racing sport spectators”, in: Journal of Media Psychology, Vol. 20, No.1, 2434. Hatcher, L. (1994), “A Step by Step Approach to Using the SAS System for Factor Analysis and Structural Equation Modeling,”, SAS Institute Inc, Cary. Horton, D. & Wohl, R.R. (1956), “Mass communication and parasocial interaction: Observation on intimacy at a distance,” in: Psychiatry, Vol. 19, No. 3, 215-229. Katz, E.; Blumler, J.G. & Gurevitch, M. (1974), “Utilization of mass communication by the individual”, in: J.G. Blumler and E. Katz (eds.): The Uses of Mass Communication: Current Perspective on Gratifications Research, Sage, Beverly Hills, 19-34. Krotz, F. & Eastman, S.T. (1999), “Orientations toward Television outside the home”, in: Journal of Communication, Vol. 49, No. 1, 5-27. Levy, M.R. (1979), “Watching TV news as a para-social interaction”, in: Journal of Broadcasting, Vol. 23, No. 1, 69-80. Loges, W.E. (1994), “Canaries in the coal mine. Perceptions of threat and Media System Dependency Relations,” in: Communication Research, Vol. 21, No. 1, 5-23. Loges, W.E. & Ball-Rokeach, S.L. (1993), “Dependency relations and newspaper readership”, in: Journalism Quarterly, Vol. 70, No. 3, 602-614. Mcquail, D. (1995), “Mass Communication Theory”, Sage, California (3rd ed.). McQuail, D., Blumler, J.G. & Brown, J.R. (1972), “The television audience: A revised perspective”, in: D. McQuail (ed.): Sociology of Mass Communications, Penguin, Middlesex, England, 135165. Mittal, B. (2006), “I, me and mine-how products become consumers’ extended selves”, in: Journal of Consumer Behaviour, Vol. 5, 550-562. Nordlund, J. (1978), “Media interaction”, in: Communication Research, Vol. 5, No. 2, 150-175. Nunnally, J.C. & Bernstein, I.H. (1994), “Psychometric Theory,” McGraw-Hill, New York. Palmgreen, P, Wenner, L.A. & Rayburn, J.D.I. (1980b), “Gratification discrepancies and news programs choice”, in: Communication Research, Vol. 8, No. 4, 451-478.

212

Curras-Pérez, Miquel-Romero, Ruiz-Mafé and Sanz-Blas

Palmgreen, P., Wenner, L.A. & Rayburn, J.D.I. (1980a), “Relation between gratifications sought and obtained: A study of television news”, in: Communication Research, Vol. 7, No. 2, 161-192. Perse, E.M. (1986), “Soap opera viewing patterns of college students and cultivation,” in: Journal of Broadcasting & Electronic Media, Vol. 30, No. 2, 175-193. Perse, E.M. (1990), “Media involvement and local news effects”, in: Journal of Broadcasting & Electronic Media, Vol. 34, No. 1, 17-36. Perse, E.M. & Rubin, A.M. (1990), “Chronic loneliness and television use, ” in: Journal of Broadcasting & Electronic Media, Vol. 34, No. 1, 37-53. Perse, E.M. & Rubin, R.B. (1989), “Attribution in social and parasocial relationships,” in: Communication Research, Vol. 16, No. 1, 59-77. Perse, E.M., & Rubin, A.M. (1990), “Chronic loneliness and television use”, Journal of Broadcasting and Electronic Media, Vol.34 No.1, pp.37-53. Rosengren, K.E. (1974), “Uses and gratifications: A paradigma outlined”, in: J.G. Blumler and E. Katz (eds.): The Uses of Mass Communications: Current Perspective on Gratifications Research, Sage, Beverly Hills, 269-286. Rosengren, K.E. & Windahl, S. (1972), “Mass media consumption as a functional alternative”, in: D. McQuail (ed.): Sociology of Mass Communications, Penguin, Middlesex, England, 166-194. Rosenstein, A.W. & Grant, A.E. (1997), “Reconceptualizing the role of habit: A new model of television audience activity,” in: Journal of Broadcasting & Electronic Media, Vol. 41, No. 3, 324-344. Rubin, A. & Step, M. (2000), “Impact of motivation, attraction, and parasocial interaction on talk radio listening”, in: Journal of Broadcasting & Electronic Media, Vol. 44, No. 4, 635-654. Rubin, A.M. (1981), “An examination of television viewing motivations,” in: Communications Research, Vol. 8, April, 141-165. Rubin, A.M. (1983), “Television uses and gratifications: The interactions of viewing patterns and motivations,” in: Journal of Broadcasting, Vol. 27, No.1, 37-51. Rubin, A.M. (1984), “Ritualized and instrumental television viewing,” in: Journal of Communication, Vol. 34, No. 3, 67-77. Rubin, A.M. & Perse, E.M. (1987), “Audience activity and soap opera involvement: A uses and effects investigation,” in: Human Communication Research, Vol. 14, No. 2, 246-268. Rubin, A.M.; Perse, E.M. and Powell, R.A. (1985), “Loneliness, parasocial interaction and local television news viewing,” in: Human Communication Research, Vol. 12, No. 2, 155-180. Rubin, R.B. & Mchugh, M.P. (1987), “Development of parasocial interaction relationship”, in: Journal of Broadcasting & Electronic Media, Vol. 31, No. 3, 279-292. Ruiz, C. and Sanz, S. (2007), “Teleshopping adoption by Spanish consumers”, in: Journal of Consumer Marketing, Vol. 24, No. 4/5, 242-250. Ruiz, C. and Sanz, S. (2008), “The impact of television dependency on teleshopping adoption”, in: Direct Marketing: An International Journal, Vol. 2, No. 1, 5-19. Sanderson, J. (2008a), “You are the type of person that children should look up to as a hero”. Parasocial interaction on 38pitches.com”, in: International Journal of Sport Communication, Vol. 1, 337-360. Sanderson, J. (2008b), “Spreading the word: Emphatic interaction displays on BlogMaverick.com”, in: Journal of Media Psychology, Vol. 20, No. 4, 157-168. Sanderson, J. (2009), “‘You are all loved so much’ Exploring relational ,aintenance within the vontext of parasocial relationships”, in: Journal of Media Psychology, Vol. 21, No. 4, 171-182. Satorra, A. & Bentler, P.M. (1988), “Scaling corrections for chi square statistics in covariance structure analysis,” American Statistical Associations Proceedings of the Business and Economic Sections, American Statistical Association, Alexandria, VA, 308-313.

The Role of Parasocial Interaction and Teleparticipation on Teleshopping Behaviour

213

Schiappa, E.; Allen, M. &. Gregg, P.B. (2006), “Parasocial relationships and television: A metaanalysis of the effects”, in: R.W. Preiss; B.M. Gayle; N. Burrell; M. Allen and J. Bryant (eds.): Mass Media Effects Research, Erlbaum, New York, 301-314. Schiappa, E.; Gregg, P.B. & Hewes, D.E. (2005), “The parasocial contact hypothesis”, in: Communication Monographs, Vol. 72, No.1, 92-115. Schramm, H. & Hartmann, T. (2008), “The PSI-Process Scales. A new measure to assess the intensity and breadth of Parasocial processes”, Communications: The European Journal of Communication Research, Vol. 33, 385–401. Schramm, H. & Wirth, W. (2010), “Testing a Universal Tool for Measuring Parasocial Interactions across Different Situations and Media”. Journal of Media Psychology, Vol. 22 No.1, 26–36. Skumanich, S.A. & Kintsfather, D.P. (1993), “Television shopping: A mediated communications perspective,” Paper presented at the annual conference of the Pennsylvania Sociological Society, Pittsburgh, PA. Cited by Skumanich and Kintsfather (1998). Skumanich, S.A. & Kintsfather, D.P. (1998), “Individual Media Dependency. Relations within television shopping programming: A causal model reviewed and revised,” in: Communication Research, Vol. 25, No. 2, 200-219. Sofres, A.M. (2008), “Resumen general de audiencias,” Sofres, Madrid. Stephens, D.L.; Hill, R.P. & Bergman, K. (1996), “Enhancing the consumer-product relationship: Lessons from the QVC home shopping Channel,” in: Journal of Business Research, Vol. 37, No. 3, 193-200. Ulaga, W. and Eggert, A. (2005), “Relationship value in business markets: the construct and its dimensions,” in: Journal of Business-to-Business Marketing, Vol. 12, No. 1, 73-99.

The Impact of e-WOM: Determinants of Influence Manuela López, University of Murcia, Spain María Sicilia, University of Murcia, Spain

1

Introduction

Marketing literature have shown personal influences have a great importance on consumer behavior. Opinions by other consumers have been found to exert a significant impact on consumers’ buying decisions (Arndt, 1967; Chatterjee, 2001; Katz and Lazarfeld, 1955), as well as on post-purchase product perceptions (Bone, 1995). This communication process is called word of mouth (WOM) and it is considered more effective than traditional marketing tools such as personal selling and advertising (Engel et al., 1969; Goldsmith and Horowitz, 2006; Katz and Lazarfeld, 1955). The expansion of new technologies has increased the importance of WOM (Bickart and Schindler, 2001; Chevalier and Mayzlin, 2006; De Bruyn and Lilien, 2008; Dellarocas, 2006; Kozinets et al., 2010; Mayzlin, 2006; Stephen and Lehmann, 2009). The advent of the Internet has brought major changes to the way consumers express their opinions about product or services (Goldsmith and Horowitz, 2006). They can interact with other consumers via e-mail, instant messaging, homepages, blogs, forums, online communities, chat rooms and review sites (Goldsmith, 2006). Recent research is showing these interactions (known as electronic WOM) also influence consumer behavior (Chevalier and Mayzlin, 2006; Park and Kim, 2008). Online consumer opinions/reviews have a direct effect on sales (Chevalier and Mayzlin, 2006; Dellarocas et al., 2007; Duan et al., 2008; Liu, 2006) and nearly 70% of Internet users trust on e-WOM (Nielsen Global Online Consumer Survey, 2009). Thus, although marketers may not have control over e-WOM the way they do other forms of marketing communication, they still have the need to understand the impact of e-WOM (Lim and Chung, 2011). The Internet’s accessibility, reach, and transparency have empowered marketers who are interested in influencing and monitoring e-WOM (Kozinets et al., 2010). However, few researchers have examined what makes certain reviews/opinions more influential than others (Xia and Bechwati, 2008). There is also need for understanding to what extent e-WOM is similar or different from offline WOM (Godes and Mayzlin, 2004; Libai et al., 2010; Liu, 2006).

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_14, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

216

López and Sicilia

In order to overcome these gaps in the literature, this paper analyzes the effect of e-WOM on consumers’ behavior. Influence of e-WOM and traditional WOM on decision making is compared by using the volume of information obtained as a predictor variable. The role of demographic factors on e-WOM is also examined. As recent research (Bickart and Schindler, 2001; Chevalier and Mayzlin, 2006; Dellarocas, 2006; Mayzlin, 2006), we focus on e-WOM communicated via consumer opinions because it represents the most widely used eWOM format (Henning-Thurau et al., 2004). According to this purpose, we first review the literature and propose a set of hypotheses, then describe the methodology and present the results, and we end with a comment about the implications derived from this research.

2 2.1

Theorical Background and Hypotheses WOM

Literature about WOM is very extensive. First studies date from the sixties (Dichter, 1966; Arndt, 1967), although even in the fifties Katz and Lazarsfeld (1955) published “Personal Influence” in order to explain how this personal influence works. WOM can be defined as “a face-to-face conversation between consumers about a product or a service experience” (Sen and Lerman, 2007, p.77). It usually occurs in a private conversation, face-to-face, constructed by two parties: the source of WOM information, and the receiver (Gilly et al., 1998). An important WOM characteristic is the source of information has not commercial interest (Smith, 1993). This aspect appears on several definitions. Arndt (1967; p.291) defines WOM as “face-to-face communication about products or companies between those people who are not commercial entities”. More recent studies have contributed with other definitions, which take into account WOM is unbiased. For example, Litvin et al. (2008; p. 460) describe WOM as “the communication between consumers about a product, service, or a company in which the sources are considered independent of commercial influence”. Consumers tend to believe their friends and relatives are more credible and trustworthy than advertising, public relations, and sales people, thus information from consumer- dominated communication can be more powerful and influential than information from other sources (Herr et al., 1991). Grewal et al. (2003) indicated that individuals are more inclined to embrance the information sent through WOM than that sent through commercial promotion, on the ground that

The Impact of e-WOM: Determinants of Influence

217

WOM information transmitters are usually believed to have no ulterior motive nor receive incentive for their referrals.

2.2

E-WOM

The advances of information technology and the emergence of online social network sites have profoundly changed the way information is transmitted and have transcended the traditional limitations of WOM (Shin, 2007). Consumer can share their product-related experiences on the Internet through e-mail, bulletin boards, chat rooms, forums, fan clubs, brand and usergroups (Goldsmith, 2006). Many terms have been introduced used interchangeably to refer to eWOM, including web-of-mouse, word of mouse, Internet word of mouth, consumers’ reviews (Shin, 2007). This communication process is defined as “any positive or negative statement made by potential, actual, or former customers about a product or company, which is made available to a multitude of people and institutions via the Internet” (Hennig-Thurau et al., 2004, p. 39). Litvin et al. (2008, p. 461) also defined e-WOM as “all informal communications directed at consumers through Internet-based technology related to the usage or characteristics of particular goods and services, or their sellers” (p. 461). More broadly, Huang et al., (2009, p. 160) describe this communication process as “the informal communications through Internet”. The effect of e-WOM on consumers is similar to that of traditional WOM (Hennig-Thurau et al., 2004). Consumers’ reviews on products and services have been an influential tool as an information source for consumers (Gogoi, 2007). As traditional WOM, e-WOM is more effective than firm-generated sources on the Internet (Bickart and Schindler, 2001). Trusov et al. (2009) have even demonstrated that it is more effective than traditional media. However, there are some differences between traditional WOM and e-WOM which should be taken into account when comparing both communication processes.

2.3

WOM vs. e-WOM

Online WOM activity differs from traditional WOM in many aspects. In traditional WOM communication, the information is exchanged in face-to-face conversation (Park and Kim, 2008), while in e-WOM consumers need only to interact with their computers to post or search consumer reviews (Sen and Lerman, 2007). Consumers do not have to directly meet other people to exchange their opinions and experiences on products, services and trades (Shin, 2007), whereby sender and receiver of information are separated by both space and time (Stefees and Burgee, 2009). Sources of information in e-WOM are anonymous individu-

218

López and Sicilia

als who have little or no prior relationship with de information seeker (Xia and Bechwati, 2008). Thus, it is a many-to-many communication in which the information exchanged is more voluminous in quantity compared to information obtained from traditional contacts in the offline world (Hennig-Thurau et al., 2004). As opposed to traditional WOM, the non-commercial focus may not be certain in e-WOM (Chaterjee, 2001). Results regarding the differences between the influence of WOM and eWOM on decision making are inconclusive. Previous researchers have shown that WOM from strong ties bear greater influence on the receiver’s behavior than WOM from weak ties (Bansal and Voyer, 2000; Brown and Reingen, 1987). As traditional WOM comes from friends and relatives while the source is anonymous in e-WOM (Chaterjee, 2001; Sen and Lerman, 2007; Xia and Bechwati, 2008), one could argue that WOM is more persuasive than e-WOM. However, a recent study has found that information gained from online consumer reviews is more influential on consumers than speaking with friends or acquaintances in person (Steffes and Burgee, 2009). Therefore, these differences of influence between the traditional and the online process may also depend on other factors. In addition of tie strength (Brown and Reingen, 1987), marketing literature has shown WOM influence depends on source experience (Bansal and Voyer, 2000; Gilly et al., 1998), on the valence of opinions (Chevalier and Mayzlin, 2006) and on the volumen of information exchanged (Chevalier and Mayzlin, 2006; Dellarocas et al., 2007; Duan et al., 2008; Liu, 2006). Volume is among the most important WOM attributes that have been examined (Mahajan et al., 1984). Previous studies have found the volume of WOM correlates significantly with consumer behavior (Anderson and Salisbury, 2003). Recent studies on eWOM have confirmed these correlations by showing a positive impact on sales (Chevalier and Mayzlin, 2006; Dellarocas et al., 2007; Duan et al., 2008; Liu, 2006). Therefore, the relative influence of WOM and e-WOM may be dependent on the volume of information obtained from both sources. When consumers obtain more information from traditional WOM than from e-WOM, WOM influence is likely to be higher than e-WOM influence. However, there may be many situations in which consumers find it difficult to obtain advice or recommendation from their friends, family or even acquaintances. In these circumstances, they are eager to use salient and more accessible information (Häubl and Trifts, 2000). As there is a huge amount of online information about any product, consumers will easily find information from others through this medium. Therefore, the reverse effect is expected for consumers who obtain more information from e-WOM than from WOM. Thus, we propose: H1: The relative influence of WOM and e-WOM on decision making depends on the volume of information obtained from each source:

The Impact of e-WOM: Determinants of Influence

219

H1a: The influence of e-WOM will be higher for consumers who have obtained more information from e-WOM than from traditional WOM. H1b: The influence of traditional WOM will be higher for consumers who have obtained more information from WOM than from e-WOM.

2.4

The Role of Demographics

Marketing literature has revealed relationships between certain demographic characteristics (e.g., gender, age, education levels) and consumer behavior (Eastlick and Feinberg, 1994). In a WOM context, previous research has studied how demographic similarities between receiver and sources affect the influence of WOM (Brown and Reingen, 1987). However, very little is known about which receiver’s demographic characteristics determine the influence of eWOM. Previous studies have highlighted gender differences in consumer behaviour. Men and women have different buying attitudes (Dittmar et al., 2004), search patterns (Cleveland et al. 2003), post-purchase behavior (Solnick and Hemenway, 1992), perceived risk (Darley and Smith, 1995), and shopping stereotypes held (Dholakia and Chiang 2003). These gender differences persist in terms of Internet-related attitudes and activities (Garbarino and Strahilevitz, 2004). Some studies have shown that females generally spend less time online, are less interested in the Internet, and are less likely to shop online than males (Garbarino and Strahilevitz, 2004). In addition, females perceive more risks associated with shopping online (Garbarino and Strahilevitz, 2004). Additionally, online shopping may be perceived less favourably by women due to the reduced social interaction as compared to the physical store (Van Slyke, et al., 2002). In fact, females not only place more emphasis on social interaction than males, but they also communicate with others more often (Tannen, 1990). In the context of online behavior, it has been observed that women view the chance to communicate with others to be among the greatest benefits of the Internet (Brunner and Bennet, 1997). As a consequence, females are more likely to use the Internet for interpersonal communications, and they have been shown to react more favourably to websites that contain more humanistic elements (Rodgers and Harris, 2003). In addition, receiving a recommendation from a friend has a greater effect on reducing risk perceptions among women than among men (Garbarino and Strahilevitz, 2004). Generally speaking, men exhibit a weaker sensitivity to the opinions of their friends (Shoaf et al., 1995) and are less prone to engage in active information provision to other consumers (Feick and Price, 1987). On the basis of the above reasoning, we propose:

220

López and Sicilia

H2: The influence of e-WOM on decision making will be higher for females than for males. Consumers’ education level is also regarded as an important factor that greatly influences consumer behavior (Chen and Dhillon, 2003). For example, consumers’ education is associated with the ability to acquire and process information to make decisions (Berges and Casellas, 2006). The higher educational levels have lower information processing costs and a higher degree of involvement (Chen and Liu, 2008). In addition, there is a positive relationship between educational levels and information search extent. The higher the educational level, the more extensive the information search (Chen and Liu, 2008), and the higher the probability of examining different information sources (Strebel et al., 2004). Thus, sources influence could be dependent on this demographic characteristic. Previous research has shown consumers with lower education level were more likely to go to a dentist who advertised (Hite et al., 1988). Additionally, Andaleeb (1994) suggested that the higher the education level of consumers, the less likely they are to favor advertising. On the one hand, consumers with a high education level are more likely to search on the Internet than those with less (Chen and Liu, 2008), but on the other hand, education negatively influences people’s trust in online information (Bomda Web survey, 1998). Then, consumers with a high education level will be more aware that different aspects of this communication process are suspect of some kind of manipulation. The lack of personal knowledge about the motivations of unseen strangers offering recommendations online could affect the perceived credibility of eWOM (Chatterjee, 2001). Marketers can control some types of e-WOM. For example, they can decide whether to allow certain consumer reviews to be shown or not (Park and Kim, 2008). In addition, firms can easily disguise their promotion as consumer recommendations (Dellarocas, 2006; Mayzlin 2006). Individuals with a high education level may be further aware that manipulation occurs, but they cannot directly distinguish honest opinions from fake opinions, so will be difficult for them to trust on this information (Dellarocas, 2006; Mayzlin 2006). Therefore, we propose: H3: The influence of e-WOM on decision making will be higher for consumers with a low education level than for those with a high education level. Previous literature has shown age affects consumer behavior. For example, old consumers are more loyal to a particular brand than young consumers (Lambert-Pandraud et al., 2005). In addition, younger individuals are likely to be innovative when adopting a new product, service or communicating with a

The Impact of e-WOM: Determinants of Influence

221

new provider, while older people are more likely to be conservative in these aspects (Im et al., 2003). They are also more conservative in adopting new channels of communication than young people. Younger are more likely to be computer literate and therefore use the Internet more effectively, that is, the ability to use the Internet as a source vary inversely with age (Ratchford et al., 2003). The Internet has surpassed television in overall time spent to become the primary medium of choice among the young (Greenspan, 2003). Young people are also more likely to use the Internet to gather information about products and services than older people (Pastore, 2001). In contrast, older consumers are more likely to use information channels that provide relatively less complex information (Streebel et al., 2004) because information processing capacities decline with age (Evanschitzky and Wunderlich, 2006). Young people also produce more WOM than old people (East, 2007). They are more likely to be influenced by WOM, friend/relative is a relatively productive source of information for them (Ratchford et al., 2003). Thus, as Internet is a preferred medium for younger as compared with older people, WOM communication development in this medium will also have more influence for young than for old people. Therefore, we propose: H4: The influence of e-WOM on decision making will be higher for young people than for old people.

3

Methodology

Data were collected from tourist services users because WOM represents the most important information source for travellers (Dey and Sarma, 2010; Gretzel and Yoo, 2008). Internet has changed tourist behavior dramatically (Mills and Law, 2004). Prospective travellers have direct access to a much greater wealth of information and make online purchase themselves instead of relying on travel agencies to undertake this process for them (Buhalis and Law, 2008). The emerge of Web 2.0 brings the concept of social networking and applies it to the tourism industry (Buhalis and Law, 2008). Nowadays, many tourists rely on eWOM to reduce the perceived risk and uncertainty before they purchase the service (Litvin et al., 2008). The sample consisted of older-16-years tourists who had searched Internet to plan their last travel. Furthermore, they had to be first-time visitors because prior experience with the destination could affect travel planners’ use of information sources (Woodside and Dubellar, 2002). We sent 735 questionnaires

222

López and Sicilia

through an university email listing. We suggested the receiver to re-send the questionnaire to other people, after inviting them to participate in the study. A very similar procedure was developed by De Bruyn and Lilien (2008). A remainder mailing was also undertaken to boost response rate. As a result of this process, we collected 165 valid questionnaires. Before starting data collection, the questionnaire was pretested to ensure it was well understood. The opening questions of the survey referred to general trip characteristics such as destination, time to plan, length of stay, travelling party composition, and trip budget. In order to measure information sources used to plan the trip, respondents were asked to rate how much information they had obtained from different offline (friends/relatives/acquaintances, advertising on TV, newspapers, magazines and travel agencies) and online sources (online travel agencies, consumer opinions on Internet and any other web sites) using a 5 point-Likert scale (from 1= nothing to 5= a lot). We next asked individuals to what extent traditional WOM and e-WOM influenced their destination decision. We used a 3-item 5-points semantic differential scale based on Mishra et al. (1993) to measure this concept. At the end of the questionnaire individuals provided their demographic information (sex, age, level of education, and occupation).

4 4.1

Results Sample Characteristics

The demographic profile of the sample is outlined in Table 1. Mean age is around 30 years old, and almost 60% of subjects are female. Most of the individuals have university studies. Over 50% are employed, while a forth part of the sample is constituted by students.

223

The Impact of e-WOM: Determinants of Influence

Table 1: Sample characteristics

Gender Male Female Education level University Secondary School Primary School No qualifications Employment Student Housewife Self-employed Civil servant/employed Unemployed Age (mean)

4.2

% 41.7 58.3 % 70.1 26.2 2.4 1.2 % 26.8 1.2 10.4 55.5 6.1 28.59

N 68 95 N 115 43 4 2 N 44 2 17 91 10

Results

A t-Student test was used to test the first hypothesis. The sample was divided into two groups depending on the volume information obtained from WOM and e-WOM communication. We have individuals who obtained more information from the traditional process than from online communication, and individuals who obtained more information from e-WOM than from offline WOM. We used a t-Student test to compare the influence of both communication processes within each group.

Table 2: Comparison between traditional WOM and e-WOM influence depending on the volume of information obtained from each source (mean values)

InfoWOM> Infoe-WOM InfluenceWOM = 4.141 Influencee-WOM=3.724 t = 2.857; p<0.01

Infoe-WOM> InfoWOM InfluenceWOM = 3.218 Influencee-WOM=3.989 t = -3.110; p<0.01

As shown in Table 2, WOM influence is higher than e-WOM influence for consumers who obtain more information from the traditional process than from e-WOM. Thus, H1a is supported. For consumers who obtained more informa-

224

López and Sicilia

tion from e-WOM than from WOM, e-WOM exerted a greater influence on decision making than traditional WOM. Therefore, H1b is also accepted. Further analyses were developed in order to extend the above results. We created two additional subgroups considering all information sources used in decision making, offline (friends/relatives/acquaintances, advertising on TV, newspapers, magazines and travel agencies) and online sources (online travel agencies, consumer opinions on Internet and any other web sites). Then, we divided individuals between those who obtained more information from offline sources and those who obtained more information from online sources. A t-test was also used to compare results between these two groups.

Table 3: Comparison between traditional WOM and e-WOM influence depending on the volume of information obtained from online and offline sources (mean values)

InfoOFFLINE> InfoONLINE InfluenceWOM = 4.417 Influencee-WOM=3.896 t = 2.871; p<0.01

InfoONLINE> InfoOFFLINE InfluenceWOM = 3.578 Influencee-WOM=3.882 t = -2.109 p<0.05

As shown in table 3, for individuals who obtained more offline information, traditional WOM is more influential than e-WOM. In accordance with results obtained for H1, for individuals who obtained more online information, e-WOM has more impact on decision making than traditional WOM. Regarding the role of demographics, the influence of e-WOM on decision making is higher for women than for men (see table 4). Thus, H2 is supported. In order to test H3 we divided the sample into two groups depending on consumers’ level of education. Through a median split procedure, we obtained a first group in which all individuals have university studies, considered with high education level; and a second group of consumers with lower education level, in which individuals do not have university studies. As proposed, consumers’ education level affects the influence of e-WOM on decision making. The higher the education level, the lower the influence of this communication process. The difference on e-WOM influence is significant (p<0.05). Therefore, H3 is accepted. Using a similar procedure, we divided the sample into two groups of age: young (under 26 years old) and old people (over 26 years old). As shown table 4, e-WOM has a higher influence on decision making for young than for old people, however these differences are not significant (p>0.05). Thus, H4 is not accepted.

225

The Impact of e-WOM: Determinants of Influence

Table 4: The influence of e-WOM depending demographic characteristics

Demographic characteristics Gender Men Women Education level High (University studies) Low (No university studies) Age Young people (under 26 years old) Old people (over 26 years old)

5

N

Mean values

F

p

68 95

3.660 4.059

6.233

0.014

115 49

3.813 4.188

5.222

0.024

79 81

3.960 3.814

0.817

0.368

Discussion

The present findings expect to clarify some researchers’ doubts about e-WOM influence. The differences between online and traditional process (Herr et al., 1991; Sen and Lerman, 2007) could affect the potential influence of e-WOM (Hennig-Thurau et al., 2004; Sen and Lerman, 2007). This study shows that eWOM has a great importance on consumer behavior. This result is in line with a recent research by Steffes and Burgee (2009) who found online process is even more persuasive than traditional communication. Our study qualifies Steffes and Burgee’s (2009) study by explaining when e-WOM has a greater influence than WOM. E-WOM increases its impact when individuals obtain more information from online consumer reviews than from friends and relatives. Similarly, eWOM is more influential than WOM for individuals who obtained more information in online sources than in offline sources. In contrast, WOM has more influence than e-WOM for consumers who obtain more information either from traditional WOM or from offline sources in general. This result is supported by previous research that has shown the influence of e-WOM depends on the volume of information exchanged (Chevalier and Mayzlin, 2006; Dellarocas et al., 2007; Duan et al., 2008; Liu, 2006). They have shown the more conversations there are about a product, the greater sales. In addition, we have also demonstrated that the influence of e-WOM on decision making is determined by demographic factors. Specifically, gender and education have impact on this communication process. We have observed that the influence of e-WOM is higher for women and consumers with a lower level of education than for men and individuals with a higher level of education. This result is in line with Garbarino and Strahilevitz (2004). They showed that WOM information was more effective in reducing perceived risk and increasing will-

226

López and Sicilia

ingness to buy online among women than it was for men. Additionally, it confirms previous studies which have shown that individuals with a lower level of education are more likely to have positive attitudes about all forms of advertising (Williams and Hensel, 1995). However, we have observed that age does not affect the influence of e-WOM. This result is in line with Durand and Lambert’s (1985) finding that criticisms of advertising were explained more by other variables than by demographics. Similarly, Forsythe and Shi (2003) found that age was a predictor of amount spent on the Web, but did not predict other online behaviours. Therefore, this study contributes to the lack of literature related to the factors that determine the influence of e-WOM communication (Dellarocas, 2006). It also extend the studies on the differences between offline WOM and e-WOM (Godes and Mayzlin, 2004; Libai et al., 2010). Finally, the results could be interesting for the tourism industry, because there are few researchers that have empirically studied e-WOM on this field (Litvin et al., 2008), despite of the fact that e-WOM is the information source most consulted among tourists (Dey and Sarma, 2010; Gretzel and Yoo, 2008).

6

References

Andaleeb, S.S. (1994), “Hospital Advertising: The Influence of Perceptual and Demographic Factors on Consumer Dispositions,” in: Journal of Services Marketing, Vol. 8(1), 48–59. Anderson, E. & Salisbury, L. (2003), “The Formation of Market-Level Expectations and Its Covariates,” in: Journal of Consumer Research, Vol. 30, 115–24. Arndt, J. (1967), “Role of Product-Related Conversations in the Diffusion of a New Product,” in: Journal of Marketing Research, Vol. 4, 291–295. Bansal, H.S. & Voyer, P. A. (2000), “Word-of-Mouth Processes within a Services Purchase Decision Context,” in: Journal of Service Research, Vol. 3, 166–167. Berges, M. & Casellas, K. (2006), “Quality Warranties and Food Products in Argentina. What do Consumers Believe in?,” in: International Association of Agricultural Economists Conference, Gold Coast, Australia, Bickart, B. & Schindler, R. M. (2001), “Internet forums as influential source of consumer information,” in: Journal of Interactive Marketing, Vol 15(3), 31-39. Bone, P. F. (1995), “Word-of-mouth effects on short-term and long-term product judgment” in: Journal of Business Research, Vol. 32, 213-223. Brown, J. & Reingen, P. (1987), “Social Ties and Word-of-Mouth Referral Behavior,” in: Journal of Consumer Research, Vol. 14, 350-362. Brunner, C. & Bennett, D. (1997), “Technology and gender: Differences in masculine and feminine Views,” in: NASSP Bulletin, Vol. 81(592), 46-51. Buhalis, D. & Law, R. (2008), “Progress in information technology and tourism management: 20 years on and 10 years after the Internet-The state of eTourism research,” in: Tourism Management, Vol. 29, 609-623.

The Impact of e-WOM: Determinants of Influence

227

Chatterjee, P. (2001), “Online Reviews: Do Consumers Use Them?,” in: Advances in Consumer Research, Vol. 28, 129-133. Chen S. C., & Dhillon G. S. (2003), “Interpreting Dimensions of Consumer Trust in E-Commerce,” in: Information Technology and Management, Vol. 4(2-3), 303-318. Chen, H. Y. & Liu, K. Y. (2008), “Web-based synchronized multimedia lecture system design for teaching/learning Chinese as second language,” in: Computers & Education, Vol. 50, 693-702. Chevailer, J. A. & Mayzlin, D. (2006), “The effect of word of mouth on sales: online book reviews” in: Journal of Marketing Research, Vol. 43(3), 345-354. Cleveland M.; Babin B. J.; Laroche M. & Bergeron J. (2003), “Information search patterns for gift purchases: a cross-national examination of gender differences," in: Journal of Consumer Behaviour, Vol. 3(1), 20–47. Darley, W. K. & Smith, R. E. (1995), “Gender differences in information processing strategies: An empirical test of the selectivity model in advertising response,” in: Journal of Advertising, Vol. 24, 41–56. De Bruyn, A. & Lilien, G. L. (2008), “A multi-stage model of word-of-mouth influence of viral marketing,” in: International Journal of Research in Marketing, Vol. 25(3), 151-163. Dellarocas, C. (2006), “Strategic Manipulation of Internet Opinion Forums: Implications for Consumers and Firms,” in: Management Science, Vol. 52(10), 1577-1593. Dellarocas, C. N., Awad, N. F. & Zhang, X. (2007), “Exploring the Value of Online Product Reviews in Forecasting Sales: The Case of Motion Pictures,” in: Journal of Interactive Marketing, Vol. 21 (4), 23-45. Dey, B. & Sarma, M. K. (2010), “Information usage among motive-bassed segments of travelers to newly emerging tourist destinations,” Tourism Management, Vol. 31(3), 341-344. Dholakia, R. R. & Chiang, K. P. (2003). “Shoppers in cyberspace: Are they from venus or mars and does it matter?,“ in: Journal of Consumer Psychology, Vol. 13, 171-176. Ditcher, E. (1966), “How word of mouth advertising works,” in: Harvard Business Review, Vol. 44(6), 147-166. Dittmar, H., Long, K. & Meek, R. (2004), “Buying on the Internet: Gender Differences in On-Line and Conventional Buying Motivations,” in: Sex Roles, 50 (516), 423–44. Duan, W.; Gu, B. & Whinston, A. (2008), “The Dynamics of Online Word-of-Mouth and Product Sales: An Empirical Investigation of the Movie Industry,” in: Journal of Retailing, Vol. 84(2), 233-242. Durand, R. M. & Lambert, Z. V. (1985), "Alienation and Criticisms of Advertising," in: Journal of Advertising, Vol. 14 (3), 9-17. East, R. (2007), “Researching Word of Mouth,” in: Australian Marketing Journal, Vol. 15, 23-26. Eastlick, M. A. & Feinberg, R. A. (1994), “Gender differences in mail-catalog patronage motives,” Journal of Direct Marketing, Vol. 8(2), 37-44. Engel, J.F., Blackwell, R.D. & Miniard, P.W. (1969), “Consumer behaviour”, Fort Worth, TX: Dryden. Evanschitzky, H., & Wunderlich, M. (2006), “An examination of moderator effects in the four-stage loyalty model,” in: Journal of Service Research, Vol. 8 (4), 330–45. Feick, L. & Price, L. (1987), '"The Market Maven: a Diffuser of Marketplace Information," in: Journal of Marketing, Vol. 51 (1), 83-97. Forsythe, S.M. & Shi, B. (2003), “Consumer patronage and risk perceptions in internet shopping,” in: Journal of Business Research, Vol. 56, 867−875. Garbarino, E. & Strahilevitz, M. (2004), “Gender differences in the perceived risk of buying online and the effects of receiving a site recommendation,” Journal of Business Research, Vol. 57(7), 768–775.

228

López and Sicilia

Gilly, M. C.; Graham, J. L.; Wolfinbarger, M. F. & Yale, L. J. (1998), “A dyadic study of interpersonal information search,” in: Journal of the Academy of Marketing Science, Vol. 26(2), 83100. Godes, D. & Mayzlin, D. (2004), “Using online conversations to study word-of-mouth communication,” in: Marketing Science, 23(4) 545–560. Gogoi, P. (2007). “Retailers take a tip from MySpace,” retrieved from Businessweek.com Web site http://www.businessweek.com/bwdaily/dnflash/content/feb2007/db20070213_ 626293.html. Goldsmith, R.E. (2006), “Electronic Word of Mouth”, in: Enciclopedia of E-Commerce, EGovernment and Mobile Commerce, Mehdi Khosrow-Pour, Ed., Hershey, PA: Idea Group Publishing. Goldsmith, R.E. & Horowitz, D. (2006), “Measuring motivations for online opinion seeking,” in: Journal of Interactive Advertising, Vol. 6(2), available in www.jiad.org. Greenspan, R. (2003), “Girl Gamers Go Online,” Gretzel, U. & Yoo, K. H. (2008), “Use and impact of online travel reviews,” In P. O`Connor, W. Höpken, and U. Gretzel (Eds.). Information and communication technologies in tourism. New York: Springer, 35-46. Grewal, R., Cline, T. & Davies, A. (2003), “Early-Entrant Advantage, Word-of-Mouth Communication, Brand Similarity, and the Consumer Decision-Making Process,” in: Journal of Consumer Psychology, Vol. 13 (3), 187–97. Haubl, G. & Trifts, V. (2000), “Consumer decision making in online shopping environments: the effects of interactive deci-sion aids, ” in: Marketing Science, Vol. 19(1), 4-21. Hennig-Thurau, T.; Gwinner K. P.; Walsh G. & Gremler D. D. (2004), “Electronic Word-of-Mouth Via Consumer-Opinion Platforms: What Motivates Consumers to Articulate Themselves on the Internet,” in: Journal of Interactive Marketing, Vol. 18(1), 38-52. Herr, P.M., Kardes, F.M. & Kim, J. (1991), “Effects of Word-of-Mouth and Attribute Information on Persuasion: an Accessibility-Diagnosticity Perspective,” in: Journal of Consumer Research, Vol. 17(3), 454-462. Hite, R.E., Bellizzi, J.A. & Faser, C. (1988), “A Content Analysis of Ethical Policy Statement Regarding Marketing Activities,” in: The Journal of Business Ethics, Vol. 7, 771-776. Huang, C., Lin, T. & Lin, K. (2009), “Factors affecting pass-along email intentions (PAEIs): Integrating the social capital and social cognition theories,” in: Electronic Commerce Research & Applications, Vol. 8(3), 160-169. Im, S.; Bayus, B.L. and Mason, C.H. (2003), “An Empirical Study of Innate Consumer Innovativeness, Personal Characteristics, and New-Product Adoption Behavior,” in: Journal of the Academy of Marketing Science, Vol. 31 (1), 61–73. Katz, E. & Lazarsfeld, P.F. (1955), “Personal Influence”, Glencoe, IL: Free Press. Kozinets, R.V.; De Valck, K.; Wojnicki, A.C. & Wilner, S.J.S. (2010), „Networked Narriatives: Understanding Word-of-Mouth Marketing in Online Communities,” in: Journal of Marketing, Vol. 74 (2), 71-89. Lambert-Pandraud, R.; Laurent, G. & Lapersonne, E. (2005), “Repeat Purchasing of New Automobiles by Older Consumers: Empirical Evidence and Interpretations,” in: Journal of Marketing, Vol. 69, 97–113. Libai, B.; Bolton, R.N.; Bügel, M.S.; Ko de Ruyter; Götz, O., Risselada, H. & Stephen A.T. (2010), “Customer to Customer Interactions: Broadening the Scope of Word of Mouth Research,” in: Journal of Service Research, 13 (3), 267-282. Lim, B.C. & Chung, C.M.Y. (2011), “The impact of word-of-mouth communication on attribute evaluation,” in: Journal of Business Research, Vol. 64(1), 18-23.

The Impact of e-WOM: Determinants of Influence

229

Litvin, S.W., Goldsmith, R.E. & Pan, B. (2008), “Electronic word of mouth in hospitality and tourism management,” in: Tourism Management, Vol. 29, 458-468. Liu, Y. (2006), “Word-of-mouth for movies: Its dynamics and impact on box office revenue,” in: Journal of Marketing, Vol. 70, 74-89. Mahajan, V., Muller, E. & Kerin, R. (1984), “Introduction Strategy for New Products with Positive and Negative Word-of-Mouth,” in: Management Science, Vol. 30 (12), 1389–1404. Mayzlin, D. (2006), “Promotional Chat on the Internet,” in: Marketing Science, Vol. 25(2), 155-163. Mills, J. & Law, R. (2004), “Handbook of consumer behaviour, tourism and the Internet,” New York: Harworth Hospitality Press. Mishra, S., Umesh, U.N. & Stem, Jr. D.E. (1993), “Antecedents of the Attraction Effect: An Information-Processing Approach,” in: Journal of Marketing Research, Vol. 30(3), 331-353. Novak, T.P., Hoffman, D.L. & Yung, Y.F. (2000), “Measuring the Customer Experience in Online Environments: A Structural Modeling Approach,” in: Marketing Science, 19(1), 22-42. Park, D.H. & Kim, S. (2008), “The effects of consumer knowledge on message processing of electronic word of mouth via online consumer reviews,” in: Electronic Commerce Research and Applications, Vol. 7, 399-410. Pastore, M. (2001) Young Internet Users Prefer EBrowsing to E-Commerce. Based on surveys from theIpsos-Reid study ‘The Face of the Web: Youth’, on young people between the ages of 12 and 24 in 16 countries. [WWW document]. URL http://www.clikz.com/stats/big_picture/demographics/article.php/707631 Ratchford, B.T., Lee, M. & Talukdar, D. (2003), “The Impact of the Internet on Information Search for Automobiles,” in: Journal of Marketing Research, Vol. 40, 193–209. Rodgers, S. & Harris, M. A. (2003), “Gender and e-commerce: An exploratory study,” in: Journal of Advertising Research, Vol. 43(3), 322–329. Sen, S. & Lerman, D. (2007), “Why are You Telling me This? An Examination into Negative Consumers Reviews on the Web,” in: Journal of Interactive Marketing, Vol. 21(4), 76-94. Shin, K.A. (2007), “Factors influencing source credibility of Consumer Reviews: Apparel Online Shopping. Thesis, Oregon State University. Shoaf, F., Robert, J., Scattone, M.M. & Maheswaran, D. (1995), "Gender Differences in Adolescent Compulsive Consumption," in: Advances in Consumer Research, Vol. 22, 500-504. Smith, V.L. (1993), “When prior knowledge and law collide—Helping jurors use the law,” in: Law and Human Behavior, Vol. 17, 507-536. Solnick S.J. & Hemenway, D. (1992), “Complaints and disenrolment at a health maintenance Organization,” in: Journal of Consumer Affair, Vol. 26(1), 90 –103. Steffes, E.M. & Burgee, L.E. (2009), “Social ties and online word of mouth,” in: Internet Research, Vol. 19(1), 42–59. Stephen, A.T. & Lehmann D.R. (2009), “Why Do People Transmit Word-of-Mouth? The Effects of Recipient and Relationship Characteristics on Transmission Behaviours”, working paper. Strebel, J., Erdem, T. and Swait J. (2004), “Consumer search in high technology markets: Exploring the use of traditional information channels,” in: Journal of Consumer Psychology, 14 (1/2), 96– 104. Tannen, D. (1990), “Gender differences in topical coherence: Creating involvement in best friends' talk,” in: Discourse Processes, Vol.13, 73-90. Trusov, M., Bucklin, R.E. & Pauwels, K. (2009), “Effects of Word-of-Mouth versus Traditional Marketing: Findings from an Internet Social Networking Site,” in: Journal of Marketing, Vol. 73, 90-102. Van Slyke, C., Comunale, C.L. & Belanger, F. (2002), “Gender differences in perceptions of Webbased shopping,” in: Communications of the ACM, Vol. 45(7), 82–86.

230

López and Sicilia

Williams, J.R. & Hensel, P.J. (1995), “Direct to Consumer Advertising of Prescription Drugs,” in: Journal of Health Care Marketing, Vol. 15(1), 35–41. Woodside, A.G. & Dubelaar, C.(2002), “A general theory of tourism consumption systems:A conceptual framework and an empirical exploration,” in: Journal of Travel Research, 41(2),120-132. Xia, L. & Bechwati, N. N. (2008), “Word of mouse: the role of cognitive personalization in online consumer reviews,” in: Journal of Interactive Advertising, Vol. 9(1), available in: www.jiad.org.

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns Shintaro Okazaki, Universidad Autónoma de Madrid, Spain María Ángeles Navarro-Bailón, University of Murcia, Spain

1

Introduction

Our business-to-consumer markets are becoming increasingly ubiquitous. Due to quickly advanced wireless communication technology, many firms have already adopted mobile-based promotions. For example, DreamWorks Home Entertainment used a viral campaign for the DVD of the film The Ring 2 that operated on mobile phones and email accounts. Visitors to the campaign site were invited to “scare their friends” by entering their email address and mobile telephone number. The site then sent them an email, inviting them to click on a link and watch the video. McDonald’s offers three different types of mobile coupons on three different platforms, from the basic text-only SMS version to the graphically rich version—that is, fully trackable and redeemable, all without any POS hardware or software requirements (iMedia Connection, 2005a, b). This study mainly addresses two issues associated with mobile-based promotion. Fist, prior research unanimously claims that ubiquity—or the usage flexibility of time and location—represents the most important feature of mobile Internet. However, it seems that our understanding of this concept has been far from optimal. Much prior only touches the surface of the issue; otherwise it appears practically unexplored—with interesting exceptions of Klenejn et al. (2007) and Okazaki et al. (2009). Second, an increasing number of mobilebased promotions generate general concerns over a misuse and intentional deception of sensitive personal information (Petty, 2003). At this stage of mobile commerce diffusion, more attention is needed with regard to the interaction between ubiquitous context and mobile users’ information privacy concerns. The objective of this study is designed to fulfill these research gaps, by conducting a quasi-experimental study.

S. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_15, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

232 2 2.1

Okazaki and Navarro-Bailón

What is Ubiquity? Time-Space Perspective

The core concept of ubiquity seems to stem from a seminal work by Hägerstrand at the University of Lund, Sweden, who proposed the so-called “time-space perspective.” whose origin can be traced to the work of. This perspective focuses on a question as to how to organize activities into temporal and spatial terms. These activities in temporal and spatial terms can serve to define the human performance. This was the very first step in so-called timegeography. Time-geography emphasizes the factors associated with the spatial and temporal spread of innovations within particular environments. Lenntorp (1999) argues that time-geography constitutes a foundation for a general geographical perspective. It represents a new structure of thought under development, which attempts to consolidate the spatial and temporal perspectives of different disciplines on a more solid basis than has thus far occurred. Time-geography is not a subject area per se, or a theory in the narrow sense, but rather an attempt to construct a broad structure of thought that may form a framework capable of fulfilling two tasks. The first is to receive and bring into contact knowledge from highly distinct scientific areas, and from everyday praxis. The second is to reveal relations, the nature of which escapes researchers as soon as the object of research is separated from its given milieu, in order to study it in isolation, experimentally, or in some other distilled way (Lentorp, 1999). Accoriding to Hägerstrand (1975), the importance of spatial factors is demonstrated by interpersonal communications, where most influence is transferred within local social systems or the "neighborhood effect." According to him, both terrestrial and social distance barriers impede diffusion, in that human activities form environments that have a hierarchical ordering to the extent that those who have access to power in a superior domain frequently use it to restrict the set of possible actions permitted inside the subordinate domains. On this basis, Hägerstrand (1970) developed the basics of time-geographic notation, in order to have a means to keep track simultaneously of both the spatial and temporal dimensions. In his theory, there is a “time-space” entity called a “domain.” A domain is defined as “a time-space entity within which things and events are under the control of a given individual or a given group” (Hägerstrand, 1970, p. 16). In a domain, activities and events are under the control and influence of specific individuals or organizations. Domains, such as a school or an office building, often serve as stations for individuals to bundle. The ability of an individual or an organization to navigate through the domain depends on the following three time-space constraints that characterize

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

233

information technology (Hägerstrand, 1970): Capability constraints, coupling constraints, and authority constraints. Coupling constraints requires the user’s presence at a specific time and place, and therefore they are instrumental, physiological, and cognitive limitations. That is, individuals must join other individuals or organizations in order to form production, consumption, social and other activity bundles. Capability constraints refer to the user’s resources and ability to overcome spatial separation at a specific moment. They circumscribe the amounts of effort needed for people to associate themselves with others and with material artifacts at specific places and times for a certain duration, in order to realize production, consumption, and transactions. Finally, authority constraints become important when several activities are pretended to be packed into a limited space. Authority constraints subsume such limited space occupation in terms of rules, laws, economic barriers, and power relations, which determine who does or does not have specific access to specific domains at specific times in specific spaces. Although Hägerstrand’s theory is derived from a different discipline, the concepts of coupling, capability, and authority constraints are very relevant to our conceptualization of the ubiquity concept. Telecommunication systems allow humans to eliminate distance for some types of activities and interactions. Transportation, along with telecommunication and settlement systems, grows and declines in response to human activities in space and time. They influence economic, social, and knowledge networks, and in turn shape human activities and their locations in time and space. No coupling constraints—presence and timing—are relevant when using Twitter or MySpace, because users are allowed to form social networks via the wireless Internet connection at any time, and in any place. Similarly, capability constraints become unimportant when search engines enable users to overcome spatial distance, and to reach almost all spatial information at any moment. Finally, mobile banking and payment functions overcome authority constraints, because the right and freedom to control specific domains at specific times is drastically extended.

2.2

Personal Extensibility Theory

In terms of theory building for ubiquity, personal extensibility theory may be more specific to human mobility. The personal extensibility concept attempts to measure the ability of a person (or group) to overcome the friction of distance through transportation or communication. Fundamentally, it pertains to “the scope of sensory access and knowledge acquisition and dispersion, and to people’s horizons as social actors.” Janelle claims that personal extensibility is

234

Okazaki and Navarro-Bailón

“conceptually the reciprocal of time-space convergence,” and argues that the rapid advances in communication and transportation technologies and their associated institutions imply a “shrinking world,” with expanding opportunities for extensibility. That is, the focal point of this theory lies in the expansion of opportunities for human interaction, rather than improved abilities for movement over greater distances. Here, of special interest to us is that personal extensibility depends on developments in communications technology, because such innovation will reduce the time required to interact with persons in distant places. Personal extensibility can be anywhere between high and low. In a high personal extensibility situation, one can satisfy important information needs, while moving away or toward a destination, by the use of a mobile device. For example, a stockbroker has just left home, and walking along a street toward a commuter train station. On the way, he turns on his mobile to check real-time Dow Jones prices, because he needs to make a buying decision this afternoon. In this case, his perceived extensibility is high, because he can access the information he needs, as he could not have done without his mobile. By contrast, in a low personal extensibility situation, one stays in a static place, and thus takes little advantage of the device’s portability. For example, a sales manager enjoys her quiet Sunday morning at home. She may or may not use a mobile device to check a weather forecast for her afternoon picnic, or simply turn on her TV or laptop in the living room. In this case, her perceived extensibility is low, because she may not feel any desperate information needs that require her to use her mobile to catch up. Such changes in the significance of distance affect economic, political, and cultural life. For example, Adam (1995) argues that “as distant connections become easier to maintain, spatial patterns of social interaction change; work and home, resources and industries, management and labor assume varying spatial configurations” (p. 267-268). This is precisely the result of the rapid advances in the Internet in general, and the mobile device in particular.

3

Information Privacy Concerns

Additional perspective this study focuses on is the issue of information privacy concerns. Often, a promotional campaign requests consumers to subscribe newsletters or information alerts by registering their personal information. This provides firms with an opportunity to create a customer database, a principal benefit of direct marketing. On the other hand, consumer information privacy concerns with respect with PC or mobile Internet have become an important issue for policymakers, trade groups, and consumer advocates. For example, in

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

235

the U.S., SPAM increased by 38 percent from 2006 to 2007, to 1.1 million, and is expected to increase by 50 percent more to 1.5 million messages in 2008 (Cloudmark, 2008). Some of these unsolicited messages request users to provide personal information, including their credit card numbers, or attempt to infiltrate mobile devices with virus programs by asking users to register for services (CNET co.uk, 2006). To alleviate consumers concerns about these potential invasions, the mobile industry has implemented several self-regulations. For example, the Mobile Marketing Association recently revised its consumer best practices guidelines for cross-carrier mobile content programs to establish rules in several areas, such as the use of wireless carrier SMTP (Simple Mail Transfer Protocol) gateways for commercial traffic; handset delivery; clarifications of the guidelines regarding promotional content and marketing to children; and the expansion and clarification of guidelines for free, standard, and premium rate messaging, mobile Web, and interactive voice responses (Mobile Marketing Association, 2008). The U.S. Federal Trade Commission held a town hall meeting in May 2008 for business executives, consumer advocates, and scholars to explore the issues of mobile privacy and consumer protection (Corbin, 2008). In Japan, police may investigate spam, and violators may be sentenced to up to one year in prison and a fine of up to 1 million yen (Ministry of Internal Affairs and Communications, 2008). Traditionally, consumers’ information privacy concerns have been a central topic in organizational practice. In this light, Smith et al. (1996) developed a 15item scale to measure multidimensional nature of information privacy concerns. In this scale, information concerns were represented as a reflective second-order factor, connecting four first-order constructs, improper access to personal information, collection, errors, and unauthorized secondary use. This scale was empirically confirmed in various offline contexts (Campbell 1997; Milberg et al., 2000; Rose 2004; Stewart and Segars 2002). Privacy has converted into one of the key issues in computer-mediated communication. For example, Zeithaml et al. (2002) examined privacy as one of the key dimensions that consumers use to evaluate the quality of a Web site (or the quality of e-service). In a similar vein, Stewart and Pavlou (2002) recognize concerns about privacy/security as being crucial for thinking about the effectiveness of interactive media. Drawn upon the Smith et al.’s (1996) scale, Malhotra et al. (2004) developed Internet users’ information privacy concerns (IUIPC) scale, with an attempt to identify Internet-specific dimensions of privacy concerns, in contrast to traditional or offline direct marketing. Based on social contact theory, they proposed three first-order factors, namely, collection, control, and awareness of

236

Okazaki and Navarro-Bailón

privacy practices, and attempted to confirm a nomological network within the trust-risk framework. Collection is defined as the degree to which a person is concerned about the amount of individual-specific data possessed by others relative to the value of benefits received. This factor was derived from the SC’s principal of distributive justice, which assumes that, in an equitable exchange, consumers give up some information in return for something of value after evaluating the costs and benefits associated with the particular transaction. Thus, individuals will be reluctant to release their personal information if they expect negative outcomes. Control is defined as the degree to which a person can exercise freedom to either accept or reject the process or decision outcome related to his or her personal information. This construct was based on SC’s principle of procedural justice, which states that consumers want to exercise process control and influence changes in organizational policies they find to be objectionable. Awareness of privacy practices is defined as the degree to which a consumer is concerned about his or her awareness of organizational information privacy practices. This construct is based on two types of justices: interactional and informational justice. In the former, violating transparency and propriety of information leads to decreased perceptions of fairness. In the latter, perceptions of fairness increased with the specificity of information used to provide justification. Based on two empirical studies, their findings provided empirical support for the scale reliability and validity. Furthermore, nomological validity was established within a trust-risk framework, confirming that IUIPC is strongly and positively related to trust beliefs, which in turn negatively and directly influences risk beliefs. The ultimate consequence of information privacy concerns is behavioral intention to give private information to online merchants.

4

Hypotheses

The present study addresses three independent variables, two of which we have discussed in the previous section: spatial mobility and timing of incentive. The construct of spatial mobility is adapted from personal extensibility theory. Here, thanks to mobile device, consumers enjoy their “extended” mobility which is posited to influence attitudinal as well as behavioral aspects. Timing of incentive has been documented as a crucial element for the successful sales promotion or reward program. Immediate rewards are the benefits consumers receive at the point of transaction. Delayed rewards are the ones consumers receive at a later date from the point of sale. For example, Yi and Jeon (2003) found that consumers are likely to be attracted by immediate incentives, such as lotteries. Similarly, Keh and Lee (2006) suggest that delayed rewards with

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

237

higher value would build higher loyalty than would immediate rewards. One additional independent variable, the level of sensitivity in personal information, is defined as the consumers’ apprehension over firms’ misuse or abuse of clients’ personal data. Consumers are often requested to offer personal data by a promotional campaign. Often, however, their personal data are misused or disclosed by the third party without consumers’ prior consent (Malhotra et al. 2004). In this line, we primarily posit that the spatial mobility has direct and negative effect on the perceived level of information privacy concerns, in conjunction with timing of incentive and the sensitivity in personal information requested. In what follows, we formulate five specific hypotheses to be tested in this study. First, if one can receive promotional message at any place and any time, and asked to participate in the campaign by registering his or her personal information, one would first feel uneasy, thus increasing information privacy concerns. Accordingly, if he or she feels uncomfortable due to the spatial context, the intention to participate in the campaign would be low. Therefore: H1: High spatial mobility, compared to low spatial mobility, would lead to greater level of information privacy concerns. H2: High spatial mobility, compared to low spatial mobility, would lead to lesser level of intention to participate in the mobile campaign. Second, if the message delivers an attractive incentive that can be redeemed immediately, it may neutralize information privacy concerns, thus increasing the likelihood of participating in the campaign. More formally: H3: Immediate incentive, compared to the delayed one, would lead to lesser level of information privacy concerns. H4: Immediate incentive, compared to the delayed one, would lead to greater level of intention to participate in the mobile campaign. Third, if the campaign asks one to register sensitive personal information, he or she would perceive the high level of information privacy concerns, because of high risk associated with misuse of personal information. In such a case, the intention to participate in the campaign would be minimal. However, if other conditions collectively help him or her overcoming the primary concerns, the whole situation could change. Taken together:

238

Okazaki and Navarro-Bailón

H5: High sensitivity in personal information requested, compared with low sensitivity, would lead to greater level of information concerns. H6: High sensitivity in personal information requested, compared with low sensitivity, would lead to greater level of intention to participate in the mobile campaign. H7: There is an interaction effect among spatial mobility, timing of the incentive, and sensitivity in personal information requested.

5

Method

We test our hypotheses using data from a survey of mobile users in Japan during fall 2009. A professional research firm recruited participants from its online panel. After applying filter questions associated with mobile Internet usage and mobile campaign participation, we chose 640 participants whose demographic distribution approximately matches that of the general Japanese population. We chose a three-way between subject design, with a scenario-creation method (Keppel, 1991), in which we created 8 types of scenarios based on three independent variables: the level of spatial mobility (high versus low), timing of incentives (immediate versus delayed) (Keh and Lee, 2006), and the sensitivity in personal information requested (Okazaki et al., 2009). A basic scenario was: “You are reading a newspaper at home (or walking outside) and found a popular convenience store’s advertisement with QR-code. If you scan the code and access the campaign Website, you are asked to subscribe an email news alert by registering your personal information. By doing so, you will receive a free shopping voucher that can be redeemed immediate (or at a Christmas sales period).” Here, a Quick Response or QR-code is a two-dimensional barcode which a user can scan and automatically jump to a Website without typing in a URL. According to trade journals, almost 70% of mobile users have accessed QR-code to visit Websites or participate in promotional campaigns. A principal dependent variable was the level of information privacy concerns (Malhotra et al., 2004) and intention to participate in the mobile campaign. Respondents see only one of the 8 scenarios. Pretests confirmed the significant difference between the two levels of spatial mobility and the two levels of sensitivity in personal information requested. Accordingly, 640 participants were randomly assigned to 8 scenario situations (80 for each), and asked to read and complete the questionnaire. Missing data and outliers were eliminated from the dataset because they tend to distort the results (Hair et al., 2006). As a result, 579 usable data were singled out for the subsequent analysis.

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

6

239

Results

As a preliminary data analysis, we performed a confirmatory factor analysis that reveals an acceptable level of reliability and validity. Both composite reliability and average variance extracted exceed a generally recommended 0.70. Further, discriminant validity was confirmed for each dependent variable (Hair et al., 2006). Before testing the effects on the dependent variables, we assessed the success of the manipulation both for the degree of spatial mobility and the level of sensitivity in personal information requested by conducting several t-tests. Results showed that the difference in the degree of spatial mobility was statistically significant for the high M=4.72 and for the low M=4.36 degree of mobility (t=4.28, p<.001). Also, significant differences were obtained for the high M=5.24 and the low M=4.65 level of sensitivity in personal information requested (t=5.13, p<.001), in the manipulated direction. Because t-tests showed significant differences between conditions in the manipulated direction for both independent variables, manipulation checks were successful. A subsequent three-way ANOVA was conducted in order to test the effects on our dependent variables. Tables 1 and 2 show the descriptive statistics as well as ANOVA results for information privacy concerns and intention to participate in the mobile campaign, respectively. First, H1 and H2 posit the main effect of spatial mobility on information privacy concerns and the intention to participate in the mobile campaign. Specifically, in H1 we postulate that the greater spatial mobility would increase information privacy concerns. Our results suggest that the level of information privacy concerns is marginally but significantly higher in the high spatial mobility condition, compared to a low degree of spatial mobility [F (1, 571) =3.09, p=0.08]. Therefore H1 was supported. In H2, a similar hypothesis was formulated with regard to the effect of spatial mobility on the intention to participate in the opposite direction: the higher spatial mobility would decrease the behavioral intention. ANOVA also produced significant direct effect of this independent variable [F (1, 571) =4.22, p=0.04], which rings true to H2. H3 hypothesizes that timing of incentive also affects the level of information privacy concerns. More specifically, this level of information privacy concerns is lesser in an immediate incentive, compared to the delayed one. However, our results revealed that the main effect of timing of incentive was not statistically significant between both conditions [F (1, 571) =0.16, p=0.70]. Therefore, H3 was not supported. Similarly, H4 assumes that timing of incentive determines the intention to participate in the mobile campaign. Thus, we postulate that an immediate incentive would lead to greater level of behavioral intention.

240

Okazaki and Navarro-Bailón

Table 1: Descriptive statistics and ANOVA results for Information Privacy Concerns Descriptives

High Mobility N

High Sensitivity Low Sensitivity

Mean

Immediate

73 5.46

Delayed Immediate Delayed

Low Mobility SD

N

Mean 70 5.26

71 5.27

.97

71 5.08

.94

75 4.95

.92

72 4.92

1.01

74 5.12

.99

73 4.98

.81

K2

p

Degree of Mobility Sensitivity

2.94

3.09

.005

<.10

1 10.78

11.32

.019

<.005

Timing

1

.15

.15

.000

.69

Mobility x Sensitivity Mobility x Timing Sensitivity x Timing Mob x Sens x Timing Error

1

.46

.48

.001

.48

1

.09

.09

.000

.75

1

3.19

3.35

.006

<.10

1

0.09

.09

.000

.75

571

0.95

df

ANOVA

SD

1.14

MS 1

F

.96

Table 2: Descriptive statistics and ANOVA results for Intention to Participate Descriptives

High Mobility N

High Sensitivity Low Sensitivity

Mean

Immediate

73 4.47

Delayed Immediate Delayed

Low Mobility SD

N

Mean 70 4.67

71 4.24

1.84

71 4.82

1.77

75 4.94

1.60

72 4.81

1.43

74 4.94

1.40

73 5.39

1.32

K2

p

Degree of Mobility Sensitivity

1 11.09

4.21

.007

<.05

1 32.22

12.24

.021

<.005

Timing

1

2.22

.84

.001

.35

Mobility x Sensitivity Mobility x Timing Sensitivity x Timing Mob x Sens x Timing Error

1

1.90

.72

.001

.39

1

8.16

3.10

.005

<.10

1

3.91

1.48

.003

.22

1

.38

.14

.000

.70

571

2.63

ANOVA

SD

1.92

df

MS

F

1.52

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

241

In this sense, our evidences give not support for H4 because the results also showed that this direct effect was not statistically significant [F (1, 571) =0.85, p=0.40]. H5 and H6 assume that the level of sensitivity of personal information requested would also affect on our dependent variables. As we postulate in H5, high sensitivity in personal information requested would increase the level of information privacy concerns. Our results supported this proposition at p<0.001 [F(1, 571)=11.33]. In contrast, H6 posits that low sensitivity in personal information requested would decrease the intention to participate in the mobile campaign. The ANOVA results were consistent with our hypothesized direction because the high sensitivity in personal information requested led to the lesser intention to participate [F(1, 571)=12.25, p<0.001]. Finally, H7 suggests that there is an interaction effect among the three independent variables on information privacy concerns and the behavioral intention. However, on the contrary to our prediction, the three-way interaction effect was not statistically significant for information privacy concerns [F(1, 571)=0.095, p=0.75] and for the intention to participate in the campaign [F(1, 571)=0.147, p=0.70]. Thus, H7 was not supported by our data. However, despite this rejection of H7, our additional analysis indicates that there is a marginally significant two-way interaction between the sensitivity of information requested and the timing of incentive on consumers’ information privacy concerns. Therefore, the level of information privacy concerns is greater in a specific situation of high sensitivity and with an immediate incentive. Similarly, results suggest a marginally significant two-way interaction between the degree of spatial mobility and the timing of incentive on the behavioral intention, so consumers’ intention to participate in the mobile campaign is greater in a low degree of spatial mobility and for a delayed incentive.

7

Discussion

The present study addresses one of the most important questions in mcommerce: Do mobile users feel that their privacy has been protected? If not, what are the consequences? Our study tries to respond to this question, operationalizing ubiquitous context by proposing spatial mobility. Although ubiquity could be a more complex construct, our conceptualization could serve as a stepping stone for future research. Our quasi-experimental study offers rather mixed results. First, spatial mobility exhibits a modest direct effect on information privacy concerns, while

242

Okazaki and Navarro-Bailón

an interaction effect of the three independent variables was not significant. However, a closer examination of two-way interaction reveals that intention to participate in the mobile campaign was higher in high spatial mobility when immediate incentive was given. By contrast, the strong intention was observed in low spatial mobility when the delayed incentive was given. How can we interpret these results? The most reasonable explanation is that, perhaps, consumers are generally more comfortable in accessing the mobile campaign (via QR-code) at a stable place such as home, since they want to access the information when feeling relaxed. Otherwise, they may feel unfocused or stressed, thus are reluctant to analyze the content with much attention. Nonetheless, when they are outside and in a highly mobile situation, an immediate (and attractive) incentive may convince them to examine the information, up to the point where they do not mind provide personal information. Managerially, we could use such findings in planning more cost-effective mobile campaign. Mobile marketers and advertisers should be aware that a QRcode-printed hand-out campaign on a street may be a feasible alternative with an attractive give-away. This kind of campaign may be used as a part of crossmedia campaign, along with other print media, such as free papers, magazines, and newspaper insertions. At the same time, a responsible and honest treatment of personal information is extremely important, because, if the information practice of sponsor firms is not perceived as trustworthy, only few users would dare to access the QR-code campaign, thus generating little success.

8

Limitations

A few limitations should be recognized to make our findings more objective. First, the operationalization of spatial mobility was somewhat limited. Although space plays an important role in defining mobility, other contextual factors may need to be taken into account. For example, social anxiety in a crowded space has been found to affect consumer psychology. Future research should expand the definition in this regard. Second, the present study only focuses on a low involvement service: supermarket. Thus, any generalization of the results should be treated with caution, as the study did not contrast the results with a high involvement service (e.g., bank).

Who’s Afraid of Ubiquitous Promotion? Exploring the Effects of Privacy Concerns

9

243

References

Cloudmark (2008). Mobile Operators Brace for Global Surge in Mobile Messaging Abuse. February 11, available at http://www.marketwire.com/mw/rel_us_print.jsp?id=819426 (accessed March 28, 2008). CNET.co.uk (2006). Mobile spam spells trouble for text-based ads. November 16, available at http://news.cnet.co.uk/mobiles/0,39029678,49285278,00.htm (accessed April 4, 2008). iMedia Connection (2005a). McDonald's Goes Mobile. Creative Showcase. available at: http://www.imedia connection.com/content/7191.asp (accessed November 18, 2005). iMedia Connection (2005b). Ringing in a New Marketing Strategy. Creative Showcase. available at: http://www.imedia connection.com/content/6672.asp (accessed November 14, 2005). Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., & Tatham, R.L. (2006). Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall. Hägerstrand, T. (1975). Space, Time and Human Conditions. In Dynamic Allocation of Urban Space, A. Karlqvist, L. Lundquist, and F. Snickars, eds. Farnborough, UK: Saxon House, 3-14. Keh, H.T., & Lee, Y.H. (2006). Do reward programs build loyalty for services? The moderating effect of satisfaction on type and timing of rewards. Journal of Retailing, 82 (2), 127–136. Keppel, Geoffrey (1991). Design and Analysis: A Researcher’s Handbook. Englewood Cliffs, NJ: Prentice-Hall. Kleijnen, M., de Ruyter, K., & Wetzels, M. (2007). An assessment of value creation in mobile service delivery and the moderating role of time consciousness. Journal of Retailing, 83(1), 3346. Okazaki, S., Li, H., & Hirose, M. (2009). Consumer Privacy Concerns and Preference for Degree of Regulatory Control: A Study of Mobile Advertising in Japan. Journal of Advertising, 38(4), 63– 77. Mobile Marketing Association (2007). Mobile Marketing Industry Glossary. Mobile Marketing Association. available at: hhtp://www.mmaglobal.com (accessed March2, 2008). Malhotra, N.K., Kim, SS., & Agarwal, J. (2004). Internet Users’ Information Privacy Concerns (IUIPC): The Constructs, the Scale, and a Causal Model. Information Systems Research, 15(4), 336-355. Ministry of Internal Affairs and Communications (2008). Policies related to annoying emails prevention. available at http://www.soumu.go.jp/joho_tsusin/d_syohi/m_mail.html (accessed March 1, 2008) (in Japanese). Multinational Monitor. Commercial Alert: Advergaming. 2008(November/December), 56. Petty, R.D. (2003). Wireless Advertising Messaging: Legal Analysis and Public Policy Issues. Journal of Public Policy & Marketing, 22(1), 71-82.

Part V

Organization, Publicity and Reputation

Determinants of the Impact of Crises on Organizational Reputation: An Experimental Test of Crisis Communication Strategies and the Moderating Impact of Locus of Control An-Sofie Claeys, Ghent University, Belgium Verolien Cauberghe, Ghent University, Belgium Patrick Vyncke, Ghent University, Belgium

1

Introduction

Academic interest in the field of crisis communication has significantly increased over the years (Ulmer et al., 2007). Corporations are becoming more and more aware that crises can and will happen to organizations just like themselves (Benson, 1988). No company can escape all crises during its life cycle, and it is during these crisis situations that the battle to protect credibility in the eyes of stakeholders is most fierce (Hobbs, 1995). It takes many years to build a favorable organizational reputation but only one single crisis to ruin it (Dean, 2004). Since a good reputation is a central goal of every organization, there is a need to study the use of crisis response communication strategies in order to defend reputations during crises (Benoit, 1995; Coombs and Holladay, 2002). During the nineties, the image restoration theory focused on conducting case studies in order to create a typology of image restoration strategies (Benoit, 1997). The lack of experimental research however, resulted in the fact that little was known about how stakeholders react to these communication strategies, and how these reactions depend on the characteristics of the crisis type (Coombs, 2007). Therefore, Coombs (2007) developed a theory based on attribution theory to match these crisis response strategies to the specifics of the crisis situation. The Situational Crisis Communication Theory (SCCT) was developed to offer guidelines for crisis communication. The SCCT empirically studies which response strategies organizations should apply for which crisis type in order to restore the organization’s reputation in the best possible way. The main goal of this study is to experimentally test the matches between crisis types and crisis response strategies proposed by the SCCT. The study also looks deeper into the relationship between the severity of a crisis and the resulting post-crisis reputation. In addition, the moderating influence of a personality trait, locus of control, is examined. Little research has been conducted on perS. Okazaki (Ed.), Advances in Advertising Research (Vol. 2), DOI 10.1007/978-3-8349-6854-8_16, © Gabler Verlag | Springer Fachmedien Wiesbaden GmbH 2011

248

Claeys, Cauberghe and Vyncke

sonality traits that may affect the respondents’ reactions to a crisis or to the crisis response strategy used to manage the crisis (Coombs, 2007).

2

Situational Crisis Communication Theory

Based on the attribution theory, the SCCT was developed in order to provide crisis managers with guidelines to match crisis response strategies to crisis types (Coombs, 2007). The attribution theory focuses on the universal concern of individuals with explanation for what happened (Weiner, 2000). According to Kelley (1973), “Attribution theory is a theory about how people make causal explanations, about how they answer questions beginning with “why?””(p. 107). Individuals act like naïve scientists; searching underlying causes for events they observe (Kelley, 1973; Dean, 2004). Organizational crises are events in which individuals tend to make causal inferences and assess a certain amount of responsibility to the organization (Coombs, 2007). According to the SCCT, the reputational threat caused by crises, can be determined by looking at the amount of responsibility consumers attribute to the organization (Coombs, 2007, 2004). The SCCT has developed a list of 13 crisis types, based upon attributions of crisis responsibility that can be divided within three crisis type clusters (Coombs, 2007; Coombs and Holladay, 2002). The victim cluster can be defined by crises with weak attributions of organizational responsibility (e.g., product tampering). The accidental cluster involves crises with a certain, but low level of responsibility attribution to the organization (e.g., technical-error product harm). The preventable cluster incorporates crises for which the organization is perceived to be responsible (e.g., organizational misdeed with injuries) (cf. Table 1). According to the SCCT, the more responsibility that is accredited to the organization with respect to the crisis, the more negative the impact on the organizational reputation will be (Coombs, 1998): H1a: The victim crisis leads to the least negative effect on organizational reputation compared to the accidental and preventable crisis. H1b: The accidental crisis leads to a moderate negative effect on organizational reputation compared to the victim and preventable crisis. H1c: The preventable crisis leads to the most negative impact on organizational reputation compared to the victim and accidental crisis.

Determinants of the Impact of Crises on Organizational Reputation

249

Table 1: Match between Crisis Types and Crisis Response Strategies

Crisis types Victim cluster ƒ Natural disaster ƒ Rumor ƒ Workplace violence ƒ Product tampering/Malevolence Accidental cluster ƒ Challenges ƒ Technical-error accidents ƒ Technical-error product harm Preventable cluster ƒ Human-error accidents ƒ Human-error product harm ƒ Organizational misdeed with no injuries ƒ Organizational misdeed management misconduct ƒ Organizational misdeed with injuries

Crisis response strategies Deny strategies ƒ Attack the accuser ƒ Denial ƒ Scapegoat Diminish strategies ƒ Excuse ƒ Justification Rebuild strategies ƒ Compensation ƒ Apology

Source: Adapted from: Coombs, 2007, p. 168 - 170

These crisis types, differing in organizational responsibility, can be matched to three clusters of crisis response strategies, based on the responsibility of the organization taken in each of them (cf. Table 1) (Coombs, 2007). When using deny strategies, the organization rejects all responsibility for the crisis. Diminish strategies minimize organizational responsibility or crisis damage. With rebuild crisis strategies organizations admit full responsibility. Research has shown that apology, a rebuild crisis response strategy, leads to a more effective reputation repair compared to deny and diminish strategies (Coombs and Holladay, 2008). Therefore, we expect: H2: The reputation of an organization using rebuild crisis response strategies will be more positive than the reputation of an organization using either deny or diminish crisis response strategies. The attribution theory provides the rationale for the relationship between crisis types and crisis response strategies (Coombs, 2007). Crisis managers should select crisis response strategies that are appropriate for the amount of potential reputational damage a certain crisis may inflict (Coombs and Holladay, 2002). Based on these assumptions, the SCCT provides specific guidelines for match-

250

Claeys, Cauberghe and Vyncke

ing responses to crisis situations (Coombs, 2007). Table one shows that deny strategies can best be used in case of a victim crisis, diminish strategies are suited for accidental crises and rebuild strategies offer a way to defend reputations in case of a preventable crisis. Coombs and Holladay (1996) found in their experimental study that when crisis communication responses match the crisis type in terms of responsibility attribution, this leads to a more positive reputation than either no response or a mismatched response. They examined the match between a crisis type and one single crisis response strategy. However, Benoit (1997) suggests that the use of a combination of multiple strategies can increase the effectiveness of the image restoration. Based on guidelines from the SCCT we expect that deny strategies match with victim crises, diminish strategies match with accidental crises and rebuild strategies match with preventable crises (Coombs, 2007). These assumptions lead to the following: H3: Matches between crisis type and crisis response strategy lead to a less negative organizational reputation than mismatches between crisis type and crisis response strategy.

3

The Relationship between Crisis Severity and Post-Crisis Reputation

According to the SCCT, more severe crises have a more negative impact on organizational reputations than crises with trivial damage (Coombs, 1998; Coombs and Holladay, 2002). Crisis severity has proven to modify individuals’ evaluations of certain crisis types (Coombs and Holladay, 2002). Former research (Coombs, 1998) showed that crisis severity has a negative impact on post-crisis reputation in case the organization is fully responsible for the crisis. If on the other hand consumers consider the company itself as a victim of the crisis, crisis damage will not negatively impact post-crisis reputation (Coombs, 1998). Based on the current typology of crisis types, offered by the SCCT (Coombs, 2007), this implies that organizations confronted with a victim crisis, do not need to fear that crisis damage will have a negative impact on post-crisis reputation. In case of an accidental or preventable crisis on the other hand, postcrisis reputations will suffer more when the crisis severity is perceived as higher by individuals. H4a: The perceived severity of a crisis has a negative impact on the organizational reputation if the organization has some (accidental crisis) or full (preventable crisis) responsibility for the crisis

Determinants of the Impact of Crises on Organizational Reputation

251

H4b: The perceived severity of a crisis is not related with the organizational reputation if the organization has no responsibility for the crisis

4

Moderating Influence of Locus of Control

The moderating impact of personality traits has been rarely investigated within the domain of crisis communication. Literature in traditional marketing and consumer behaviour shows a variety of research in the field of personality traits, and a growth of interest in the identification of personality variables that influence consumer behavior and attitudes (Dutta and Vanacker, 2000). Crisis communication research equally needs to consider personality traits that might be of interest considering stakeholders during crises. Ahluwalia et al. (2000) claim that there has been no systematic research of how consumers process negative information about brands. In order to extend the knowledge of SCCT, research must determine how stakeholders’ perceptions of organizational reputations after crises are moderated by their personality characteristics. An individuals’ locus of control is of interest given that this personality trait builds on the attribution theory, on which an important part of the SCCT framework is based (Beretvas et al., 2008). Deny strategies make consumers believe that a crisis is caused by an attribute of the environment (external attribution) (Coombs, 2007). Rebuild strategies on the other hand argue that a crisis is caused by an attribute of the actor/company (internal attribution) (Collins, 1974). Diminish strategies are somewhere in between, at the same time taking some of the responsibility and rejecting part of it. Locus of control on the other hand, relates to whether an individual has the perception to have power over what happens to him or her (internal locus of control), or attributes it to external factors (external locus of control) (Lefcourt, 1966). Rotter (1966) stated that a person’s locus of control may correlate with the value the subject places on either internal or external control in others. If Rotter’s (1966) presumption is true, internals would assign more value to organizations that take full responsibility for their own actions (using rebuild strategies), thus stating that they control their own actions. This would equally imply that externals might assign more value than internals to organizations that reject responsibility for what happened (using deny strategies), thereby stating that the environment controlled what happened more than the organization itself. This leads to the following hypotheses:

252

Claeys, Cauberghe and Vyncke

H5a: When an organization uses deny strategies, its reputation will be perceived as less negative by respondents with an external locus of control than by respondents with an internal locus of control H5b: When an organization uses diminish strategies, its reputation will be perceived as equal by respondents with an external locus of control and respondents with an internal locus of control H5c: When an organization uses rebuild strategies, its reputation will be perceived as less negative by respondents with an internal locus of control than by respondents with an external locus of control

5 5.1

Method Design and Stimuli

A 3 (crisis type) x 3 (crisis response) between-subjects factorial experimental design was used to investigate the hypotheses. Crisis type and crisis response strategies were manipulated using different scenarios. Crisis type was manipulated by the selection of one crisis from each of the three clusters (victim cluster, accidental cluster, preventable cluster). To manipulate the response strategy, the combination of the two strategies within each cluster was used. Only one crisis response strategy was used from the deny cluster, since all three deny strategies tend to conflict with one another. Combining strategies will only enhance their individual impact when the responses are compatible (Huang, 2006).

5.2

Participants and Procedure

Data were collected from 316 respondents using an online questionnaire. The respondents were randomly divided across the 9 experimental conditions and were instructed to read a scenario for a fictitious juice company. Participants were Dutch-speaking Belgian men and women with an average age of 35 (S.D. = 14.46; range 13 to 70 years). About 47% were male and 53% were female.

5.3

Measures

Organizational responsibility for a crisis was measured using the four items 10point Likert scale of Griffin et al. (1992) (e.g., “How responsible was the organization with respect to the crisis?”) (α = .81).

Determinants of the Impact of Crises on Organizational Reputation

253

Organizational reputation was measured by a combination of the Reputation Quotient (RQ) scale of Fombrun et al. (2000) and McCroskey’s scale of credibility (Cited in: Coombs and Holladay, 1996) (α = .97). Crisis damage was held constant across scenarios by telling respondents that each crisis caused the death of two adults. To measure the possible effects of the perceived crisis severity, respondents were asked to answer one question on a 10-point scale, ranging from one (not at all severe) to 10 (very severe): “How severe do you consider the damage caused by this crisis?” Locus of control was measured using the internal-external (I-E) locus of control scale of Rotter (1966). The I-E scale had a Cronbach α of .73 and consisted of nine factors with eigenvalues higher than 1.0. One of these factors, the misfortunes dimension, considers the perceived causes of peoples’ misfortunes. Since the I-E scale has shown to oversimplify the multidimensionality of the locus of control construct and since the misfortunes dimension is closely related to the attribution theory, this dimension was taken separately into analyses (Duffy et al., 1977).

5.4

Pre-tests

Two pre-tests were conducted to test the manipulated variables, crisis response and crisis type. In the first pre-test, 12 respondents were instructed to read each of the three crisis response scenarios, and order them based on a list containing all five crisis response strategies. Some adaptations were made in the scenarios based on these results. The second pre-test assessed the manipulation of corporate responsibility for each crisis type. Twenty-one respondents participated in a within-subjects design. A seven-point Likert scale (Griffin et al., 1992) was used. In the victim crisis (M = 4.31, S.D. = 1.49) the company was perceived as less responsible than in the accidental crisis (M = 5.31, S.D. = .98, t (40) = 2.56, p = .015) and the preventable crisis (M = 5.79, S.D. = 1.01, t (40) = 3.75, p = .001). However, no significant differences occurred between the accidental (M = 5.31, S.D. = .98) and preventable crisis (M = 5.79, S.D. = 1.01, t (40) = 1.55, p = .13). All scenarios were slightly adapted.

254 6 6.1

Claeys, Cauberghe and Vyncke

Results Manipulation Check

A one-way ANOVA was performed to check for the manipulation of crisis type, measured by the amount of crisis responsibility (Griffin et al., 1992). The mean difference in crisis responsibility between the three crisis types was significant (F (2, 313) = 82.93, p < .001). The Scheffé follow-up procedure showed that the victim crisis (M = 4.78, S.D. = 2.11) differed significantly from the accidental crisis (M = 6.81, S.D. = 1.87) (p < .001) and from the preventable crisis (M = 8.07, S.D. = 1.62) (p < .001). Furthermore, the accidental crisis differed significantly from the preventable crisis (M Accidental crisis = 6.81, S.D. = 1.87; M Preventable crisis= 8.07, S.D. = 1.62) (p < .001).

6.2

Impact of Crisis Type and Response Strategy on Reputation

To address the main effects of crisis types (H1) and response strategies (H2) on organizational reputation and to test the interaction effect of crisis type and crisis response strategies as hypothesised in H3 on reputation, a univariate twoway ANOVA (general linear model) was used. Two main effects occurred. Both crisis type (F (8, 307) = 94.72, p < .001), and crisis response strategy (F (8, 307) = 4.68, p = .01), had a significant main effect on organizational reputation. In order to detect the significant differences between the crisis types and crisis response strategies pairwise, one-way ANOVA’s with Scheffé follow-up tests were conducted. The results of a separate one-way ANOVA (F (2, 313) = 95.33, p < .001) show that individuals perceive the organization’s reputation as less favourable in case of a preventable crisis (M = 2.72, S.D. = 1.42), compared to both the victim (M = 5.95, S.D. = 1.97) (p < .001) and accidental crisis (M = 5.48, S.D. = 2.04) (p < .001) (cf. Figure 1). No significant difference in reputation occurred between the victim crisis (M = 5.95, S.D. = 1.97) and the accidental crisis (M = 5.48, S.D. = 2.04; p = .18). These results support H1c, and partially support H1a and H1b.

Determinants of the Impact of Crises on Organizational Reputation

255

Figure 1: Main Effects of Crisis Types on Post-crisis Reputation

The main effect of crisis response strategy on organizational reputation was confirmed (F (2, 313) = 4.52, p = .012) (cf. Figure 2). The Scheffé procedure revealed that rebuild strategies (M = 5.26, S.D. = 2.36) significantly lead to a more positive reputation than diminish strategies (M = 4.35, S.D. = 2.24) (p = .018). The difference between the rebuild strategies and deny strategies was only marginally significant (M Rebuild strategies= 5.26, S.D. = 2.36; M Deny strategies = 4.57, S.D. = 2.28, p = .087). Therefore, H2 is partially supported. The interaction-effect of crisis type and crisis response strategies on reputation was not significant (F (8, 307) = 1.28, p = .28). Therefore, H3 could not be supported. The effects of the crisis response strategies did not depend on the specifics of the crisis situation and therefore no proof was found for the guidelines offered by the SCCT to match crisis response strategies to crisis types.

256

Claeys, Cauberghe and Vyncke

Figure 2: Main Effects of Crisis Response Strategies on Post-crisis Reputation

6.3

Relationship between Crisis Severity and Post-Crisis Reputation

The fourth set of hypotheses concerned the relationship between the severity of a crisis and the organization’s post-crisis reputation. As expected, a significant negative correlation appeared between the severity of the crisis and organizational post-crisis reputation in case of both an accidental (r (103) = - .312, p = .001) and a preventable crisis (r (104) = - .308, p = .001). These results support H4a. However, the perceived severity of a crisis is not related with the organizational reputation in case of a victim crisis, for which the organization bears no responsibility what so ever (r (109) = - .04, p = .68). These results support H4b.Therefore, the severity of a given crisis will only correlate negatively with the subsequent organizational reputation, in case consumers find that the organization itself was at least partially responsible for the events.

6.4

Moderating Influence of Locus of Control

Hypotheses 5a, 5b and 5c were tested by means of a univariate two-way ANOVA (general linear model). A significant interaction-effect occurred between crisis response strategy and the locus of control (misfortunes dimension) (F (5, 310) = 3.82, p = .023). The plot (Cf. figure 3) shows that when a company uses a deny response strategy, its reputation is less negative for individuals with an external locus of control (M = 5.30, S.D. = 2.15) than for individuals with an internal locus of control (M = 3.90, S.D. = 2.20) (t (105) = 3.33, p = .001). Therefore, H5a is supported. When an organization uses diminish strategies

Determinants of the Impact of Crises on Organizational Reputation

257

however, its reputation is perceived as equally positive by both externals (M = 4.27, S.D. = 1.91) and internals (M = 4.44, S.D. = 2.53) (t (99) = .38, p = .70), which supports H5b. H5c was not supported.

Figure 3: Interaction between Crisis Response Strategies and Locus of Control on Postcrisis Reputation

7

Discussion and Conclusions

Two main effects occurred of crisis type and crisis response on organizational reputation. Corporate reputation was least favourable when organizations are confronted with a preventable crisis, as compared to an accidental or victim crisis. These results support the SCCT (Coombs, 2007), since preventable crises lead to more attributions of responsibility and subsequently d