18.Education

18• Education

18• Education Computer-Aided Instruction Abstract | Full Text: PDF (266K) Computer Engineering Education Abstract | Full Text: PDF (87K) Continuing Education Abstract | Full Text: PDF (111K) Educational Trends Abstract | Full Text: PDF (122K) Electrical Engineering Curricula Abstract | Full Text: PDF (71K) Electrical Engineering Education Abstract | Full Text: PDF (122K) Intelligent Tutoring Systems Abstract | Full Text: PDF (168K) Management Education Abstract | Full Text: PDF (68K) Professional Preparation Abstract | Full Text: PDF (155K)

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20E...AND%20ELECTRONICS%20ENGINEERING/18.Education.htm17.06.2008 21:40:13

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2901.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Computer-Aided Instruction Standard Article Marion O. Hagler1 and William M. Marcy2 1Texas Tech University, Lubbock, TX, 2Texas Tech University, Lubbock, TX, Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2901 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (266K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2901.htm (1 of 2)17.06.2008 21:41:12

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2901.htm

Abstract The sections in this article are Learning Environments on Desktop Computers Learning Environments on Networks Appropriate Application of Computer-Aided Learning Environments Creation of Learning Environments | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2901.htm (2 of 2)17.06.2008 21:41:12

648

COMPUTER-AIDED INSTRUCTION

COMPUTER-AIDED INSTRUCTION Learning engineering, as a student or as a practitioner, has always required intense participation by the learner. Engineering faculty traditionally have assigned homework problems, conducted problem sessions, answered questions, and tutored students one-on-one, and of course given quizzes, all as means of providing feedback for their students, and thereby increasing the intensity of interactivity in the learning environment. Despite the recurring impression that laboratories mainly develop practical technical skills in students,

the fundamental purpose of laboratories arguably is to provide students with an intensely interactive learning environment in which feedback occurs realistically and, often, immediately. Although these traditional routes to providing interactive learning environments have proved effective (some more than others), it is easy to recognize two difficulties. First, most of these interactive approaches, especially those that provide prompt feedback, require the learner and the teacher to meet together for the approaches to be effective. This requirement for spatial and temporal coincidence of learners and teachers is particularly inconvenient because much, if not most, learning in typical engineering courses occurs outside the classroom and beyond the presence of the teacher. A second difficulty, not independent of the first, is that providing extensive interactivity to students is labor intensive and hence quite expensive. In view of these constraints, engineering educators began to envision how to provide on-site on-demand interactivity through the use of computers almost as soon as practical computers became available. Looking into the future of engineering education in 1962, for example, W. L. Everitt, Dean of Engineering at the University of Illinois, foresaw the time when every student and practicing engineer could access interactive and adaptive learning environments through inexpensive personal computers of moderate capacity connected, as needed, to more complex computers by wire or radio (1). Dean Everitt’s vision was not developed in isolation. In December 1961, the IRE Transactions on Education had published a special issue, guest edited by Mager (2), on automated teaching with contributions that addressed the questions, ‘‘Why Automate Instruction?’’ and ‘‘How Effective Are the New Auto-instructional Materials and Devices?’’ and presented ‘‘A Rational Analysis of the Process of Instruction’’ and ‘‘A Method for Preparing Auto-instructional Programs’’ (2). These questions and issues remain contemporary. Doubtless, Dean Everitt was influenced especially by the programmed logic for automatic teaching operation (PLATO) project on his own campus, led by Professor Donald L. Bitzer in the Department of Electrical Engineering (3). In 1960, Bitzer demonstrated the first version of PLATO as one terminal connected to Illiac I (2–6). The initial stated purposes of the project were to investigate the potential role of the computer in the educational process and to design an economically and educationally feasible educational system. Bitzer and his colleagues were therefore among the first to address what remains the fundamental question in computer-aided instruction: ‘‘How can we use computers to improve education effectively and inexpensively?’’ (In this article, computeraided instruction is a broad term that encompasses almost any reliance on computers and networks in learning environments. It encompasses such terms as computer-assisted instruction and computer-based learning.) From the beginning, PLATO served both on-campus and distance learners. By 1975, the PLATO IV system consisted of over 900 terminals at 146 different sites, some across the United States and Canada and some on campus. The PLATO terminals consisted of a special transparent plasma display panel (512 ⫻ 512 dot matrix) with a touch screen that could superimpose computer-generated output on photographic slides or movies projected through the rear. Audio and even laboratory apparatus could be incorporated, as well. PLATO

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

COMPUTER-AIDED INSTRUCTION

649

Figure 1. A learner interacts with the seminal PLATO network (PLATO IV) through a graphical interface and a touch screen, circa 1975. (Courtesy of Prof. Donald L. Bitzer.)

IV, funded in part as a demonstration project for computerassisted learning by the National Science Foundation (NSF) and commercialized by Control Data Corporation, ultimately offered approximately 8,000 h of instructional material, prepared by about 3,000 authors, in subjects that included electrical engineering, computer science, classical mechanics, accounting, astronomy, geometry, biology, chemistry, algebra, foreign languages, law, medical sciences, library science, agronomy, and elementary reading (Fig. 1). The scope of the PLATO vision is indicated by the proposal for 1,000,000 terminals, most in elementary and secondary schools, in a PLATO V system by 1980–81. To attract enough users to be economically feasible, PLATO V was to include email, on-line library card catalogs, graphics, and games, as well as access to on-line computation and interactive learning environments. As far as functionality is concerned, the PLATO V proposal resembles an early version of the World Wide Web (WWW), though confined to a few mainframes and, therefore, tiny in size by comparison and more starlike in its connective topology. PLATO V was never implemented, however, because of two evident concerns. The simplest to address is cost. From very early in the PLATO project, the goal had been to provide learning through the use of computers at costs comparable to the cost of classroom instruction. The PLATO strategy was to employ centralized mainframes for the necessary computing power and rely on relatively simple terminals, connected to the mainframes by high-speed telephone links, to give students access to the PLATO learning environments. This strategy sought to exploit the economies of scale available at that time in purchasing and maintaining large computers in comparison with those costs for smaller computers. The PLATO strategy became less appropriate, however, as communication costs proved larger than expected and minicomputers, first,

then personal computers undermined the cost and capability advantages of mainframes. Simply stated, PLATO proved to be too expensive. The second concern was that lessons for use on PLATO, and another large computer-aided learning NSF demonstration project known as time-shared, interactive, computer-controlled information television (TICCIT), took too long to prepare. When completed and implemented, most failed to achieve the results anticipated for interactive computer learning environments (5,6). Despite notable exceptions (6,7), much of the educational material developed for PLATO and TICCIT used the computer mainly to check specific answers entered as a learner moved along an inflexible learning path. Such a format, mainly drill and practice, was straightforward to program, but achieved little of the complex interactive learning environments good teachers were accustomed to providing, albeit labor intensively, for their students. For their part, learners found much of the software for PLATO and TICCIT boring. The most worrisome problem in developing computer-aided instructional materials for PLATO and TICCIT was that psychological and educational theories at that time gave little guidance about how to use computers to construct intensively interactive learning environments. The disappointing results from these large and well-funded projects, as well as disappointment in the richness of the educational materials they produced, was thus doubly frustrating because the available theoretical framework offered few suggestions about how results might readily be improved. After the reported investment of more than $900 million by Control Data Corporation, PLATO never became profitable (5). PLATO nevertheless represented a grand vision of what might be. Some forces and concepts it spawned and the continuing elaboration and development of its vision were central

650

COMPUTER-AIDED INSTRUCTION

to developments in computer-aided instruction decades later. Some, for example, point to the interaction between thousands of PLATO authors and the PLATO staff as the first sizable on-line community (8). As far as computer-aided instruction is concerned, however, the main legacy of PLATO (and TICCIT) is the beginning of focused and continuing efforts to address the fundamental question, ‘‘How can we use computers to improve education effectively and inexpensively?’’, still the central question for computer-aided instruction. The question never has been answered definitively and, even if it had been, the answer would have been short lived, outdated by rapid developments in technology and in the improving understanding of the nature of learners who use computers in the hope of better progress. Grappling with a question whose answer changes constantly is a task not all would favor to address. A confluence of economic, social, and political factors, however, makes the application of computers in education increasingly inevitable and therefore makes addressing the fundamental question, first addressed by the PLATO workers, inevitable as well (9,10). Contemporary approaches to the fundamental question, fortunately, benefit considerably from the very factors, developments in technology and improving understanding of learners, that drive changes in the answer. The following sections explore developments on these fronts.

LEARNING ENVIRONMENTS ON DESKTOP COMPUTERS The emergence of desktop computing can be viewed as a firestorm much like the one for automobiles and roads initiated by Henry Ford’s Model T automobile. Early desktop computers relied upon inexpensive mass-produced components to achieve modest functionality at low cost. Just as people often preferred independent control of a personal automobile to typically speedier mass transportation, people often preferred a desktop computer that they controlled to the inconvenience of accessing the power and speed of a mainframe. The firestorm effect occurred as increasing demand for desktop computers drove prices down and functionality up, which in turn attracted numerous talented authors of software whose software increased demand even more. Perhaps the most important impact of desktop computing on computer-aided instructional was to increase dramatically the number of workers and hence the amount of experimentation. The availability of increasingly powerful and inexpensive desktop computers led to almost startling penetration of computing into industries and universities, as well as elementary and secondary schools and homes. The sheer number of desktop computers, far larger than the seemingly incredible 1,000,000 terminals in the unimplemented PLATO V proposal, presented unprecedented opportunity for computeraided instruction. However, development of computer-aided instruction with desktop computers, beginning in the late 1970s near the end of the PLATO and TICCIT projects, also benefited little from psychological and educational theories as far as guidance about how to construct effective learning environments with computers was concerned. As a consequence, the learning environments produced for desktop computers suffered basically the same complaints and criticisms as those directed at mainframe learning environments: they were either ineffective or effective only unpredictably, they were

mainly drill and practice and hence neglected a large number of important interactive learning possibilities and, perhaps most damning, both students and teachers often found them boring once the newness wore off. Earlier efforts to strengthen the theoretical foundations of computer-aided instruction continued and gained momentum with the participation of more people, attracted by the tantalizing possibilities of desktop computers (11). In practice, much of the work that produced materials for computer-aided instruction on desktop computers, especially in engineering, simply ignored theory and relied instead on authors’ intuitions, not necessarily a bad approach under the circumstances. In engineering education, two broadly applicable tactics emerged: drill and practice and simulation. Drill and Practice The drill and practice approach, popular during the PLATO and TICCIT projects, uses the computer for direct checking of specific responses entered by learners. The sheer volume of computer-aided instructional material that abuses the drill and practice technique, admittedly of limited power and applicability, has given the drill and practice approach a reputation that is worse than it deserves. Drill and practice exercises often can help learners achieve rudimentary competence in a subject before moving on to its more challenging aspects. Contrary to abundant folklore, drill and practice exercises need not be boring. In electrical engineering, a drill and practice program that is widely used with good effectiveness is CircuitTutor
, developed by Oakley and marketed by a commercial publisher (12). CircuitTutor integrates drill and practice exercises into tutorials on various aspects of elementary circuit theory, as typically taught in the first course on circuits in an electrical engineering undergraduate curriculum. After selecting the tutorial on writing node equations, for example, learners either can choose to enter numerical answers to a number of problems with parameter sets generated by the computer or can choose a step-by-step approach that begins with their entering the appropriate node equations in symbolic form. In the step-by-step route, the tutorial leads the student to write the equations in symbolic form, to find numerical solutions to the equations and, perhaps, to use the results to find the maximum power that can be supplied by a pair of terminals of the circuit. In the step-by-step approach, incorrect answers must be corrected before proceeding to the next step. Even this brief description of CircuitTutor reveals a degree of flexibility and sophistication that helps account for its widespread acceptance and success despite the inescapable limits of its drill and practice approach. In a portion of ELECSIM, a learning environment for undergraduate electrical engineering students enrolled in the first course in analog electronics, Marcy and Hagler have extended the drill and practice approach to evaluation of answers to problems that do not have unique answers (13). Specifically, learners enter component values for a specified circuit (such as a voltage amplifier) that cause the circuit to meet computer-generated performance specifications (frequency response, gain, and input and output impedances, for instance) and to satisfy certain design rules available to the learner. In one example, the computer checks that the transistors in an audio amplifier are appropriately biased (according to the design rules), that the open-circuit voltage gain

COMPUTER-AIDED INSTRUCTION

and the output resistance for the amplifier are appropriate for the design constraints, and, again according to design rules, that coupling capacitors are neither too large nor too small. If the learner enters parameter values that lie outside ranges required by the design rules, advisory messages to the learner appear on the screen. Checking the consistency of entries with a set of design rules in this manner requires no more than simple branching constructs in the software. Simulation The simulation approach to computer-aided instruction often has relied on simulation software to simplify a complex process so that a learner could concentrate on only a few salient aspects during the learning process. In engineering, laboratory simulations have proved useful in preparing students for physical laboratories, although it is possible to use them apart from a physical laboratory as well. Mosterman, Campbell, Brodersen, and Bourne developed the electronic laboratory simulator (ELS) system, for example, which includes simulations of nine laboratories for a course in electronics (14). Topics range from Thevenin’s theorem to operational amplifier implementations of multiple-pole filters. In each laboratory, learners assemble circuits for simulation by dragging and dropping graphically realistic components onto a graphically realistic breadboard. Learners make measurements on the circuit by connecting graphically realistic instruments, such as an oscilloscope, to the circuit. Learners use their mouse to adjust the instrument controls on a closeup image of the instrument. Each laboratory includes an introductory tutorial and context-sensitive on-line help. More generally, powerful and realistic simulators can enrich the learning environment by helping the student to learn self-assessment of solutions to open-ended problems. The circuit simulator PSpice, for example, is used widely in both industry and academia for simulation of analog and mixed analog and digital circuits. A free evaluation version limited to small circuits is available for use by students on their personal computers. Using PSpice, a student can check the performance of a design attempt against design specifications and revise the design as necessary to achieve the specified performance without direct involvement of the teacher. The learner not only receives immediate feedback about the success of the design (at least within the accuracy limits of the simulator), but also gains invaluable experience in selfevaluation of work on open-ended problems that have no unique solution. The iterative interplay between the learner and the simulator creates a powerful real-time interactive learning environment without real-time participation by the teacher. Any topic for which a useful simulator is available is a good candidate for interactive instruction in engineering education. Possibilities beyond circuit and logic simulators include any numeric simulation software, as well as compilers (for computer language instruction), which, strictly speaking, are not simulators but indeed the real thing. Symbolic manipulation programs are, in a sense, symbolic simulators in that learners can compare their results from symbolic calculations with the results from the symbolic software. In circuits, for example, a Thevenin equivalent circuit might be derived analytically by the learner and checked with a symbolic manipulation program. If the result is incorrect, then the learner can

651

reprise the derivation until the disagreement is resolved. The learners thus take responsibility for assessing and correcting their work, a most valuable and productive activity. Engel developed SPLICE software that combines the schematic capture capability of PSpice and the symbolic manipulation capability of Waterloo Maple V to provide a simple means for learners to obtain analytical answers with which to compare their own results (15). By simply drawing a circuit schematically and selecting a pair of terminals, SPLICE generates mathematical expressions for the Thevenin voltage and impedance of the circuit for those terminals, for example. SPLICE thus provides learners a convenient means of evaluating their own symbolic analyses of circuits. Graphical User Interfaces and Multimedia In addition to multiplying the number of people exploring the use of computers in teaching, the emergence of powerful inexpensive desktop computers with graphical user interfaces (GUIs) and multimedia capabilities has transformed the learner interface to computers. The learner interaction offered by contemporary desktop computers was simply unthinkable during PLATO’s heyday. Furthermore, the extensive application of desktop computers for presentations in the business community has stimulated the development of powerful and inexpensive software tools for easy creation of complex interactive presentations. The difficulty is that this power can be dangerous for creators of learning environments. Just as in the early days of desktop computing the newfound ability to experiment with countless fonts led to rampant publication of nearly unreadable newsletters in the proverbial ransom note font, the multimedia capabilities of contemporary desktop computers can ruin the effectiveness of learning environments by confusing and distracting, rather than supporting, both creators and learners. Avoiding such pitfalls in developing a learner interface is a matter of designing to provide as much richness as the learning environment can support without confusing the learner. Fortunately, principles of graphical user interface design have emerged that can guide the development of effective interfaces (16,17). Perhaps the best simple rule in interface design is to make an interface to the learning environment that focuses the learner’s attention on the learning environment, not the interface. If the interface either distracts the learner with unnecessary bells and whistles or obstructs attention by frustrating or confusing the learner, then the learner interface is a failure and must be reworked. Electronic Books and Tutorials Electronic books and tutorials for desktop machines typically rely on some combination of drill and practice, simulation, and multimedia to provide learners a considerably more complex learning environment than is possible with printed material. Doering, for example, has combined circuit simulation, video, audio, graphics, and text in a tutorial that helps learners visualize the dynamic behavior of circuits (18). Harger constructed a hyperlinked interactive book in which Mathcad, an inexpensive but powerful and widely used commercial mathematical analysis package, is used to achieve an interactive learning environment for a course in digital signal processing (19). Learners can experiment by modifying the examples and seeing the consequences immediately. Wood

652

COMPUTER-AIDED INSTRUCTION

developed interactive tutorials for digital logic design, digital signal processing, and engineering mathematics that aim specifically at helping learners integrate conceptual fragments and thereby achieve a deeper understanding of basic concepts (20).

LEARNING ENVIRONMENTS ON NETWORKS Although a number of successful interactive engineering learning environments for stand-alone desktop computers became widely available, the collection of available environments failed to exhibit anything like the variety in interactive strategies traditionally provided to students by teachers of engineering. On a more pragmatic level, developers of interactive learning environments for desktop computers encountered two inhibiting limitations. First, a learning environment on an isolated desktop machine, or even one on a local area network, could access and incorporate only a limited variety of resources in comparison with a mainframe environment. Second, the mixture of MS-DOS Windows, Macintosh, and UNIX operating systems on desktop computers required preparation of multiple versions of a learning environment if it were to be readily accessible to the majority of desktop machines. These limitations significantly impeded the development of interactive computer-aided instructional environments for desktop computers. The explosive development of networks, especially the Internet and the WWW, provided a widely available means of addressing the limitations of both isolation and operating system incompatibility for desktop computers. For example, the Internet permits users of Oakley’s CircuitTutor tutorial software described earlier to submit homework assignments and quizzes electronically to an Internet server, which automatically runs a grading program and returns the results to the user (12). Bulletin board software permits students, faculty, and teaching assistants to discuss problems and questions related to the class and therefore function as a virtual learning community. The availability of unlimited access to the Internet at low monthly rates, of inexpensive powerful desktop computers, and of multimedia WWW browsers for all popular operating systems means that an increasing fraction of students and practicing engineers have ready access to the WWW, regardless of the operating system for their machine. The WWW thus provides a de facto standard for writing and providing access to computer-aided learning environments, independent of the operating system. Such a cross-platform standard provides access for a large number of potential users and hence establishes a strong incentive for the development of learning environments (21). The number of potential users on the WWW dwarfs even the erstwhile incredible 1,000,000 terminals envisioned for PLATO V. Moreover, it makes available a variety of resources for incorporation into computer learning environments that is larger even than that available, before, on mainframes. At the same time, the WWW restores the promise, dating from the days of PLATO, of efficiency that results from using the same computer-aided instructional materials to provide on-site on-demand access to complex learning environments for both on-campus and distance learners.

Because of these advantages, WWW-based tutorials and interactive learning environments began to appear in the mid-1990s. Schodorf, Yoder, McClellan, and Schafer, for example, established a WWW home page for a digital signal processing course taken as the first course in electrical engineering by students at Georgia Institute of Technology (22). The home page gives students access to demonstrations with video and audio files, Matlab quiz problems for review and drill, and interaction with each other via a newsgroup devoted entirely to the course. Material from the WWW site for the course, recorded on a CD-ROM that can be viewed with a WWW browser, accompanies a textbook (23). Sears and Watkins developed a multimedia manual for a telecommunications setup and placed it on the WWW (24). The manual makes extensive use of hypertext markup language (HTML) image maps to permit users to click on a component in a photograph to access, for example, a close-up view of a printed circuit board from a different perspective, as well as obtain further technical details about the component. An interesting feature is the possibility of incorporating on-line information made available by the manufacturer of the equipment. CyberProf was initiated at the University of Illinois by Hubler in 1994 with the notion of updating and expanding the functionality of PLATO through WWW technology and a more robust and intelligent human-computer interface based on complexity theory (25). Specifically, CyberProf is designed to support teachers’ efforts to integrate lecture notes, laboratories, and homework into a cohesive package. Students, who access the learning materials with a standard WWW browser configured to accommodate common multimedia file formats, can solve problems and receive immediate feedback ‘‘from a sophisticated grading package that makes use of the latest complex systems data analysis tools to handle ambiguous input in an intelligent manner.’’ The grading package analyzes the learner’s work and determines errors in arithmetic, sign, and units and then provides hyperlinks to appropriate help. The grading package also can check symbolic expressions including, for example, differential equations. The pages with problems for students contain hyperlinks to relevant lecture notes and help files. Special tools permit the students to draw images and create animations. Integrated conferencing software provides communication among the students, the teacher, and teaching assistants. Controlled access to an online grade book indicates the student’s progress. CyberProf provides specialized HTML editors for constructing lecture notes and problem sets. CyberProf is implemented on a server as a package of Perl scripts and C routines that respond to input from students and teachers submitted via ordinary HTML forms. By 1997, courses at the University of Illinois that use CyberProf included offerings in physics, chemistry, biology, and economics, with courses under development in several other areas, including electrical engineering. In 1997, commercial publishers began to offer on-line course packages for several computer languages. The purchaser of the package, sold at bookstores, receives a textbook, an ancillary CD-ROM, and, at no additional charge, entry to a WWW site that provides access to peer discussion groups, a list of frequently asked questions and responses, a tutor (who answers a fixed maximum number of additional questions for each subscriber), and on-line examinations. Some universities began to offer degree credit for courses offered over the

COMPUTER-AIDED INSTRUCTION

WWW. In the late 1990s, a few complete degree programs became available over the WWW and offerings were increasing rapidly. The National Engineering Education Delivery System (NEEDS) maintains a database of noncommercial curricular materials in electronic form on the Internet and hence provides a means of finding and obtaining materials for computer-aided instruction in engineering as they are developed. NEEDS includes material for stand-alone desktop computers and, increasingly, materials for use on the WWW (26). The World Lecture Hall, maintained on the WWW, contains links to pages created by faculty worldwide who are using the WWW to deliver class materials. Information is classified according to subject and includes course syllabi, assignments, lecture notes, exams, class calendars, and multimedia textbooks (27). Perhaps the most important conceptual contribution of the WWW to computer-aided instruction is the perspective of an open learning environment in which there are few limits on the types, quantity, or location of materials that can be incorporated. Before the WWW, it was easy to think of learning environments as circumscribed by a few computers on a local network or even by a single machine, but this is no longer true. APPROPRIATE APPLICATION OF COMPUTER-AIDED LEARNING ENVIRONMENTS Widespread availability of desktop computers and pervasive access to the Internet and the WWW give teachers unprecedented choice about how to deploy computer-aided learning environments. Once, computers at universities were available mainly in special on-campus computer laboratories. This location limited the use of computers both in the classroom and at home. Access to desktop and laptop computers and to the Internet removes these constraints and presses the recurring question about where and how computers best can aid learning. These dramatic changes in technology, the demands for more student-oriented approaches to learning on campus and indeed fundamental changes in the structure of engineering education, as well as calls for improved opportunities for distance learners, all mean that the fundamental question of computer-aided instruction, ‘‘How can we use computers to improve education effectively and inexpensively?’’, must be approached with a careful understanding of basic strategies in engineering education. Otherwise, it is easy to lose perspective and, as a consequence, spend lots of time and money to achieve disappointing results. Responsibility for Learning Teachers at the college level adopt, intentionally, a fundamentally different strategy of instruction than teachers in secondary school, or high school. In high school, the teacher lays out during class what the students are expected to learn and then directly supervises the learning, most of which occurs also during class meetings. Although work outside class (homework) is a part of high school education, most learning occurs during class. In college, the teacher still lays out during class what students are expected to learn but expects that only a small part of the learning takes place during the class meetings. Teachers expect college students to accept the re-

653

sponsibility for learning most of what they need to learn outside normal class meeting times. This approach is a major reason why college students typically spend less than half the time in class than high school students spend. In college, therefore, the teacher spends much of the class time explaining to students what they should learn. Then the students leave class and learn the material, usually with the help of other students and perhaps with the help of the teacher or an assistant, study groups, and, perhaps, interactive tools such as computer-aided instruction. This approach leads to the oft-stated rule of thumb that college students should spend approximately two hours outside class studying and learning the course material for each hour spent in class. The principle that underlies this approach is that post-secondary-school students should learn to take responsibility for learning the specified material as a step toward lifelong learning in the real world. In the world of practice, they must assume not only the responsibility of learning whatever they need to know, but of deciding what they should learn as well. From this perspective college serves students as a kind of half-way house between learning in the highly structured high school environment and learning in the real world after they leave college. Note that focusing on changing engineering education by changing the classroom (for which the need for improvement seems inarguable) misses the largest target for reform: the two-thirds of the course time that students spend studying and learning the material outside class. It is this time outside class that should be a prime target for interactive instructional environments, computer-aided instruction, collaborative learning, teamwork, and many other approaches that focus on the learner. It is, after all, outside class that most learning occurs during college courses. An obvious rejoinder is that teachers in college should spend less class time laying out the material to be learned and more time conducting learning activities related to the content of the course. Such a change suffers from two difficulties: one pragmatic and one philosophical. From a pragmatic perspective, encroaching on class time for extensive in-class learning activities means decreasing course content or interfering with other functions of the class meeting, discussed later, unless either efficiency in presenting the material to be learned increases dramatically or the time spent in class increases. Universities are unlikely to invest more resources in providing increased time in class in view of widespread calls on them from many of their customers to increase institutional productivity. Even dramatic gains in efficiency of presenting the material are unlikely to make available enough time to reduce significantly the two hours or so of time students need to spend on learning outside the classroom after each class meeting. From a philosophical perspective, implementing these learner-centered approaches to a significant degree during class time emphasizes learning supervised by the teacher rather than helping learners accept responsibility for their own learning activities. Directing instructional reform primarily at class meetings, therefore, not only misses the larger part of the course (outside class) during which most learning in college occurs, it also jeopardizes the success of efforts to help students learn to accept responsibility for a lifetime of learning on their own.

654

COMPUTER-AIDED INSTRUCTION

The Class Meeting If most learning in a collegiate course occurs outside the classroom, a logical question is, ‘‘Why ask the students to meet together at all?’’ Perhaps the major attractive feature of a class meeting is the efficiency it affords the teacher in dealing simultaneously with a large number of students. As mentioned earlier, the class meeting permits the teacher to describe to the students, simultaneously, what it is they should learn. In addition, class meetings can offer efficient means of communicating to all students in the class answers to questions of common interest, of administering examinations simultaneously, of collecting student work, and of handling necessary organizational details. The class meeting also provides special opportunities for motivating students through, for example, presentation of supplemental material by the teacher, class participation by the students, and interaction with the teacher. Class participation can range from informal group discussion to the once-popular approach of requiring several students to work example problems, simultaneously, in front of the class and the teacher (as a reasonably efficient, if somewhat intimidating, means of demonstration, correction, and motivation). The fundamental purpose of the class meeting, however, is to prepare students to learn efficiently after they leave the class meeting. From this perspective, the class meeting is analogous to a staff meeting in the business world: the meeting itself accomplishes little of the work that needs to be done, but permits communication, planning, and coordination that help the participants accomplish their work after the meeting. From this viewpoint, nothing is fundamentally wrong with the oft-criticized lecture method. Until recent times, the lecture method arguably has been one of the most time-efficient means available to teachers for communicating, within a limited time, to students what they are to learn. Much of the criticism directed toward the lecture method apparently stems from belief that the primary purpose of class meetings is to provide a time for students to learn rather than a time to prepare them to learn outside the classroom. From another perspective, much of the criticism of the lecture method seems to be based on confusion between the supervised learning strategy practiced by teachers in high school and the strategy, used by teachers in college, of helping students (indeed requiring students) to become independent learners. The recent widespread availability to teachers of inexpensive, and easy-to-use, software for word processing, graphics, presentations, simulation, spreadsheets, symbolic manipulation, and multimedia and that of desktop and laptop computers to students means that placing material to be learned in files on disks or on networks (especially the WWW) permits distribution of more detailed, accurate, and timely information than could possibly be accomplished through the traditional chalk and blackboard use of the lecture method. The traditional lecture method’s days as the predominant teaching mode, therefore, are threatened not so much by its lack of interactivity in comparison with collaborative learning and other student-centered approaches to learning, but by its recent relative inefficiency in presenting the material to be learned in comparison with techniques that exploit information technologies. Exclusive use of the traditional lecture method during class meetings is, therefore, increasingly dif-

ficult to justify. In particular, it is exceedingly difficult to justify the use of class time for the teacher to transcribe notes on the board as the students transcribe them (incompletely and inaccurately) into their notes. Any modest cognitive benefit learners might derive from the transcription process surely can be outstripped by suitably challenging problems or projects they pursue outside the classroom. It is important to note, however, that a judicious combination of the lecture method and the distribution of information electronically can permit the teacher to devote much more time during the class meeting to motivating students, discussing relevant issues, and answering questions than use of the traditional lecture method alone. The class meeting time saved also can permit the teacher to introduce some studentcentered learning approaches during the class meeting time. The objective in doing so, however, is not to make it unnecessary for students to use these approaches outside the class meeting time, but to improve the efficiency with which students use them outside the classroom. Accessing and reviewing some, but certainly not all, of the course material located on servers can prove useful during the class meeting in a suitably equipped classroom. Given that most learning occurs outside the classroom, however, investment in classrooms that provide a computer in front of each student is difficult to justify, especially given the continuing expense of maintaining up-to-date computers in such classrooms. Students’ computers, in contrast, are a renewable resource. Learning Outside the Class Meeting If most of the learning in a collegiate course occurs during the two-thirds of the course that lies outside class meetings, teachers must consider how to design learning experiences for students outside the classroom that help them make the best use of that time. Traditionally, most engineering teachers have paid far less attention to designing, carefully, learning activities for students outside the class meeting time than they have to designing the class meeting time, despite the fact that students are supposed to spend twice as much time on the course outside the classroom as they do within it. Common approaches to fostering learning by the students outside the classroom include assignments of reading and homework. Unembellished, both of these approaches fail to provide intensely interactive learning experiences that students appreciate and, increasingly, demand. In practice, homework assignments often turn out to be largely a waste of time for both the student and the person who evaluates them. Consider the following situation. The teacher assigns a homework problem, due one week later. The night before the assignment is due, students consider the problem for a time, take a shot at providing a solution and submit it the next morning to see if they got it right. The teacher carefully studies, corrects, annotates, and grades the papers in time to return them to the students during the very next class meeting. By that time, the students have focused their attention on other matters and lost almost all interest in the problem. They probably never read the careful corrections and admonitions provided by the teacher. In short, few students assume ownership of typical homework to the point of committing to self-evaluation of their work. Instead, they make a quick pass at a solution and send it off for evaluation by someone else

COMPUTER-AIDED INSTRUCTION

whose evaluation cannot possibly reach them in time for due consideration. What students need to improve their efficiency in learning outside the classroom are highly interactive learning environments in which they receive feedback about their work immediately, while their interest is high. How can such environments be achieved in practice? Formal and informal problem sessions have been conducted by engineering educators or their surrogates for years as a means of providing interactive learning environments in which students can ask for, and receive, immediate response to more questions than the teacher can entertain during the class meeting. In contrast to an apparently widespread assumption that laboratory activities are mainly a means of inculcating practical skills in students, the main purpose of laboratories is to provide a highly interactive learning environment in which learners receive immediate feedback about their work directly from experimental apparatus. Especially for electrical engineers, the cost of sophisticated electronic components has become low enough that construction and testing of surprisingly complex circuits and systems assembled from parts that the students purchase at local stores can be required as hardware homework that demands no allocation of laboratory space or personnel (beyond a grader for the homework) by their academic department (28). In recent years, engineering educators have begun to encourage formation of study groups among students of a class. Such groups provide a potentially highly interactive, and effective, learning environment at little cost to the institution. One-on-one learning sessions with the teacher, or a hired tutor, can be quite effective in providing a highly interactive learning environment but, unfortunately, at a high cost. Few of these approaches offer as much promise for improving the learning environment outside the classroom as computeraided instruction. Current widely available computer and network technology afford intensely interactive learning environments at any time on desktop computers almost anywhere and, via networks, offer learners opportunity to interact with the teacher and each other, without the necessity of overlapping exactly in either time or space. In the mid-1990s, the Alfred P. Sloan Foundation initiated a program specifically directed at learning outside the classroom to explore ‘‘new outcomes in science and engineering higher education’’ made possible by affordable technology, including desktop computers, network access, CD-ROMs, and video tape (29). They began with the perspective that lectures and study groups, and indeed most on-campus learning activities, nowadays could occur without the learners and the teacher gathering at the same place at the same time through e-mail and conferencing, for example. They termed this concept asynchronous learning network (ALN). ALNs are intended to serve both on-campus and distance learners. The Sloan Foundation does not envision asynchronous learning networks to require specially constructed software learning environments of the kind often associated with computer-assisted instruction, but considers ALNs mainly as a means of facilitating connections among teachers and learners. Possible outcomes of asynchronous learning networks include selfpaced learning, lower cost to the learners, and pursuit of degrees or certifications at home. The Foundation also is interested in the effects of ALNs on the time required to complete a degree and on student retention.

655

Implications for Distance Learning If two-thirds of a typical on-campus course occurs outside the class meetings, the good news for distance learning is that much of the material prepared to help on-campus students learn outside the classroom applies to distance learners as well. Computer-assisted learning environments, in particular, can be readily available and useful to on-campus and distance learners alike. Especially since on-line study groups are becoming common, much of a course designed to provide interactive learning environments outside the classroom for oncampus students should be useful to distance learners with little adaptation. The main difference for distance learners is in accomplishing the functions that are carried out during class meetings for on-campus students. Specifically, the question is how to accomplish (1) dissemination to the students of the information that they are to learn during the course, (2) communication of answers to questions of common interest to all students in the class, (3) administration of examinations, (4) collection of student work, (5) organization of the course, and (6) motivation of the learners, perhaps through interaction with the teacher and other students. It is easy to see how the increasingly ubiquitous Internet and WWW (in combination with surface mail, fax, and telephone) can provide all of these functions without re-creating a classroom environment for the distance learner. Indeed, the Internet is profoundly changing distance learning (30). Perhaps providing motivation is the most difficult function to furnish distance learners. Fortunately, distance learners, by the virtue of the fact that they have taken the trouble to become distance learners, usually are highly motivated and may not need as much classroomlike motivation as typical on-campus students. The on-line experience itself provides motivation for some students. Given the perspective that the class meeting is not the primary occasion for learning during a course and given that most functions of the class meeting can be accomplished by other means, it seems difficult to justify heavy investment in two-way video links to recreate a classroom environment for distance learners. Investment in enriching the environment in which they learn on their own seems more appropriate and productive. Recreating a classroom environment is not only expensive and accomplishes little that cannot be accomplished by alternative means, it also demands that distance learners congregate at a specific time and place—a severe disadvantage for many distance learners. Distance learners seem to view on-site, on-demand, highly interactive learning opportunities as the ideal. With computers on the desktops (or laps) of most distance learners, on-site on-demand, highly interactive learning seems much more nearly achievable via the Internet and surface mail materials such as videos and printed items than via real-time video and/or audio links. Student and Teacher Responsibilities Student responsibilities in a course of study are simpler to describe than to accomplish. Student responsibilities are threefold: (1) find out what is to be learned during the course of study, (2) assume responsibility for doing whatever is necessary to learn it, and (3) learn it. Teacher responsibilities are more complicated to describe and, perhaps, to accomplish. The most visible responsibility

656

COMPUTER-AIDED INSTRUCTION

of a teacher in a college course is to introduce students enrolled in the class to what is to be learned. This responsibility may be carried out, for example, through delivering lectures, assigning readings in textbooks and supplemental material, distributing handouts, posting files on network servers, and providing interactive computer-aided learning environments. Explaining the material to the members of the class, discussing it, and answering questions are other important responsibilities. A most important related responsibility for teachers is the following: do not assume responsibility for a student’s learning. That responsibility must lie squarely on the student to avoid subverting perhaps the most important single objective for college students: learning how to learn without supervision. Another obvious responsibility of the teacher is to evaluate the achievements of students enrolled in the class. Arguably the most important single responsibility of the teacher, however, is to plan, actually to design, the course of instruction for the class. Although the most obvious part of this responsibility is planning the class meetings, the most critical, most demanding, and most easily neglected part is designing that portion of the course of instruction that involves the students when they are not in class meetings but learning on their own. Unfortunately, most experienced teachers have the distressing impression that the time students spend outside of class is singularly unproductive. Thus, an important responsibility of the teacher is to design learning activities for students that help them make the best use of their efforts outside the classroom instead of wasting precious time. Indeed, the design of that part of a course that occurs outside the classroom is potentially the most productive single opportunity that is available to contemporary teachers for improving learning by students. Elements available to teachers for designing this part of the course of instruction include, as already mentioned, study groups, student-teacher conferences, problem-solving sessions, hardware homework, interactive computer software, and network communication between students and the teacher. Faculties of education at universities have long advocated explicitly designing a course of instruction subject to constraints such as time, financial resources, student capability, and accreditation requirements. The general process consists of (1) determining what students already know, (2) deciding what students should learn, (3) identifying specific instructional approaches (lectures, collaborative learning approaches, interactive software, and laboratories, as examples) that may be useful, (4) synthesizing a coherent plan for student learning that exploits these approaches, (5) selecting and/or developing appropriate materials, (6) developing examinations, projects, portfolio requirements, and other means of assessment that measure the effectiveness of the plan in practice, (7) trying in practice what has been conceived, and (8) modifying the course of study, based on the assessments, to improve its effectiveness (31). This process of designing a course of instruction subject to constraints can be conveniently termed course design to distinguish it from the term instructional design, encountered later. Although engineering faculties at universities certainly can recognize the parallels between course design and what they term engineering design, only a few seem to apply the design process, consciously and routinely, in developing

courses of study for engineering students. In the absence of widely accepted theories of learning and instruction, it might seem likely that engineering teachers would adapt familiar approaches from engineering design to develop courses that, subject to constraints, help students achieve such fundamental objectives as problem solving and learning to work with others as a team. Perhaps, however, the relative stability, until recently, of the instructional paradigm for engineering education that emerged and persisted during the decades that followed World War II permitted the design of a course of study to mean little more than ensuring that students are exposed to a minimum set of technical topics. Approaches based on folklore and customs learned during the teacher’s student days too often substituted for careful explicit design of courses of instruction. Such a naive approach to designing courses of instruction is always risky, but is especially likely to be ineffective during times of rapid changes in educational approach, student preparation, industry expectations, and student expectations. Careful course design is vitally important as courses rely more on computer-aided instruction, an unfamiliar tool that, without careful forethought and planning, can fail either by running amok or alienating learners. Widespread application of the course design process to develop courses of study in which the learning activities outside, as well as inside, the classroom form a coherent and effective whole could improve the resulting course designs dramatically by (1) focusing faculty attention on the underlying fundamental learning objectives for their students instead of simply the topics to be covered and (2) incorporating a greater variety of learning approaches in engineering courses than has been customary. Indeed, one answer to the fundamental question, ‘‘How can we use computers to improve education effectively and inexpensively?’’, is to use computers as an excuse to convince engineering faculty to plan carefully the entire course of study, within and without the classroom, as a coherent whole. If such a ploy were successful, computeraided instruction would have succeeded nicely even in courses in which computers play no explicit role. However, successfully incorporating computer-aided instruction, with the additional variables related to the relatively unexplored attendant networking and multimedia environments, into courses of instruction becomes very hard without some explicit design strategies suggested by theories of learning and instruction.

CREATION OF LEARNING ENVIRONMENTS The major impediment to progress in computer-assisted instruction is less and less the cost of networking and powerful hardware, but rather incognizance about how to design effective interactive learning environments. From a naive perspective, psychology would provide a theory of learning on which to base a theory of instruction that would prescribe how to design a successful computer-aided instructional environment. Reality is far different. Theories of learning from psychology certainly are available (32). Indeed, a major difficulty is too many theories of learning. Most are rooted in movements in psychology, such as behaviorism or cognitive science. Seemingly endless conflicts appear among the various theories, however, mainly from uncertainties about the domain of their respective applicability. Such uncertainty

COMPUTER-AIDED INSTRUCTION

clearly hampers adequate verification, and widespread acceptance, of the theories. Imagine the conflict and confusion that would result among circuit theorists and electromagnetic theorists if the domain of applicability of each theory were not understood. Circuit theory is a simplification of electromagnetic theory that applies only when the wavelength that corresponds to the highest frequency at which the circuit will be used is much larger than the largest linear dimension of the circuit. Without this knowledge, the violation of Kirchhoff ’s current law along conductors in antennas, for example, could produce endless confusing arguments. Doubtless, all of the available learning theories apply under some circumstances. Not understanding what those circumstances are means that discovering conflicting conclusions from the different theories sheds little light on how to improve the theories, or on the question as to whether one or the other is incorrect in some fundamental way. Perhaps because of the seemingly countless, inconsistent, unverified (unverifiable?) theories of learning and instruction, engineers traditionally have been largely ignorant, indeed skeptical, of theories of learning and instruction. As a consequence, the application of theories of learning and instruction to the development of instruction in engineering and science at the university or professional level has received scant attention compared with the development of instruction in kindergarten through grade 12. Fortunately, theories of learning and instruction can provide considerable insight during course design by engineering educators, despite the absence of a single widely accepted theory. Exploring the available insights and exploiting them in addressing the fundamental question, ‘‘How can we use computers to improve education effectively and inexpensively?’’, however, requires familiarity with the various approaches, vocabularies, and patterns of thought characteristic of these areas. Learning Theories and the Creation of Learning Environments PLATO and TICCIT, the large computer-aided instruction pilot experiments boosted by substantial funding from the National Science Foundation and other sources, followed quite different approaches to designing the enormous amount of computer-aided instructional materials that these large demonstrations required. Authors of learning materials for the PLATO project had complete freedom as far as the types of instructional strategies and design procedures they employed (33). A basic assumption initially was that developing computer-aided instructional materials was just a matter of automating instructional techniques commonly used at that time. As a consequence, learning materials developed for PLATO included a diverse combination of drill and practice exercises, simulations, games, and tutorials. Although the basic assumption ultimately proved wrong, the collection of PLATO materials showed that diverse design approaches could yield useful environments and that there is no best approach, even for a particular discipline. Authors also found it convenient that PLATO permitted small modules to be constructed and evaluated easily and then, later, combined with others to form a larger unit. Not surprisingly, results for a specific PLATO lesson depended dramatically upon how the teacher who used the lesson felt about it and upon how the teacher chose to implement the lesson.

657

In contrast, the TICCIT project, conducted by the Mitre Corporation, used a much more structured production approach to developing large quantities of learning materials. Individual teams consisting of a subject matter expert, a psychologist, an instructional designer, an evaluator, and a packaging specialist designed learning environments that carefully controlled learner activities (5,33). Partly to achieve production efficiency, the project chose, initially, a rules-examples-practice pattern as the single instructional strategy for developing materials designed to ensure that learners mastered the information in the lessons. This approach made it easy to generate templates that permitted new lessons to be developed merely by adding subject matter to the template rather than spending the time and effort to create a new design for each lesson. In TICCIT, again, the attitudes and approaches of the teachers strongly influenced the success of the lessons. The approach to lesson development in TICCIT clearly was more systematic than that for the PLATO project and hence offered the possibility of further development into a broadly applicable procedure that authors could rely on for guidance in developing computer-aided instructional materials. The TICCIT approach, which ultimately led to what is called instructional systems design or simply instructional design, relied heavily on concepts from the movement in psychology known as behaviorism for its initial development. Behaviorism. Before the mid-1950s, behaviorism was a dominant force in psychology (5,32). Behaviorists held that psychology should concern behavior without consideration of consciousness or mental models and constructs. Specifically, they maintained that psychology should deal only with prediction and control of observable behavior. Without much oversimplification, behaviorism can be viewed as a reaction against not only the earlier psychological concepts of mind and consciousness, which, to behaviorists, seemed too close to the religious idea of the soul to be the subject of proper scientific inquiry, but also a reaction against Freud’s preoccupation with the unconscious, the id, and the libido, which seemed too fanciful to be the subject of scientific research. From investigations with animals in simple experimental configurations, behaviorists came to believe that essentially all human behavior, including learning, is the result of conditioning. Classical Pavlovian conditioning sought the means of achieving a desired response after application of a stimulus. Specifically, the approach was to begin with an existing stimulus-response pair, such as a dog’s salivating in response to the stimulus of seeing food, and then to add a simultaneous neutral stimulus, such as the ringing of a bell that initially would not produce the response. After sufficient repetition of the stimulus pair (food-bell) and the ensuing response (salivation), application of the initially neutral response (bell) alone, without the original stimulus (food), stimulated the original response (salivation). B. F. Skinner, whose views came to dominate psychology in the United States for several decades, introduced an alternative approach known a operant conditioning. In operant conditioning, the frequency of a desired result, called an operant, is increased if it is followed by positive reinforcement, or alternatively, if undesired results are followed by negative reinforcement. In animal experiments, a rat that pressed a bar (operant behavior) would receive a pellet of food (positive

658

COMPUTER-AIDED INSTRUCTION

reinforcement) or a pigeon that pecked a dot (operant behavior) would receive some grain (positive reinforcement). In operant conditioning, notice that a stimulus can be absent or, if present, may be unknown or ignored. From a Skinnerian perspective, traditional instruction emphasized providing stimuli, through content, to the learner. In contrast, operant behaviorism, often called simply behaviorism because of its ultimate dominance of the behavioral perspective in psychology, shifted the emphasis to reinforcing desired operants, or behavior, of the learner. With this perspective, Skinner developed programmed instruction, in which competence is developed in a learner by dividing the learning process into steps sufficiently small to be easily achievable and by providing reinforcement when each small step is accomplished successfully. The size of the steps is chosen to be small so that the learner experiences positive reinforcement as frequently as possible. The small steps in the programmed learning approach also mean that the information can be presented and the learner response can be checked and correct responses reinforced automatically by what Skinner called teaching machines. Skinner viewed teaching machines as more effective in providing reinforcement than teachers because they could provide reinforcement more quickly. The concept of teaching machines shifted the focus of applying technology to instruction from presentation alone, as with films, for example, to reinforcement as well. Empirical results for programmed instruction compared with those from conventional approaches were disappointing. Moreover, many students found the programmed instructional materials boring. As an illustration of the dynamics that typify the interplay between psychological theories of learning and their consequent theories of instruction, however, the programmed instruction movement persisted until the late 1960s, more than a decade after adherence to behaviorism in psychology had waned. Thus, the concepts of programmed instruction and the teaching machine, obviously ready-made for implementation on digital computers, were available to influence significantly early large computer-aided instructional projects, such as PLATO and TICCIT, despite rising dissatisfaction with these concepts among psychologists at that time. Another behavioral theory of instruction that attracted the attention of some engineering educators during the 1970s was the Keller plan, which focused not on programmed instruction or teaching machines, but on personalized, self-paced, mastery-oriented instruction (5). Although performance on final examinations by students who used the Keller approach typically exceeded that for students in traditional courses and students liked the flexibility of self-paced instruction, critics charged that it inevitably taught students a subservient approach to learning that, in the long run, is quite unfortunate because it inhibited later independent learning. Some studies indicate increased time requirements for learning and higher dropout rates for students as well. Cognitive Science. In contrast to behaviorism, cognitive psychology, or cognitive science, emphasizes understanding the role of consciousness, thinking, and reasoning in behavior (5,32). The consequent development of cognitive models for mental processes considerably expanded the possibility of providing insight useful in understanding and constructing successful learning environments. Jean Piaget, as an early exam-

ple, developed models of cognitive development that helped teachers understand the cognitive readiness of students for different types of learning, such as dealing with abstractions and hypotheses (5,32). During a decade beginning in the mid-1950s, the cognitive perspective eclipsed behaviorism in psychology, and cognitive theories of learning therefore, in time, became dominant. In contrast to behaviorism, the cognitive perspective not only emphasizes the study of mental constructs and organization of knowledge, it concentrates on knowing rather than responding and considers people to be active, problem-solving learners rather than passive subjects of conditioning. Although psychologists such as John Dewey (5) and Kurt Lewin (5) had advocated cognitive views earlier, the influences that favored eventual predominance of the cognitive perspective included the translation of most works of Piaget into English by the 1960s, the influence of information theory as developed by Claude E. Shannon and Norbert Wiener, and the advent of the computer and artificial intelligence (5). The architecture of computers, for example, suggested to cognitive scientists that cognitive processes were as real as physiological processes. Such information processing analogies led to early models of memory and of cognitive algorithms for making sense of sensory information. The cognitive perspective took longer to affect theories of instruction than it did theories of learning, however. Even those who embraced at least some elements of cognitive science constructed theories of instruction that produced learning environments that emphasized learning by conditioning rather than learning by problem solving and exploration. Benjamin S. Bloom, for example, proposed a classic learning taxonomy based on cognitive concepts but developed a mastery learning approach for instruction that was essentially behavioral and hence proved ill-suited for encouraging divergent thinking and creativity (5). Robert M. Gagne also developed a taxonomy starting from cognitive concepts and from it developed an approach for accomplishing learning by achieving carefully predetermined behavioral objectives (5,32). Despite conceptual cognitive influence, the initial emphasis on behavioral objectives resulted in learning environments that suffered the limitations typical of behavioral approaches. David P. Ausubel developed a cognitive theory that foreshadows a number of later developments and, based on the manner in which he believed learners build cognitive structures, indicates that learning environments should begin by introducing learners to general concepts and then proceeding to the specific (5,32). Over time, the choice of learning strategies has expanded far beyond those available in the initial restrictive templates of TICCIT by relying more and more on cognitive perspectives and relying less and less on pure behaviorism. Developments based on cognitive concepts have proceeded along several different paths, however. Instructional Design. Much of the particular design procedure known as instructional design or instructional systems design stems from the work of Gagne, and hence includes cognitive aspects together with a substantial behavioral component (34–37). Over the years, Gagne and others have incorporated more and more cognitive concepts into the instructional design process to achieve additional dimensions in the learning environments developed from this perspective. One detailed account of instructional design, the most widely used

COMPUTER-AIDED INSTRUCTION

single approach to designing computer-aided instruction, is given by M. David Merrill, who originally worked on the TICCIT project (37). Beginning with concepts from Gagne and Ausubel, authors Mengel and Adams more recently have modified the conventional approach to instructional design by adapting and including concepts from software engineering to develop a design methodology specifically for hypertext computer-aided instructional materials (38). Incorporating hypertext (a concept sometimes attributed to Vannevar Bush, an electrical engineer) (39) offers the possibility of increasing the flexibility of learning environments created through instructional design and yet maintaining some of its best features. Ausubel’s approach of moving from the general to the specific is compatible with the top-down design approach central to software engineering. Despite impressive expansion of its scope, critics of instructional design complain that the resulting environments achieve too little of the complexity and diversity that is readily envisioned for interactive learning environments after, say, browsing the WWW. Perhaps the difficulty is that instructional design emphasizes highly controlled and therefore in some sense closed, or at least highly circumscribed, learning environments. Intelligent Tutors. Developers of intelligent tutors, based on applications of artificial intelligence in learning environments, seek to apply cognitive science directly to computeraided instruction (5,6,16,33,40–42). The tutors typically maintain separate complex internal models of the learner and an expert and include a tutoring component that relies on the learner and expert models in selecting a course of interaction with the learner. In one simple approach, the learner is modeled as knowing a strict subset of what the expert knows. Learning progress is indicated by the size of the subset in comparison to the set of what the expert knows. Such a model obviously cannot take into account the mistaken knowledge that the learner ‘‘knows.’’ One means of avoiding this and other problems in constructing a model for learners is to eliminate the learner model and permit the learner to interact with the expert model through a mutual exchange of questions and answers. An early classic example of this approach, the sophisticated instructional environment (SOPHIE), sought to teach troubleshooting of relatively complex electronic circuits as a means of transforming classroom knowledge about electronics into intuitive, experiential knowledge (6,33,42–44). SOPHIE I and SOPHIE II combined a powerful natural language interface, an inference engine and an early version of the circuit simulator called simulation program with integrated circuit emphasis (SPICE) to realize a learning environment that contributed significantly to the credibility of intelligent tutoring systems. The circuit simulator SPICE functioned as the expert module. To SOPHIE I’s capability for addressing natural language questions posed by learners, SOPHIE II added an articulate expert to give demonstrations. With this capability, a learner could insert a fault (replacement of a good component with a faulty component) into the circuit and watch the articulate expert explain its strategy as it tracked down the problem. During this troubleshooting process, the learner was asked to make qualitative predictions (higher, lower, or roughly the same) about the result of each measurement on the circuit into which the fault had been inserted, in comparison with the circuit without the

659

fault, before the expert actually took that step. If the learner made a wrong prediction, SOPHIE took steps to help the learner understand the result of the measurement. SOPHIE II also included a game in which two learners took turns inserting faults for each other to find. The learner who was looking for the faulty component was assessed a cost for each measurement. That cost increased approximately according to the difficulty of conducting the measurement in a practical setting. The learner who inserted the fault, in contrast, was called upon to predict (higher, lower, or roughly the same) the result of each measurement on the circuit into which the fault had been inserted, in comparison with the circuit without the fault, before the other learner actually made that measurement. The game was complex in that the score of the learner who inserted the fault depended upon the product of the percentage of successful predictions (made by the learner who inserted the fault) about the measurement and the cost of the measurements (made by the learner who must find the fault). If you were the learner who inserted the fault, the winning strategy was to insert faults with the most complicated consequences that you could understand and, of course, increase your understanding of the circuit so that you could insert faults with more complicated consequences to better thwart scoring by the other learner. When teams of two rather than individuals played the game, debates between the partners regarding the next move provided valuable insight into the thinking strategies of the learners that otherwise was difficult to obtain. SOPHIE III represented an attempt to move away from simulation-based (SPICE) expertise about circuits to a more flexible representation of such expertise. For pedagogical purposes, explaining the basis of a result to a learner can be critically important, but simulations provide little information about the causality on which their inferences lie. SOPHIE III explored replacing the circuit simulator with qualitative, or causal (rule-based), models so that it could better follow the activities of a learner and then provide coaching rather than merely answering specific questions posed by the learner or by giving demonstrations. In the end, the expertise for electronics troubleshooting proved to be difficult to accommodate successfully in a causal model, and workers pursued application of such models to develop coaches in other fields. SHERLOCK, a more recent intelligent tutoring system for electronic troubleshooting, employs software object techniques for managing complexity in the expert component of the tutor (45). Although SOPHIE II was used in an actual short course, application of SOPHIE and subsequent intelligent tutoring systems for instruction has been limited. While some educators question whether intelligent tutors, by their very nature, can provide the learner-centered environments that promise greater effectiveness in learning, the use of intelligent tutors in actual learning environments appears to have been limited mainly by the immense effort required to write the necessary software. The rare deployment of intelligent tutors in actual learning environments also reflects the circumstance that studies of intelligent tutors are directed as much towards improved understanding of certain aspects of cognitive science as towards successful instruction. Learning Styles. Beginning with the work of Piaget, Dewey, and Lewin, David A. Kolb developed the concept that different learners prefer different learning styles in building the

660

COMPUTER-AIDED INSTRUCTION

mental or cognitive constructs associated, in the cognitive perspective, with learning (46,47). For example, some students find it easier to relate to facts and data while others prefer to deal with theories and mathematics. Some find it easier to deal with information visually, in pictures or diagrams, for instance, although others learn more easily from written or spoken information. Some students prefer to learn by interacting with other students. Some prefer to learn alone. Kolb holds that learning improves when learners pursue all styles of learning, not merely their individually preferred styles. Some engineering educators conclude that, in practice, professionals must learn in all of the styles (48). From either perspective, it follows that learning environments should help students develop learning skills in the learning styles they do not prefer as well as in those that they do. This presumption leads to the concept of teaching around the cycle to help learners experience learning in the various styles. Computer-aided instruction seems to offer the promise of applying, and then evaluating, the Kolb approach by structuring learning environments that accommodate various learning styles and by measuring and studying the response of learners to the approach of teaching around the cycle. In principle, the learning environments even could be changed adaptively, depending upon the performance of the learner in the environments corresponding to the different learning styles. That promise has not yet been realized in practice, however. Biological Bases of Learning. Just as knowledge about computers stimulated developments in cognitive science, so too, has increasing knowledge about the biology of the human brain. Neurophysiology, while still maturing, identifies several characteristics of the brain that relate to cognition (32). Modularity and plasticity, for example, may have important consequences for learning. Modularity means that different parts of the brain correspond to memory for different cognitive functions. The left and right sides of the brain, for instance, seemingly correspond to different functions, and certain functions seem correlated with even more specific regions. The different regions, or modules, seem to function with some degree of autonomy and appear to be stimulated, or accessed, by different senses (vision, hearing, and so forth). The ease with which the various modules can be accessed seems to vary significantly from individual to individual. A consequence of modularity for learning and instruction is that different individuals have preferred modes of processing information, depending upon which brain modules are most easily accessible to them through corresponding sensory stimulation. Some may prefer to learn through listening, for example, and some through seeing. Thus, a learning environment that involves several sensory stimuli probably helps learners access more of the brain’s different modules and, presumably, enhances learning through construction of more complex cognitive structures. Plasticity means that brain structure continues to develop even after birth, rapidly at first, but continuing throughout life. Continued development seems to mean increased capacity for learning. Moreover, development appears spurred by complex interactive environments. Thus, plasticity opens the possibility that appropriate learning environments can stimulate increased capacity for learning, even in adults. Modularity and plasticity of the human brain, taken together, may imply increased effectiveness of complex inter-

active learning environments that encompass manifold stimulation of the learner. Elaborate theories of instruction ostensibly based on current knowledge of brain biology, however, are premature and not supported by experimental understanding of the brain (49). Specifically, classification of learners as either right-brain (analytic-verbal) or left-brain (holistic-spatial) is simplistic, although learning environments designed to stimulate and exercise both sides of the brain may be useful merely because of the inevitable complexity they involve. Neurophysiology holds great promise for ultimately clarifying and providing bases for theories of learning and instruction. For now, however, the links between it and theories of learning and instruction are tenuous. Constructivism. Although instructional design provides authors with fairly specific design procedures for computeraided instruction, the resulting lessons tend to lack the flexibility and cognitive complexity that are sometimes realized in lessons composed with less systematic approaches. Adherents of constructivism, a cognitive perspective that traces its origins back to Piaget, Dewey, and Vygotsky (32), criticize the instructional design process and even other cognitive approaches as being fundamentally flawed by their basis in objectivism, a traditional philosophical perspective that ultimate reality exists, independent of any observer. In objectivism, reality is absolute. Constructivism, in contrast, considers ultimate reality to be the cognitive interpretations or mental constructs that each individual builds as a consequence of perceptions and learning. In constructivism, therefore, ultimate reality can be different for different individuals (16,32,50). For constructivists, learning becomes the construction of individual interpretations of stimuli experienced by learners. Teaching becomes the creation of environments that help learners construct their individual interpretations of what they perceive rather than transmission of information to the learners and reinforcement of what the teacher views as appropriate responses. Because different learners construct different interpretations, learning necessarily becomes learner centered. Indeed, because each learner must construct an individual interpretation, learning is impossible unless the learners take direct action to construct the interpretations. Constructivists often speak of student-centered learning and active learning. When teaching is viewed as the transmission of information and reinforcement of responses that the teacher considers appropriate, learning is much more teachercentered and the learner assumes a less active role. Constructivists call such an approach passive learning. Constructivists emphasize situated, or authentic, learning environments (learning based on the real, or at least realistic, settings in which the learner will apply it) because abstractions or simplified models of reality may distort or otherwise impede the development of the individual mental constructs that are the ends of learning. Because realistic settings are complex, constructivists stress learning from multiple perspectives to promote more complete understanding. Encountering abstract concepts in several different contexts, for example, can help learners advance their understanding. Constructivists advocate collaborative learning both to assist the learner in achieving multiple perspectives through interactions with others and, perhaps more fundamentally, to accomplish a

COMPUTER-AIDED INSTRUCTION

kind of social validation of the learner’s perspective that is necessary in the absence of an objective reality that objectivists rely on as a comparative standard for learners. Because situated learning should occur in a realistic environment, constructivists assert that evaluation and testing should be an integral part of the learning process, not a separate activity. Such integrated evaluation and testing might be accomplished through projects or portfolios, for example. From the constructivist point of view, numerous shortcomings of traditional instructional design become apparent. Breaking up the material to be learned into incremental steps small enough that learners can complete them successfully with little probability of error requires methodical dissection of the material and careful reassembly into logical paths of learning. By design, learner involvement mainly requires the persistence to follow a learning path developed by the teacher. Thus, the environments that result from traditional instructional design are more teacher-centered rather than learner-centered. As a consequence of their limited involvement, learners in such environments fail to gain experience in developing their own approaches to learning unfamiliar complex material. In short, environments that result from traditional instructional design fail to help the learner learn how to learn independently, a central goal of learning, certainly at the university level. Moreover, environments produced with traditional instructional design tend to avoid the complexity of situated learning environments and use simplified, often prematurely abstract, models of what is to be learned to make tractable the dissection of the material into small steps and its reconstruction into paths of learning. Being based on simplified models of reality, what the learner learns may be distorted significantly by convenient but ultimately misleading simplifications. Reassembling small learning steps into multiple learning paths requires considerable imagination, and perhaps even more time and effort, by the teacher. Thus, learning environments produced with traditional instructional design often offer the learner few alternative paths through, and hence perspectives of, the material to be learned. Because the learning steps are so small and the paths of learning are so well-defined in learning environments that result from traditional instructional design, little collaboration among learners is needed, nor indeed is possible, during use of the environment. Learners in these environments therefore miss out on possible enrichment of their learning by the ideas of other learners. Integrated evaluation and testing of a very narrow kind is certainly a part of learning environments created with traditional instructional design, but implementation of the concept as advocated by constructivists usually is impossible because the learning is not situated in sufficiently realistic environments. Constructivism certainly suggests several routes of escape from the somewhat delimited domain of traditional instructional design. Constructivism itself, however, is open to criticism on several points, not the least of which is its sacrifice of the objectivism that lies at the heart of modern engineering and science to accommodate a diversity of perspectives constrained only by social negotiation. A socially negotiated approach to electromagnetics that is inconsistent with Maxwell’s equations ultimately would prove counterproductive, perhaps at considerable cost to the learners. Situated learning environments can degenerate into a mimetic apprenticeship approach that neglects development of abstract concepts

661

necessary for applying learning in alternative topical domains and yet consumes and distracts the attention of the teacher. The complexity of realistic situated learning environments can hamstring beginning learners. Practical collaborative learning environments can allow uneven participation by learners, and hence uneven learning, to a degree that makes evaluation and testing difficult and the results fruitless. Integrating evaluation and testing into situated learning environments can make focusing assessment on learners’ understanding of concepts that are implicit in the task difficult. For example, a teacher who assesses a fuzzy-controlled robot that successfully balances a vertical rod by appropriate compensating movements may find it difficult to decide if the learners who collaborated on the design and implementation of the robot understood important concepts in fuzzy control or merely implemented a ready-made algorithm that they found. Supplemental means of assessment are necessary. Review of the criticism by constructivists of learning environments created with traditional instructional design reveals that many of their complaints are directed at behaviorism rather than at objectivism. As a result, the insights provided by constructivists are being adopted by both constructivists and cognitive objectivists to enrich the approaches in contemporary instructional design (36,51). Modification and replacement of conventional instructional design by constructivist concepts began in the mid-1990s (52–54). Although computer-aided learning environments developed specifically from a constructivist perspective for use by engineering students are not yet common, existing environments exhibit some of the attributes that constructivists advocate. ELECSIM, already mentioned, illustrates one approach to structuring a computer learning environment that includes features advocated by constructivists. In the basic approach, a learner encounters a collection of connected simulated ‘‘rooms.’’ Each room concerns a topic with scope roughly comparable to a subsection in a conventional textbook. Learners basically are free to interact with the material in each room and indeed the entire collection of rooms, in whatever sequence they choose. If teachers wish, however, they can limit possible destination rooms accessible from each room and can insist that learners successfully complete a quiz or other work before leaving one room for another. A collection of rooms is called a simulation implementation of multifaceted peripatetic learning environments (SIMPLE). A SIMPLE room contains, at the pleasure of the designer, informative notes and explanations, drill and practice exercises, project assignments, (software) tools, and network access. While touring the complex of rooms, the learner can develop understanding, accomplish and document tasks, learn new tools, and demonstrate competence. A major difficulty with implementing any computer learning environment is, of course, software development. The strategy in constructing SIMPLEs is to rely primarily on existing commercial software that is widely used and thus authentic to some degree. ELECSIM, a SIMPLE that concerns a course in analog electronics for undergraduate electrical engineers, can include readily available software components such as circuit simulators (PSpice), math packages (Mathcad, Matlab, Maple, Mathematica, and Macsyma, for example), and logic simulators, as well as WWW browsers, programming languages, word processors, spreadsheets, and databases. Custom software and multimedia also are readily ac-

662

COMPUTER-AIDED INSTRUCTION

commodated. The ELECSIM software serves mainly as ‘‘glue’’ to join disparate existing software. The rooms can be easily replicated, edited, modified, and combined to form other rooms or SIMPLEs on a particular subject, whether or not they relate to analog electronics. The rooms thus are a kind of structural template into which existing materials, such as notes and examples of almost any sort, can be integrated, without undue effort, to form flexible interactive learning environments. The rooms in ELECSIM rely on strong visual images for two purposes. First, the learner develops a strong mental picture of where everything in the environment resides. The central screen in an ELECSIM room displays a graphic view of an office and helps learners recall the location of the various available resources. For example, it is easy for a learner to remember to click the mouse on the file drawer to find supplemental notes about the subject of this room, click on the index file on the desk to find interactive drill and practice exercise problems and open-ended homework problems, click on the computer to find software tools needed to do work in the room, click on the pile of mail to send work to the teacher, click on one of the shelved books to find references or links to related material, and click on the road in a picture on the wall to leave the room. The strong graphical context therefore permits the learner to access directly much greater functionality, without maneuvering through several levels of screens or menus, than would be convenient with a more conventional interface. The second purpose served by the visual environment is to give each screen in the room a very definite mental context. When the learner is concentrating on a particular screen, the teacher can know with some certainty what the learner is thinking about and doing at the moment. The sequence of screens chosen by the learner therefore tells much about the learning strategy and process that the learner is using. The teacher can choose to record automatically the learner’s activity to any level of detail. All of this information can be available to the teacher on-demand over a network or from a diskette for processing and interpretation. As far as constructivism in ELECSIM is concerned, learners control their learning paths and utilization of the various elements in each room and construct individual solutions to complex open-ended problems. The environment is thus manifestly learner-centered and requires active learning. ELECSIM provides a situated, or authentic, learning environment by utilizing the tools of practice, rather than tools built especially for the learning environment. ELECSIM stimulates learning from multiple perspectives by providing ready access in the room to notes and other resources, as well as requiring the learner to move beyond drill and practice exercises to synthesis of concepts through design. Video, other multimedia, and hypertext could provide perspectives beyond those possible with text and simple graphics. Because ELECSIM is essentially a consistent user interface to available resources, e-mail and chat rooms can be incorporated easily, if the computer is networked, to provide a framework for collaborative learning. Learners, for instance, could collaborate on a design without being coincident in space or time. Submission of documents that record and describe the learner’s activities, on openended homework problems, for example, provide integrated evaluation and testing in ELECSIM.

The SIMPLE paradigm, illustrated in ELECSIM, provides a consistent graphical user interface to tools and other resources related to particular topics and collects them in ‘‘rooms’’ that can be visited at the learner’s convenience and provides, as well, one metaphor for creating flexible learning environments that can incorporate the elements advocated by a variety of cognitive and constructivist theories of instruction. A key to the flexibility is incorporation of existing software. From a different perspective, the scope of the rooms in the SIMPLE approach is much larger than the small Skinnerian steps of instructional design, and the learner in a SIMPLE environment has much more choice of learning paths. On the other hand, a teacher who develops a SIMPLE dissects the material and exerts control over the learner more than most constructivists deem appropriate. In this sense, SIMPLE represents an intermediate approach to that of extreme behaviorism and constructivism. Sun and Chou describe the role of constructivist ideas in the design of the cooperative remotely accessible learning (CORAL) system, a multiyear effort in distance education funded for development by the National Science Council in Taiwan (55). The specific constructivist ideas considered during the design of CORAL were to (1) make courseware learning materials as rich as possible, (2) give students authentic tasks on which to practice (building a local area network system in their computer laboratory, for example), (3) encourage students to provide different solutions to given problems (brainstorming innovative methods for preventing computer viruses, for example), (4) encourage students to navigate through instructional nodes and construct their own learning paths, (5) encourage students to interpret new learning situations based on their existing knowledge and experiences, and (6) encourage students to discuss, debate, and work cooperatively. Access to the hypermedia CORAL courseware is possible with a special browser that permits video conferences, exchanges of text and graphical information via an electronic whiteboard, and discussions, all as means of promoting communication and interaction between teachers and learners. Students access CORAL with standard personal computers, although video conferencing requires a high-speed network card and extra equipment such as a camera, microphone, and so forth. The CORAL system records details of learner activities that permit construction of student models. Such records open the possibility of rewarding students for helping each other. Ultimately, CORAL will include a student tutor system (56). Collins advocates cognitive apprenticeship specifically as an approach for developing cognitively rich computer-aided instruction that helps learners develop the mental constructs that are the essence of learning from a constructivist perspective (7). This approach, based on a generalization of the traditional apprenticeship concept, identifies six components that leverage and exploit available computing technology in implementing learning environments. Situated learning involves reliance on actual, or at least realistic, representations of the environments in which what the learner learns will be applied. For example, a circuit simulator used in industry, such as PSpice, could form the basis of a situated learning environment, as could a C⫹⫹ software development environment. Beyond the domain of software, computers can help create situated learning environments with multimedia to bring realism to the desktop.

COMPUTER-AIDED INSTRUCTION

Modeling represents a complex process in simpler terms for the purposes of explanation. In addition to utilization of mathematical models and simulators, computers can deploy animation and other forms of multimedia to help learners grasp complex material. Computers can even help learners understand how to solve problems by showing (modeling) how experts solve problems. The articulate expert in SOPHIE, mentioned earlier, is an example of modeling. Coaching provides personalized hints or assistance to learners as they need them. Relatively simple rule-based branching programs can provide coaching in limited contexts, such as within a single example, without the difficulties associated with authoring and implementing a more ambitious traditional intelligent tutoring system. Reflection requires learners to reconsider their activities and, in so doing, evaluate their own performance. Comparison of the simulated performance of a learner’s design with performance requirements can stimulate reflection. Submission to the teacher of word processing files, with attachments, that document and explain a learner’s work can stimulate and document reflection, as can certain types of examinations. Articulation requires learners to describe and explain their learning activities as a means of making their tacit knowledge explicit. In addition to word processing files with attachments mentioned earlier, chat rooms and e-mail via networks permit learners to interact with others and thereby practice articulation and, in addition, gain new perspectives from their peers. Exploration requires learners to try out different hypotheses, methods, and strategies to see their effects. Because exploration puts the learners in control, they must learn how to explore productively. Computers offer the advantage of permitting rapid examination, in limited time, of wide-ranging alternatives, through simulators, for example. As far as cognitive apprenticeship is concerned, ELECSIM provides situated learning through the use of tools, such as the circuit simulator PSpice and other software, that are widely used for electronic design and communication in industry. As discussed earlier, PSpice permits learners to carry out realistic iterative solutions to open-ended design problems of the type encountered in practice and, thereby, develop the important ability to assess their own work, with only modest direct involvement of the teacher. As appropriate, additional realism can be incorporated into the environment with video clips or other multimedia. In the stand-alone version of ELECSIM, modeling is provided to some degree by drill and practice exercises in which a learner views a circuit and must enter component values that cause the circuit’s operation to meet certain design rules. The exercise thus presents a simplified, or modeled, view of the circuit’s operation and thereby permits the learner temporarily to focus entirely on the design rules. The answers entered by the learner are checked to determine if they satisfy appropriate design rules, as explained earlier. Violation of the rules produces informative messages to the learner, as described later. The entries into the exercises are not checked with the circuit simulator PSpice, which does not operate on the basis of the design rules. Instead, branching comparisons are used to check whether or not the answers satisfy the design rules. In the open-ended homework problems, learners use the circuit simulator (a different model) to evaluate the performance of their design with respect to the given perfor-

663

mance specifications. Neither audio nor video were included in the learning environment for analog electronics, although they certainly could be central to SIMPLEs on other topics and incorporating them is straightforward. In the stand-alone version of ELECSIM, modeling in the sense of showing how an expert solves problems is present only in the explanatory notes available in the rooms. In a networked environment, email, chat rooms, or conferencing could show the teacher solving problems in action, complete with pursuits down blind alleys and mistakes. Over time, selected transcripts of these interchanges could be posted as notes in appropriate rooms as one means of preserving the teacher’s initial approach to solving an unfamiliar problem, an approach that likely will be lost in an inevitable polishing process, otherwise. The implementation of coaching in ELECSIM is kept simple by limiting the context to which coaching applies to the particular context to which the learner is directing attention at the moment. Specifically, consider as an example a drill and practice exercise from an ELECSIM room that deals with the design of simple bipolar junction transistor (BJT) audio voltage amplifiers. In a standard four-resistor bias configuration for a typical BJT, the learners are required to specify values (not unique) of the resistors, the power supply voltage, and the coupling capacitors that will bias the BJT at a specified operating point and achieve a specified open-circuit voltage gain over the audio range (57). A simple branching structure checks learners’ entries against design rules explained in notes available in the room in which the exercise appears and provides either a message that the design is acceptable, in that it satisfies those rules, or if it does not, provides onscreen messages that indicate which components have inappropriate values and whether the values entered by the learner seem too high or too low. Coaching in ELECSIM is thus fairly specific but easy to implement because the rules for coaching apply only in a limited context. Although even drill and practice exercises such as the one just described can stimulate reflection in learners to some degree, solution of open-ended homework problems that require iterative solution absolutely demand reflection. In the room that deals with the design of simple BJT audio voltage amplifiers, learners are required to design a BJT common emitter audio amplifier that gives a certain voltage gain to a specified load, subject to the constraints of a given Thevenin impedance for the driving circuit and a given power supply voltage. This problem involves enough variables and constraints to require an iterative approach to the design. No simple set of equations employed step-by-step suffices. The learner employs approximate analytical results and design rules for an initial design and then evaluates the design by simulating its performance with the circuit simulator and comparing the simulation results with the performance specifications. If the performance specifications are not met, the learner can then refer back to the approximate analytical results and reflect on how the design might be changed to bring its performance within the specified ranges. This iterative reflective process may be repeated several times before the learner achieves a successful design. In the process, the learner practices, as mentioned earlier, self-evaluation and, through reflection, experiences prompt feedback about relatively complex open-ended problems without direct involvement of the teacher. Articulation in ELECSIM is required when learners prepare and submit documents, using a standard word processor,

664

COMPUTER-AIDED INSTRUCTION

that describes their learning activities. For the open-ended audio amplifier design problem described above, the teacher might require the learner to submit a document file that includes (perhaps as software attachments) the PSpice schematic capture file to show the circuit of the design, design calculations, samples of the simulation output (values, graphs and so on) for the circuit, and perhaps most important a discussion about how the simulation results demonstrate that the design satisfies performance specifications, or if it does not, a discussion of why the performance specifications could not be achieved. Such a document provides considerable insight to the progress of a learner and provides the teacher a convenient means of investigating the learner’s work in more detail. For example, a double-click on the imbedded PSpice schematic capture file by the teacher or an assistant not only displays the schematic file, but makes possible an immediate simulation of the schematic displayed and subsequent investigation of the output from the learner’s design. Exploration of the various rooms and their topical contents is the essence of ELECSIM, which can be viewed as one instance of a consistent graphical user interface to a diverse collection of tools, notes, problems, media, and network resources assembled by the teacher to assist the learner. From a different perspective, solutions to the open-ended problems in ELECSIM require learners to formulate hypotheses and test the consequences. Jonassen (58) discusses how teachers can implement cognitive apprenticeship on stand-alone desktop computers, without specialized computer software, but with widely available and inexpensive software such as databases, spreadsheets, semantic networks, and expert systems. In addition to selfassessment documentation, summary statistics about performance, and portfolios for assessing learning outcomes, his suggestions include learning logs, student rankings of course objectives, think-aloud protocols, documented problem-set solutions, brief autobiographical essays on a specific learning experience, cognitive interviews, directed paraphrasing, analytical memos, classification/decision matrices, diaries and journals, experiments, concept maps, and debates. Specific applications of spreadsheets in engineering that can employ elements of cognitive apprenticeship include simulation of computational and sequential logic circuits and the solution of ordinary and partial differential equations, as well as easy evaluation of complicated equations and generation of graphs that display the results for various parameter values as part of, for example, an iterative process (59–61). Evaluation of Learning Environments Development of evaluation criteria for learning environments, electronic or not, is complicated by the absence of consensus about an underlying theoretical framework for theories of learning and instruction, as well as by several different expectations for the evaluation process (62). A particular difficulty is that approaches to evaluation that adopt, explicitly or implicitly, the viewpoint of a particular theory of learning or instruction can give negative results for a project developed from a different theoretical perspective. That is, the result of the review may follow more from the nature of the learning environment than from the degree of success it achieves according to the theoretical perspective with which it was developed and implemented. At first glance, focusing evaluation

directly on the learning accomplished by participant would seem to circumscribe this problem. Alas, deciding what should have been learned depends directly on the theoretical perspectives chosen. The matter is usually not as simple as deciding whether the focus of learning should have been facts or process, but that dichotomy illustrates the point. Moreover, application of evaluation approaches developed for a different context are not always easily adapted to the evaluation of materials for computer-aided instruction, especially those for engineering education. Evaluation, therefore, demands careful planning and work. In practice, it is not easy. To the extent that evaluation is neglected, however, the iterative approach to design that is so traditional in engineering endeavors is not possible in the development of interactive learning environments for engineering education. The National Engineering Education Delivery System (NEEDS), mentioned earlier, developed criteria for the review and classification of software it receives for posting on its WWW site, as reported by Eibeck (63), which can serve as a useful guide. The evaluation and classification criteria used at first span nine categories. Engineering content deals mainly with whether or not the material is free of errors and corresponds to the level of the intended users. Engagement is an assessment of the appeal of the courseware to the intended users. Impact on learning indicates whether or not different learning styles are accommodated and whether or not feedback is provided to the learner. User interface evaluates consistency, clarity, and ease of use as well as the effectiveness of help features available to the learner. User interaction assesses to what extent the software involves the learner and whether the involvement is active or passive. Multimedia design considers the quality of the multimedia and whether the media effectively support the learning process or merely distract the learner, instead. Instructional use concerns how easily the software can be incorporated into a course by a teacher other than an author. Performance appraises how well the software runs on the specified computer platform. Accessibility from NEEDS deals with operational issues of finding the files in the NEEDS database and downloading them. Although it would be difficult to argue that these categories can be neglected by authors of useful software, the review process based on them proved unwieldy, in practice, to reviewers. NEEDS therefore simplified the review process to focus on ensuring that the content is error free, that the package is complete and includes descriptions and recommendations for use, that the software is appealing to users, and that it is potentially useful to teachers other than the author. Projects for developing computer-aided instruction, as well as complete courses designed with computer-aided instructional components, require a broader approach to evaluation than considering existing computer-aided instructional material for possible use. Fortunately, some widely accepted approaches to project evaluation (62) have been adapted for projects in engineering and to the development of computeraided instruction projects, as well (16,33,64). Project evaluation consists of three basic stages: (1) planning evaluation, (2) formative evaluation, and (3) summative evaluation. During development of a project, planning evaluation, although often neglected, helps focus attention on goals of the project as well as on strategies and schedules for achieving them. While the project is underway, formative evaluation identifies opportu-

COMPUTER-AIDED INSTRUCTION

nities to improve the project. After the project is completed, summative evaluation assesses the success of the project. Forsyth, Jolliffe, and Stevens discuss application of a multilevel evaluation model that permits concentration on (1) the learner’s feelings and about the course, (2) learning achievement during the course, (3) behavioral changes in the learners during the course, or (4) overall impact of an innovation in an organization (65). Worthen, Sanders, and Fitzpatrick describe and contrast several alternative approaches to evaluation that can be adapted for application to computer-aided instruction (62). Careful project evaluation is important in achieving credibility necessary for widespread use of the results of any project. It is almost essential, however, for projects in an emerging area such as computer-aided instruction in engineering where the lack of familiarity and confidence among potential users may breed skepticism that prevents widespread interest in the results of the project. Pragmatic Development of Computer-Aided Learning Environments As the power and sophistication of hardware and software available for computer-aided instruction continue to increase, contemporary answers to the fundamental question ‘‘How can we use computers to improve education effectively and inexpensively?’’ amount to discoveries of strategies for developing environments in which learners can proceed effectively. Strategies for Conceptual Design. The likelihood that teachers of engineering can discover carefully and coherently designed ready-to-use courses of study that match the needs of their students always has been low. At best, teachers can hope to find elements and components that they can incorporate into their own designs. Incorporating computer-aided instruction into the design of courses of study changes that picture but little. An analogy with engineering textbooks suggests that some teachers of engineering will play a dominant role in developing computer-aided instructional components just as they do in authoring engineering textbooks. The market for engineering instructional materials is just too small to attract full-time authors with the appropriate expertise. Clearly, the teachers of engineering who develop components for widespread use must understand principles of course design very well. If their students are to realize substantial benefits from the materials available, however, even teachers of engineering who mainly integrate computer-aided instructional elements produced by others into learning environments for their own students must understand and apply the principles of course design as well. If they do not, the risk of producing a flood of poor-quality courses that can damage the success of computer-aided instruction in engineering for a long time is great. The stakes are high. From one perspective, finding a suitable strategy for teaching a particular topic or designing a course seems confusing and, even worse, unlikely. Despite substantial developments in theories of learning and instruction, no consistent approach to designing learning environments (computer-aided or not) is widely accepted. Candidates for an overall theory of design suffer from (1) poor understanding of their domain of applicability and (2) scarcity of empirical verification. Perhaps with the availability of powerful, inexpensive computer systems

665

and networks, the development and verification of design approaches based on theories of learning and instruction eventually will be forthcoming, but not now. That pessimistic view comes easily to engineers who are accustomed to working with experimentally verified theories, such as those of electromagnetics, thermodynamics or signal processing, whose power, marvelous accuracy, and domain of applicability are well understood and documented. But engineers have long worked successfully where theories are far less robust. Management is just one example. A manager responsible for some particular effort finds no powerful universally sanctioned theoretical approach to managing the activities of a particular project. Indeed, a manager easily can become bewildered by the multiplicity of diverse and inconsistent approaches advocated by hosts of management theorists. And yet, individuals find ways to manage complex projects successfully. How? The situation is not quite as complicated as first it seems. Certain fundamental principles of management have become widely accepted and understood (66). These principles, alone, do not give very specific guidance to a manager, but they are ignored at great peril. More detailed guidance is provided by more specialized theories, such as Theory Z or total quality management (TQM), whose domain of applicability usually is not clear. The manager must develop a specific management approach for a particular effort based on management fundamentals and on insights provided by the specialized theories. That the specialized theories seem to come and go complicates matters, of course. A designer of computer learning environments faces a similar situation. Certain fundamental ideas about learning are becoming accepted and understood by people with diverse perspectives. A teacher must combine a knowledge of these principles with insights from more specialized theories that seem to fit the situation, topics, and learners at hand to develop successful learning environments. What are some points (67) of consensus that seem to be emerging? First, concepts are best learned when students encounter them in a variety of contexts rather than from a single perspective. Even if a learner retains a concept experienced from a single perspective, that concept is likely to be isolated and unavailable for linking with others to build related or more complex concepts. Applying the concept in a different context is an important means of understanding it. Second, realistic experiences are extremely effective in helping learners learn, especially in grasping abstractions. An abstraction not linked to several real situations is unlikely to be accessible for building understanding of diverse contexts in which it might apply. Third, learners learn effectively when they take action and then something happens in turn from which they can learn. Giving and receiving feedback in a peer group is one example. Physical experiments are another. Interaction with simulations offers a third possibility. In short, an emerging consensus is that learning should be active and experiential. Tools for Implementation. Although careful conceptual design of a computer-aided learning environment is essential, the realization of the environment in practical and robust software may require more effort and produce greater frustration. The PLATO system included TUTOR, perhaps the first widely used tool designed specifically for helping teachers to become authors of interactive learning environments (3). Cy-

666

COMPUTER-AIDED INSTRUCTION

berProf provides a set of tools for the same purpose (25). More generally, almost any programming language can serve, in principle, as a tool for constructing interactive learning environments. In practice, the complexities of building an acceptable graphical user interface alone require powerful software tools if mere mortals are to succeed. Fortunately, the business market has stimulated the development of numerous powerful, easy-to-use, and relatively inexpensive tools for potential authors of computer-aided instruction. Unfortunately, the large number of authoring programs available and the variety of features included (and omitted) can make the choice bewildering. Helpful perspective is provided by Schwier and Misanchuck, who describe a number of features that should be considered in selecting an authoring program appropriate for construction of multimedia learning environments (17). Portability determines what fraction of desktop machines can use the learning environment constructed. Although the WWW has simplified the problem of portability to some extent, the basic hypertext markup language (HTML) environment on the WWW is not as rich as many platform-specific environments. Two trends are ameliorating this problem. First, HTML itself is being continually upgraded to provide a richer environment. Second, WWW browsers now accommodate special file-type-specific plug-ins that permit browsers to display, or play, files of almost any type, provided only that an appropriate plug-in is available and it has been installed in the browser on a particular machine. Just as browsers are platform specific (Windows, Macintosh, UNIX), so too are plugins, and it might seem that plug-ins accomplish little. The number of computers on the WWW is so large, however, that competitive pressures have led many vendors of software, including vendors of authoring programs, to make available, free of charge and for all major platforms, plug-ins that accommodate file types special to their products. With this approach, multimedia environments can be created with an authoring program that runs only on a single platform, although the resulting files can be displayed by browsers on any platform. Which file types ultimately may prove popular enough that capability for handling them is built into browsers and which file types will continue to be supported by special plugins is decided by complex market processes the outcome of which is difficult to foresee. Nevertheless, the WWW clearly is becoming an increasingly rich environment for computeraided instruction. Licensing agreements must be purchased before software produced with some authoring programs can be distributed. WYSIWYG (what you see is what you get) is an almost essential, but not universal, feature for contemporary authoring programs. Not being able to see an environment that you are designing without interrupting the design process specifically to display the environment wastes time and precludes convenience. Flexibility means that the author, not the authoring tool, should determine the kind of learning environment to be constructed, although advanced author support that provides suggestions to help the author maintain instructional integrity according to some particular design paradigm may be beneficial. The ideal authoring tool should accommodate animation, video, and audio, as well as text and graphics. User control concerns the degree to which the authoring program supports, for example, a keyboard, mouse, graphics tablet, touch screen, light pen, speech recognition interface, barcode

reader, or virtual reality interface. The authoring program also should be extensible through programming features that give ready access to a high-level programming language that, among other things, accommodates bridges to external software and permits authors to add custom features to the authoring program. Performance tracking includes features such as answer judging and activity reporting. Networkability is a measure of how well the authoring program itself, and the software it produces, works on networks. If the learning environment constructed is for deployment on the WWW, for example, the availability of suitable plug-ins is an important consideration. Overall, the desirable features of an authoring program amount to the requirement mentioned earlier for a successful graphical user interface: transparency. An author should be able to concentrate on designing the learning environment without distraction or frustration by the authoring program. SIMPLE, a Windows authoring program available on an archived CD-ROM and used to create ELECSIM, is especially designed for constructing interactive learning environments (13). It provides a WYSIWYG authoring environment and straightforward extensibility through Visual Basic and incorporation of external software, includes performance-tracking features and configuration management tools for network deployment of the learning environments, incorporates multimedia and simple animation, and carries no license fee for educational use. Although learning environments for the WWW offer unprecedented portability, the available authoring tools (apart from those for CyberProf) at first offered little more than capability for constructing graphical user interfaces for the learning environments. Platform-specific authoring environments were unrivaled in power and flexibility. Emergence of the JAVA programming language, however, promises development of authoring environments for the WWW that provide, in addition to the boon of portability, the power and flexibility previously available only with platform-specific tools. JAVA, an updated and improved version of the powerful object-oriented C⫹⫹ programming language, is designed specifically to achieve (1) seamless incorporation of the WWW into software and (2) cross-platform portability far greater than provided by C⫹⫹. JAVA programs, like HTML documents, require only a machine with a suitable WWW browser for use. Like HTML documents, JAVA programs (or applets), can be written to function perfectly well even on non-networked machines, although hyperlinks to sources on the WWW and calls to network servers are not possible, of course. Thus, as the huge WWW market stimulates the development of powerful and sophisticated HTML authoring programs that embody JAVA, it seems likely that the WWW (or a WWWlike environment on non-networked machines) will become the environment of choice for most computer-aided instruction, even that intended for use mainly on non-networked machines. Early applications of JAVA in interactive learning environments began to appear in 1997 (68). Scripting languages such as JavaScript, supported by most WWW browsers, offer the possibility of including and executing simple program procedures in HTML documents, although they provide far less capability than JAVA (13). Current Status. Engineering teachers only have begun the difficult task of sorting through a multiplicity of conflicting

COMPUTER-AIDED INSTRUCTION

approaches to find useful ways of relying on computer-aided instruction (69–71). The current best practice in assessing the fundamental question of computer-aided instruction, ‘‘How can we use computers to improve education effectively and inexpensively?’’, for a particular learning situation is to apply the familiar approach of engineering design to the development of interactive learning environments, together with tactics suggested by available theories of learning and instruction, principles of interface design, and intuition as guides. This approach avoids the limitations inevitable in commitment to a single approach and gives creativity reign within constantly changing constraints. Only time will tell how satisfactory the best answers to the fundamental question prove in practice.

BIBLIOGRAPHY 1. W. L. Everitt, Engineering education—circa 2012 A.D. Proc. IRE, 50: 571–572, 1962. 2. R. F. Mager (ed.), Special issue on automated teaching, IRE Trans. Educ., E-4: 1961. 3. D. L. Bitzer, The wide world of computer-based education. In M. Rubinoff and M. Yovits (eds.), Advances in Computers, New York: Academic Press, 1976, Vol. 15. 4. D. L. Bitzer, The million terminal system of 1985. In R. J. Seidel and M. L. Rubin (eds.), Computers and Communications: Implications for Education, New York: Academic Press, 1977.

667

17. R. A. Schwier and E. R. Misanchuk, Interactive Multimedia Instruction, Englewood Cliffs, NJ: Educational Technology Publications, 1993. 18. E. R. Doering, CircuitViz: a new method for visualizing the dynamic behavior of electric circuits, IEEE Trans. Educ., 39: 297– 303, 1996. See especially material on the accompanying CD-ROM in folder ⶿ieeezips⶿005⶿cd⶿. 19. R. O. Harger, Teaching in a computer classroom with a hyperlinked interactive book, IEEE Trans. Educ., 39: 327–335, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿010⶿cd⶿. 20. S. L. Wood, A new approach for interactive tutorial software for engineering education, IEEE Trans. Educ., 39: 399–408, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿026⶿cd⶿. 21. M. O. Hagler et al., Standards, the Virtual University and CDROM/WWW Technology, 1996 ASEE Int. Conf. Proc., CD-ROM, file /PAPERS/HAGLER.PDF. 22. J. B. Schodorf et al., Using multimedia to teach the theory of digital multimedia signals, IEEE Trans. Educ., 39: 336–341, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿011⶿cd⶿. 23. J. McClellan, R. Schafer, and M. Yoder, DSP First: a multimedia approach, Upper Saddle River, NJ: Prentice-Hall, 1998. 24. A. J. Sears and S. E. Watkins, A multimedia manual on the World Wide Web for telecommunications equipment, IEEE Trans. Educ., 39: 342–348, 1996. See especially material on the accompanying CD-ROM in folder ⶿ieeezips⶿012⶿cd⶿.

5. P. Saettler, The Evolution of American Educational Technology, Englewood, CO: Libraries Unlimited, 1990, Chaps. 1, 3, 10–12, and 14–16.

25. A. W. Hubler and A. M. Hassad, CyberProf: An intelligent human-computer interface for asynchronous widearea training and teaching. In Proc. 4th Int. WWW Conf., pp. 231–238, 1995. See also http://www.w3.org/pub/Conferences/WWW4/Papers/247/.

6. T. O’Shea and J. Self, Learning and Teaching with Computers, Englewood Cliffs, NJ: Prentice-Hall, 1983, Chaps. 3, 4, 6, and 7.

26. The URL of the NEEDS home page on the WWW is http:// www.needs.org.

7. A. Collins, Cognitive apprenticeship and instructional technology. In L. Idol and B. F. Jones (eds.), Educational Values and Cognitive Instruction: Implications for Reform, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1991, pp. 121–138.

27. The URL for the World Lecture Hall is http://www.utexas.edu/ world/lecture/index.html.

8. D. R. Woolley, PLATO: the emergence of on-line community, Comput.-Mediated Commun. Mag., 1 (3): 5, 1994. 9. J. S. Daniel, Why universities need technology strategies, Change, 29 (4): 11–17, 1997. 10. J. S. Daniel, Mega-Universities and Knowledge Media: Technology Strategies for Higher Education, London: Kogan Page, 1996. 11. D. H. Jonassen (ed.), Instructional Designs for Microcomputer Courseware, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1988. 12. B. Oakley II, A virtual classroom approach to teaching circuit analysis, IEEE Trans. Educ., 39: 287–297, 1996. See especially material on the accompanying CD-ROM in file ⶿ieeezips⶿002⶿cd⶿tutorial.htm.

28. M. Hagler, Hardware homework for courses in circuits and electronics. In Lawrence P. Grayson (ed.), Proc. 1994 Frontiers Educ. Conf., Piscataway, NJ: IEEE, 1994, pp. 557–561. 29. The URL of the Sloan Foundation home page on the WWW is http://www.sloan.org. 30. D. Minoli, Distance Learning Technology and Applications, Boston: Artech House, 1996. 31. R. A. Reiser and W. Dick, Instructional Planning: A Guide for Teachers, 2nd ed., Boston: Allyn and Bacon, 1996. 32. M. P. Driscoll, Psychology of Learning for Instruction, Boston: Allyn and Bacon, 1994. 33. E. R. Steinberg, Computer-Assisted Instruction: A Synthesis of Theory, Practice, and Technology, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1991, Chaps. 3, 9, and 10.

13. W. M. Marcy and M. O. Hagler, Implementation issues in SIMPLE learning environments, IEEE Trans. Educ., 39: 423–429, 1996. See especially material on the accompanying CD-ROM in file ⶿ieeezips⶿033⶿cd⶿elecsim⶿qselecsm.htm.

34. R. M. Gagne, L. J. Briggs, and W. Wager, Principles of Instructional Design, 4th ed., New York: Rinehart and Winston, 1992.

14. P. J. Mosterman et al., Design and implementation of an electronics laboratory simulator, IEEE Trans. Educ., 39: 309–313, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿007⶿cd⶿.

36. M. D. Merrill, Constructivism and instructional design. In T. M. Duffy and D. H. Jonassen (eds.), Constructivism and the Technology of Instruction: A Conversation, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1992, pp. 99–114.

15. T. G. Engel, Splice: an analytical network analysis software, IEEE Trans. Educ., 39: 394–398, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿023⶿cd⶿.

37. M. D. Merrill, Instructional Design Theory, Englewood Cliffs, NJ: Educational Technology Publications, 1994.

16. T. Boyle, Design for Multimedia Learning, London: PrenticeHall, 1997.

35. W. Dick and L. M. Carey, The Systematic Design of Instruction, 4th ed. New York: Harper Collins, 1996.

38. S. A. Mengel and W. J. Adams, The need for a hypertext instructional design methodology, IEEE Trans. Educ., 39: 375–380, 1996.

668

COMPUTER-AIDED PRODUCTION PLANNING

39. J. M. Nyce and P. Kahn (eds.), From Memex to Hypertext: Vannevar Bush and the Mind’s Machine, Boston, MA: Academic Press, 1991. 40. J. H. Larkin and R. W. Chabay (eds.), Computer-Assisted Instruction and Intelligent Tutoring Systems: Shared Goals and Complementary Approaches, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1992. 41. C. Frasson and G. Gauthier (eds.), Intelligent Tutoring Systems: At the Crossroads of Artificial Intelligence and Education, Norwood, NJ: Ablex, 1990. 42. E. Wenger, Artificial Intelligence and Tutoring Systems: Computational and Cognitive Approaches to the Communication of Knowledge, Los Altos, CA: Morgan Kaufmann, 1987. 43. J. S. Brown and R. R. Burton, Multiple representations of knowledge for tutorial reasoning. In D. G. Bobrow and A. Collins, Representation and Understanding: Studies in Cognitive Science, New York: Academic, 1975, pp. 311–349. 44. J. S. Brown, R. R. Burton, and J. de Kleer, Pedagogical, natural language and knowledge engineering techniques in SOPHIE I, II, and III. In D. Sleeman and J. S. Brown (eds.), Intelligent Tutoring Systems, London: Academic, 1982, pp. 227–282. 45. A. Lesgold et al., SHERLOCK: a coached practice environment for an electronics troubleshooting job. In J. H. Larkin and R. W. Chabay (eds.), Computer-Assisted Instruction and Intelligent Tutoring Systems: Shared Goals and Complementary Approaches, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1992, pp. 201–238. 46. D. A. Kolb, Experiential Learning: Experience as the Source of Learning and Development, Englewood Cliffs, NJ: Prentice-Hall, 1984. 47. B. G. Davis, Tools for Teaching, San Francisco: Jossey-Bass, 1993. 48. R. Felder, Matters of style, ASEE Prism, 5 (4): 18–23, 1996. 49. S. F. Chipman, Integrating three perspectives on learning. In S. L. Friedman, K. A. Klivington, and R. W. Peterson (eds.), The Brain, Cognition and Education, Orlando: Academic Press, 1986, pp. 203–229. 50. T. H. Duffy and D. H. Jonassen (eds.), Constructivism and the Technology of Instruction, Hillsdale, NJ: Lawrence Erlbaum Assoc., 1992. 51. R. A. Schwier, Issues in emerging interactive technologies. In G. J. Anglin (ed.), Instructional Technology: Past Present and Future, 2nd ed., Englewood, CO: Libraries Unlimited, 1995, pp. 119–130. 52. A. Collins, Design issues for learning environments. In S. Vosniaadou, E. De Corte, R. Glaser, and H. Mandl (eds.), International Perspectives on the Design of Technology-Supported Learning Environments, Mahwah, NJ: Lawrence Erlbaum Assoc., 1996, pp. 347–361. 53. B. Wilson, J. Teslow, and R. Osman-Jouchoux, The impact of constructivism (and postmodernism) on ID fundamentals. In B. B. Seels (ed.), Instructional Design Fundamentals: A Review and Reconsideration, Englewood Cliffs, NJ: Educational Technology Publications, 1995, pp. 137–157. 54. D. LeBow, Constructivist values for instructional systems design: Five principles toward a new mindset. In B. B. Seels (ed.), Instructional Design Fundamentals: A Review and Reconsideration, Englewood Cliffs, NJ: Educational Technology Publications, 1995, pp. 175–187. 55. C. T. Sun and C. Chou, Experiencing CORAL: design and implementation of distant cooperative learning, IEEE Trans. Educ., 39: 357–366, 1996. 56. S. R. Hiltz, The Virtual Classroom: Learning without Limits via Computer Networks, Norwood, NJ: Ablex, 1994. 57. M. O. Hagler et al., Publication of archival journals accompanied by CD-ROMs, IEEE Trans. Educ., 40 (4): 1997, CD-ROM file ⶿CDROM⶿HTML⶿JAVASCRP⶿JAVASCRP.HTM.

58. D. H. Jonassen, Computers in the Classroom: Mindtools for Critical Thinking, Englewood Cliffs, NJ: Merrill, 1996. 59. M. O. Hagler, Spreadsheet solution of partial differential equations, IEEE Trans. Educ., E-30:130–134, 1977. 60. L. P. Huelsman, Electrical engineering applications of microcomputer spreadsheet analysis programs, IEEE Trans. Educ., E27:86–92, 1984. 61. A. Kharab and R. Kharab, Spreadsheet solution of hyperbolic partial differential equations, IEEE Trans. Educ., 40: 103–110, 1997. 62. B. R. Worthen, J. R. Sanders, and J. L. Fitzpatrick, Program Evaluation: Alternative Approaches and Practical Guidelines, New York: Longman, 1997. 63. P. Eibeck, Criteria for per-review of engineering courseware on the NEEDS database, IEEE Trans. Educ., 39: 381–387, 1996. See also material on the accompanying CD-ROM in folder ⶿ieeezips⶿021⶿cd⶿. 64. F. Stevens, F. Lawrenz, and L. Sharp, In J. Frechtling (ed.), UserFriendly Handbook for Project Evaluation: Science, Mathematics, Engineering and Technology Education, Washington, D. C., National Science Foundation, NSF 93-152, 1993. 65. I. Forsyth, A. Jolliffe, and D. Stevens, Evaluating a Course: Practical Strategies for Teachers, Lecturers and Trainers, London: Kogan Page, 1995. 66. P. F. Drucker, Management: Tasks, Responsibilities, Practices, New York: Harper and Row, 1974. 67. F. J. Rutherford and A. Ahlgren, Science for All Americans, New York: Oxford University Press, 1990, pp. 185–188. 68. M. Chiorico et al., The real experiment execution approach to networking courseware, IEEE Trans. Educ., 40: 1997 (CD-ROM folder 15). 69. M. O. Hagler (guest ed.), Special issue on the application of information technologies to engineering and science education, IEEE Trans. Educ., 39: 285–454, 1996. See also material on the accompanying CD-ROM. 70. T. E. Batchman (ed.), IEEE Trans. Educ., 40: 1997. See especially the CD-ROM Rapid Publication Supplement. 71. M. F. Iskander (ed.), Computer Applications in Engineering Education, past and current issues, New York: Wiley.

MARION O. HAGLER WILLIAM M. MARCY Texas Tech University

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2902.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Computer Engineering Education Standard Article James G. Harris1 1California Polytechnic State University, San Luis Obispo, CA, Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2902 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (87K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2902.htm (1 of 2)17.06.2008 21:41:32

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2902.htm

Abstract The sections in this article are Early Development of Computer Engineering Education Development of Accreditation Activities Organization of Computer Engineering Education Basic Computer Engineering Curriculum Future Trends | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2902.htm (2 of 2)17.06.2008 21:41:32

COMPUTER ENGINEERING EDUCATION

COMPUTER ENGINEERING EDUCATION Just like the area of computers, the area of computer engineering education is a fast changing field. This discussion will limit itself to the development of accredited undergraduate computer engineering education programs in the United States, and their relation to accredited programs in electrical engineering and computer science. Fortunately, because the United States has been a recognized leader in the development of computers, this narrow discussion on computer engineering education will address most of the issues regarding computer engineering education in the global context. This approach is not taken to deprecate the contributions to computer engineering education outside the United States but rather to allow this discussion to succinctly identify and address the issues that are intrinsic to all computer engineering education programs. Furthermore, since accredited undergraduate computer engineering programs now are the principal source for providing engineers who work in the discipline of computer engineering, this discussion will not address graduate education in the area of computer engineering. It is sufficient to state that graduate education in either computer science or electrical engineering will accommodate all the areas of further study in computer engineering. To understand the nature of computer engineering education, it is necessary to relate it to the education of the computer scientists and the electrical engineers. While electrical engineering education is over 100 years old, computer science education is less than 40 years old. Accredited electrical engineering programs were established in 1936 by the precursor to the current Accreditation Board of Engineering and Technology (ABET) and accredited computer science programs were established in 1986 by the Computing Sciences Accreditation Board (CSAB). Even though computer engineering education evolved out of the computer science and electrical engineering programs, initial accreditation for computer engineering programs was administered by ABET and first granted in 1971, prior to the computer science accreditation activity. With establishment of CSAB, there appeared a dual accreditation designation for a program that educated computer engineers which was named computer science and engineering; the dual accreditation required both ABET and CSAB review of the program. Thus accredited undergraduate computer engineering education is performed at universities with computer engineering, or similarly named programs, or with a computer science and engineering program. It should be noted that there are university programs leading to a degree in computer engineering, or similarly named degrees, that are not accredited; however, without reviewing the specific curriculum and other aspects of the program, it is not possible to say with certainty that it is a program that successfully prepares students for the practice of computer engineering.

EARLY DEVELOPMENT OF COMPUTER ENGINEERING EDUCATION Computer engineering education has developed commensurately with the development of the computer. Initially the education of the people who worked as engineers in the com-

701

puter field was in other disciplines, electrical engineering, mathematics, and physics, to name a few. The first electronic computers were developed at universities. John Atanasoff (1) designed and constructed a digital computer at Iowa State at the start of World War II, and J. Presper Eckert and John Mauchly designed and built their computer, the ENIAC (Electronic Numerical Integrator and Calculator) at the Moore School at the University of Pennsylvania at the end of World War II (2). Further development of the digital computer occurred at universities such as Princeton, Harvard, Cambridge, and MIT among others, in the late 1940s (3). The activities of the graduate and postgraduate students at these institutions provided the education of ‘‘computer engineers,’’ if you assume that the ability to design and build a computer is a central component of an education of a computer engineer. Claude Shannon’s M.S. thesis at MIT in 1938 (4) related the work of George Boole and the Huntington postulates to switching circuits. In the 1950s this material began to appear in undergraduate electrical engineering curricula as a technical elective course in digital logic design. During this same era, courses in computer programming and computing theory that would evolve into the computer science curriculum were beginning to be offered in electrical engineering and mathematics programs. As the computer science curriculum developed, there occurred a bifurcation in the development of the computer science curriculum. The direction that the curriculum took depended on the disciplines the faculty that taught the curriculum came from, electrical engineering or mathematics. In addition the discipline of the faculty who taught the early computer science programs, eventually determined whether the new departments of computer science resided in the college of engineering, or in the college of letters and sciences, or, in some cases, resided as a department in each college, one being engineering oriented and the other mathematics oriented. The engineering-oriented computer science faculty, along with the computer-oriented electrical engineering faculty contributed to the development of the first computer engineering programs.

DEVELOPMENT OF ACCREDITATION ACTIVITIES One of the first recognized model computer science curricula was published by the Association for Computing Machinery (ACM) in 1968 (5). With this event there was established a nationally recognized computer science curriculum that evolved with subsequent model curriculum versions in 1978 and 1991 (5). With the standardization of the computer science curriculum, there was a movement to accredit computer science programs. Using the model of the Accreditation Board of Engineering and Technology (ABET), the Computing Sciences Accreditation Board (CSAB) was formed and accredited the first computer science programs in 1986. ABET granted accreditation to a program named Computer Engineering at Case Western Reserve University in 1971. Prior (and subsequent) to this event, there were a number of reports on curriculum for computer engineering education; the reader is referred to the references in the Bibliography for details. The ABET accreditation of a program called Computer Engineering preceded the computer science accreditation activities. Subsequent to this initial ABET accreditation action for programs named computer engineering, Taylor

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

702

COMPUTER ENGINEERING EDUCATION

Accredited programs

250 200 150 100 50 0 1930

1940

1950

1960 1970 Year

1980

1990

2000

Figure 1. Accredited programs in electrical engineering (closed diamond), computer engineering (open diamond), computer science and engineering (closed square), and computer science (open square), cumulative from the initial year of their accreditation (8,9).

Booth of the University of Connecticut led an effort to define a model Computer Science and Engineering curriculum based on both the computer science and the electrical engineering curricula. This activity lead to the publication of a model program in Computer Science and Engineering in 1983 (6). This model curriculum was used by many universities to establish accredited computer engineering programs using ABET, as well as the dual accredited ABET/CSAB criteria. Currently there are two criteria that apply to every accredited computer engineering program: the ABET criteria, called computer engineering or similarly named programs, and the dual ABET/ CSAB criteria for programs named computer science and engineering. Figure 1 shows the relative growth of the four different accredited programs: electrical engineering, computer engineering, computer science and engineering, and computer science over the period of accreditation. ORGANIZATION OF COMPUTER ENGINEERING EDUCATION Undergraduate computer engineering education is a work in progress. The state of the curriculum for undergraduate computer engineering education is changing as fast as the field that its majors enter upon graduation. The state of the computer engineering curriculum also is affected by the administrative organization in which it resides. Most everyone will agree that foundation of the computer engineering curriculum is based on the disciplines of electrical engineering and computer science. The determination of the point of balance between the two disciplines for the foundation of the computer engineering curriculum is a matter that is determined by each university’s faculty and is influenced by the organizational structure that administers the computer engineering program. There are examples of every possible organization: separate computer engineering departments within either a college of engineering or another college; computer engineering programs within departments of electrical engineering or within departments of computer science; interdisciplinary computer engineering programs between separate electrical engineering and computer science departments, where the two departments may be within the same college or in separate colleges; and computer engineering programs within administrative units that contain both the electrical engineering and computer science programs. The different organizational structures, of course, determine the computer

engineering faculty administrative unit and hence the local implementation of the curriculum for computer engineering. In 1989 the members of the National Electrical Engineering Department Heads (NEEDHA) at their annual meeting held in San Diego, California, discussed the issue of the administrative structure of computer engineering programs. Their conclusion was that all logical structures are possible and that there was no recommended structure. Each university had to determine a structure to fit local needs (7). That conclusion appears to hold true in the late 1990s. BASIC COMPUTER ENGINEERING CURRICULUM All of the accredited computer engineering programs require four years of university education and include at least one year of basic mathematics and science, a year and one-half of engineering and computer science topics, and one-half year of social science and humanities courses. This leaves one year open for university faculty to specify other courses. For the dual accredited programs, there are additional computer science courses required. The year of mathematics and science usually contains science courses in chemistry, physics, and biology and mathematics courses in calculus through differential equations, discrete mathematics, and statistics. The engineering and computer science courses are usually balanced between electrical engineering and computer science offerings. Typically the computer engineering major will be required to take at least the same lower division courses as do each of the computer science and electrical engineering majors, and then a number of their required upper division courses. In addition there are usually a number of specialized courses that address the software/hardware interface, computer architecture, computer networks, and embedded microprocessor system design as well as advanced courses in computer science such as computer graphics and artificial intelligence, and in electrical engineering such as VLSI design and system theory. Table 1 shows a typical accredited computer engineering program for reference. Also presented in the Table 1 data is a summary of the approximate units in each of the accreditation categories yielding a total number of 128 units. The term used for this presentation is the 15 week semester, which generally has a one-hour lecture for each unit and three-hour laboratory for each unit. It should be emphasized that while this display is typical, there are many ways to satisfy the criteria for an accredited computer engineering program that deviate from this display. FUTURE TRENDS At the undergraduate level there are defined criteria for undergraduate computer engineering programs. However, in the proposed ABET Criteria 2000 to be instituted in the year 2000, the program criteria for electrical engineering and computer engineering have been merged into one program criteria. Thus, except for the name of the organizational unit in which the computer engineering program resides, there could be little distinction between the electrical engineering and the computer engineering programs. If this proposal is ratified, then after the year 2000 the only distinct accrediting criteria for computer engineering programs may be the dual ABET/ CSAB accreditation criteria for programs named computer

COMPUTER ENGINEERING EDUCATION

703

Table 1. Typical Accredited Computer Engineering Curriculum First Year Term 1 Computer Science I(cs) Discrete Mathematics(ms) Chemistry(ms) English Communication I(o) Digital Logic(cpe,ee)

Term 2 Computer Science II(cs) Calculus I(ms) Biology(ms) English Communications II(o)

Second Year Term 1 Data Structures(cs) Circuit Analysis(ee) Calculus II(ms) Physics I(ms) Social Science and Humanities(ssh)

Term 2 Software Engineering(cs) Electronics(ee) Differential Equations(ms) Physics II(ms) Social Science and Humanities(ssh) Third Year

Term 1 Operating Systems(cs) Computer Architecture I(cpe,cs) Digital Electronics(ee) Statistics(ms) Social Science and Humanities(ssh)

Term 2 Programming Languages(cs) Computer Architecture II(cpe,cs) Linear Systems Theory(ee) Engineering Science Elective(ee) Social Science and Humanities(ssh) Fourth Year

Term 1 Microprocessor-Based System Design(cpe) Computer Networks(cpe,cs) Technical Elective(cs,ee) Technical Elective(cs,ee) Social Science and Humanities(ssh)

Term 2 Senior Project(cpe) Engineering Profession(o) Technical Elective(cp,cpe,ee) Technical Elective(cs,cpe,ee) Social Science and Humanities(ssh)

Approximate Unit Distribution Category Basic Mathematics and Science(ms) Social Science and Humanities(ssh) Engineering and Computer Science Topics(cs,cpe,e) Other(o) Total

science and engineering. Therefore curriculum for programs called computer engineering after the year 2000 could be the same as those called electrical engineering and, in fact, could merely be options within the electrical engineering program, or the electrical engineering curriculum could be an option within the computer engineering program. In addition a task force of ABET and CSAB members was formed in 1995 with the charge to plan the merger of CSAB into ABET. The results of such efforts are expected to have an effect on the accreditation of computer engineering programs in the future. So the accreditation process for computer engineering is in a state of flux at this time. Obviously computer engineering education is also in a state of flux, and will be evolving rapidly in the next few years. Given the state of computer engineering education and its program, this discussion has to be considered a snapshot in time. This material provides some background for what will follow in the next few years, but the story of computer engineering education is far from over.

Unit Total 35 18 66 9 128

BIBLIOGRAPHY 1. N. Stern, Who invented the first electronic computer?, Annals of the History of Computing, 2 (4): 375–376, 1980. 2. H. H. Goldstine, The Computer: From Pascal to von Neumann, Princeton, NJ: Princeton Univ. Press, 1972. 3. W. A. Burks, H. H. Goldstine, and J. von Neumann, Preliminary discussion of the logical design of an electronic computing instrument, Report to the US Army Ordinance Department, p. 1, in W. Aspray, A. Burks (ed.), Papers of John von Neumann, Cambridge, MA: MIT Press, 1987, pp. 97–146. 4. Claude E. Shannon, A symbolic analysis of relay and switching circuits, Trans. AIEE, 57: 713–723, 1938. 5. Computing Curricula 1991 Report of the ACM/IEEE-CS Joint Curriculum Task Force, Association for Computing Machinery, Incorporated, ACM, 1515 Broadway, New York, NY 10036-5701, 1991. 6. The Model Program Committee, J. T. Cain, G. B. Langdon, Jr., M. R. Varanasi (co-chairs), The 1983 IEEE Computer Society Model Program in Computer Science and Engineering, New York, IEEE

704

COMPUTER EVALUATION Educational Activities Board, 1983 (updated a 1977 committee report).

7. Keith Carver, History of the National Electrical Engineering Department Heads Association: A comprehensive look at NEEDHA’s development, June 1992, available from http:// www.needha.org/history.shtml. See Fifth Annual NEEDHA meeting: San Diego (1989). 8. 1996 Annual Report for Year Ending September 30, 1996; Computing Sciences Accreditation Board, Incorporated, Two Landmark Square, Suite #209, Stamford CT 06901. See Programs in Computing Accredited by the Computer Science Accreditaton Commission (CSAC) of the Computing Sciences Accreditation Board (CSAB) as of October 1995 and Followed by Year of Initial Accreditation and the Engineering Accreditation Commission (EAC) of the Accreditation Board for the Engineering and Technology (ABET). 9. 1996 Annual Report for the Year Ending September 30, 1996, Accreditation Board for Engineering and Technology, Incorporated, 111 Market Place, Suite 1050, Baltimore, MD 21202, pp. 14–34, 40–41. Reading List Note: To learn more about the disciplines and accreditation, the following website homepages can be visited at the URL listed. Disciplines http://www.acm.org Association for Computing Machinery http://www.ieee.org Institute of Electrical and Electronics Engineers http://www.needha.org National Electrical Engineering Department Heads Association http://computer.org IEEE Computer Society Accreditation http://www.csab.org/앑csab Computing Sciences Accreditation Board http://www.abet.ba.md.us Accreditation Board for Engineering and Technology

JAMES G. HARRIS California Polytechnic State University

COMPUTER ETHICS. See SOCIAL AND ETHICAL ASPECTS OF INFORMATION TECHNOLOGY.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2903.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Continuing Education Standard Article Edwin C. Jones Jr.1 1Iowa State University, Ames, IA Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2903 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (111K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2903.htm (1 of 2)17.06.2008 21:41:49

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2903.htm

Abstract The sections in this article are Defining Continuing Education Need for Continuing Education Topics for Discussion Graduate Study Continuing Education Programs Distance Education Career Development and Lifelong Learning Conclusion | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2903.htm (2 of 2)17.06.2008 21:41:49

CONTINUING EDUCATION

247

CONTINUING EDUCATION One of the outstanding engineering educators of all time, William L. Everitt, former Dean of Engineering at the University of Illinois, often said, ‘‘Engineering is not a learned profession, it is a learning profession.’’ This statement captures the essence of this article, which discusses the need for engineers to continue to learn throughout their professional careers, and discusses some of the ways engineers find useful to engage in this pursuit. The need has existed for a long time, and engineers have utilized the best available technologies and techniques to meet the need. The importance of continuing education continues to increase as the pace of technological change increases, and as more knowledge is available and used. The focus of this article will be on programs designed by industry, government, and the university. DEFINING CONTINUING EDUCATION In its most general sense, the phrase continuing education usually refers to a formal program of post-baccalaureate-degree education established for the purpose of maintaining and enhancing knowledge and skills. Often there is a connotation of a job-related activity. The program may lead to a graduate or professional degree, or it may lead to a certificate of accomplishment or attainment. It may be part of a legal requirement to maintain a license in a profession. The outcome may be personal satisfaction. The program may last for a few hours or for several years. The term ‘‘continuing education’’ is extremely broad, and so are the ways in which engineers accomplish the goals of continuing education. There are, however, some related terms that help clarify the ideas. Many of the terms have connotations that are significant to the user. Some universities are now using the phrase extended education to describe continuing education activities that are designed to enable engineers to earn post-baccalaureate degrees on a part-time basis while fully employed in industry or government. This term is an attempt to distinguish between types of activity within the university, but at present it does not seem to have widespread acceptance. A closely related term is lifelong learning. This latter phrase applies to an individual who engages in continuing education, so that the term describes the participation more than the process. NEED FOR CONTINUING EDUCATION Engineers, indeed practitioners of all professions, have always sought better ways for accomplishing their goals. They seek to improve performance, efficiency, reliability, and aesthetics, and to reduce costs. Where possible they share ideas. They envision new ways to improve the quality of life for people. They perform research and try new ideas continually. In short, they learn from each other in discussions, in classes, at J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

248

CONTINUING EDUCATION

conferences, by reading, and in formal programs designed for this purpose. While the need for continuing education has always existed, it takes on increasing importance in today’s world. It is generally agreed that knowledge is ever increasing, and that the challenge of keeping abreast of changes, while at the same time evaluating the information available, poses a significant challenge to everyone. Today’s engineer must be aware of changes in the practice of engineering close to her or his field, and must be aware of changes away from this field, as engineering is becoming more interdisciplinary. Formal programs and self-discipline are required for engineers to be able to maintain and enhance a high level of competence in the profession. There is another reason for continuing education. Nearly all engineers simply want to continue learning all of their lives, both within their profession and for personal self-satisfaction. To achieve the latter, reading is necessary, but often formal programs are needed for effective learning. TOPICS FOR DISCUSSION This article will discuss several topics that are closely related to continuing education. The first important topic will be graduate study in engineering. Graduate study is a domain entirely of the universities, but may take place either on or far away from the campus. If the students are located away from the university, then some form of distant education technology is involved, and this topic is discussed. Continuing education programs offered with a goal of advanced learning, often with certificates but not formal degrees, are offered by a wide variety of agencies, and these are discussed. Finally, some of the terms frequently used by engineering educators, especially those involved in some form of continuing education, are discussed. GRADUATE STUDY What Is Graduate Study? The phrase graduate study refers to formal educational programs offered by universities that provide additional study and degree programs beyond the baccalaureate (bachelors) degree programs. These programs have two major foci, though the dividing lines between the two are not rigorously drawn. First, the primary purpose of graduate study traditionally has been to prepare students for careers in research in their respective disciplines, though of course not all recipients have pursued such careers. Typically programs include study of a discipline at an advanced level, high-level study in closely related disciplines, and preparation of a thesis or dissertation. The latter is invariably documentation of an independent, though carefully guided, research effort by the candidate. Not only must it be defended within the university, but significant parts of it must normally be submitted for publication in peerreviewed professional journals as well. The highest level of these degrees is almost always called Doctor of Philosophy (Ph.D.). The degree intermediate between the Ph.D. and the baccalaureate degree is usually the Master of Science (M.S.) degree. Often, but not always, the M.S. also requires preparation of a thesis.

A second type of post-baccalaureate graduate study leads to a professional degree. In a few cases, these may be bachelors degrees, for example, Bachelor of Engineering or equivalent, but more common titles, in engineering, are Master of Engineering, Doctor of Science, or Doctor of Engineering. A professional degree is no less rigorous than the more common research degree, but the focus is different. The research project becomes a major development project, leading to a significant advance in an engineering application, though not necessarily basic research in engineering. Publication is less important; projects and products are more important. As a recipient of either of these degree types matures and contributes, the distinction between the types of degree diminishes. Because both programs tend to emphasize high quality, creativity, originality, and rigor, the similarities are more important than the differences. Graduate Study in the United States Graduate degree programs in the United States and in most of the world serve several important functions. One is to provide degree programs for students who want to continue their education beyond the bachelors degree, in order to prepare for careers in research, in universities, and in high-level advanced development. Another is to provide programs in which senior research scholars are able to enhance their research by attracting students who, in addition to studying in a formal sense of the word, also serve as research assistants. In this capacity they learn to do research by actually working under the close supervision of an established scholar. The faculty members who teach the advanced, graduate-level courses gain in another important way. They remain involved at the leading edge of developments in their disciplines, and continually evaluate current journal articles and products for their students. They involve their students. Many outstanding graduate-level instructors take their students to a knowledge frontier during each class period, and help them survey the scenes for future ideas. In this way, teaching graduate-level courses is actually a form of continuing education. Each idea is discussed further in the following paragraphs. The engineering profession is an old one. Until the nineteenth century, however, nearly all of the practitioners of engineering learned as apprentices rather than as students. Engineering advanced empirically in virtually every aspect. The interaction between the scientist and the engineer was minimal. Today, however, the scientist and the engineer interact on a regular basis, to the mutual advantage of both. Individuals now move freely in both directions over the vaguely defined boundary between science and engineering. Engineering advances today are almost always based in basic sciences and modern information technology and have a firm theoretical foundation. This fact alone is a major reason for the rapid advance of knowledge all seen today. This fact also leads to the need for graduate study. Engineers who want to do research or advanced development, either in a university, in a government laboratory, or in industry, now find it necessary to have an education that is firmly grounded in basic science, mathematics, and the information sciences. Such an education simply cannot be gained in a four-year baccalaureate program. Engineers whose primary motivation is teaching also find such an education absolutely necessary. Virtually all universities require

CONTINUING EDUCATION

their faculty members to have earned a Ph.D. degree before being employed as regular faculty members. They further require that their faculty continually use and extend their knowledge, whether strictly as research or in related forms. Engineering research is now a team effort. The projects are usually large and require the talents and ideas of many people. Engineering research in a university setting is usually characterized by group leaders who are senior faculty members, less-senior faculty working on closely related projects, and graduate students, who are commonly called research assistants (RAs). The RA is an important part of the team. The RA typically performs the research needed for a thesis or dissertation in this setting, and the RA’s work serves to advance the knowledge produced within the group to the benefit of all. The RA learns to research by doing it, and eventually becomes a research leader, whether in government, industry, or a university. The process is occasionally criticized as a ‘‘cloning’’ operation. Those who disagree with this criticism point out that it has led to a world-wide rapid expansion of knowledge and a concomitant expansion in the quality of life for many people. Engineering educators regularly state well-founded opinions that continuing education is necessary for engineers in government and industry. Continuing education is also necessary for engineering educators. Some of the common characteristics of a person who has completed advanced degrees are a thirst for knowledge and the willingness, self-discipline, and ability to continue learning throughout a lifetime. Knowledge expands so rapidly that relatively little of what the engineer teaches is material studied as an undergraduate or, for that matter, as a graduate student. When one knows how to do research and how to continue learning, doing it is a feasible and realistic expectation for all concerned. An important conclusion that one can draw from comparing the expansion of knowledge in engineering with the expansion in both the number and quality of engineering graduate programs in the United States and around the world is that there is a close correlation between the two activities, which has served the profession and society well. Graduate Study as Continuing Education The education that is offered to students seeking to earn advanced degrees is usually appropriate for engineers in industry who want to continue their studies. The relationship between engineers in industry and their managers has a major influence on how programs develop. Some companies encourage, even require, their engineers to pursue masters degrees on a part-time basis. Other companies encourage their engineers to take specific courses in order to gain enhanced knowledge, though there is no major incentive for those engineers to earn advanced degrees. In both cases, a major motivation for the engineer is to increase individual educational level, which is also the goal of continuing education programs in general. Universities around the country have worked closely with industry to provide these opportunities. The techniques vary, but all have the common feature of a highly qualified faculty, usually regular but occasionally adjunct, presenting the courses to engineers in appropriate formats. Faculty members have ridden on trains, driven cars, and flown in university aircraft to industrial sites. In some cases engineers have come

249

to the campus, though today this is less common. Large industries located near major universities have built special educational facilities. More recently, television is being used. The process of televised education is discussed later, in the section titled ‘‘Distance Education.’’ Some universities do not distinguish between the degrees earned by on- and off-campus students, whereas others use different names for the degrees the two groups of students earn. Engineers in industry who are students in these programs have important needs that must be considered. One question for the faculty to consider is whether or not to require a thesis. The students might like to write a thesis on a workrelated topic, but this desire often leads to proprietary concerns. A thesis is a contribution to the public domain, which conflicts with industry’s need for protection of its trade secrets. Two solutions to this concern are to have degrees conferred solely on the basis of the completion of course work, or to replace the thesis requirement with a project requirement. The advantage of a project is that it is not in the public domain, yet it can be held to standards comparable with those of the thesis. Typically the project is advanced development vis-a`-vis research. The disadvantage of these solutions is that, should the recipient later decide to pursue a Ph.D., credit for the earlier work may be deflated or even denied. The combination of these pressures and concerns is a major factor in the mixture of types of masters degrees mentioned earlier. An important program that is meeting the needs of students in industry is offered by the National Technological University (NTU). (1). The NTU is a consortium of approximately 50 major universities across the United States that supply their graduate courses to engineers in industry. (The delivery method is discussed in the ‘‘Distance Education’’ section.) NTU serves a number of major functions in this masters degree program: It grants the degrees and it serves as a broker or ‘‘clearinghouse’’ to select courses and make them available. The universities themselves both cooperate and compete in this process. They compete with each other in order to get their courses listed in the catalog and offered on the network, but they also cooperate in order to have comparable standards, to compile logical selections of courses, to advise students, and to see to the other details that are essential to presenting a good degree program. The degree offered by the NTU is a Master of Science (without thesis, though the option of writing a thesis exists). With few exceptions, universities find these educational programs made available to industry to be valuable. They frequently lead to industry-sponsored research contracts. They may lead to consulting and other work relationships between industry and the faculty. They may open opportunities for undergraduates to enroll in cooperative programs. Obtaining information on these programs has become quite easy with the rapid development of the World Wide Web. The Bibliography gives addresses for a few of the more than 50 universities offering such programs (2).

CONTINUING EDUCATION PROGRAMS A strict but not universally accepted use of the phrase continuing education refers to programs designed to give students additional knowledge, but not intended to lead to degrees.

250

CONTINUING EDUCATION

The phrase short courses often describes the activity. Continuing education programs are offered by many institutions. These include universities, industries, vendors, professional societies, and educational institutions founded for the primary purpose of providing continuing education events for engineers and other professional people. The programs may lead to certificates of completion, continuing education units (CEUs) (3), or the knowledge the participant needs to accomplish a needed task. Many licensing agencies require licensed professionals to earn CEUs or equivalent before license renewal. The lines between these types of programs are not distinct. One program that has attracted engineers for 25 years or more is the University of Wisconsin Professional Development Program (4). Many universities offer continuing education events in conjunction with their degree programs, while others concentrate primarily on short courses. The faculty for the short courses may be the regular university faculty, or may be outside experts brought in for that purpose. Often the faculty is a mixture of these two types. The short courses are almost always carefully planned, they present material that is very up-to-date, and they are practically oriented. Students in these courses tend to be, in a good sense of the word, demanding. They have high expectations for rapid, effective learning. Students tend to return to universities that present outstanding short courses. Not all short courses are presented on-site at the university. Many are taken to the industrial sites, and some are taken to convenient, attractive venues in various parts of the country. Industries themselves organize and present short courses for their own employees. The topics may be proprietary in nature, so that the industry does not find it necessary to release data on new products to people outside the company. The topics may be concerned with new information that has not yet made its way to the universities or to other sources. The topics may be company specific, such as studies of company management systems or changes in manufacturing methods. Many companies find these programs to be somewhat more expensive than courses provided by outside agencies, so that they use them only when necessary. Others find them to be cost and educationally effective. Vendors often provide short courses for their new products and services. In some cases, these are offered at nominal cost to the participants, but in other cases, they are quite expensive. These courses train people to use new equipment and software, so that the users have the knowledge needed to enhance their own job performance. Sometimes these programs lead to certifications for new skills for engineers and other technical people who operate as independent consultants. An example might be the programs used by several computer networking companies to train installers and maintenance personnel for their networks. Professional societies, such as the Institute of Electrical and Electronics Engineers (IEEE) (5), frequently include short courses in the programs for their conferences, often at the beginning or the end of the conference (the reason for this is that the short courses tend to be longer than typical conference sessions). Nominal extra fees are usually charged, and the courses are generally presented by outstanding scholars and presenters who are closely associated with the conference itself. These conferences are particularly effective venues of learning for many people. Professional societies, such as

IEEE, also may produce home study courses for their members. As delivery tools these courses use a mixture of videotapes, computer disks, and books for the engineer. CEUs may be awarded for participation in these activities. Professional societies play another important role in continuing education. In the United States, this is exemplified by the American Society for Engineering Education (ASEE), especially its Continuing Professional Development (CPD) Division (6). Other organizations include the International Association for Continuing Engineering Education (IACEE) (7), with headquarters in Finland, and the International Association for Continuing Education and Training (IACET) (3). These organizations bring together educators in industry, government, and universities. These people together study the needs for continuing education, do research in effective techniques, publish results, hold conferences, and provide all concerned with research-based knowledge in effective techniques for continuing education. Two important documents published by the CPD Division of ASEE are listed in the bibliography (8,9). Another type of institution that provides continuing education is a company organized for just this purpose. These companies tend to compete with universities. They serve industry by making all of the arrangements for company courses, whether on company premises or at convenient locations. They present short courses at professional conferences. Because of the high expectations of both industry and students, these companies find that they must provide high-quality presentations of state-of-the-art material. The National Technological University (1) has an extensive short-course program in addition to its degree program. The courses are delivered from a variety of sources around the world, and are available by subscription to participating industries and universities around the country. Some universities have found it desirable to produce highquality videotapes to meet some types of continuing education needs. Some 35 of these universities have banded together to form a marketing organization for such tapes. It is known as the Association for Media-Based Continuing Engineering Education (AMCEE) (10). Closely related to this is the Web site maintained by the International Association for Continuing Engineering Education (IACEE) (7).

DISTANCE EDUCATION The phrase distance education has arisen in the past few years to describe a long-standing process. The phrase describes the situation in which students and instructors are geographically separated, and it has also been used to describe the situation in which students and instructors are separated in time, though proximate in location. This is reasonable, as the characteristics of education when instructor and student are separated by space or by time have much in common. The relation of the concept of distant education to continuing education, whether in the narrow or broad sense, is that the process of distance education is being heavily used to enable continuing education. Distance education is also being used for undergraduate education and for K–12 education, though these uses are not a focus of this article. Early forms of distance education include correspondence courses or involved travel by the instructors, who used the

CONTINUING EDUCATION

best available transportation methods. The archives of land grant universities include pictures of engineering faculty riding trains to teach engineers the latest techniques for road construction or electric power installations. In the 1950s and 1960s, faculty switched to cars and airplanes as they traveled to meet with their students. These techniques are still used, but in the late 1960s, engineering faculty began to experiment with the use of television in a variety of forms, in order to effectively and economically give their students the education they wanted. The experiments were successful, and today several forms of television play a major role in graduate education and in short courses. Today engineering faculty members are experimenting with use of the World Wide Web in continuing education, either in conjunction with television or as a totally new technique. Synchronous and Asynchronous Delivery A useful classification of distance education techniques is to consider whether they are synchronous or asynchronous. Some add an intermediate classification of semisynchronous or partially synchronous. A synchronous method is one in which the students and the faculty are in full, two-way communication throughout the educational event. The students in the distant class may ask questions and, what is most important, all proceed at the same pace, usually set by the faculty member. Ideally there is a lot of student-to-student interaction. In addition to in-class interaction, students and faculty may interact by e-mail, telephone, facsimile, and other techniques, so that there is as much interaction in a distance education students class as in a conventional class. An asynchronous method is one in which students set their own paces, within limits, and student-to-teacher interaction occurs when needed by either party. Students work at different speeds, with the requirement being that all finish by the specified deadline. There is little if any student-to-student interaction, and none is expected except when there is a group of students at the distant site. Student-to-faculty interaction is entirely by e-mail, facsimile, and telephone. Semisynchronous techniques have some of the characteristics of synchronous systems. All students proceed at the same pace, and usually, though not always, there is considerable student-to-student interaction. The distinction is that the students are receiving the learning at a time different from the instructor’s presentation and, if there are several groups of students enrolled in the same class, the various groups receive the learning at different times. The delay is virtually constant throughout the learning period. Students in individual groups interact readily, but not with students in other groups. Student-to-faculty interaction again takes place by email, telephone, and facsimile, but because additional presentations are likely, one student’s question can serve to give instructors ideas to present to all students. Synchronous and semisynchronous techniques usually have the characteristic that the faculty is presenting material to a group of students on the university main campus, while at the same time presenting to distance education students. Asynchronous presentations may or may not have an oncampus student group. Television and Continuing Education Television links students and faculty in a variety of ways. Each has advantages and disadvantages, and thus adherents

251

and critics. The early systems would today be described as semisynchronous. They used videotape in sparsely populated areas, and microwave links in heavily populated areas. Satellite systems appeared at this time. The first videotape uses were reported by Colorado State University and by Iowa State University, beginning in 1969. Microwave links were reported in Texas, Florida, and California at about the same time. The cameras were black-and-white, large, and intrusive in the classrooms. They inhibited on-campus students, especially from asking questions in class, but the off-campus students were eager to get the knowledge, and so accepted the system and quickly helped to guide its improvements. Both videotape and microwave transmission continue to be important today, though improvements in equipment have led to studio classrooms that combine television equipment and computers in ways that really enhance the instructor’s capability to present material. Stone (8) studied the performance of graduate engineering students in televised class at about 10 major universities, and concluded that those students who were seriously pursuing advanced degrees performed as well as and, in many cases, slightly better than the on-campus students. An early change made to microwave broadcasting was to add audio feedback from students to the instructor by providing a telephone callback feature. This makes the system synchronous, provided the students at the receiving site view the broadcast at the same time as the class is being transmitted. Often the students desire the time-shifting that videotape provides, so they record the class and watch the tape at a more convenient time. This also presents great advantages when the engineers in industry must be in transit. In practice, most instructors report little use of the callback feature. One important reason for this is the fact that in these modes there is usually a group of students on campus, and more and more they have become willing to ask questions in class, usually very similar to those the off-campus students would like to ask. Since the questions are recorded, all students benefit from the responses. The National Technological University In 1984, the National Technological University (NTU) (1,11) began using satellite delivery of graduate courses transmitted from more than 20 universities around the country. In due course each university had its own transmitter, so that a larger population could be addressed, and the students in industry around the country had the potential to receive classes as they were being presented on the university campus. By the late 1990s, NTU had nearly 50 member universities, 900 participating company sites, was broadcasting some 200 courses per academic term, and awarding more than 200 masters degrees annually. In most cases, the system is best described as semisynchronous because of the fact that, while substantial student-to-faculty interaction occurs, it does not occur during the broadcast time. If, say, students at ten or more receiving sites were to try to contact the instructor in the class period, then keeping a class going could become quite a challenge. Student surveys indicate that the majority of students record the classes for viewing at a convenient time. In addition to its credit courses and degree programs, the NTU also broadcasts several thousand hours per year of continuing education events.

252

CONTINUING EDUCATION

In the United States, NTU is the first ‘‘virtual university,’’ an institution that is not located in one specific place, but is available to students anywhere provided they have the technological equipment needed to receive the knowledge. (In the United Kingdom, the Open University began in the 1970s.) As this is written (1998), NTU grants degrees in 12 disciplines: Aerospace Engineering, Chemical Engineering, Computer Engineering, Computer Science, Electrical Engineering, Engineering Management, Hazardous Waste Management, Health Physics, International MBA, Material Science and Engineering, Manufacturing Engineering, and Software Engineering. It also has a Special Majors program. Two-Way Audio and Video Systems In many situations, instructors and students alike wish to have real-time, two-way interaction during the class presentation. Such interaction appears to be more important in classes that require a lot of discussion, such as those in which case studies and active or cooperative learning play an important role, and in classes presented to undergraduates. The importance of real-time interaction certainly depends on the instructor’s style, the content, and the characteristics of the students. Major developments have made such programs possible, and the capability is achieved in a variety of ways. In this section, discussion is limited to systems that permit twoway audio and video interactions. In principle, any one-way system can be converted to a two-way system by simply duplicating facilities, thereby providing a communications channel in each direction. This nearly doubles the cost, but works well from a technical standpoint when the instructor is in contact with one or a very small number of student groups. Two-way microwave links, which are typically links between a university and a single industry, work well, and are found in many locations. Two-way satellite links work well when used between one instructor and one or a small number of student groups, but in practice are rarely implemented because of cost and technical complexity. A practical two-way audio/visual system (full duplex system) is provided by video-conferencing. This system uses digitally compressed video signals that may be transmitted over high-grade telephone lines, including the various microwave and satellite links that are included. The systems permit full duplex operation between students in one group and an instructor at a distant site. With suitable switching, more than one student group is possible. The Iowa Communications Network The State of Iowa has built what is probably the most extensive network for distance education in the United States. The state is nearing completion of a set of more than 800 facilities equipped for full duplex television with central switching, so that one transmitting site can send to as many sites as desired. The connection medium is fiber optic cable. The network was originally conceived as being appropriate for enabling smaller, more remote high schools to join with larger schools for advanced instruction in such topics as foreign languages and advanced mathematics. Because of its great capability and flexibility, however, it is being used more widely for distance education. Courses in all disciplines, including engineering (12,13), are being exchanged between universi-

ties, between universities and community colleges, and from universities and community colleges to libraries, hospitals, National Guard armories, and public schools. Ways for connecting this network to systems in adjoining states are being developed. Many states, including North Carolina and Indiana, have networks that connect a variety of institutions. More will undoubtedly be developed in the future. Television and the World Wide Web The idea of transmitting full-motion television over the World Wide Web (WWW) attracts research and development. Already televised images are transmitted, but the quality in general is, so far, inadequate for educational purposes. This will change in the near future, as the bandwidth of the internet increases and compression techniques are further developed. The opportunities that this capability will provide will be virtually unlimited, and distance education involving the entire world is a likely result. World Wide Web and Internet Courses The advent of the WWW has challenged educators to find ways to use it for distance education. Many universities have active development groups working on courses in computer programming, mathematics, biology, climatology, history, electric circuits, and many other topics. Some use the WWW to deliver materials to students regardless of location, while others are programmed so that the students interact with the subject matter on the Web, taking ‘‘mini-tests,’’ asking questions, and moving through the material as needed to learn effectively. Some use audio, while others depend entirely on visual images. Short segments of video are included, and still images are a major part of the attractiveness of the technique. The technique is growing so rapidly that students can or soon will be able to find educational material, either continuing education or credit courses, on almost any subject of interest. Issues and Concerns in Distance Education Distance education courses raise some interesting concerns. Students must be attracted and their qualifications evaluated. One technique for WWW courses is to allow students to take one or two of the beginning parts of a course in order to evaluate both the technique and the specific content. After that, the student must formally register and gain access to the course content and credit for completing the course only under control of a password or similar security provision. The same password is needed for testing, along with a password from a proctor, whose responsibility it is to ensure that the student taking the exam is the one actually registered. Some early experiments of controlling experiments on the WWW have been reported. These are not simulations but actual experimental apparatuses that would always be computer controlled, but are now controlled at a distance (14). When they are being graded and earning credits toward a degree, students want to be able to interact with an instructor in a flexible, economical, and effective manner. Students in continuing education events also have a similar desire. In these situations, students interact with instructors via electronic mail, facsimile, telephone, and over the WWW. These development projects are greatly expanding the capability and responsibility of the faculty in engineering and in all dis-

CONTINUING EDUCATION

ciplines, and the challenges and opportunities are very significant. Another significant issue is the faculty reward system. Developing such a learning experience requires creative scholarship and a lot of hard work, but these alone do not qualify the work for typical faculty rewards such as promotion and tenure. Televised and WWW courses do provide one important feature, in that they project the faculty member beyond the campus, so that the faculty member potentially has impact across the country and around the world. There are also questions of financial rewards for such activity, especially of how to compensate the faculty member over time. The issue of copyright is an important though not fully resolved one. Who owns the copyright to material on the WWW, and how is this protected? Who owns copyright of videotapes that are produced and then copied at a receiving site? These issues have generated a lot of discussion and opinion pieces. Reasonably effective solutions are in place for most of the televised material, as agreements consistent with the laws and also the policies of the universities can be written in advance, but more research and legislation is needed for WWW activities. Another issue is that of cooperative learning (see EDUCATIONAL TRENDS). Engineering faculty members are finding that presenting classes using advanced techniques of a cooperative and guided active learning greatly increase their effectiveness as well as learning on the parts of students. Extending these techniques to distant education is a challenge most easily accomplished in fully synchronous modes. Cooperative learning in semisynchronous and asynchronous modes is a topic just beginning to be studied. Degree Programs and Nontraditional Students The phrase nontraditional student is a loosely used term to describe students who are older than the typical student found on a university campus. Some of the characteristics of nontraditional students are that they are older, perhaps 24 to 40 years old, employed, have family and community responsibilities, and want or need additional education, either for personal growth and satisfaction or for job-related reasons, such as job security and advancement within the company. While for some of these people short courses are sufficient, others want degree programs, and they are willing to do the work and spend the time (12). The demands for providing high quality undergraduate programs present some challenging problems. A typical masters degree requires 30 to 36 semester hours, or 10 to 12 courses of three credits each. At a rate of three courses per year, a student earns a degree in four years. Accelerated programs, sometimes called executive programs, may be completed in two years. However, a typical baccalaureate program requires 120 to 128 semester hours. At 15 to 16 hours per year, an eight-year commitment is needed, and a longer commitment is almost always going to be necessary. Many universities in larger metropolitan areas have met this need for years with evening programs, and they have been quite successful. In lesser populated areas, however, evening programs do not work, and so distant education is meeting this need. As with evening programs, some universities will award graduates exactly the same degree as the oncampus student receives, while others create special degree designations.

253

CAREER DEVELOPMENT AND LIFELONG LEARNING Industry 2000 Conference and Career Development Continuing education is certainly a fact of life in the modern world. Engineers find continuing education necessary both for personal growth and satisfaction, and for reasons that may be described as career enhancement or career development. In 1994, the Institute of Electrical and Electronics Engineers (IEEE) held a very significant conference devoted to this topic. The conference venue was Denver, Colorado, and approximately 100 people from engineering groups in industry, continuing education operations in industry, and universities attended. The conference theme was Technical Vitality Through Continuing Education. A conference Proceedings was published (15). The conclusions of this conference substantiate those of other conferences and studies, but this conference brought together people with knowledge, experience, and vision, and synthesized ideas and actions from a wide variety of sources. Thus it is a rich source of knowledge and ideas on which to base future actions. The conclusions of this conference are presented extensively here. Challenges of Advancing Technology. One important question considered was that of how engineers and their employers should face the challenges of advancing technology, the competencies needed, and how to get the necessary education and training. Those present recognized that both employers and engineers (employees) have a responsibility here. As professionals, engineers have long had a major responsibility for their own careers, and this will increase. Some of the reasons for this increase are the world-wide nature of engineering practice today, and the fact that highly talented engineers are found around the world. Another is the trend in some industries to employ more outside consultants and thus to have fewer engineers on their staffs. Engineers will be evaluated more for their capability than for their knowledge, more for their ability to work in teams and to think, work, and produce in a systems context, and will be expected to work and produce in a process orientation vis-a`-vis a product orientation. Engineers must keep themselves ‘‘lifetime employable.’’ The responsibility for professional development is not limited to the engineer. Employers have a responsibility, especially to people employed by the employer for a substantial time. These people have talent and experience, but they can also cause employers major problems if regular professional development does not occur. Time must be made available by the employer, and the employer must make such an activity an expectation or job requirement. The employer must view continuing education as an investment, not as an expense. The employer must link continuing education for its professional employees to long-term company goals. Hindrances to Technical Vitality. An important concern in continuing education programs is to identify hindrances to maintaining vitality. Industry 2000 participants considered this question, and identified eight factors to address. They apply to engineers and employers alike and, in order of importance, are: 1. Availability of time 2. Lack of motivation

254

3. 4. 5. 6. 7. 8.

CONTINUING EDUCATION

Inability to justify a return on investment Inaccessibility Lack of resources Lack of a plan Quality of materials Organizational culture

The time problem has been discussed earlier in this article. The engineer may deal with pressure from the job at hand and from home and family responsibility. Continuing education usually has a long-term focus and therefore is easily delayed. Closely related is the issue of motivation. In general, both the individual and the company must be motivated for effective continuing education to occur. Engineers and companies that see continuing education as an expense will seek to minimize it, while those seeing it as an investment will seek to maximize their return over a reasonable period. All of these factors are manifestations of the corporate or organizational culture being considered. To be effective, continuing education must be convenient. This is especially true of short courses, which optimally will come to the work site and, in the future, must come to the desktop. The material must be high quality, sharply focused, and tailored to the audience. The course must be affordable and must fit in with the plans of the engineer and the employer. Probably the important factor is that the leaders of the organization must set examples for their employees by engaging in continuing education themselves. Essential Principles for Success. The conference defined success as having a technically vital organization. The first principle was important work—work that makes a difference and generates good reputations among peers. A second, which repeats a point from a previous section, is the personal example set by the management. The third is a customer focus— maintaining contact with customers and being sure that the products and processes being developed are what the customers want and need. The fourth is establishing and clarifying a link between business success and technical vitality. Ideal Continuing Education Scenarios. The conference attendees took a general, long-range view of the question of what constitutes ideal continuing education or career-long professional development. The scenario is more a question of what constitutes an effective program than of immediate delivery and content questions, though those were considered also. There is a need for interdisciplinary efforts, as well as for sharply focused events. Interdisciplinary ideas include a range of engineering topics, as well as business topics such as marketing and globalization. Regardless of topic, programs must be short, and well designed by recognized experts, both in content and in delivery. The programs must meet perceived needs and be relevant. They must be tailored to the abilities of the learner, and be cognizant of the learner’s previous background. Education for needed knowledge must be linked to project planning. If the provider is a university or other outside agent, then the planning must involve both, with due regard to issues of ownership of intellectual property, short life cycles of materials, and the rigidity of university calendars. Finally, the programs must be cost effective, even when viewed as investments.

Both employer and employee must see the programs as having a favorable cost–benefit ratio. Professional Society Recommendations. The Industry 2000 conference was organized by the IEEE. It addressed the question of what the IEEE should be doing for its members with respect to continuing education. The conclusions came in the form of four principal recommendations, and IEEE is implementing these recommendations for its members (5) (presumably similar conclusions would apply to other professional societies). They are: 1. The IEEE should develop a set of career planning tools for use by each member. 2. The IEEE should develop a centralized database of continuing education materials, which informs all members about the opportunities available to them. 3. The IEEE should establish and support a forum for ‘‘Best Practices’’ in continuing education. 4. The IEEE should explore appropriate measures for recognizing participation by engineers in lifelong learning. For each engineer, a career plan is essential. The plan should help the engineer define success and the values and behaviors that lead to success. It should then lead to a set of goals and the projects to accomplish the goals. It should raise awareness about emerging technologies, in the broadest sense of the term. A lot of continuing education material is available from a wide variety of sources, but it is not always easy to find out what is available or, what is more important, where the material might be that meets a particular need. There is a need for some evaluation of this material. Both engineers and employers need to know how to perform continuing education effectively. There is need for sharing ideas, though there is no more incentive for a company to share its practices than its knowledge. There is a role for professional societies in this regard. Finally, there is very little way to formally recognize continuing education, as discussed earlier in this article. A meaningful measure established by one society would slowly but surely win acceptance across the profession. Conference Conclusions. The conference conclusions are of value. Some are ideas that have been around for a long time. Others are quite new. They are: 1. Engineers have a personal responsibility to maintain their own technical vitality, but they must be supported by their employers in this endeavor. 2. It is especially important to focus on continuing education during business cycle expansions. While time is short then, the pressures during recessionary cycles are more severe. Also, such training can enhance competitiveness during expansion periods and may mitigate future down cycles. 3. There is substantial correlation between company size and continuing education effectiveness, with the larger companies being more effective. 4. Senior management must be committed to and engaged in continuing education.

CONTINUING EDUCATION

5. Tools for assessing the quality of continuing education and plans for continuing education are needed. 6. Colleges of engineering need to do a better job of convincing undergraduates of their forthcoming need for continuing education. Interestingly, this recognition across the profession has led to an accreditation criterion that will be discussed in the next section, Lifelong Learning. 7. Evaluation criteria for nonuniversity courses are needed. 8. The IEEE should elevate the visibility and effectiveness of its continuing education activities. Lifelong Learning In the United States, engineering programs are accredited by the Engineering Accreditation Commission of the Accreditation Board for Engineering and Technology. New criteria based on outcomes assessment techniques have been developed and are being implemented, with full implementation scheduled for the year 2001 (16). One criterion states that graduates of an accredited engineering program must demonstrate that their graduates have a recognition of the need for and the ability to engage in lifelong learning (Criterion 4.i) (16). Lifelong learning is the result of effective continuing education, both the interest in doing it and actually accomplishing it. Faculty members in colleges of engineering across the country are trying to design learning experiences for undergraduates that will develop lifelong learning skills and attitudes in students. Lifelong learning is also an expectation for faculty members. The accreditation criteria also have a requirement for professional development of the faculty. Continuing Education Resources The engineer who is seeking continuing education will find a wide variety of opportunities. These resources may be found by contacting education and training offices of employers, who usually have a lot of information available. Often these people know who provides good opportunities and can assist the engineer in finding programs to meet needs. Most engineers belong to at least one of the more than 50 engineering professional societies in the United States alone, such as IEEE and the American Society for Mechanical Engineers. Nearly all of these societies present programs related to their specialty. A telephone call, e-mail message, or World Wide Web (WWW) search will provide the needed information. Many of the approximately 400 colleges of engineering in the United States provide continuing education events, both degree programs and short courses. The distance education technologies described make distance a minor factor in choice, so that the student can choose based on subject matter and reputation of the provider. Again, finding the information is relatively easy. The bibliography lists 10 such university sites (2), and search engines lead to many more. Data show, however, that the programs are quite variable in many respects, and finding good evaluations of the programs is difficult. The person who wants continuing education needs to put effort into evaluating the opportunities. The student should carefully consider the reputation of the offering institution and of the faculty who present the material. The material must be up to date. This is true for all events, but

255

especially for the short courses. The student should consider price, availability of written materials and computer software, access to the instructor, location of the events (when travel is required), and teaching styles of the faculty. Many courses are presented on videotapes and the student is advised to obtain a sample tape for, usually, a two-week evaluation at minimum cost. Are CEUs available, and needed? Many licensing agencies require CEUs. Time and effort spent in evaluating the opportunities will pay big dividends to the student. CONCLUSION Lifelong learning is an essential activity for engineers today and continuing education methods enable lifelong learning. Continuing education includes programs that are based in academic credit and that usually lead to university degrees as well as noncredit or short courses that are targeted toward particular knowledge that usually has an immediate application on the part of the recipient. Continuing education for engineers is an important part of the practice of engineering in the United States and around the world today. BIBLIOGRAPHY 1. Current online information on the National Technological University is available on their World Wide Web Site: http:// www.ntu.edu 2. Introductory or home page World Wide Web addresses for a few of the more than 50 colleges of engineering that offer graduate degrees in engineering in the United States are: a. The Johns Hopkins University: http://www/jhu.edu/앑wse1 b. Stanford University: http://www.stanford.edu/home/ academics/continuing.html c. University of Wisconsin: http://www.engr.wisc.edu/general/ depts.html d. Iowa State University: http://www.eng.iastate.edu/ede/homepage.html e. Purdue University: http://www.ecn.purdue.edu/cee f. University of Illinois at Urbana–Champaign: http:// www.conted.ceps.uiuc.edu g. George Washington University: http://www/gwu.edu/앑ceep h. University of Southern California: http://www.usc.edu i. University of California at Los Angeles: http://www.ucla.edu j. University of Minnesota: http://www.technology.umn.edu 3. The CEU is defined by the International Association for Continuing Education and Training (IACET). The definition of the unit and the processes for granting and receiving CEUs is available online on the World Wide Web at http://www.iacet.org 4. A complete description of this program may be found online at the University of Wisconsin World Wide Web site: http:// www.engr.wisc.edu/general/depts.html 5. The IEEE World Wide Web site is found at: http://www.ieee.org 6. The Continuing Professional Development World Wide Web site is found at: http://www.asee.org 7. The IACEE World Wide Web sites are found at: http:// www.dipoli.hut.fi/org/IACEE and http://www.dipoli.hut.fi 8. H. R. Stone, Video Delivery of Graduate Engineering Instruction, Washington, DC: Continuing Professional Development Division, American Society for Engineering Education, 1990. 9. J. S. Greenberg and M. S. Bonhomme, Compendium on Uses of Distance Education Technologies, Washington, DC: Continuing

256

CONTRACTS Professional Development Division, American Society for Engineering Education, 1996.

10. The AMCEE World Wide Web site is found at: http://www. amcee.org 11. Annual Report, Fort Collins, CO: National Technological University, 1996–1997. 12. R. Horton, L. Van Brocklin, and E. C. Jones, Jr., Challenges of delivering undergraduate engineering programs to place-bound members of the Technical Work Force, Proc., Frontiers Educ. Conf., Pittsburgh, 27: 588–592, 1997. 13. D. Jacobson et al., A course exchange using television, Proc., Frontiers Educ. Conf., San Jose, 25: 586–589, 1995. 14. C. Knight and S. DeWeerth, A distance learning laboratory for engineering education, Poster session, 27th Annu. Frontiers Educ. Conf., Pittsburgh, PA: 1997. 15. Proceedings, Industry 2000: Technical Vitality Through Continuing Educ., Piscataway, NJ: IEEE Educational Activities Board, 1995. 16. The criteria are known as ‘‘Criteria 2000’’ or ‘‘EC2000’’ and are available on the World Wide Web at: http://www.abet.org

EDWIN C. JONES, JR. Iowa State University

CONTINUOUS MEDIA. See DISTRIBUTED MULTIMEDIA SYSTEMS.

CONTINUOUS PHASE MODULATION. See MINIMUM SHIFT KEYING.

CONTINUOUS PROCESSES. See PAPER INDUSTRY. CONTINUOUS-TIME CONTROL SYSTEM DESIGN. See CONTROL SYSTEM DESIGN, CONTINUOUS-TIME. CONTRACTING. See OUTSOURCING.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2905.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Educational Trends Standard Article Edwin C. Jones Jr.1 1Iowa State University, Ames, IA, Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2905 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (122K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2905.htm (1 of 2)17.06.2008 21:42:19

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2905.htm

Abstract The sections in this article are Studies of Engineering Education The National Science Foundation Coalitions Outcomes Assessment in Engineering Education and Accreditation Changes Factors Affecting Engineering Education Issues in Engineering Education Active Learning and Cooperative Learning Integrated Curricula Learning Styles Enrollment, Degrees, and Retention Faculty Development Promotion and Tenure | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2905.htm (2 of 2)17.06.2008 21:42:19

EDUCATIONAL TRENDS

197

EDUCATIONAL TRENDS This article on trends in engineering education considers some of the important influences on engineering education during the last fifty years. It examines the studies of engineering education commissioned over the years, starting with the Grinter report of the 1950s. It looks at changes in educational philosophy, industry–university interaction, student issues, such as enrollment and diversity, and concludes with a short discussion of a very significant issue of the times, the role of academic freedom, tenure, and posttenure reviews. STUDIES OF ENGINEERING EDUCATION At the end of World War II, the US Congress made a decision that was to have far-reaching effects on engineering education, indeed on all of higher education. This decision was popularly called ‘‘The G.I. Bill.’’ It gave all returning veterans the opportunity to receive four years of university education. The veterans took advantage of the program, and soon engineering education classrooms were crowded. The students were mature, motivated, and set very high standards for themselves and the educational institutions. The effects of the G.I. Bill are probably exceeded only by those of the Morrill Act which established the land grant system of colleges and universities in the USA, when educational legislation is considered. By 1950, the veterans were completing their degrees, and, as enrollments dropped and stabilized, engineering educators began to address the concerns that were so apparent during World War II. Too many engineers were unable to extend their knowledge to do the basic work needed to develop new technologies needed for war. Instead, the fundamental work was done by scientists. However, the engineers were able to take the basic work done by others and turn it into devices and systems needed for the war efforts. The Grinter Report Engineering education has always been introspective and willing to evaluate itself. The result in this case became known as The Grinter report (1). This study set the stage for nearly four decades of engineering education, both graduate and undergraduate. It led to programs with a strong engineering science content, a well-defined base in mathematics and basic science, and a clear emphasis on the social sciences and humanities. It led to programs that attracted engineering students to North America from around the world. It led to a period of rapid economic development on nearly every continent. As engineering educators implemented the Grinter Report’s recommendations, the Space Age began in 1957. The J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

198

EDUCATIONAL TRENDS

first satellites and orbital vehicles, some with human occupants, were launched. The first Moon landing occurred on July 20, 1969. For engineering education, the decade was marked by rapid implementation of the Grinter provisions and expansion of engineering research by universities across the United States. This combination led to a rapid expansion in technological knowledge and development and a wide variety of new products and services. The QEEP Report Concerns with the programs began to emerge in the 1980s. In the early 1990s, some of the tenets of the Grinter report and its implementations were being questioned, and new studies were undertaken. The 1986 study, known as the ‘‘Quality of Engineering Education’’ (QEEP) study (2), makes recommendations in four major areas. In faculty development, the study recommends more industrial experience for faculty and recommends that faulty bring such experience to the classroom. It also recommends increasing the relevance of an engineering education to the demands of the modern world and calls on the Accreditation Board for Engineering and Technology (ABET) to strengthen its faculty criteria with these factors in mind. A second part of the study calls for making faculty development a structured process, not the ad hoc process it has always been. Universities, ABET, the American Society for Engineering Education (ASEE), Government, the National Academy of Engineering, the Foundations, Professional Societies, and Employer Organizations were charged with contributing to this process. Many of the agencies took the recommendations seriously and began to develop responsive programs. The third part of the study deals with educational technology. It considers the role of the computer and of television and makes some very important and far reaching recommendations. Many of these have been implemented. It did not, however, foresee the rise of the Internet and the World Wide Web and their effects on education and the university in general. The fourth section deals with the undergraduate laboratory. It points out two major problems, inadequate funding for equipment, and the heavy reliance on teaching assistants for laboratory teaching. These problems persist today. It foresaw the role of laboratory instruction in developing communications and teaming abilities, which have really become important in the past few years.

THE NATIONAL SCIENCE FOUNDATION COALITIONS Engineering Coalitions In 1989, the National Science Foundation called for and received proposals for programs to implement the recommendations, especially of the QEEP project. The call was for a group of universities to work together to effect the types of change called for in the report. The universities were to be a group diverse in size, support base, and clientele. The historically black colleges and universities (HBCU) were to be included. The institutions could be geographically proximate or nationally dispersed. The projects were to last five years.

By 1996, eight coalitions involving 60 colleges and universities were in place (8). Each had its goals or themes. The first two were formed in 1990. The Synthesis Coalition (California State Polytechnic University, San Luis Obispo; Cornell University; Hampton University; Iowa State University; Southern University; Stanford University; Tuskegee University; and the University of California at Berkeley) engaged in pioneering work in the delivery of educational materials and in the synthesis of knowledge for problem solving. The ECSEL Coalition (City College of New York, Howard University, Massachusetts Institute of Technology, Morgan State University, Pennsylvania State University, University of Maryland, and the University of Washington) emphasized design across the curriculum. Two more coalitions were formed in 1992. The SUCCEED Coalition (Clemson University, Florida A&M University and Florida State University; Georgia Institute of Technology; North Carolina A&T State University; North Carolina State University; University of Florida; University of North Carolina, Charlotte; and Virginia Polytechnic Institute and State University) took on the responsibility of developing ‘‘Curriculum 21,’’ which is intended to bring together the engineering and engineering education processes. The Gateway Coalition (Case-Western Reserve University, Columbia University, The Cooper Union, Drexel University, Florida International University, New Jersey Institute of Technology, The Ohio State University, University of Pennsylvania, Polytechnic University, and the University of South Carolina) is charged with doing research and development in integrative curricula. The fifth and sixth coalitions were approved in 1993. The Greenfield Coalition includes Central State University, Lawrence Technological University, Lehigh University, the University of Detroit, the University of Michigan, Wayne State University, and a virtual university operated at HOPE’s Center for Advanced Technology in Detroit, called FOCUS. It also includes industrial partners and is intended to study new methods and to develop new programs for manufacturing education. The Foundation Coalition is composed of Arizona State University, Maricopa Community College District; Rose–Hulman Institute of Technology; Texas A&M University, College Station; Texas A&M University, Kingsville; Texas Woman’s University; and the University of Alabama. Its vision is to create an enduring foundation for continuing development and lifelong learning by students. Its goal is to integrate subject matter within the curriculum, incorporate cooperative and active learning, and use technology-aided education and continuous assessment of outcomes. The final two were formed in 1994. One is known as the Academy (or the Engineering Academy of Southern New England), and it includes Central Connecticut State University; Connecticut Community Technical College System; Hartford Graduate Center; the University of Connecticut; the University of Massachusetts, Amherst; the University of Massachusetts, Lowell; and the University of Rhode Island. Its primary goal is to elevate the position and perception of manufacturing in both universities and industries. It focuses primarily on postbaccalaureate education. The final coalition, SCCEME (The Southern California Coalition for Education in Manufacturing Engineering) includes California State University, Fullerton; California State University, Long Beach; California State University, Los Angeles; the University of California at Los Angeles; and the University of Southern California. It

EDUCATIONAL TRENDS

also involves manufacturing and features development of interuniversity programs of undergraduate manufacturing engineering education. Coalition Impact The National Science Foundation has taken several steps to ensure that the knowledge developed by the coalitions is widely disseminated to and evaluated by the engineering educational community. It has organized plenary sessions and poster presentations at the annual ‘‘Frontiers in Education’’ conference (8) and at the annual conference of the American Society for Engineering Education (9). It has sponsored other conferences and involved coalition participants (3). These events have always been well attended, and the ideas presented have been widely discussed and considered by those in attendance and their colleagues with whom they shared the ideas. Few if any of the engineering programs in North America, especially the United States remain unaffected by the coalitions and the educational research and development done by coalition members. OUTCOMES ASSESSMENT IN ENGINEERING EDUCATION AND ACCREDITATION CHANGES Three Studies in 1994 Three additional studies were released in 1994 (3–6). Additional studies were made by other groups and are cited in the bibliographies of the references cited. The three 1994 studies have much in common, though they use different clientele and different methods. They were performed by the National Science Foundation (NSF), the Deans Council of the American Society for Engineering Education, and the National Research Council (NRC). The NSF study encouraged development of more diversity in engineering programs, defining this word broadly to include diversity of people and of experiences. It pointed out the centrality of students to the venture and encouraged faculty to be more active in designing the totality of educational experiences. It urged moving from predominately lecture classes to active learning. It encouraged engineering educators to develop broad, flexible curricula. The Deans Council Study pointed out that engineering programs must be relevant, attractive, and connected. The programs must be relevant to the lives and careers of students and prepare them to contribute around the world over a lifelong, changing career. The programs must attract the best students from all groups of society. The programs must be connected to the needs, issues, and concerns of the broader community. There must be substantive partnerships between colleges of engineering and other educational institutions and with government and industry. The NRC study predicted that future engineering programs will be designed to meet the demands of present and future engineering workplace challenges and life in an increasingly complex society. Such programs will include all necessary fundamentals but exclude redundant material, integrate design with fundamentals, be practice-oriented, emphasize both teamwork and individual effort, build a sense of social and business context, prepare graduates for entry into professions, such as law and medicine, instill a desire for lifelong learning, and prepare students for graduate study.

199

The 1996 Study The 1996 study deals primarily with assessment (7). In this sense it follows the previous studies and takes steps toward providing the engineering education community with tools and methods for self-evaluation. It brought together five organizations with mutual interests in assessment activity. These include the American Society for Engineering Education (ASEE), the Accreditation Board for Engineering and Technology (ABET), the National Council of Examiners for Engineering and Surveying (NCEES), the ASEE Engineering Deans Council (EDC), and the National Society of Professional Engineers (NSPE) (8). The report recommends that the following qualities be considered in designing an assessment program for any degreegranting engineering educational unit: 1. 2. 3. 4. 5.

institution-specific mission and goals institutionwide, longitudinal assessment programs professional (ABET) accreditation broader career goals of students and graduates cost factors.

The report proposes assessment ideals, including the improvement of student learning and development, a focus on undergraduate (and, separately, graduate) education, educational breadth, relevance to practice and citizenship, validated measures of desired outcomes, comparisons with other programs, accommodation to future needs, and cost effectiveness. The report goes on to point out that no one assessment tool is likely to suffice. Rather, an array of devices is needed. The report suggests that three independent measures of most qualities may be needed. Measurement tools may include student data, such as transcripts, portfolios, and videotapes of presentations. Other tools include performance of graduates on national examinations, though these are seen as having limited value with the present examinations. The most powerful tools are surveys designed for this purpose as part of self-analysis. The same conclusion is reached for assessing the performance of students after graduation. ABET Criteria 2000 As the various studies were being conducted and preliminary analyses made of the data, it became apparent to the leaders of the Accreditation Board for Engineering and Technology (ABET) that the criteria for professional accreditation of engineering programs in the USA need major changes. Because of the international ties of ABET, the effect would extend beyond the United States. E. W. Ernst discusses accreditation substantially in a companion article. The discussion here is confined to points deemed essential to putting the process into the context of this article. The concerns with the existing criteria, which substantially followed from the Grinter Report, were that they were too prescriptive, the process was too expensive, the criteria were becoming obsolete, and the process itself needed major changes. Although the process was changing in ways to reduce the intensity of these criticisms, more rapid change was needed. Universities and industry both supported change. The result was a draft of a totally new set of criteria, which

200

EDUCATIONAL TRENDS

have become known as ‘‘Criteria 2000’’ or ‘‘EC 2000.’’ ABET is testing the criteria experimentally in 1996 and in 1997 and will use the new criteria optimally in 1998, 1999, and 2000, with full implementation in 2001. Criteria 2000 are written primarily in terms of outcomes assessment. The criteria place the responsibility on an institution for identifying its mission, goals, and objectives, and for showing that its program leads to baccalaureate engineers who exhibit the desired characteristics. The curricular specifications are minimal, though they do not allow an institution to develop a program with an engineering title that is not engineering. Criteria 2000 are a major reason for the impetus for studying outcomes assessment techniques in the late 1990s. Reference (9) describes one successful technique for outcomes assessment in an undergraduate program. Reference (10) describes an effective use of alumni in outcomes assessment, and (11) describes very carefully a model for institutional planning for outcomes assessment. An Assessment Plan The following eight-step approach to developing an effective assessment plan is given by Rogers and Sando (12): 1. Identify goals. (What is to be achieved?) 2. Identify specific objectives for each broad goal, and state the circumstances under which you will know whether or not the goal has been achieved. 3. Develop performance criteria for each objective. (What will students be able to do, to be, or to possess when the goal is attained?) 4. Determine the practices to achieve the goals. (What will be done to achieve the goals and objectives? How might practices be modified in response to feedback?) 5. Select assessment methods for each objective, and choose data collection methods. 6. Conduct assessments. Use specified methods to collect the evidence, and analyze the evidence in comparison with performance criteria. 7. Determine feedback channels that provide timely information to enable continuous improvement, decision making, and evaluation. 8. Evaluate whether or not performance criteria were met and objectives achieved. Typically, this last step occurs during the continuous improvement process (formative evaluation) and at the end of the project (summative evaluation). Assessment Action Recommendations The assessment study concludes with the following five recommendations to all engineering programs, though it recognizes that there are still many unanswered questions with regard to assessment. It appears that it will be necessary to apply the continuous improvement process to the assessment activity itself. 1. Each engineering program should develop an appropriate assessment program using the ABET Engineering Criteria 2000 in conjunction with criteria specific to individual institutional and program goals.

2. NSPE and NCEES should actively encourage and participate in the continuing discussion of the relationships of engineering education to licensure and practice. 3. ASEE and ABET, in cooperation with the Deans Council, should coordinate the efforts of selected institutions and the major employers of the graduates of those institutions to identify and report the possible relationships between performance as a student and subsequent professional performance. 4. ASEE should seek resources to act on the resolution of the Deans Council which ‘‘calls for a continuing forum for the development and analysis of assessment methods and measures of learning appropriate to the stakeholders in engineering education.’’ Such a forum could help the engineering community implement program assessment by disseminating best practices, identifying measures associated with educational outcomes, and sharing experiences of specific institutions and programs. 5. ASEE should establish a clearninghouse for a nationally shared database on engineering educational program assessment measures. Data should be collected, aggregated, and reported at the program (or discipline) level.

FACTORS AFFECTING ENGINEERING EDUCATION Institutional Changes The preceding paragraphs point out the vast changes that have occurred in engineering education in the last decade. The changes largely result from the self-analysis in which engineering education has always engaged. If the goal of outcomes assessment is continuous improvement, then engineering education has been following the tenets of continuous improvement for nearly a century, and the last decade is no exception. Although the changes result from self-analysis, they are not independent of institutional, technological, and international forces at work throughout society. In the United States and around the world, universities have become large, multifaceted, and quite visible. This is true of both public and private universities. The clientele of the universities, including government, taxpayers, donors, students, and industries, are demanding increasing accountability by the universities for their expenditures. Mandatory retirement has been eliminated, though no evidence exists that very many people are working longer than they should. In fact, many valuable faculty are choosing early retirement. The institution of tenure itself is under scrutiny, especially in professional schools, such as engineering. One result of these pressures is the drive to reduce the number of credit hours required to earn an engineering degree, or for that matter, degrees in other disciplines. Part of the pressure comes from parents who wish to reduce the cost of an education. Some comes from taxpayers and governing bodies. This seems counterproductive in the face of rapid technological development, but it does require institutions to focus on their missions and to articulate carefully their goals and objectives. This drive is consistent with outcomes assessment. It forces faculty to be sure that topics studied are really needed, and that they will serve the student well. It probably

EDUCATIONAL TRENDS

is also true that the amount of study required to earn a unit credit has slowly risen over the years. Technological Changes Technological changes continue at a rapid pace. Much of the change has been caused by the graduates of the engineering programs in the United States and elsewhere, so it is difficult to say that the programs of recent years are ineffective. The technological changes, in fact, have led to major societal and engineering education changes. One, alluded to in the previous paragraph, is the fact that it simply is not possible to discuss in the engineering classroom all of the technological developments nor to predict accurately what is likely to come in the near future. It always makes a class interesting to discuss current developments, but it may not be the most effective educational method. It is also true that students are often more comfortable with some technologies than the faculty, which leads to a certain tension in the classroom. Students also have grown up in a period heavily emphasizing technological systems, such as television, computers, instantaneous communications, and rapid transportation. No doubt this affects their learning styles substantially. International Considerations One of the most apparent changes in engineering education in the last decade has been the rapidly increasing importance of international factors. Engineering itself has become international, as students travel to a variety of countries to study and engineers practice around the world. The companies that employ engineers work around the world, and they recruit engineers from many nations. Products and engineering services are designed and developed for a world market. Products designs are completed in one country and transmitted electronically halfway around the world for manufacture, and the completed products are then marketed worldwide. Engineering with Information Traditionally, engineers have worked primarily with energy and materials. The knowledge that they accumulated was stored in handbooks, and current information was available in manufacturer’s literature. Knowledge and information have always been important to the engineer, but the study and the practice of engineering have focused on materials and energy. The digital computer has changed this characteristic of engineering. Engineers now practice as much with information as with energy and materials. Although most evident in electrical and computer engineering, it is true in all engineering fields. Much attention must be given to the design of information systems so that complete and accurate information is readily available early in the design phase of a product or service, and this is fully as important as proper choice of materials and use of energy. In fact it is essential to the proper choices. ISSUES IN ENGINEERING EDUCATION Cooperative Educational Programs The pressure on universities to reduce the number of credits required for an engineering degree has been mentioned. Some

201

of this pressure has come from industry. Concurrently, industries have expanded their cooperative programs (coop). In its basic form, a coop program (not to be confused with cooperative learning) includes a traditional four-year engineering program interleaved with structured industrial experiences. The length of industrial experience is normally a year or more. A typical pattern might be a summer assignment following the sophomore year, a semester and a summer assignment after the first semester of the junior year, and a summer assignment in the senior year. Many other patterns are possible. Included in this expansion are international coop experiences, and these are proving to be a valuable part of the engineering education enterprise. Much of this change has been apparent starting in 1996. The anecdotal data available at present suggest that institutions which traditionally have had about 10% of their students involved in coop programs suddenly have 60–75% participation, and employers who would like to have even more coop students.

Industrial Partnerships Besides coop programs, universities and industry are forging other partnerships. Industries are encouraging their senior people to serve on Industrial Advisory Committees of Colleges of Engineering, and these people are making major programmatic contributions. Their ideas are being carefully considered and often implemented. Their ideas relate to course and curricular content, space and facility use, finances, research development, and helping to develop public support (13–15). Research projects in the university supported by industry are becoming particularly important and often include undergraduates. One reason is that the resources supplement the public monies available to universities, funds that have become scarcer in recent years. The projects undertaken are challenging and, in most cases, involve leading-edge technology. Major needs are being considered. One difficulty with many of the projects is the need for industry to keep its proprietary information confidential, which contrasts with the need of the university for openness and publication. As faculty advancement depends in large measure on publication, this is a major problem. It is being resolved in a case-by-case, university-by-university method. Another reason for the importance of industrially sponsored research is the fact that it brings the faculty into close contact with industry. This contact enhances their engineering background and enables them to be more effective in the classroom and learning laboratory as they educate the next generation of engineers. Many of the research projects are closely tied to the coop programs mentioned earlier and to the graduate coop programs emerging on some campuses. Many of the industry–university partnerships are characterized as joint ventures. The two agencies agree to cooperate to develop a marketable product or service, with appropriate provisions for sharing the risk and gain. Many of the ventures involve students. Activities such as these provide an invaluable opportunity for students to experience the thrills and frustrations of engineering while still students, and usually they indicate that their motivation to continue study toward baccalaureate and advanced degrees is strengthened.

202

EDUCATIONAL TRENDS

ACTIVE LEARNING AND COOPERATIVE LEARNING William L. Everitt, former Dean of Engineering at the University of Illinois, said on many occasions that ‘‘engineering is not a learned profession, it is a learning profession.’’ It is true that the successful engineer today must continue to learn throughout a career. Often this is called ‘‘lifelong learning.’’ Being a lifelong learner requires that the engineer wants to continue to learn and to have the skills to accomplish the task. Helping students develop this skill is, in part, a responsibility of the engineering faculty. Active Learning The lecture/recitation/laboratory method has been used in traditional classes in many disciplines, including engineering. In this classroom, the faculty member presents material and guides discussion. Further discussion takes place in the recitation section. Laboratory work serves the dual purpose of teaching students to become experimentalists and to reinforce the lecture material. As the class grows larger, students are engaged but often only to a limited extent. Much of their learning takes place individually outside the classroom. Formerly, the laboratory was probably the principal learning arena for the engineering student, but, as programs have reduced credits and as laboratories have become more expensive, the time and effort devoted to laboratory instruction has decreased. Active learning, sometimes called interactive learning or interactive instruction, is a process of developing a framework in which students interact with the material in the classroom. The faculty functions more as a manager and resource than a presenter of information. Students are encouraged to seek information on their own or in groups, with guidance from the faculty. In some institutions, faculty are learning these techniques from colleagues in the Colleges of Education, where substantial research into learning methods has taken place (16). The most important characteristic of active learning is that the students are and must be actively involved or engaged with the material while in the classroom, understanding as much as possible why they are doing what they are doing, and seeking help from other sources, including the faculty, when they lack a specific item of knowledge (17). Cooperative Learning The most effective way to achieve the goal of active learning is to use the technique of cooperative learning. Sometimes called ‘‘teaming’’ or ‘‘collaborative learning,’’ cooperative learning also meets a need of contemporary employers for graduate engineers who are skilled in working with other people. The basic idea of cooperative learning is quite simple, but its implementation requires a lot of skill and practice (18). Cooperative learning requires grouping students in a class into groups of 2 to 4 people. These students sit together and usually work together inside and outside the classroom. The students work cooperatively on the assignments given and shared roles of leadership, recording of results, and other necessary tasks. In many cases, the group turns in written assignments, and each member receives the same grade. In some cases, students comment on colleagues, and this information modifies grades to some extent. To encourage teamwork, many instructors give bonus points on examinations to

all members of a group when all earn grades above some threshold value. Because grading ‘‘on the curve’’ tends to encourage competition among students, it is necessary to grade on ‘‘absolute standards.’’ A collection of students is not necessarily a team. Faculty find that they must put some effort into helping the students function as a team. Reference (18) suggests the following five basic elements of cooperative learning: 1. Positive interdependence. The students must be convinced that they need each other and that it is in their best interests to work together. With beginners, students need to be assigned roles, and in some cases, faculty make information available only to one member in a group. 2. Face-to-face promotive interaction. Students are encouraged to help each other by sharing and encouraging each other. When possible, they pass on their knowledge to classmates outside the group. Students talk through solutions to problems. It is self-defeating for each person to work alone and to come to a meeting to present individual solutions. 3. Individual accountability. Though the students work together, each is responsible for learning. Individuals may be tested regularly or called on to recite when the class size permits. 4. Interpersonal and small group skills. All members of a group need to have a basic set of skills, including time management, communicating ability, willingness and ability to resolve conflicts, and decision making. There must be mutual trust and respect among all members. If these are not present, the group will not function and must be reorganized. 5. Group processing. Members need to put some effort into discussing how well they are achieving their goals, and the instructor must be involved in this activity. Feedback must be given. Students have always worked together, but optionally. In most cases, these new techniques work well for most of the students. Students, however, are busy, and finding common meeting times and places is a chore for some. This turns out to be especially difficult when several instructors are using the techniques and a student is a member of several groups. There are always a few students who resist the idea, and they require individual consideration. INTEGRATED CURRICULA A traditional engineering program has been composed of 32 to 40 courses (semester system) or 48 to 60 courses (quarter system). With few exceptions, engineering curricula are characterized by a rigid prerequisite structure. Instructors assume a consistent background for all of the students in a class and design a course to build on that background. This structure achieves a high level of integration, but to do this, it requires that students put a lot of effort into the integrating effort, though they are often unsure how the pieces fit together. Because, increasingly, students demand a clear indication why study of particular topics is important, much research has gone into developing programs and curricula

EDUCATIONAL TRENDS

designed to improve the cohesiveness of a program and the learning by the students. One term that describes this effort is the ‘‘integrated curriculum’’ (19–22). The basic idea of an integrated curriculum is simply described. In a traditional program, for example, physics and calculus are studied independently. The mathematics faculty presents ideas, such as the derivative and the integral of a function. Following that, the physics faculty defines concepts, such as velocity or current, in terms of the derivative, and concepts, such as voltage or work, in terms of the integral. When syllabi are such that the topics are covered closely together, and the physics follows the mathematics, then the process works well. But if the physics precedes the mathematics, or the topics are well separated in time, then full integration becomes much more difficult. The same problem continues in basic mechanics or electrical courses, which then depend on the physics. In an integrated curriculum, the physics and mathematics faculty work together, presenting physical concepts and the mathematical tools needed to understand them. Often the classes are team taught, but, in other cases, the topics are very closely coordinated. Simultaneously, the engineering faculty is presenting the core ideas of the various branches of engineering, working closely with the physics, mathematics, and chemistry faculties, and others if necessary. Sometimes the phrases ‘‘just in time’’ or ‘‘need to know’’ are used to describe the process. Another term is ‘‘holistic thinking.’’ The idea can be extended. In senior design, students often need economic ideas and are faced with ethical decisions. They need to study reliability. This may be the time to introduce basic ideas of engineering economy, professional ethics, and statistics. Again, in the early years, engineering students are taught some of the basics of design, and, more important, given the opportunity to express their creativity, although they may not yet have all of the tools needed for industrial level design. By being given such problems, the studies indicate that they are encouraged to go beyond their basic classroom assignments and study material appropriate to their design problems. The integrated techniques are probably most effective in the early part of an engineering curriculum but find application throughout the program. It is important to note that the ideas are not limited to the technical parts of a curriculum but apply to all components of an engineering course of study.

203

their individual learning styles, students in any of the groups can become outstanding engineers. Kolb Theory The Kolb theory (23–25) considers the ways in which people perceive and process information. Some people perceive information primarily by concrete experiences (CE), others by abstract conceptualization (AC). (These terms, and those that follow, represent continua between extremes, not absolute conditions.) Some people process information primarily by active experimentation (AE), others by reflective observation (RO). Kolb calls those ‘‘divergers’’ who perceive CE and process by RO. Those who perceive as AC and process as RO are called assimilators. Those, who perceive as AC and process as AE, are called convergers. Finally, those who perceive as CE and process as AE are called accommodators. Though no individual fits neatly into one of these four categories at all times and may move around significantly in different situations, the four characterizations do provide some useful insights. Divergers. Divergers, also called imaginative learners, are given this description because they see concepts from different perspectives and generate ideas readily. They learn through discussion and want to interact personally with the faculty. Feelings are important, and they need to be convincingly shown why material being studied is important, to themselves and to others. Assimilators. Assimilators like order, are detail-oriented, and follow directions carefully. They are also called analytic learners. They learn well in traditional classrooms and prefer to work alone rather than in groups. They see the instructor as the expert and authority figure. They like lectures, especially those that are well presented, organized, and complete. Because so much of education is organized in a way that assimilators like, other students adopt many of the characteristics of assimilators to succeed.

LEARNING STYLES

Convergers. Convergers, also called common sense learners, quickly move (converge) to the essence of a problem or situation. They test information and ideas and are interested in the practicality and usefulness of the information presented. They are active and are less interested in lecturing as a class style. They prefer laboratory classes and prefer to work alone rather than in groups. Their preferred teacher plays the role of a coach, one who permits the students to take an active role in their own learning.

Not all students learn the same way. Recognition of this fact has received new emphasis in the last fifteen years. Some of the reasons for this include the desire to increase retention of students, especially the outstanding students who are uncomfortable in a traditional engineering program, and the desire to attract and retain women and students from underrepresented minorities. Of course the most important reason is simply to be more effective as faculty and to educate better engineers. The basic idea is for the faculty member to understand one or more of the different models that explain how students learn and to design instructional experiences to meet the needs of all students. The similarities in the models are more important than the differences among them. Despite

Accommodators. Accommodators are so called because they take what they have learned and adapt it to new problems, usually showing a lot of creativity. Often they are called dynamic learners. They like interaction and like to take an active role in their own learning and self-discovery. They resent too much structure. Their ideal instructor is one who remediates, encourages, and evaluates, but also who remains in the background as much as possible. Personal, individual inventories have been administered to many engineering students in both public and private universities. The data show that about 10% of the students are divergers, 40% are assimilators, 30 to 40% are convergers, and 10 to 20% are accommodators. The significance of these data

204

EDUCATIONAL TRENDS

is that, whereas 40% of the students are assimilators, for whom the traditional lecture is designed, 60% may be better served by extensive use of other techniques, especially those which promote active learning. This observation explains the rising importance of cooperative learning which, if used in conjunction with other techniques, allows educational experiences intended for all the students. As people mature, they have more interest in active learning experiences. This fact is important as engineering classes have more and more ‘‘nontraditional’’ students. Myers–Briggs Type Indicator The Myers–Briggs Type Indicator (MBTI) (26,27) is an instrument originally developed to measure, in part, whether an individual prefers to learn by sensing or intuition. Now the term usually refers to a measurement instrument that measures five axes of learning, including perception, information reception, reasoning progression, preferred learning processes, and preferred method of presenting materials. It thus has five axes, each with two extremes, and the possibility of 32 learning styles. While this does not suggest that faculty needs to segment their teaching into 32 distinct styles, it does suggest to many today that they need to alter presentations to reach many of the styles. Sensing and Intuitive Learners. Sensing involves gathering of data through the senses, especially seeing and hearing, whereas intuition involves indirect perception by way of the unconscious, including hunches, imagination, and speculation. Most undergraduate engineering students are, at this stage of their development, sensors, wherereas most of the faculty are intuitors. Intuitors prefer use of symbols, theories, and principles, whereas sensors prefer facts, data, and experimental work. Laboratory courses appeal to sensors. Visual and Auditory Learners. Engineering students may prefer to receive information verbally or visually. (There is a third method, kinesthetic, which plays little if any role in engineering education.) Visual learners prefer demonstrations, videos, computer animations, and diagrams. Auditory learners prefer lecturing and discussion. Because the process of writing equations on a blackboard is nearly always accompanied by speech, such a technique is considered primarily auditory. Engineering students are visual learners, though most engineering education today is auditory. Recognition of this fact is one reason for the continuing research into the subject and the significant amount of time being invested in World Wide Web learning and other computer-based techniques.

Unfortunately they may be easily discouraged and drop out prematurely. Because much of engineering education is structured for sequential learners, teaching global learners is a challenge to the faculty. Active and Reflective Learners. Information must be processed into knowledge by the learner. Some do this by active experimentation, using the ideas, testing them, working with them. They like experimental work. They do not like lectures, but work well in groups. Reflective learners prefer to work alone and need time to think about the information presented. It is important to recognize that this distinction differs from that between sensors and intuitors. Some sensors process information reflectively, others actively, and the same is true for intuitors. Cooperative learning, as the term is used earlier, incorporates some of the features of both active learning and reflective learning, which explains why it is such a powerful tool in the modern engineering classroom. Inductive and Deductive Learners. Inductive learning is a reasoning process that proceeds from the specific to the general, whereas deductive learning proceeds from the general to the specific. Inductive learners look at data, measurements, and observations and from this try to determine the underlying structure. They infer principles. Deductive learners start with theories and principles and try to infer consequences. Deduction is the natural teaching style, as it enables the faculty to start with basic principles and study the application of the principles. It gives the students a foundation for future study. But it is not necessarily the natural human learning style, especially for college-age students. People learn more by observation and experimentation and from that draw basic inferences. Much effort today is going into combining these techniques in the classroom, to accommodate all students and still provide a strong foundation for future learning by all of the engineers. One of the difficulties of deductive teaching for students, is that, when a book author or a faculty member starts with a principle and proceeds through a long derivation to a conclusion, the student often believes that the process is automatic, not one with many stops and starts along the way. Certainly the first person to do the work did not find it automatic. The process can be intimidating for students, yet attempts to show students the stops and starts often come across as if the instructor does not know how to do the derivation. The inductive-deductive dilemma may be one of the most difficult for the engineering faculty in today’s classroom. ENROLLMENT, DEGREES, AND RETENTION

Sequential and Global Learners. Some engineering students learn in a logical progression of principles, data, hypotheses, and new ideas, at a pace controlled in time. This is called sequential learning, and the students learn the material as it is presented. Such students can work with partial knowledge, going back over the ideas repeatedly until they master them. In contrast, global learners absorb and assimilate ideas in large blocks. They are characterized as confused until the ‘‘aha moment.’’ Suddenly, it seems, they understand the material well and are able to use it, often applying the ideas to quite difficult problems, and using the knowledge creatively.

Undergraduate engineering enrollment in the United States was relatively constant from 1966 until 1976, when it increased rapidly, reaching a peak in 1986 (28,29). Since then, enrollment has declined but has been relatively constant for the last ten years. It is often more interesting to look at degrees granted annually. The number of baccalaureate degrees was substantially constant at about 40,000 from 1966 until 1976, when it rose rapidly to a peak of nearly 80,000 in 1986. Since then, the degree numbers have declined to a relatively constant value of just over 60,000 per year, and this number

EDUCATIONAL TRENDS

does not appear to be changing significantly. One effect of this constancy has been for colleges and universities to focus more effort on retention, increasing the fraction of the students starting an engineering program who finally earn an engineering degree. At the graduate level, the number of engineers earning masters degrees has steadily increased over the interval 1966 until 1993 (there were two minor declines in this period). This number has gone from about 14,000 in 1966 to 28,000 in 1993. Many of these degrees are earned by students who combine work in government or industry with advanced study, taking advantage of more than 50 distance education and evening programs across the country. In electrical and computer engineering, the National Technological University now grants more than 200 masters degrees per year. Its primary delivery mechanism is satellite television, using courses from more than 40 major universities across the country. The number of engineers earning Ph.D.s first surpassed 1000 in 1962, and rose quickly to about 3500 in 1972. Then it fell to about 2200 in 1979, but since then has risen to nearly 6000 in 1993. Demographics The demographics of the baccalaureate graduates have changed. In 1966, approximately 1% of the graduates were female. By 1993, this number had risen to about 17%. A lot of effort has gone into developing programs to encourage young women to prepare for careers in engineering science while they are still in public schools. Summer workshops, careful and special advising, and programs to educate the faculty and majority group students how to avoid words and actions that make women feel uncomfortable have all played a major role. The Society for Women Engineers has been a major factor, as it provides role models for colleges and universities. Employers have also made special efforts to enhance the attractiveness of an engineering career to young women. At the graduate level, the fraction of women earning masters degrees has increased from 1% in 1966 to about 15% in 1993, whereas at the Ph.D. level, the number has increased from less than 1 to 9%. Three groups traditionally underrepresented in undergraduate engineering programs are Native Americans, Hispanic Americans, and African Americans. Much effort has been devoted to increasing these percentages in recent years, including the coalitions of the National Science Foundation, the National Action Council for Minorities in Engineering (NACME), and special programs on the campuses of many colleges and universities across the country. Because of their efforts, the fraction of engineering degrees going to these three groups has increased from 2.9% in 1972–73 to 9.2% in 1994–95. However, these three groups comprise about 21% of the U.S. population, so the groups remain underrepresented. Retention Many engineering educators are focusing attention on retention or graduation rates, comparing numbers of graduates with numbers of entering first year students (3–6,8,29,30). Retention rates vary widely across the country and with the type of institution. Some of the private institutions report graduation rates as high as 95%, whereas the rates in public universities vary from 50 to 70%. Students drop out of engineering for many reasons, but the focus is on those students

205

who, though they have the ability, lose interest for any of a variety of reasons. This problem may be more acute for underrepresented minorities and women, but not significantly so. Many of the new techniques employed in engineering education have as one of their main goals that of improving graduation rates. This includes cooperative learning, cooperative programs (with industry), attention to learning styles, and more efficient curricula. Other techniques shown to be important are being sure that students have adequate financial resources, sometimes said to be the most important single reason for students dropping out. Special programs must be carefully targeted. Programs designed to provide academic support for ‘‘at risk’’ students must not seen as programs for underrepresented minorities, or vice versa. Programs must be accessible, and the people involved must be available when the students need them, not necessarily during normal working hours. FACULTY DEVELOPMENT The preceding discussions have identified many new techniques and skills needed by engineering faculty. Faculty need time and must expend much effort to learn cooperative learning, teaming, skills for working with underrepresented minorities, women, and international students, how to do outcomes assessment, how to teach larger classes effectively, and how to do scholarly research in education and in technical disciplines. Many universities are making these opportunities available and are positively encouraging all ranks of their faculty to participate. At the same time, the faculty of our engineering schools and their graduates at all degree levels are doing exactly what they are expected to do, to advance knowledge and the level of technological development. This advancement is occurring rapidly, and the faculty must also keep up with their profession in addition to improving their effectiveness as teachers. These challenges are real, but meeting them is the reason most if not all faculty choose the academic career. Industrial Experience One of the concerns described in the studies mentioned earlier (3–5) is that a large portion of the engineering faculty have little or no industrial experience, especially at a level comparable to their university responsibilities. The faculty may have had summer assignments and have done some consulting, but many have not had the major responsibility for a complete design from the conceptual stage to production and marketing. Many see this as a weakness, and there is no doubt that those with good industrial experience are equipped to share these ideas with their students. Industrial leaders recognize this concern, and many are working to improve the quality of engineering education by several efforts. The most important is to give selected faculty members opportunities to work in industry. Some assignments may be a summer in length, whereas others extend well beyond a year. The assignments are challenging and give the faculty member opportunity to work in several phases of the industrial process. Some are sending their outstanding engineers to the campus for a semester or a year. These programs are not without problems. A faculty member or an engineer in industry, who finds employment rela-

206

EDUCATIONAL TRENDS

tively far from the home campus or corporate office, must move a family. In an era of two-career families, this is a major challenge. Assignments within commuting distance sometimes meet this need, but not all universities are located in or near industrial centers. Research programs and product developments may suffer. This problem is acute for young faculty members striving to earn tenure. Many of the programs are designed for faculty who have earned tenure and for engineers in industry who have advanced equivalently far along the ladder of achievement. A major concern is the academic emphasis on individual excellence, which contrasts with the industrial emphasis on teamwork. Universities are wrestling with this issue, but it is far from resolved, even though much research today requires teamwork and interdisciplinary activities. Scholarship An engineering faculty member must have, among other qualities, a desire for and a record as a scholar. Since World War II, the primary way a faculty member demonstrates scholarship is through research. Research has been important for the university, the nation, and the world, and for the agencies sponsoring it. Government agencies have sponsored most of the research, which has been targeted toward a variety of national needs. Research reports, conference presentations, and refereed publications are available for peer evaluation. To a faculty member, peer acceptance is crucial. The increase in engineering knowledge over this period has been remarkable. The universities, however, have been severely criticized over this period by students, ruling bodies, and the public. The charge has been that the faculty have been involved in research to the detriment of their teaching and undergraduate teaching. No study has demonstrated that the leading researchers are ineffective teachers. On many campuses, teaching awards voted by students go to those who are active in research. But the criticism continues. Changes are emerging in the research picture. More research is supported by industry and relatively less by government agencies. Industrial research is often proprietary, which inhibits timely publication of results. Industrial research has generally shorter periods for delivery of results than government agencies require. Universities are learning how to do industrial research and how to evaluate products, patents, processes, and publications. This major change has effects on tenure, to be discussed in the next section, and, when combined with public pressures, is requiring universities to rethink what they mean by scholarship. The most significant effort dealing with scholarship was a book published by Ernest L. Boyer (31) in 1990. Boyer studied universities and the concept of scholarship. He defined the work of the faculty, so as to reflect the wide variety of responsibilities that the faculty has, and showed that all of the major activities have a strong element of scholarship to them. He defines four separate functions, namely the scholarship of discovery, integration, application, and teaching. After defining them, he goes on to suggest ways in which all forms of scholarship might be evaluated. These ideas are being considered across the country as faculty and university leaders become accustomed to them. Scholarship of Discovery. The scholarship of discovery is defined as disciplined investigative efforts within the academy.

This is research, which should be undertaken systematically but with the freedom of inquiry to pursue knowledge for its own sake, wherever the trail may lead. Such scholarship invigorates the academy and is central to the academic mission. It must be cultivated, supported, and defended. It is vital to our society and to all people in the world. Scholarship of Integration. The scholarship of integration is characterized as interpretive, integrative, and interdisciplinary. It follows, in a sense, from the scholarship of discovery and includes doing research at the boundaries between fields (‘‘overlapping academic neighborhoods’’). It includes interpretation, or fitting one’s own research and the research of others into larger intellectual patterns. The scholarship of integration looks at the meaning of results. It is quite difficult and challenging. It rarely can be done by individuals, as it requires collaboration in most instances. Its evaluation is difficult, but it is important to the academy and to all of society. Scholarship of Application. The scholarship of application refers to scholarly activities proceeding out of research to make the results useful and uplifting for all of society. It may refer to applications of computers to medical systems to improve health or to electric power systems to increase reliability of such systems. Such scholarship is very demanding and usually requires a team effort to effect results. It is exemplified in today’s college of engineering by the efforts to encourage technology transfer, the movement of results from the laboratory to the marketplace. Technology transfer requires careful attention on the part of the academy to ensure that the rights of the sponsors and the public are properly considered. While a few universities have had such activities for many years, many universities are now learning how to encourage the scholarship of application. Scholarship of Teaching. The scholarship of teaching includes all of the activities a faculty must do to promote student learning (some have suggested it should be called the scholarship of learning). The teacher must be informed about the subject at hand and of interrelationships. The teacher must transmit knowledge and ideas and must also tranform and extend knowledge. The teacher must motivate the students to become learners themselves. The faculty does this by being learners themselves and exhibiting this behavior. In addition to the activities in direct contact with students, the scholarship includes preparation of textbooks, educational research, preparation of software, and laboratory materials. All of these are essential parts of the art and science of teaching. In Boyer’s book, the author goes on to discuss ways of evaluating these forms of scholarship and suggests that evidence must be gathered from at least three sources, self assessment, peer assessment, and student assessment. Across the country, promotion and tenure committees are struggling with these ideas and learning techniques for collecting the right data and performing rigorous assessments of the four forms of scholarship. Portfolios for faculty are emerging as one of the components of the evaluation and assessment process. PROMOTION AND TENURE A faculty member who is either doing research or teaching in a potentially controversial discipline may need a form of

EDUCATIONAL TRENDS

protection. Research may lead to results that challenge accepted positions or authorities. Teaching at the frontier of knowledge may require the faculty member to express ideas and research results that are unpopular for whatever reason. To protect the academic person from reprisals in such situations, the concept of academic freedom has developed, especially since the late 1930s. All faculty members have disciplines in which they are recognized as expert or authorities. When such faculty members express ideas, opinions, or research results within this discipline and do it in an academically responsible manner, academic freedom gives them the needed protection. Academic freedom is closely related to tenure, the granting of indefinite employment to a faculty member upon demonstrating appropriate scholarship and other characteristics. Tenure Granting indefinite employment to a person is a decision that the academy takes very seriously. In most cases, faculty members serve a probationary period of six years, near the end of which their records are examined by peers away from the campus, by students, and by colleagues on the campus. A few universities have probationary periods as long as nine years. Regardless of the length, all forms of scholarship are carefully evaluated, and, usually, secret votes among the faculty colleagues are taken. Recommendations are reviewed at several administrative levels before tenure is finally granted. Usually, but not always, the faculty member is also promoted in rank to Associate Professor. No good statistics on the fraction of nominees who actually earn tenure exist, but there is no question that the process today is long, arduous both for candidates and colleagues, and involves a lot of time and effort on the part of many people. Posttenure Review Not surprisingly, people not closely connected with the university have great difficulty understanding the concept of tenure. Few other professionals have or need a similar status. Many people see tenure as simply a lifetime contract and fear that many in the academy will ‘‘retire in place’’ after earning tenure. The elimination of mandatory retirement policies has exacerbated this concern, though no evidence exists that larger numbers of people are abusing the system. In public universities, governing bodies are responding to this concern by mandating more stringent reviews of faculty than they believe have been taking place. The details vary widely across the country. Nearly all faculty already receive periodic (usually annual) performance reviews, and they also are continually receiving peer reviews from their colleagues on and off of the campus, who carefully read their latest research papers and textbooks. Governing bodies are asking for more. The process is called ‘‘posttenure review.’’ The term refers to a formal, periodic process for reviewing faculty members who have earned tenure, using standards comparable to those for earning tenure. Faculty committees on many campuses are studying the process with great care. At the University of Minnesota, the Board of Regents has approved a new Promotion and Tenure document which contains a posttenure review section (32). This plan builds on annual reviews, and prescribes a process for working with those faculty members who receive less than favorable annual

207

reviews. Each such person is to receive further evaluation by a peer group. This group may conclude that the unfavorable review is not unwanted, or it may recommend a variety of courses of action having to do with work assignments and performance. In extreme cases it can recommend reductions in salary and even the initiation of dismissal procedures. The process is very structured and contains many safeguards for all concerned, especially the affected faculty member. It is too soon to assess the impact of the new procedure. As this is one of the first, committees all across the country are studying their document, and it is likely that many universities will have such a process in place in the near future. Some may be more, some less stringent. Tenure Concerns As mentioned earlier, colleges of engineering are striving to enhance their industrial contacts. One technique used is to add to faculties engineers who have outstanding records in industry, records of patents, product development, project management, and systems design. These people do not typically exhibit normal faculty credentials. Although many have taught in-plant short courses, they have not often taught university courses. They are discouraged or even prohibited from publishing in refereed journals for proprietary reasons, and they have little incentive because publication is not a part of their reward structure. Promotion and tenure committees are struggling with ways to recognize such contributions so that these new faculty members are properly evaluated but a tenured position is not immediately made available. Another concern for people in some disciplines is how to properly work with tenured faculty when enrollment in disciplines becomes very small and programs are discontinued. Although all governing bodies allow dismissing faculty members in such situations, the universities try very hard to find appropriate new situations for the people. Arguing that a scholar in one discipline can become a scholar in a different discipline, they universities are designing programs for ‘‘study in a second discipline,’’ granting leaves, and providing other appropriate opportunities. BIBLIOGRAPHY 1. Report of the committee of evaluation of engineering education, J. Eng. Educ., 46 (1): 25–60, 1955. Reprinted, J. Eng. Educ., 85 (1): 74–94, 1994. 2. Quality in engineering education. Executive summary of the final report, quality of engineering education project, Eng. Educ., 77 (1): 16–24, 49–50, 1986. 3. Restructuring Engineering Education: A Focus on Change, Report of an NSF Workshop on Engineering Education. Division of Undergraduate Education, National Science Foundation, April 1995. 4. Engineering Education for a Changing World, a Joint Project of the Engineering Deans Council and the Corporate Roundtable of the American Society for Engineering Education. Washington, DC: American Society for Engineering Education. 5. Major Issues in Engineering Education, A Working Paper of the Board on Engineering Education. Washington, DC: National Research Council, 1994. 6. Engineering Education, Designing an Adaptive System, National Research Council Study, Washington, DC: National Academy Press, 1995. (This report follows from Item 6.)

208

EIGENVALUES AND EIGENFUNCTIONS

7. A Framework for the Assessment of Engineering Education, The Joint Task Force on Engineering Education Assessment. Washington, DC: American Society for Engineering Education, 1996.

32. Available from the University of Minnesota, World Wide Web site http://www.umn.edu/usenate/faculty_senate/facultytenure.html, as of July 15, 1997.

8. The engineering education coalitions, Prism, 6 (1): 24–31, 1996. 9. J. Shaewitz, Outcomes assessment in engineering education, J. Eng. Educ. 85 (3): 239–246, 1996. 10. D. Soldan, Alumni assessment in the ABET 2000 environment, Proceedings, Frontiers in Education Annual Conference, Pittsburgh, PA, IEEE/ASEE, November, 1997. Vol. 27. 11. D. Aldridge and L. Benefield, A planning model for ABET Engineering Criteria 2000, Proceedings, Frontiers in Education Annual Conference, Pittsburgh, PA, IEEE/ASEE. November, 1997, Vol. 27. 12. G. M. Rogers and J. K. Sando, Stepping Ahead: An Assessment Plan Development Guide, Rose–Hulman Institute of Technology, Terre Haute, IN, 1996. 13. R. Payne, Communication conduit, Prism, 6 (5): 16–17, 1997. 14. V. Hendley, The basics of successful joint ventures, Prism, 6 (5): January 1997, pp. 18–21. 15. A. Dessoff, Profiles in Collaboration, Prism, 6 (5): January 1997, pp. 22–28. 16. Project LEA/RN (Learning Enhancement Action/Resource Network), a joint effort between the faculties of the Colleges of Engineering and Education at Iowa State University; and others. 17. S. Scrivener, K. Fachin, and G. Storey, Treating the all-nighter syndrome: Increased student comprehension through an interactive in-class approach, 85 (2): 152–155, 1994. 18. D. Johnson, R. Johnson, and K. Smith, Active Learning: Cooperation in the College Classroom, Edina, MN: Interaction Book Company, 1991. 19. J. Bordogna, E. Fromm, and E. Ernst, Engineering education: Innovation through integration, J. Eng. Educ., 82 (1): 3–8, 1993. 20. R. Quinn, Drexel’s E4 program: A different professional experience for engineering students and faculty, J. Eng. Educ. 82 (4): October 1993, pp. 196–202. 21. J. Shaewitz et al., The holistic curriculum, J. Eng. Educ., 83 (4): 343–348, 1994. 22. The Drexel Engineering Curriculum, Faculty, College of Engineering, Drexel University, Philadelphia, PA, 1995. 23. J. Sharp, J. Harb, and R. Terry, Combining Kolb learning styles and writing to learn in engineering class, J. Eng. Educ., 86 (2): 97–102, 1997. 24. D. Kolb, Experiental Learning: Experience as the Source of Learning and Development, Englewood Cliffs, NJ: Prentice Hall, 1984. 25. Learning Styles: Putting Research and Common Sense into Practice, American Association of School Administrators, Arlington, VA, 1991. 26. R. Felder and L. Silverman, Learning and teaching styles in engineering education, Eng. Educ., 78 (7): 674–692, 1988. 27. E. Godleski, ‘‘Learning style compatibility of engineering students and faculty, Proceedings, Frontiers in Education Annual Conference, IEEE/ASEE, 1984, Vol. 14. 28. National Science Foundation, Science and Engineering Degrees: 1966–94, NSF 96-321 Arlington, VA, 1996. 29. M. Reichert and M. Ashber, Taking another look at educating African American engineers: The importance of undergraduate retention, J. Eng. Educ., 86 (3): 241–254, 1997. 30. C. Moller–Wong and A. Eide, An engineering student retention study, J. Eng. Educ., 86 (1): 7–16, 1997. 31. E. Boyer, Scholarship Reconsidered: Priorities of the Professoriate, Princeton, NJ: The Carnegie Foundation for the Advancement of Teaching, 1990.

EDWIN C. JONES, JR. Iowa State University

EEG. See ELECTROENCEPHALOGRAPHY.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2904.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Electrical Engineering Curricula Standard Article Bruce A. Eisenstein1 1Drexel University, Philadelphia, PA Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2904 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (71K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

Abstract The sections in this article are | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20EL...ECTRONICS%20ENGINEERING/18.%20Education/W2904.htm17.06.2008 21:42:45

238

ELECTRICAL ENGINEERING CURRICULA

ELECTRICAL ENGINEERING CURRICULA Prior to the 1880s, courses began to appear in the area of electrical engineering, initially taught in the physics or the mechanical engineering department (1). These early offerings fell far short of a course of study or curriculum. However, in 1882 a program in electrical engineering was started in the physics department of the Massachusetts Institute of Technology. In the same year many European universities began to offer a variety of courses in circuits and machinery. The first separate and dedicated department of electrical engineering was formed at Cornell in 1883. Over the next two decades there was rapid growth in both numbers of electrical engineering departments as well as the number and variety of courses taught. As the number of departments giving degrees in electrical engineering increased, the need for a curriculum, or a canon, grew at an even faster rate. Engineering education at that time was a quasi-apprenticeship, with the student working in close proximity to the professor in order to learn the ‘‘art of engineering.’’ The professor, in turn, was closely tied to industry, frequently getting problems as well as experience from neighboring companies. Thus the early courses offered in the nascent departments paralleled the topics of interest to the professors and industry. Between 1884 and 1889 the vast majority of technical papers published in the AIEE Transactions (2) were in the areas of machinery, lighting, instruments, and circuit devices. By the early 1900s the number of papers and the intensity of the work in machinery was an order of magnitude greater than the next areas (lighting and instruments). The curricula in the early 1900s reflected the needs of industry with courses in electric power generation, transmission, and distribution, electrical measuring instruments, machinery, and an intensive laboratory for experimentation. From the catalog of Drexel Institute (the predecessor to Drexel University), in 1903 one finds that for the first two years of their education, students received training in algebra, basic English, social studies, geography, etc. In the third year, students took: calculus, chemistry—qualitative analysis, physics, mechanics of materials, principles of mechanism [sic], electricity—general theory mechanical engineering laboratory, electrical engineering laboratory, and noncredit courses in English and engineering seminary [sic]. The total number of scheduled class hours per week was 30, composed of 15 classroom hours and 15 laboratory hours. The student yearbooks from that time indicate that in fact students spent much more than 15 hours in the laboratory. In the fourth year, students would get courses and laboratories in machinery, telephone systems, dynamo design, instrumentation, machinery, thermodynamics and the steam engine, electrochemistry, building construction, telegraphs and signal systems, and noncredit courses in business, English, engineering seminary, and monthly visits of inspection. Again, students spent 30 hours per week at school divided equally between laboratory and classroom work. In 1914, on the eve of World War I, the curriculum for electrical engineering students had begun to look surprisingly similar to that in the 1960s with the exception of the tuition, J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

ELECTRICAL ENGINEERING CURRICULA

which was $70 per year at a private school like Drexel. Students still spent 30 hours per week at school; however, now more time was spent in class and less time in laboratories. Senior year consisted of 18 hours per week in subjects like dynamo design, ac circuits, telegraphy, and another four hours in civil engineering—hydraulics and structures. Economics and English are now given course credit, and the English course is on writing contracts and specifications. In 1929 the curriculum began to become more streamlined. Total hours per week at the university ranged from 25 to 28, and some of those hours were in required military training. Required courses in public speaking, poetry and prose, and composition and literature are standard. New courses include separate courses in synchronous and induction machinery, but steam turbines remains part of the electrical engineering requirements. In 1941, on the eve of World War II, one can scarcely find a change in the curriculum from 12 years earlier. There is one course on the theory of vacuum tubes and a new course on the differential equations of the electric circuit. Otherwise, electrical engineers went through the same curriculum as their predecessors did 10 to 20 years earlier. World War II was a watershed for electrical engineering education in a number of ways. I heard a speech by Vannevar Bush, who had been the science advisor to President Harry Truman and who was deeply involved in the war effort, in which he said that electrical engineers were unprepared for the rapid pace of change and new technologies that arose during the intense period of the war. Looking back, through a prism forged in the 1990s, the pace of electrical engineering from its birth in the 1880s to the start of World War II in 1941 appears to be glacial. Of course to the people alive during that period, the pace was frenetic—electric lights, telephones, household appliances, traction motors, bright signs, and neon lights. However, it was the physicist and the mathematician that enabled us to develop radar and automatic firecontrol systems. Major efforts such as the Manhattan Project to develop the atomic bomb and the jet engine work at Cal Tech’s Jet Propulsion Laboratory required sophisticated instruments that had never been seen before. And the many efforts in automatic computation required a new look at how to solve problems. Those colleges and universities that had been heavily involved in the war effort were able to seamlessly transform their curriculum in the postwar period to include the new subjects such as electronic circuit design, microwaves, modern physics, advanced calculus which had proved so valuable to the Department of War. The newly formed Department of Defense rewarded these universities with substantial (at least at that time) research grants to continue work in these important new areas, thus positioning those schools to advance the technology and bring it into the classroom sooner. By contrast, schools that were not so involved found themselves with a now antiquated faculty and an old-fashioned curriculum. The problems in the immediate postwar period were compounded by a surge of new technical developments ranging from the transistor in 1947, the first stored-program digital computer, ENIAC, in 1948, and the beginnings of the information revolution with the publication of The Mathematical Theory of Communications by Claude Shannon and Warren Weaver in 1948. Coupled with the war technologies such as radar, servomechanisms, and new materials, the old curriculum for electrical engineers was, simply

239

stated, inadequate. Yet, at a number of institutions, there were an insufficient number of ‘‘modern’’ faculty members to make the necessary changes. Looking at the catalog of the Drexel Institute of Technology in 1951, six years after the end of the war, one finds, in addition to the courses that have been in the catalog for 40 to 50 years, some new courses in electronic industrial control, servo-mechanisms, radio communications, and ultra-high-frequency circuits. By 1961 there were courses in electronics (I am told by those who taught the courses that there was no mention of solid-state devices), introductory atomic physics, electromagnetic fields, and radio electronics. At this time there were graduate courses in circuit synthesis and switching theory, the forerunner to digital design. In 1946, prompted by their perception of the inadequacy of the educational programs for electrical engineers, Vannevar Bush and Gordon Brown, the latter head of the electrical engineering department at MIT, began a major effort to revamp the curriculum for EEs by the addition of engineering science courses, basic science courses, particularly advanced calculus, and modern physics and by deleting many of the ‘‘practical’’ courses that had been in the curriculum for 40 to 50 years. It was rumored that the intention was to give EEs as much physics as the physics major and as much calculus as the mathematics majors. Much of the material taught to undergraduates was distilled from research and graduate programs and trickled down to the undergraduate courses. Many schools adopted this curricular model—current advanced research work is brought immediately into graduate courses and then introduced into undergraduate courses. Textbooks that flowed from MIT and other schools that adopted this model began to look less like the pre-war handbooks and more like research papers organized like monographs. These books provided the means for every school, with or without their own research program, to have current topics in the curriculum. In 1958 MIT received a large grant from the Ford Foundation to ‘‘modernize’’ electrical engineering education. The immediate effect was to put a ‘‘t’’ after every course number indicating a transition course, as in 6.01t Introduction to Circuit Theory. Another part of the grant was used to upgrade the laboratories and to provide each student with a take-home kit of electronic components so that circuits and devices could be designed and built outside the laboratory. There were new courses introduced such as fields, forces, and motion that replaced the traditional machinery course and molecular engineering which was a wholly new course. Even in the basic circuits courses, new concepts, previously taught only to graduate students, were introduced to the undergraduates at the sophomore level. Convolution, La Place transforms, and twodimensional signal processing put the second-year students ahead of the fourth-year students in many ways. As textbooks began to emerge from this ambitious effort, the new way of teaching percolated throughout the electrical engineering education enterprise. In 1932, the seven major technical societies, including the American Institute of Electrical Engineers, formed the Engineers Council for Professional Development (ECPD) for the purpose of developing criteria for accrediting undergraduate engineering programs. In 1936, ECPD implemented a method of visiting engineering schools and awarding accreditation to programs that meet the minima criteria set by the sponsoring

240

ELECTRICAL ENGINEERING CURRICULA

societies. ECPD became the Accreditation Board for Engineering and Technology (ABET). In the 1960s and 1970s the continuing evolution of the digital computer prompted many electrical engineering departments to change the department name by including ‘‘computer’’ in some form (most commonly as electrical and computer engineering) and to add courses in digital electronics, switching theory, computer languages, and operating systems. As the number of computer courses proliferated, many of the computer-intensive departments split into separate departments of electrical engineering and computer engineering (or computer science). With the split, and the identification of computers as a separate discipline, a new computer curriculum has evolved. A new professional accreditation board emerged, Computer Science Accreditation Board (CSAB), as a parallel to ABET. The curriculum for computer-oriented programs is discussed elsewhere. For 30 years the research/science curricular model described above persisted in nearly every engineering school in the country. By the mid-1980s, however, some engineering educators and industrialists began questioning whether the engineering curriculum was appropriate for the new workplace. Until the 1980s the workplace for electrical engineers consisted of a number of very large electronics firms that employed thousands of engineers. These engineers were almost entirely white, male, and, in this country, American. There were of course many electronics companies outside the borders of the United States, but there was little interaction among them nor was there any substantial interactions between the companies outside the United States and inside. The fact that for the entire postwar period, these electronics companies were heavily involved in defense work accounted for this separation. A typical electrical engineering graduate would begin work as a junior engineer under the supervision of a project engineer and would be working in, what is now seen to be, a homogeneous environment. Work was for a lifetime with most engineers changing jobs only once or twice throughout their careers. These large companies had marketing departments that got business for the company and was responsible for the proposals. These companies also had extensive research and development capabilities, which meant that the typical bench engineer did not have to deal with customers nor deal with the demands of new technology. The curriculum and the education of electrical engineers reinforced this paradigm, and it was that fact that started the questioning about the relevance of the engineering curriculum. The workplace was changing and the curriculum had to change with it, they felt. In the mid-1990s, a trend became apparent that more graduates were going to work for small, start-up companies than for the established large companies. The downsizing and layoffs of the early 1990s served to accelerate this trend in the latter part of the decade. In working for the small companies, engineers find themselves involved with business, marketing, personnel problems, aesthetics, maintenance, etc., in addition to the new range of engineering challenges. Thus the curriculum had to change to accommodate the new demands of the workplace. Furthermore, many university administrators were clamoring for a reduction in required hours of instruction, catching the curriculum in a four-way bind: industry is

demanding more and different skills from the graduates; administrators want time in class reduced; many universities have immutable core requirements; and ABET maintained a large number of required courses and areas of study. Something or many things had to give. The first breakout was the new curriculum initiatives, discussed later, which made the curriculum more efficient. Then ABET implemented Engineering Criteria 2000 (EC 2000), which focused on outcomes assessment and continuous quality improvement from feedback from the assessment process to inform the accreditation process. This was a major departure from what some had called a ‘‘bean counting’’ approach to accreditation, forcing a sea change in the way schools go about the business of establishing a curriculum. The National Action Agenda for Engineering Education (3) called for a radical change in the way engineers were taught. Some of the issues were: • improving the content of undergraduate programs • role of manufacturing • career-long development of the engineer In response to this call for action, the National Science Foundation issued a call for proposals that would address the following areas: the overburdened curriculum, design and manufacturing, practice-oriented graduate programs, faculty development, laboratories, career-long learning, and pre-college education. The NSF received 197 proposals to reform engineering education, and they funded a large number of these. Many of the 197 proposals represented ideas that were formed into the engineering coalitions funded by NSF over the next several years. By 1989 there was a revolution started in the curriculum for engineering that was unparalleled in the previous 100 years of engineering education. The new curricula that have emerged from these NSFfunded projects have the following features in common (4). Students have more choice in the courses they will take; nontechnical content of the curriculum has been increased; students will do more computer modeling and less paper-andpencil computation; communication skills are developed and enhanced; teamwork is facilitated; understanding of diversity, social issues, economics, and aesthetics are all encouraged. The new accreditation criteria for engineering programs, ABET 2000, stresses outcomes assessment and self-evaluation both of which resonate with the enhanced skills that the modern curriculum offers. BIBLIOGRAPHY 1. E. Weber and F. Nebeker, The Evolution of Electrical Engineering, New York: IEEE Press, 1994. 2. J. Ryer and D. Fink, Engineers and Electrons, New York: IEEE Press, 1984. 3. National Action Agenda for Engineering Education, Engineering Education, November 1987. 4. J. A. Orr and B. A. Eisenstein, Summary of innovations in electrical engineering curricula, IEEE Trans. Educ., 37: 131–135, 1994.

BRUCE A. EISENSTEIN Drexel University

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2906.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Electrical Engineering Education Standard Article Andrew P. Sage1 1George Mason University, Fairfax, VA Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2906 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (122K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2906.htm (1 of 2)17.06.2008 21:43:06

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2906.htm

Abstract The sections in this article are A Brief History of Electrical Technology Evolution Objectives for Higher Education and Engineering Education Student Educational Needs, Outcomes, and Metrics–and Program Accreditation Summary | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2906.htm (2 of 2)17.06.2008 21:43:06

ELECTRICAL ENGINEERING EDUCATION This article provides an overview of education, engineering education, and electrical engineering education. The historical evolution of electrical engineering and electrical engineering education is briefly traced. Then recent trends in industry and education, as they impact on engineering education, are described. Objectives for higher education and engineering education are described, as are student educational needs, outcomes, and associate metrics. Various attempts at educational reform, including engineering education reform, are discussed. An overview of quality assurance through accreditation and the current and evolving trends in accreditation activities conclude the article.

A BRIEF HISTORY OF ELECTRICAL TECHNOLOGY EVOLUTION Humans have used such natural tools as sticks and stones in order to develop simple products for many thousands of years. In the Old Stone Age, or from the beginning of civilization until about 15,000 years ago, human development was primarily dependent upon hunting and fishing using simple stick- and stone-based tools. The New Stone Age was made possible by the development of primitive practices involving animal husbandry and agriculture, and the evolution of building technologies that led to such constructions as the Egyptian pyramids and Stonehenge. The beginning of manufacturing, using baked clay and soft metals, enabled the development of trade and commerce. The New Stone Age led to the Metal Age and to the development of wind power as a substitute for human muscle power. The printing press, the steam engine, the telescope, metallurgical and mining advances, and continuing agricultural innovations led to the Industrial Revolution. The Industrial Revolution was an advance in power and control technology. Toward the latter portion of the Industrial Revolution, advances in electrical and electronics engineering led to the discovery of the computing machine and the beginning of the Information and Knowledge Revolution. Many conventional definitions of engineering suggest that it is the application of scientific principles to the effective and efficient conversion of natural resources into products and systems for the benefit of humankind. The notion that engineering is concerned with effective and efficient use of resources for the betterment of humankind is certainly correct. There are many constraints affecting this use and engineering is much concerned with developing solutions under resource constraints. Initially, these resources were considered to be natural resources. Today they are considered to be any of the four major resources or capital, as unspent resources are often now called:

1. 2. 3. 4.

Natural resources, or natural capital Human resources, or human capital Financial resources, or financial capital Information and knowledge resources, or information and knowledge capital

This enlarged concept of resources enables the inclusion of such important contemporary knowledge-intensive efforts as biotechnology and biomedical engineering. Science, on the other hand, is primarily concerned with the discovery of new knowledge. There is no inherent notion of purpose in scientific discoveries, although obviously many scientific investigations are directed at knowledge that will be of use to humanity. Knowledge of the principles of the natural and mathematical sciences is very necessary, but not at all sufficient for engineering practice. Much more is needed. In the beginning there were two divisions of engineering: (1) military engineering and (2) nonmilitary, or civil, engineering. With increasing knowledge, civil engineering became more and more specialized to static structures and mechanical engineering emerged as the field of engineering interested in dynamics. There are four primary and traditional engineering disciplines: (1) civil, (2) mechanical, (3) electrical, and (4) chemical engineering. Each of these has several more specialized branches. There are many other important engineering disciplines, such as industrial, mining, environmental, biomedical, aerospace, and systems engineering. Electrical engineering is concerned with the practical applications of electricity. Electronics engineering is the branch of electrical engineering that is particularly concerned with use of the electromagnetic spectrum and with such electronic devices as integrated circuits. It might be possible to make a distinction between electrical and electronics engineering on the basis of the comparative magnitude of the flowing electric currents. This is hardly, if ever, done today and the term “electronics” is infrequently used to describe academic program titles, as contrasted with the more generic term: electrical engineering. If it were done, electrical engineering would be partitioned into electrical power engineering and electronics engineering. The history of electrical engineering (1) is a very interesting and inspiring one. Much of the world has been transformed by technology, as evidenced in an excellent work by Hughes that describes the history of American invention and innovation over the century from 1870 to 1970 (2). In particular, a tremendous growth in electrical technology occurred in the last fourth of the nineteenth century. By the early 1880s, telegraph wires largely covered the United States and underwater cables connected Europe and America. Rudimentary forms of arc lights were in use in several cities and the Pearl Street Station of Thomas Edison was supplying power for the then new incandescent lighting. There were many organizations involved in the manufacture of electrical equipment. The telephone was rapidly growing in importance as a communications instrument. While this period of time could hardly be called the information age, Beniger (3) indicates that it was actually during this period that the essence of the contemporary information age began in America. The major increase in production and use of electrical equipment that was experienced in the latter part of the nineteenth century encouraged the Franklin Institute, a private philanthropic institution, to sponsor an 1884 International Electrical Exhibition at Philadelphia. In 1884, civil engineers, mining engineers, and mechanical engi-

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright © 2007 John Wiley & Sons, Inc.

2

Electrical Engineering Education

neers had each formed national professional engineering societies. There was no national electrical engineering organization at that time. A “call” for an organization of electrical engineers, signed by 25 prominent engineers of the time, was placed in an 1884 issue of the then major electrical engineering journal for the purpose of initiating formal actions to accomplish this. One of the outcomes of this exhibition was the formation of the American Institute of Electrical Engineers (AIEE), which held its first technical sessions during the exhibition. Some papers presented there were published in the first volume of the Transactions of the AIEE. The first paper in these transactions concerned the “Edison Effect,” a phenomenon which became one of the foundations of electronics. In 1902, AIEE student branches were first organized at engineering schools, with the first at Lehigh University. The AIEE was very much electrical-power-oriented and those in the “radio” world often did not feel comfortable within the AIEE. To accommodate these interests, an Institute of Radio Engineers (IRE) was founded in 1912, and the first issue of their journal, the Proceedings of the Institute of Radio Engineers, was published in January 1913. Before World War II, the IRE was small compared with the AIEE and other engineering societies. The major growth of electronic communications during and just after WW II led to an IRE membership that was much larger than the older AIEE. The then new area of electronics was attracting most electrical engineering students and the majority of new jobs for electrical engineers were in electronics, rather than in electrical power. This led to merger discussions between the AIEE and the IRE, and the Institute of Electrical and Electronics Engineers (IEEE) was officially founded in 1963. Initially, it was a “learned society” of electrical engineering professionals. The Proceedings of the IEEE was the official journal of the new institute. In 1964, the IEEE Spectrum became the new core publication of the IEEE and the Proceedings became a separate publication devoted to more technical issues, including special issues on new and emerging electrical technologies. In 1973 the IEEE relinquished its then exclusive role as a learned society concerned only with the advancement and dissemination of knowledge. It took on the role of a professional society that was concerned with nontechnical and with technical interests. As of the late 1990s, the IEEE membership was approximately 250,000, not including student and affiliate members, and it is now the world’s largest professional society. In July 2006, the IEEE has more than 365,000 members, including 68,000 students, in over 150 countries. There are 311 sections in ten worldwide geographic regions. It is comprised of 39 professional societies and 5 technical councils representing a very wide range of electrical and electronics interests. It publishes 128 transactions, journals, and magazines. There are approximately 900 active IEEE standards and about more than 400 under current development. Thus, it can be seen that the spectrum of interests of the IEEE, and generally of electrical engineering education as well, is broad. These interests are represented by the very large number of IEEE “Professional Societies”

Aerospace and Electronic Systems Society Antennas and Propagation Society Broadcast Technology Society Circuits and Systems Society Communications Society Components Packaging, and Manufacturing Technology Society Computational Intelligence Society Computer Society Consumer Electronics Society Control Systems Society Dielectrics and Electrical Insulation Society Education Society Electromagnetic Compatibility Society Electron Devices Society Engineering Management Society Engineering in Medicine and Biology Society Geoscience & Remote Sensing Society Industrial Electronics Society Industry Applications Society Information Theory Society Intelligent Transportation Systems Society Instrumentation and Measurement Society Lasers & Electro-Optics Society Magnetics Society Microwave Theory and Techniques Society Nuclear and Plasma Sciences Society Oceanic Engineering Society Power Electronics Society Power Engineering Society Product Safety Engineering Society Professional Communication Society Reliability Society Robotics & Automation Society Signal Processing Society Society on Social Implications of Technology Solid-State Circuits Society Systems, Man, and Cybernetics Society Ultrasonics, Ferroelectrics, and Frequency Control Society Vehicular Technology Society The IEEE Councils include: Council on Electronic Design Automation Council on SuperConductivity Nanotechnology Council Sensors Council Systems Council There are a plethora of transactions and magazines published by these societies and councils.

Electrical Engineering Education

Some of these subject areas are very basic ones in virtually all electrical engineering educational curricula, and many educational programs offer specialized study in some of the other society and council areas. For example, the subjects of electric circuit theory, field theory, communications theory, and automatic control theory are considered generally very basic and fundamental subjects and would be found in all undergraduate programs, as generally would be power. Areas such as vehicular technology are quite specialized and would not usually be found in typical undergraduate electrical engineering curricula. The early electrical engineers had very different backgrounds: some were formally educated, often in fields very different from engineering, and others were informally educated through experiential learning. Leaders in electrical engineering soon recognized that a new kind of education for professional practice would be needed if electrical engineering was to progress as a profession. The initial electrical engineering programs were established in the early 1880s, roughly at the same time that electrical engineers began organizing themselves professionally, and their rapid growth in following decades was very beneficial, both for the profession and for the nation. The Massachusetts Institute of Technology (MIT), then in Boston, founded the first US-based electrical engineering program in 1882, two years before the founding of the AIEE. The initial program was a program within the physics department at MIT and, partially as a result, was heavily influenced by physics. Initially, of course, there were no electrical engineering textbooks. Development of these, as well as the development of laboratories and a cadre of electrical engineering specialists, as professors and as practitioners, were among the very early and important tasks for the new profession. Initially, the electrical engineering curriculum was comprised almost totally of electrical power engineering courses. Some of the initial electrical engineering professors distinguished themselves in innovation and research, as well as in education. Others devoted their full-time efforts to education. The great importance of radio and the many possible opportunities for utilization of vacuum tubes and telephonic communications suggested that these subjects deserved a place in the curriculum of even the most strongly electrical-power-focused electrical engineering department. The resulting “communications option” began in an effort to make electrical engineering education more responsive to changing technology and changing needs of industry and society. These changes also led to establishment, generally in the 1920s, of “cooperative electrical engineering” programs, which allowed students to alternate periods of employment with periods of study. The first part of the twentieth century, and the significant impact that engineering and engineers had on society, found engineers in new societal roles, and this led to debates concerning the role of the engineer in society. Such issues as control of the radio spectrum were of great governmental, organizational, and societal importance, and led to establishment of the Federal Radio Commission in 1928, as a cooperative venture between radio engineers and government regulators.

3

The importance and role of the engineer in society has increased markedly since that time. The discovery, in 1947, of the transistor by scientists at Bell Laboratories provided the electrical engineer with the ability to exploit an entirely new electronic world that was then unknown. It was the semiconductor that was responsible for the evolution of the digital computer, even though these computers had certainly existed earlier through relay (which is where the term “computer bug,” as a real physical hazard, was first used) switching circuits and vacuum tube switching circuits. Without the ubiquitous digital computer, the various space adventures of the mid twentieth century would not have been possible. However, the decline of the space program and military programs in the 1970s led to a period of unemployment for electrical engineers, approximating 6%. While not high compared with nonprofessional laborers, this was so highly unusual in the engineering profession that it led to considerable declines in engineering enrollments. It was during this period of unrest that the IEEE reformed itself as a professional, rather than only as a strictly learned, organization. Microelectronics and integrated circuit related efforts, including digital computers and communications, became the “glamour” technologies of the 1970s and 1980s. These technologies have produced profound impacts on society and on the electrical engineering profession. The ease of development and the power of integrated circuits have actually changed the way electrical circuits are architected and then designed, and have led engineers to actively search for digital solutions to problems that are not themselves inherently digital. For example, the simulation of continuoustime dynamic physical systems, such as aircraft, is now accomplished almost totally digitally, even though the physical systems themselves are continuous-time systems for which much analog computer technology had been developed in the 1950s and 1960s. This “digital everything” trend has resulted from the major developments in semiconductors, abilities at very large scale integration of electronic circuits, the resulting microprocessor-based systems, and associated major reductions in size and cost of digital computer components and systems. The digital revolution (4) has led to a death of distance (5), through the merging of telecommunications technology and computer technology into information technology, and the emergence of networked individuals and organizations. The major characteristics of this change include great speed (6) and the major necessity for engineering, including electrical engineering, to be especially concerned with social choice and value conflicts issues (7) that surround strategic management of the intellectual capital (8) brought about by the information technology revolution and the use of information technology for organizational and societal improvement. The initial focus on data in the early days of computers shifted to a focus on information and information technology in the decade of the 1990s. Now the imbedding of information concerns into greater concerns, which affect knowledge resources and the need for transdisciplinary issues of knowledge integration, must be addressed well in order to ably handle the concerns of the early twenty-first century. This is bringing about major changes (9) and needs for engineering education to adapt

4

Electrical Engineering Education

programs to these changes such that the customers of engineering education—students and employers—remain satisfied with educational product quality. In the early days of human civilization, development was made possible through the use of human effort, or labor, primarily. Human ability to use natural resources led to the ability to develop based not only on labor, but also on the availability of land, which was the classic economic term that implied natural physical resources. At that time, most organizations were composed of small proprietorships. The availability of financial capital during the industrial revolution led to this as a third fundamental economic resource, and also to the development of large corporations and resulting steep hierarchies. This second wave is generally associated with centralization, mass production, and standardization. Major availability of information and knowledge has led to information and associated knowledge as a fourth fundamental economic resource for development. This is the era of total quality management, mass customization of products and services, of reengineering at the level of product, process, and systems management, decentralization and horizontalization of organizations, and integration at the level of products, processes, and organization. While information technology, a product of electrical engineering effort, has enabled these changes, much more than just information technology is needed to bring them about satisfactorily. The transition from the mainframe and minicomputers of the mid-twentieth century to client server computing and networking of the very late twentieth century, and the associated changes in organizations and society, have not come without concerns for growth and progress. During the decade of the 1980s and early 1990s, many asked what has happened in the past quarter century relative to this continued progress. Concerns were expressed in the 1980s and early 1990s on the subject of declining US innovation and productivity. Generally, the papers and books of that period furnished suggestions to enhance technological and management efforts in such a way that would lead to renewed efforts to enhance competitiveness through innovative research and development, associated technology transfer, and better use of human resources. Some of these writings suggested that America was in potentially deep trouble, being devalued (10) through a perceived unmaking, or “dumbing-down,” of education and associated declines in many social and political institutions. Illiberal education (11) was claimed to result from “politically correct” expediencies that were felt to be rapidly eroding the long-standing traditional mores of scholarship and individual distinction throughout much of American higher education. These have purportedly been replaced with a dogmatic, intolerant, and repressive new-liberalism (12). This is nourished by rapid increases in entitlements to individuals and groups, but with fewer and inadequate resources available to assure their availability. The net result was perceived to be a decline on virtually all fronts. ProfScam (13), in which many American university professors arrange teaching schedules, often through substitution of full course teaching efforts with very modest sponsored research charges, such that low-paid and often marginally qualified surrogates in the form of teaching as-

sistants and adjunct faculty cope with a large part of the student credit-hour production in the classroom, was cited as a contemporary reality. This allows professors to engage in more highly preferred and self-indulgent activities that are generally funded by the inadequately serviced student credit-hour earnings of the university. This is claimed to result in high cost and low quality for much of American higher education. These inferior solutions in higher education were claimed to contribute to many related difficulties concerning American productivity and innovation, primarily through a failure to provide the emphasis on human resources needed for a competitive America. Elementary and secondary education have not escaped similar criticisms. A 1983 report espoused the view that any external imposition of the situation then extant in education would likely have been proclaimed as “an act of war” (14). The report goes on to describe what is called acts of “unthinking, unilateral educational disarmament.” In the minds of many, the situation had deteriorated and the American educational system deserved “failing grades” (15) on a variety of important performance facets. Schlesinger (16) claimed that a disuniting of America has resulted from a shift in the traditional concept of America as a melting pot of individuals from all nations who join together and seek new lives. This has been replaced by an ethnicity upsurge in which America is viewed as distinct diverse groups with indelible cultural characteristics. While healthy consequences have resulted from this, such as overdue recognition of equality of opportunity across race and gender, it is also claimed to have resulted in a sacrifice in traditional American beliefs in individual responsibilities and subordination of these responsibilities to notions of group rights and entitlements. Indeed, warrants and backings have been provided (17) to support the claim that contemporary programs to ensure rights and entitlements at the expense of responsibilities have been genuinely counterproductive to the progress of the very individuals whose rights are presumably protected, but whose opportunities for emergence from their present state are thereby ultimately and significantly degraded. Related inquiries have focused more on technology, economic, and managerial issues supporting innovation and productivity. It was argued that the laws, regulations, and practices that are associated with corporate management and governance, investment practices, and executive compensation all favor a short-term, myopic perspective. This focus on “short term America” (18) was said to occur at the expense of the long-term view, leading to disinterest in developing the technologies and manufacturing the products and systems that will ensure increasing market share and the associated long-term accomplishments and subsequent and affiliated affluence of a productive America. Various contemporary investigations have suggested improvement strategies, such as for reinventing the factory (19) in order to enhance US participation in the ongoing quest for world markets. Competitiveness through increased advancement of emerging information technologies was a major thrust of many of these works. It was argued strongly (20), that information technology studies and developments extend much beyond the neoclassic engineering of data processing to incorporate intel-

Electrical Engineering Education

lectual property laws, public- and private-sector policy considerations, and economic and systems management considerations. Four major strategies were suggested in this effort: 1. Redefining the technology base to include information as an essential ingredient 2. Leading in the development and application, often through technology transfer, of new and emerging technologies 3. Capturing market share and dominating the commercial production sector 4. Controlling the use of information technology in national security applications. The major substance of this, and other works, was that there are indeed strategies that can restore what once did exist concerning productivity and quality and that properly managed information technology could represent a major supporter of these developments. In the late 1980s and early 1990s, there was an abundance of studies associated with strategies to regain the productive edge, such as an excellent MIT Commission on Industrial Productivity study (21), which identified four then existing adverse facets that detracted from American productivity in a changing world economy: 1. Technological weaknesses in development and production, especially in terms of design for manufacturability and quality 2. Neglect of human resources in terms of formal education and training, and neglect also of on-the-job reeducation and retraining 3. Failures of cooperation within individual organizations, in an interorganizational sense, and with respect to labor–management relations 4. Government and industry at cross-purposes, especially with respect to regulatory policies, technological infrastructure issues, and the lack of technologytransfer mechanisms to capture as many direct and indirect benefits of military research and development This discerning work suggested a number of strategies for industry, labor, government, and education that would potentially lead to a more productive America. The four identified critical success factors for students and faculty are of particular interest here. Stated as objectives, these were: 1. To encourage strong interest in and knowledge of real problems from economic, social, and political perspectives 2. To encourage team efforts in the creation of new processes, systems, and products 3. To develop the ability to function effectively beyond the confines of a single discipline 4. To develop the ability to integrate a deep understanding of science and technology with practical knowledge, including the necessary human and experimen-

5

tal skills and insights A major objective espoused in studies of this sort was increased recognition of the critical issues and interactions associated with university education and national productivity. Contemporary efforts at revitalization of engineering education are based in large part on such studies as this. There were such other studies as a 1991 report from the National Research Council (22), which identified five major categories of national need: 1. Systems management for technology development, including management of product-development processes throughout the entire systems life cycle 2. Management of complex large-scale processes, such as highly automated flexible manufacturing or broad band telecommunication systems that necessarily have much different management requirements from those of management of low-technology enterprises 3. Using technology for competitive advantage, in particular, including the use of such modern information technology developments as management information systems and decision-support systems that offer major potential for successful integration of technology-management and enterprisemanagement strategy 4. Technology–organization interactions, including concerns of successfully integrating technology into the organization, and having technological solutions accepted by those charged with using them 5. Social impacts of technology, including developing technologies that are societally acceptable and environmentally acceptable Studies such as these have led to major reengineering efforts across industry and, perhaps to a potentially lesser extent, government. Such efforts as total quality management, reengineering, strategic sizing, organizational learning, and associated information technology and knowledgemanagement efforts have led to major improvements in systems-management capabilities. To a great extent, systems engineering and information technology (23), now much associated with knowledge management (24) and knowledge sharing (48), as enabled by information technology, have been the catalysts for these changes. While the changes in the industrial and government sectors occurring over the last decade and a half have been remarkable and have led to the US regaining the competitive edge (25), changes in the university are generally judged to be less appreciable and less effectual, and suggestions for change continue to abound. For example, in 1996 Anderson (26) suggested ten critical points for university revitalization: 1. Prohibit student teaching by the thousand of graduate students, generally pursuing doctoral degrees, who routinely teach undergraduates. 2. Cease rewarding spurious research.

6

Electrical Engineering Education

3. Change the nature of doctoral programs such that they emphasize other than research. 4. End faculty tenure in favor of a performance-based system. 5. Reorganize faculty titles and responsibilities such as to distinguish primarily teaching efforts from those involving mostly research and publication, for whom the title of Fellow might be more appropriate. 6. Return to the four-year undergraduate degree, as contrasted with the approximate five-year degree which has become the norm in many institutions. 7. Take sexual harassment seriously. 8. Ban political discrimination. 9. Stop athletic corruption. 10. Crack down on institutional corruption. These were suggested as strategies and action items that could do much to restore the university as a place of teaching and learning. That a decade has elapsed and many still consider them clallenges speaks to a major need. Donald Kennedy, president emeritus and professor at Stanford University, wrote in 1997 about contemporary university issues and suggests that meaningful reform will not occur until more rigorous standards of responsibility are accepted by administrators and by faculty. He suggested eight major areas for reform, stated in the form of objectives (27): 1. To teach, as the core responsibility of a university whose major products are educated people 2. To mentor advanced students in a one-on-one interactive and ethical manor such as to best develop inquisitive minds 3. To serve the university through active participation in governance and outreach efforts 4. To discover new knowledge through research in a manner that deals both with sponsors of research as well as with questions of authorship and credit in a responsible and ethical manner 5. To publish in a manner appropriate for the academic area of the scholar and to provide appropriate and proportional credit to the authors of the work in question 6. To tell the truth and avoid all forms of impropriety and misconduct 7. To reach beyond the walls of the institution to accomplish technology-transfer and other service efforts while avoiding conflicts of interests 8. To change in such a manner as to reengineer the university and its faculty to make education continually meaningful, to make appropriate use of new technologies for instruction, and to ensure quality management of the educational enterprise While each of these were here addressed to the university in a general context, they have obvious applicability to engineering programs, including electrical engineering programs.

In an insightful article (28), it was noted that a 1994 National Education Goals Report indicated that fewer than half of adults in the US have the literacy skills to compete successfully in the global economy or to exercise the rights and responsibilities of citizenship. The question raised was how can the United States have the finest college-level education system in the world and at the same time have a K–12 (kindergarten through twelfth grade) system that is often mediocre or worse? Three major Improvements were suggested: 1. The needed improvements are comprehensive ones that address all parts of the education system, from public policies to classroom practices. 2. The hoped-for improvement must start with the development of challenging, rigorous standards. This should be coupled with a system of testing that provides measures of success and which also suggests the path to improvement and how students, educators, and the community can be made accountable for fulfilling these needs. 3. Organizations outside of the schools need to provide real-world incentives for students to work hard and do well in class. Four “waves” of organizational involvement in school reform were cited: 1. Individual school partnerships, adopt-a-school programs, and similar stand-alone efforts 2. Transfer of such management principles as total quality management to schools 3. School choice and higher academic standards seen as quick-fix silver bullets 4. Abandoning ad hoc programs and addressing interrelated educational conditions, from public policies to classroom practices It was felt that the fourth wave, the most recent, offered the maximum possibilities for true and lasting success. Concerns have been expressed with efforts that relate to engineering education and science education as well. In a 1996 report of the National Research Council (29) it was encouraged that all students be provided with “opportunities for access to supportive, excellent undergraduate education in science, mathematics, engineering, and technology, and that all students learn such subjects by direct experience with the methods and processes of inquiry.” The report specifically recommended that all US undergraduates attain a higher level of competence in science, mathematics, engineering, and technology. Universities were encouraged to provide these opportunities for all students, and not just for those seeking an education in engineering or science. Important recommendations were produced for faculty and for administrative units. Faculty members were encouraged to believe and affirm that every student can learn, and to emulate good practices that increase student learning. They were encouraged to have high expectations, to provide a supportive environment that encouraged inquiry, and to stress the excitement of

Electrical Engineering Education

7

discovery, communication, and teamwork, critical thinking, and the development of life-long learning skills. Administrative units and governing boards were encouraged to set goals and accept responsibility for undergraduate learning, to use technology to enhance learning and encourage an active learning environment, to reexamine institutional missions in terms of undergraduate needs, to develop reward systems that stress the importance of science and engineering education for all students, and to provide strong programs for faculty development and to reward faculty who demonstrably facilitate student learning. Accrediting agencies were encouraged to fuse principles of sound undergraduate education into accreditation criteria, and to focus on student learning and not just on organizational and process issues. Comments on the changing environment for engineering and engineering education are commonplace, and issues such as the following seven are often cited (30) today, just as in 1997 when this writing was first published:

undergraduates and to recommend ways to improve undergraduate education in SME&T. Four of the essential conclusions of this study were as follows:

1. Availability of many new engineered materials, and an associated much larger “design space” from which the engineer must choose 2. Pervasive use of information technology in the products and process of engineering 3. Increasing number and complexity of constraints on acceptable engineering solutions—where cost and functionality were once the dominant concerns, ecological and natural resource concerns, sustainability, safety, and reliability and maintainability are now also major concerns 4. Globalization of industry and the associated shift from a nationally differentiated engineering enterprise to one that is far more global 5. Major increases in the technical depth needed in manufacturing and service sectors, both in terms of absolute specific technical knowledge and breadth of knowledge needed 6. Expanded role of the engineer as part of integrated product and process teams, and the broad business knowledge required 7. Increased pace of change in which there appears to be less time to assimilate and adapt

These conclusions were strongly influenced by the reality that, in an increasingly technical and competitive world, with information and knowledge as a major determinant of competitive advantage, the lack of a properly educated citizenry places a society at significant risk. In one notable and particularly relevant 1994 work (34), relevant, attractive, and connected engineering education was outlined as education that results from engineering programs that undertake several important action items, eight of which are listed below:

Each of these, individually and especially in combination, has lead to many new challenges for engineering education. On the basis of concerns such as these, there have been many proposals for reshaping graduate education in science and engineering (31), and designing an adaptive system for engineering education (32). There have been a number of other recent studies of engineering education, in large part prompted by the challenges and needs denoted earlier. Under the auspices of the Education and Human Resources (EHR) Directorate of the National Science Foundation, a committee of the Advisory Committee to EHR has conducted an intensive review of the state of undergraduate education in science, mathematics, engineering, and technology (SME&T) in the United States (33). The purpose of this 1995 review was to consider the needs of all

1. All students should have access to supportive, excellent undergraduate education in science, mathematics, engineering, and technology. 2. All students should learn these subjects by direct experience with the methods and processes of inquiry. 3. All undergraduates in the United States must attain a higher level of competence in science, mathematics, engineering, and technology. 4. Students should learn not only facts concerning science, but also the methods and processes of research, what scientists and engineers do, how to make informed judgments about technical matters, and how to communicate and work in teams to solve complex problems.

1. Establish individual missions for engineering colleges, such that an effective planning process that enacts a clear vision supportive of excellence drives each program. 2. Reexamine faculty rewards, such as to identify incentives that assure commitment and which support the programmatic mission. 3. Reshape the curriculum to enable relevance, attractiveness, and connectivity. 4. Ensure life-long learning of all, supported in part by new and innovative technologies for education (35). 5. Broaden educational responsibility, such that engineering programs provide support for elementary and secondary education. 6. Accomplish personnel exchanges, such that faculty are able to obtain relevant experience in industry and government, and such that industry and government experience are able to contribute their talents to programs in engineering education. 7. Establish across the campus outreach, such that highquality and relevant courses in engineering are made available throughout the university. 8. Encourage research/resource sharing, open competition based on peer review, and enhanced technology transfer. The attributes associated with reshaping the curriculum are of special importance, in that these are directly focused on educating students for careers as professional en-

8

Electrical Engineering Education

gineers, for research, for planning and marketing, and for the many other functions performed by engineers. The major ingredients associated with reshaping the curriculum were suggested as:














Team skills, and collaborative, active learning Communication skills Systems perspective Understanding and appreciation of diversity Appreciation of different cultures and business practices, and understanding that engineering practice is now global Integration of knowledge throughout the curriculum Multidisciplinary perspective Commitment to quality, timeliness, and continuous improvement Undergraduate research and engineering work experience Understanding of social, economic, and environmental impact of engineering decisions Ethics

Sections of this important report are addressed to each of these important ingredients. That attention still needs to be paid to these desiderata is emphasized in a 1998 writing by the president of the US National Academy of Engineering that discusses the stillcompelling urgency of engineering reform (36) relative to such important issues as:


The need for much more design synthesis in the curricula—under economic, quality, integration, and other constraints—as contrasted with the current focus on scientific-based analysis
The need for focus on the rapidly increasing role of new innovations in such areas as information technology and biotechnology
The need for all, including the “liberally” educated, to be technologically literate
The need to seriously examine whether a realistic 120semester-hour bachelors degree in engineering can be considered as a first professional degree Other relevant works examine the role of technology and values in contemporary society (37). Still others stress the need for engineering to become more integrated with societal and humanistic concerns, such as to enable engineers to better cope with issues and questions of economic growth and development, and associated concerns regarding sustainability and the environment (38). These issues are continually being addressed in the international engineering education community. Two recent reports (49, 50) describe the situation facing engineering and engineering education in the year 2020. The first of these studies (49) suggests that the engineering profession should take the initiative in defining its future. However, to do this successfully, the report presents cogent arguments that the profession must

1. Come to agreement on a vision for its future; 2. Transform engineering education to help achieve this vision; 3. Establish a clear image of the resulting new roles for engineers, including becoming broad scope technology leaders and establishing this image in the perceptions of the public and prospective students; 4. Accommodate innovative developments into engineering that arise initially in non-engineering areas; and 5. Find approaches that will focus the energies of the different disciplines of engineering toward common agreed upon objectives that will ensure world sustainability and progress. While the results of this study indicated that there is no consensus on strategies and tactics at this time, it was agreed that innovation is the key driver and that engineering is essential to enabling innovation. However, it was stressed that engineering will only be able to contribute to success if it is able to continue to adapt to emerging new trends and to educate the next generation of students by providing them with knowledge principles, practices and perspectives needed for the world as it will be tomorrow, and not as it is today. In the second report (50), specifically on engineering education, a dedicated effort was made to answer the question, “What will or should engineering education be like today, or in the near future, to prepare the next generation of students for effective engagement in the engineering profession in 2020?” The report is very concerned with identifying approaches to enrich and broaden engineering education so that the products of the educational process will be better prepared to work in today’s constantly changing economy. The report does discuss education after the baccalaureate degree; however, its major focus is on undergraduate education, and not graduate level education or academic research. There were fourteen major recommendations of this study relative to reengineering engineering education. 1. The baccalaureate degree should be recognized as the “pre-engineering” degree or “bachelor of arts” in engineering degree. 2. Engineering schools should create accredited “professional” masters degree programs intended to expand and improve the skills and enhance the ability of engineers to practice engineering. In support of this, the Accreditation Board for Engineering and Technology (ABET) should change their present policy and allow accreditation of engineering programs of the same name at both the baccalaureate and graduate levels in the same department. 3. Engineering schools should exploit flexibilities inherent in the outcomes-based accreditation approach of ABET to experiment with novel models for engineering education. ABET should ensure that evaluators look for innovation and experimentation in curricula and not hold to a strict interpretation of the present guidelines as perceived by individual eval-

Electrical Engineering Education

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

uators. In this way, each college and university is allowed and encouraged to develop their own plans and programs that best suit their stakeholders and then be evaluated on whether the plans are efficacious and whether the desired outcomes of this planning are achieved. The iterative process of designing, predicting performance, building and testing–should be taught from the earliest stages of the curriculum, including the first year. This supports the emergent, evolutionary, and adaptive nature of an engineering education system of systems, as noted earlier. It also encourages a broad interpretation of these iterative process activities to include early attention to associated educational benefits analysis and assessment. The engineering education establishment, potentially through the Engineering Deans Council, should endorse research in engineering education as a valued activity for engineering faculty. Colleges and universities should develop new faculty qualification standards such as, for example, to require experience as a practicing engineer. They should adapt their faculty development programs to support professional growth of engineering faculty. Engineering schools must teach engineering students how to learn, and work along with professional organizations in facilitating life long learning. Engineering schools should introduce interdisciplinary learning in undergraduate programs, rather than only having it as a possible feature of graduate programs. Engineering educators should explore the development and use of case studies of engineering successes and failures and should encourage appropriate use of case studies in undergraduate and graduate curricula. In this connection, we note current INCOSE efforts to develop systems engineering case studies. Four-year engineering schools should work with local community colleges to assure effective articulation, in as seamless as possible a manner, with 2-year community college programs. Graduate students from all over the world have flocked to the U.S. for years to take advantage of the excellent graduate education available. At the same time, they should not to neglect domestic students. Thus, U.S. engineering schools must develop programs to encourage/reward domestic engineering students to aspire to the MS and/or Ph.D. degree. Engineering schools should support national efforts at improving math, science and engineering education at the K-12 level. The engineering education profession should participate in coordinated national efforts to promote public understanding of engineering through public technology literacy. The National Science Foundation (NSF) should collect comprehensive data concerning engineering department/school program philosophy and student outcomes, such as student retention rates by gen-

9

der and ethnicity, percent of entering freshman that graduate, time to degree, and information on jobs and admission to graduate school. The purpose of this would be to provide marketplace information, knowledge, and understanding across programs. While this particular report is devoted in large part to USA based engineering education, analogous statements can generally be made relative to international engineering education, including education in electrical engineering. OBJECTIVES FOR HIGHER EDUCATION AND ENGINEERING EDUCATION Engineering education is a professional activity and an intellectual activity. It is necessary that the faculty responsible for this educational delivery in engineering remain at the cutting edge of relevant technologies, including emerging technologies, as technology does change rapidly over time. Research is, therefore, an absolute essential in engineering education. It is possible through relevant research, and associated knowledge principles, to develop new engineering knowledge principles and practices that are relevant to societal improvements that result from better use of information and technological innovations. Research is exceptionally important for engineering education, as it is strongly supportive of the primary educational objective of the university. It is vital to remain vigilant relative to the educational mission and, again, this requires that faculty remain at the cutting edge of technology in order that they be able to provide education, meaning teaching and meaningful learning by students, at that forefront. It is because of the need to remain current in the classroom in order to deliver education for professional practice that the strong need and a mandate for faculty research in engineering necessarily emerges. This suggests that research activities in engineering education should generally be very student oriented. It suggests that students are an inseparable and integral part of faculty research. It suggests a major role for students in development and cooperative/internship ventures with industry and government. This creates the strong need for sponsored research and internships that assure the needed industry–government–university interactions. In addition to being intimately associated with the educational process, sponsored research also provides faculty with released time from exclusively teaching efforts for and enables them to engage in scholarly pursuits necessary to retain their currency in the classroom. Also needed is the innovative effort that transfers research in emerging technologies to engineering practice with potential marketplace success. These each have a mutually enhancing and beneficial effect when properly and ethically associated with the educational function, and when this research is attuned to the needs associated with knowledge integration for professional practice. The knowledge and skills required in engineering, and in engineering education, come from all of the sciences, and from the world of professional practice. This suggests

10

Electrical Engineering Education

that faculty in a professional school of engineering need to keep abreast of progress in relevant sciences, both the natural sciences and the economic and social sciences, and the mathematical and engineering sciences. Taken together, these comprise knowledge principles. It suggests also that engineering educators must keep abreast of and contribute to industrial practices in relevant professional practice areas. It is for this reason that engineering schools are and must remain professional schools. This is also why close industry–university and government–university interactions, become a most desirable and, in fact, essential part of successful, high-quality engineering education programs. Efforts in engineering must necessarily involve likely future technological developments as well, if the customers for electrical engineering education are to be satisfied. Thus, there is a need for knowledge practices, knowledge principles, and knowledge perspectives in engineering education. These knowledge components, and the necessary learning to enable transition and natural evolution of one form of knowledge into the other, are very important for both technology transfer and for engineering education. This leads to success attributes for faculty in electrical engineering education that include quality teaching, quality scholarship and sponsored research, and quality professional service. This suggests that faculty should be productive in terms of high-quality teaching, research, and professional service. Also it suggests that they should have a quality culture and orientation that indicates continued productivity and trustworthiness over time relative to performing one of societies more important functions—that of education. Thus, the five Cs—competency, commitment, communications, collaboration, and courage—are very important success attributes for faculty. As noted in a previous section, there has been much recent concern that teaching has become a neglected ingredient in these educational performance attributes, rather than the dominant one that it should be. This is the major theme of a noteworthy work by Boyer (39), in which the author suggests four dimensions of a reconsidered and redefined scholarship. These are:

1. The scholarship of knowledge discovery, which is the classic form of research in one well-established and defined discipline 2. The scholarship of knowledge integration, which is interdisciplinary or cross-disciplinary or multidisciplinary or transdisciplinary scholarship and research that involves the blending and infusion, or transfer or integration, of knowledge across several disciplinary areas 3. The scholarship of knowledge application, which is the utilization of knowledge to support such worthy endeavors as policy analysis, program evaluation, and professional service 4. The scholarship of teaching, which involves the communication of one’s knowledge to others, generally in the classroom, perhaps an extended- or distancelearning classroom.

No responsible educator could deny the importance of these four facets, nor should there be disagreement that the fourth facet listed by Boyer, the scholarship of teaching, should really be the first-priority facet for education. These four forms of scholarship are not at all dramatically different from the mix that results from consideration of the interaction of knowledge principles, practices, and perspectives across the three traditional university faculty efforts of (1) teaching, (2) research and scholarship, and (3) public and professional service. Nevertheless, the notion of each of these four as important is vital, as well as the uncommon assertion that teaching and application are each forms of appropriate scholarship for university faculty. What Boyer denotes as scholarship of discovery appears equivalent to research and breakthroughs relative to knowledge principles. Discovery can, of course, relate to knowledge perspectives as well. Discovery can surely involve only a single established area of inquiry or, as is much more likely to be the case when knowledge perspectives about future developments are concerned, inter-, cross-, multidisciplinary efforts. Thus, research into and knowledge of integration, or the scholarship of knowledge integration, appears to be much a blend of knowledge principles and knowledge perspectives and the transition of these into knowledge practices. The scholarship of knowledge application would, in a similar way, seem quite equivalent to knowledge practices. Teaching is, of course, the primary activity involved in communicating knowledge practices, knowledge principles, and knowledge perspectives to others. It is often suggested that much academic research is too narrowly focused to truly support the educational function, which needs to be more broadly based and integrative. Doubtless this assertion is correct. What this suggests however, is not neglect of research and knowledge principles, but a renewed focus on the large-scale and broad-scope facets of integrative knowledge-based efforts. This suggests a simultaneous broad-narrow perspective on knowledge acquisition and management efforts, especially as they relate to engineering practice and engineering education for professional practice. The real thrust of the Boyer message is indeed critical: Teaching is important, even and especially including undergraduate teaching. It has been neglected in some sectors of higher education. Its importance clearly needs to be reconsidered and reestablished. The faculty reward structure must recognize, now and for all time to come, teaching as a very important ingredient in the performance of an educator. Whether or not teaching is a form of scholarship is thus completely irrelevant to the central and much needed message in this important work. A 1997 work, denoted as an Ernest L. Boyer project of the Carnegie Foundation for the Advancement of Teaching (40), set forth six standards that can be used to develop metrics to assess the four scholarship divisions:

1. Clear goals for scholarly work 2. Adequate preparation in terms of understanding existing scholarship and having appropriate resources to enable progress

Electrical Engineering Education

3. Appropriate methods that have been properly adapted to enable success 4. Significant results that significantly impact the scholarship division in question 5. Effective presentation and communication of the results of the scholarly effort 6. Reflective critical personal evaluation of the work by the scholar performing it The three essential characteristics of a scholar are represented to be (1) integrity, (2) perseverance, and (3) courage. To these, one might add competence, commitment, communications, collaboration, and much tolerance and humility. There are a number of economic and accounting issues that need only be mentioned briefly here. There are faculty earnings necessarily associated with student credit-hour production, and charges for and earnings from sponsored program activities. These are essential as an educational enterprise must support itself. These represent the earnings of the institution from the academic enterprise. Two other important financial measures are (1) billings, or the accounts to which university personnel charge time, and (2) activities, or the productive efforts of the faculty. While generally associated directly with earnings and billings, there is not necessarily a one-to-one correspondence among these three entities: (1) earnings, (2) billings, and (3) activities. Quality educational enterprise management must necessarily be concerned with all three of these. This article will not be especially concerned with these issues, important as they are for efficient and effective operational management of the enterprise supporting engineering education. STUDENT EDUCATIONAL NEEDS, OUTCOMES, AND METRICS–AND PROGRAM ACCREDITATION Notions of what comprises an appropriate education are not at all new. Many have written, and for a very long time, about educational requirements and the characteristics of educated people. In 1852, for example, John Henry Cardinal Newman identified ten distinctive traits of university education (41). An interpretation of these is as follows: 1. It encourages one to identify and place values upon personally held views and judgments. 2. It encourages search for the truth. 3. It encourages eloquence in expressing truths. 4. It encourages clarity and integrity in observing and valuing judgments. 5. It encourages coherency of expression and communication. 6. It encourages one to solve problems through critical thinking skills and by analyzing courses of action, such as to be able to retain meaningful alternates and to discard spurious ones. 7. It encourages cultural and cross-cultural understanding. 8. It encourages compassion, forbearance, and comprehension of the views of others.

11

9. It encourages basic feelings for and understandings of events in both a social and a historical context. 10. It encourages preparation for work and a lifetime of continued learning.

These are as relevant today as they were when initially written in a somewhat different style, one-and-a-half centuries ago. Five study areas: (1) humanities, (2) social and behavioral sciences, (3) natural sciences, (4) mathematical sciences, and (5) engineering sciences are absolutely essential for provision of the general education background and perspective needed for professional practice in engineering, and for development of abilities to use knowledge principles and to successfully convert them to knowledge practices as well. Any professional educational program should, unless it is to ultimately suffer any of several impediments to its integrity and trustworthiness, become accredited by the appropriate accrediting agency. Accreditation has, for a very long time, been recognized as a mechanism for quality assessment of educational programs. It has had an especially significant history with respect to professional programs, such as engineering and computer science. Programs that have engineering in their title are potentially subject to accreditation by the Accreditation Board for Engineering and Technology (ABET) which, prior to 1980, was called the Engineers’ Council for Professional Development (ECPD). ABET is a federation of some 28 engineering professional societies and is recognized by the US Department of Education (USDoE) and the Council on Postsecondary Accreditation (COPA) as the sole agency responsible for accreditation of all educational programs leading to engineering degrees in the United States. ABET does not currently accredit programs outside of the United States unless these are at an institute that has intimate educational connections with a US university. ABET will evaluate programs outside the U.S., by institutional request, to determine if they are “substantially equivalent” to ABET-accredited programs and to make recommendations for program improvement. The initial accreditation planning for, and implementation of, computer science accreditation was accomplished through ABET, and the Institute of Electrical and Electronics Engineers (IEEE) within ABET. This became a separate but related accreditation effort, for computer science but not for computer engineering, within the Computer Science Accreditation Commission (CSAC) and Comptter science Accreditation Board (CSAB). Currently, CSAB serves as a participating body of ABET with two members on the ABET Board of Directors. It is the lead society within ABET for accreditation of programs in computer science, information systems, and software engineering, and is a cooperating society for accreditation of computer engineering. Accreditation activities once belonging to the Computer Science Accreditation Commission (CSAC) are now conducted by the Computing Accreditation Commission (CAC) of ABET with program accreditation responsibilities in computer science and information systems. The Engineering Accreditation Commission (EAC) is responsible for the accreditation of pro-

12

Electrical Engineering Education

grams in software engineering and computer engineering. Current member societies of CSAB are the Association for Computing Machinery, Inc. (ACM), the Institute of Electrical and Electronics Engineers, Inc.–Computer Society (IEEE-CS), and the Association for Information Systems (AIS). The objective of accreditation is to determine that an educational program meets minimum quality standards. It is not, in any sense, a warrant or guarantee of high quality, regardless of the multiattributed approach one might take to quality definition. A standard ABET definition of engineering is that (42) “Engineering is that profession in which knowledge of the mathematical and natural sciences gained by study, experience, and practice is applied with judgment to develop ways to utilize, economically, the materials and forces of nature for the benefit of mankind. A significant measure of an engineering education is the degree to which it has prepared the graduate to pursue a productive engineering career that is characterized by continued professional growth.” It may appear at first glance that there is not much of a focus on the humanities and social science components of general education in this definition. However, the ABET annual reports proceed to delineate outcomes of an engineering education in a very meaningful manner that strongly include this focus. These five outcomes, clearly valuable for outcome assessment purposes, are as follows: 1. An engineer should have the ability to formulate and solve, in a practical way, those problems of society that are amenable to engineering solution. 2. An engineer should have a sensitivity to those socially relevant technical problems that confront the engineering profession. 3. An engineer should have an understanding of the ethical characteristics of the engineering profession, and professional practice. 4. An engineer should have an understanding of the engineer’s responsibility to protect both occupational and public health and safety. 5. An engineer should have the ability to maintain professional competency through a lifetime of learning. Thus, the ABET criteria are generally cognizant of the ingredients of a sound general education. It is interesting that the word design does not appear in either the definition or the listing or desirable outcomes, nor does the ubiquitous role of the computer and information technology. Design abilities are clearly needed, however, to achieve the five outcome objectives and are, therefore, implicitly included. Moreover, this is a very critical focus for engineering education and ABET accreditation effort. A major contemporary ABET trend is certainly that of providing an enhanced focus on engineering design for innovation (43). To become accredited, ABET indicates that engineering programs must demonstrate that students in these programs attain abilities to: 1. apply a knowledge of mathematics, science, and engineering

2. design and conduct experiments, and analyze and interpret data 3. design a system, component, or process to meet desired needs within appropriate constraints 4. function on multi-disciplinary teams 5. identify, formulate, and solve engineering problems 6. use the techniques, skills, and modern engineering tools necessary for engineering practice 7. communicate effectively and 8. engage in life-long learning. In addition, ABET suggests that students should have an understanding of professional and ethical responsibilities, a broad education necessary to understand the impact of engineering solutions in a global, economic, environmental, and societal context; and a knowledge of contemporary issues associated with these such as to be professionally competent. Further, it indicates that accreditable programs must include: a. one year of a combination of college level mathematics and basic sciences (some with experimental experience) that is appropriate to the major discipline studied b. one and one-half years of engineering topics, consisting of engineering sciences and engineering design that is appropriate to the field of study. There is a specific focus on engineering design-as a process of devising a system, component, or process to meet desired needs–here. a general education component that complements the technical content of the curriculum and which is also consistent with the program and institution objectives. All of the program is expected to provide depth and balance. Faculty quality, institutional commitment to the program, laboratory facilities, library faculties, and other concerns are also evaluated. Almost all development of curricula for undergraduate programs in engineering in the United States is based on these criteria. These lead to a more or less standard electrical engineering curriculum that is relatively ubiquitous in terms of specific required courses across electrical engineering programs in this country. As of early 2006, for the 2005 accreditation year, ABET has accredited 2,700 programs at more than 550 colleges and universities nationwide in applied science, computing, engineering, and technology education, Some 28 professional societies are ABET members in early 2006. For the 2005 accreditation year, there were a total of 1,759 ABET basiclevel, or undergraduate, accredited programs of engineering, and 23 graduate programs, at 357 institutions in the United States. Of these, the largest number, 299, are undergraduate programs in electrical (and electronics) engineering. These are 3 accredited graduate level programs. Computer engineering had 180 accredited undergraduate programs and 2 accredited MS programs. The very small number of accredited graduate level programs is not unusual

Electrical Engineering Education

when it is realized that a department with both a graduate and undergraduate program would naturally much prefer to seek undergraduate-level accreditation, and ABET will presently accredit a program at only one level. This makes separate graduate-level accreditation proscribed, as few institutions would willingly not have their undergraduate program denoted as being accredited by ABET. ABET has undertaken a number of initiatives, such as Engineering Criteria 2000 (44), which is composed of eight criteria that are intended to emphasize quality and preparation for professional practice, to enhance engineering educational efforts. For example, The criteria retain the traditional core of engineering, math, and science requirements. However, the work also places importance on formal efforts that stress teamwork, communications, and collaboration, as well as global, economic, social, and environmental awareness. There was also an outcomes-assessment component requiring each engineering program seeking accreditation or renewed accreditation to establish their own internal assessment process. The criteria associated with the basic level accreditation in Engineering Criteria 2000 address several basic areas of concern in engineering education (44): students, program educational objectives, program outcomes and assessment, professional components, faculty, facilities, institutional support and financial resources, program criteria, and cooperative education criteria. Thus it can be seen that the ABET accreditation process is a voluntary system of accreditation that assures that graduates are prepared to enter and continue the practice of engineering. It also stimulates improvements in engineering education, encourages new and innovative approaches to engineering education, and identifies these programs to the public. Each degree program has specific requirements that are to be satisfied in addition to the general criteria just stated. The program criteria for electrical, computer, and similarly named engineering programs were submitted by the Institute of Electrical and Electronics Engineers, which is the responsible participating professional society. They apply to engineering programs that include electrical, electronic, computer, or similar modifiers in their titles. The structure of the curriculum must provide both breadth and depth across the range of engineering topics implied by the title of the program. Also, graduates must have demonstrated a knowledge of probability and statistics, including applications appropriate to the program name and objectives. In addition, graduates must have a knowledge of mathematics through differential and integral calculus, basic sciences, and engineering sciences as necessary to analyze and design complex devices and systems containing hardware and software components, again as appropriate to the objectives of the program name. The curricular specifications indicate that graduates of programs that contain “electrical” in their title must have demonstrated a knowledge of advanced mathematics, typically including differential equations, linear algebra, complex variables, and discrete mathematics, and that graduates of programs that contain “computer” in the title must have demonstrated a knowledge of discrete mathematics.

13

These accreditation criteria are based on the premises that


Technology has been a driver of many of the changes occurring in society over the last several years


It will take on an even larger role in the future
The engineering education accreditation process must promote innovation and continuous improvement to enable institutions to prepare professional engineers for exciting future opportunities These criteria are focused on ensuring competence, commitment, communications, collaboration, and the courage needed for individual responsibility and integrity. These, augmenting the usual listing of competence and assumption of individual responsibility as the two traditionally accepted key characteristics of a professional, might be accepted as the new augmented attributes of a mature professional. They should truly support the definition, development, and deployment of relevant, attractive and connected (quality) engineering education that will:


Include the necessary foundations for knowledge principles, practices, and perspectives


Integrate these fundamentals well through meaning








ful design, problem-solving, and decision-making efforts Be sufficiently practice oriented to prepare students for entry into professional practice Emphasize teamwork and communications, as well as individual efforts Incorporate social, cultural, ethical, and equity issues, and a sense of economic and organizational realities— and a sense of globalization of engineering efforts Instill an appreciation of the values of personal responsibility for individual and group stewardship of the natural, technoeconomic, and cultural environment Instill a knowledge of how to learn, and a desire to learn, and to adapt to changing societal needs over a successful professional career

The three classic steps leading to the designation of a person as a “professional” are: 1. A comprehensive and appropriate education that enables mastery of a body of specialized and relevant knowledge 2. A period of apprenticeship followed by a title or license to enable at least restricted professional practice in the area of specialization, according to a code of ethics 3. A professional organization with the power to regulate itself and impose standards of conduct and sanctions against incompetent or unethical persons or groups who would practice the profession These three classic characteristics of a professional lead to several functional characteristics of professionals. Engi-

14

Electrical Engineering Education

neering is a professional activity, and standards and practices in engineering education must conform to these notions. Thus the content in this section becomes an inherent part of the objectives for an engineering school. They lead to such important activities as curriculum design. Additionally, they imply a subset of criteria for evaluation and outcomes assessment of programs and they become critical success factors for students in these programs. Importantly also, they bring into focus the need for a balance among studies involving knowledge practices, principles, and perspectives, and for the need to ensure that graduates are capable at knowledge integration as well as understanding of subject matter in depth.

BIBLIOGRAPHY

SUMMARY

1. J. E. Brittain (ed.), Turning Points in American Electrical History, New York: IEEE Press, 1977. 2. T. P. Hughes, American Genesis: A Century of Invention and Technological Enthusiasm, New York: Viking Penguin, 1989. 3. J. R. Beniger, The Control Revolution: Technological and Economic Origins of the Information Society, Cambridge, MA: Harvard Univ. Press, 1986. 4. D. Tapscott, Growing Up Digitally: The Rise of the Net Generation, New York: McGraw-Hill, 1998. 5. F. Cairncross, The Death of Distance: How the Communications Revolution Will Change Our Lives, Boston: Harvard Business School Press, 1997. 6. S. Davis, C. Myer, Blur: The Speed of Change in the Connected Economy, Reading, MA: Addison-Wesley, 1998. 7. R. Kling (ed.), Computerization and Controversy: Value Conflicts and Social Choices, San Diego: Academic Press, 1996.

We have presented a wide-scope discussion of engineering education. We have discussed the emergence of electrical engineering as a discipline, the evolution of electrical engineering education as preparation for professional practice as well as for the development of knowledge principles through research, contemporary concerns relative to educational quality and responses to these, and educational needs and accreditation standards for the 21st century. A flowchart of interactions of electrical engineering education would show a very large number of linkages across many related elements, thereby indicating that engineering education itself is a system of large scale and scope. Our discussion is necessarily wide scope in that electrical engineering education itself is necessarily wide scope. An electrical engineer must surely understand the principles of the natural and mathematical sciences. They must have this understanding in order to know how to use these sciences to produce cost-effective electrical systems and also to have the background necessary to retain intellectual currency throughout a lifetime of continued learning. The purpose behind the engineering of electrical systems is the development of products that are successful in the marketplace through fulfillment of societal needs. Technological, organizational, and societal change are the order of the day, just as they have been throughout history. If these changes are to be truly effective, over the long term especially, they must serve societal needs. This suggests that change needs necessarily to be guided by principles of social equity and justice as well as by concerns for sustainable development and marketplace competition. There is strong evidence that this needed guidance does not always occur and that the hoped for productivity gains from technological advances may be elusive (45–47). This evidence provides the mandate for a major component of the social and behavioral sciences, and the political and policy sciences, in engineering education and in engineering practice as necessary ingredients for success. It also provides a mandate for major integrative knowledge components in engineering education and for educational accreditation standards that reflect these needs, as recognized in the reengineering efforts for education and engineering education suggested by a large number of the sources cited here.

8. D. A. Klein, The Strategic Management of Intellectual Capital, Woburn, MA: Butterworth-Heinemann, 1998. 9. M. Detrouzos,What Will Be: How the New World of Information Will Change Our Lives, New York: HarperCollins, 1997. 10. W. J. Bennett, The Devaluing of America: The Fight for Our Culture and Our Children, New York: Summit Books, 1992. 11. D. D’Souza, Illiberal Education: The Politics of Race and Sex on Campus, New York: Free Press, 1991. 12. T. Sowell, Inside American Education: The Decline, The Deception, The Dogmas, New York: Free Press, 1993. 13. C. J. Sykes, ProfScam: Professors and the Demise of Higher Education, New York: Regnery Gateway, 1988. 14. U.S. Department of Education,National Commission on Excellence in Education, A Nation at Risk, Washington, DC: Government Printing Office, 1983. 15. P. Bigler, K. Lockard, Failing Grades, Arlington, VA: Vandamere Press, 1992. 16. A. M. Schlesinger, Jr.,The Disuniting of America: Reflections on a Multicultural Society, Knoxville, TN: Whittle Direct Books, 1991. 17. S. L. Carter, Reflections of an Affirmative Action Baby, New York: Basic Books, 1991. 18. M. T. Jacobs, Short Term America; The Causes and Cures of Our Business Myopia, Boston: Harvard Business School Press, 1991. 19. R. L. Harmon, L. D. Peterson, Reinventing the Factory: Productivity Breakthroughs in Manufacturing Today, New York: Free Press, 1990. 20. D.H. Brandin, M. A. Harrison, The Technology War: A Case for Competitiveness, New York: Wiley, 1987. 21. M. L. Dertrouzos, R. K. Lester, R. M. Solow, Made in America: Regaining the Productive Edge, Cambridge, MA: MIT Press, 1989. 22. National Research Council, Research on the Management of Technology: Unleashing the Hidden Competitive Advantage, Washington, DC: National Academy Press, 1991. 23. A. P. Sage, Systems engineering and information technology: Catalysts for total quality in industry and education, IEEE Trans. Syst. Man Cybern., 22: 833–864, 1992. 24. W. B. Rouse and A. P. Sage, Information Technology and Knowledge Management, in A. P. Sage and W. B. Rouse, Handbook of Systems Engineering and Management, New York: Wiley, 1999.

Electrical Engineering Education 25. C. S. Small and A. P. Sage,“ Knowledge Management and Knowledge Sharing: A Review” Information, Knowledge, and Systems Management, Vol.X, No. Y, 2006, pp. –. 26. R. K. Lester, The Productive Edge: How U.S. Industries Are Pointing the Way to a New Era of Economic Growth, New York: Norton, 1998. 27. M. Anderson, Imposters in the Temple: A Blueprint for Improving Higher Education in America, Stanford, CA: Hoover Inst. Press, Stanford Univ., 1996. 28. D. Kennedy, Academic Duty, Cambridge, MA: Harvard Univ. Press, 1997. 29. N. R. Augustine, A new business agenda for improving U.S. schools, Issues Sci. Technol., 27 (1): 1997. 30. From Analysis to Action: Undergraduate Education in Science, Mathematics, Engineering, and Technology, Washigton, DC: National Academy Press, 1996. 31. W. A. Wulf, The changing nature of engineering, Bridge, 27 (2): 1997. 32. P. A. Griffiths, Reshaping graduate education, Issues Sci. Technol., 11 (4): 74–79, Summer 1995. 33. Engineering Education: Designing an Adaptive System, Washington, DC: National Academy Press, 1995. 34. L. S. Williams, Charge to the Subcommittee for Review of Undergraduate Education in Science, Mathematics, Engineering, and Technology, Advisory Committee to the Directorate for Education and Human Resources, National Science Foundation, June 1995. 35. Engineering Education for a Changing World, American Society for Engineering Education, 1994. 36. A. Bourne, A. Broderson, M. Dawant (eds.), The Influence of Information Technology on Engineering Education, Boca Raton, FL: CRC Press, 1995. 37. W. A. Wulf, The urgency of engineering education reform, Bridge, 28 (1): 4–8, 1998. 38. L. A. Zadeh, UC-Berkeley computer science commencement address, IEEE Trans. Syst. Man Cybern., 28A: 7–8, 1998. 39. G. Bugliarrello, Engineering at the crossroads of our species, Bridge, 28 (1): 9–15, 1998. 40. The Engineer of 2020: Visions of Engineering in the New Century, National Academy Press, Washington DC, 2004. 41. Educating the Engineer of 2020: Adapting Engineering Education to the New Century, National Academy Press, Washington, DC, 2005. 42. E. L. Boyer, Scholarship Reconsidered: Priorities of the Professorate, Princeton, NJ: Carnegie Foundation for the Advancement of Teaching, 1990. 43. C. E. Hlassick, M. T. Huber, G. I. Maeroff, Scholarship Assessed: Evaluation of The Professorate, San Francisco: Jossey Bass, 1997. 44. J. H. Cardinal Newman, The Idea of a University, Westminster, MD: Christian Classics, 1973, pp. 177–178. 45. ABET Annual Report, Available at the ABET URL http://www.abet.org/. Accessed 26 July 2006. 46. D. Christiansen, Accrediting novelty, IEEE Spectrum, 29 (8): 17, 1992. 47. Engineering Criteria 2000, Accreditation Board for Engineering and Technology, New York: ABET January 1998. 48. R. H. McGuckin, K. J. Stiroh, Computers can accelerate productivity growth, Issues Sci. Technol., 14 (4): 41–48, Summer 1998. 49. S. R. Roach, No productivity boom for workers, Issues Sci. Technol., 14 (4): 49–56, Summer 1998.

15

50. M. Moll (ed.), Tech High: Globalization and the Future of Canadian Education, Halifax, Canada: Fernwood, 1997.

ANDREW P. SAGE George Mason University, Fairfax, VA

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2907.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Intelligent Tutoring Systems Standard Article John Connelly1 and Alan Lesgold1 1University of Pittsburgh, Pittsburgh, PA Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2907 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (168K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2907.htm (1 of 2)17.06.2008 21:43:30

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2907.htm

Abstract The sections in this article are Some Background Systems Categorized by Domain/Skill Discussion | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2907.htm (2 of 2)17.06.2008 21:43:30

INTELLIGENT TUTORING SYSTEMS

529

Such tools have appeared over the last two decades under various labels, including: intelligent computer-assisted instruction (ICAI) systems; intelligent tutoring systems (ITSs); microworlds or discovery worlds; coached apprenticeship systems; reactive learning environments; and, more broadly, intelligent learning environments (ILEs). Different theories of learning or instruction underlie the various systems, from production-system models of individual instruction (7) to theories of cognitive apprenticeship and situated cognition, often involving groups (9). Among the most salient differences between the various pedagogical approaches are the type, amount, timing, and structure of the feedback from the system’s intelligent agent(s) to the user(s). In general, the availability of artificial intelligence tools has permitted a major shift in the nature of computer-based training and education. Prior to the appearance of intelligent tools, computer-based instruction consisted mostly of either pure didactic exposition or simplistic scoring of student answers to verbally posed questions. In recent years, simulations driven by numerical algorithms and using fancy graphics have also appeared (e.g., variations on the SimCity game series). Intelligent (i.e., knowledge-based) systems now support several improvements in computer-based learning: • The computer can more deeply evaluate a student’s performance. • The computer can solve a problem itself and compare its solution to the student’s, offering advice, critique, and modeling of better ways to perform. • The computer can assess a student’s strengths and weaknesses and select learning opportunities (e.g., problems to solve) that best fit the student’s current knowledge state. • The computer can explain why an expert might attack a problem differently.

INTELLIGENT TUTORING SYSTEMS The advent of increasingly powerful and inexpensive computer hardware late in this century has enabled the development (and, in some cases, deployment) of advanced, computer-based instructional software tools based on the principles of artificial intelligence. Unlike the early computerassisted instruction (CAI) systems that preceded them, most of which were offshoots of Skinnerian ideas about programmed instruction (1,2) and contained canned knowledge of experts in a given domain, these newer, intelligent systems aimed to embody the domain expertise itself (3). That is, rather than simply deliver instruction, they aimed to generate instruction (4), tailoring it to individual students’ needs (5). Indeed, many such systems have aimed to use the computer as a kind of ‘‘cognitive microscope’’ (6), revealing the processes occurring in the mind of the human learner. Although such systems are more costly to produce than CAI systems, in time, money, and effort (7), their benefits often outweigh ‘‘the costs of the I in the ITS’’ (8).

Because intelligent computer-based training and education is in a rather early stage of evolution, there are many cases of prototype and demonstration systems but few cases of fielded and practical systems. Of those, few have undergone substantial evaluations. This article surveys a subset of intelligent tools that have undergone some form of evaluation. The list of systems reviewed, while certainly not exhaustive, is representative of the existing tools for which evaluation results were readily available. Overall, these systems provide a good sense of the intelligent learning tools developed to date, and the evaluations we summarize help establish the state of the art in intelligent learning system technology. Some of the questions we consider in this article include the following: To the extent that individual systems have been deemed helpful for teaching the target domain or skills, in what ways do they help? What aspects of the system’s various components help or hinder the user’s learning? To what extent does feedback from the system help? How well does the system foster individual learning versus collaboration among multiple users, when appropriate? Given the range of disciplines and tasks for which ILEs have been built, it is difficult to compare approaches to ILE design and delivery of system feedback without accounting for the types of skills they support. Clancey (10) describes how problem solving operators and inference procedures differ

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright # 1999 John Wiley & Sons, Inc.

530

INTELLIGENT TUTORING SYSTEMS

across various domains. McKendree (11) suggests that more complex or ambiguous tasks may require a greater degree of informative feedback than more constrained ones, for which more directive feedback often suffices. Others in the field have described the process of learning from an ILE as a four-way interaction of learner style, desired knowledge outcome, type of instructional environment, and subject matter (12). Choice of subject matter has also been shown to influence the relative effects of human tutoring (13), and may also affect the results of intelligent system evaluations (6). Thus, our review categorizes systems by subject matter domain, wherever possible. SOME BACKGROUND Tutoring Many of the intelligent learning environments developed to date, most notably the ITSs, focus on the benefits of one-onone tutoring. The reasons for this are simple: studies of human tutoring have shown achievement advantages of up to two standard deviations over traditional classroom instruction (14). While effect sizes vary with the subject matter [e.g., higher for mathematics and lower for reading (13)], human tutoring is superior not only to traditional instruction but also to other classroom improvements upon it such as mastery learning (14). In an effort to reap the same benefits, early ITS developers sought to build systems that mimicked, equaled, or even surpassed human tutors (5,7,15). For example, the CIRCSIM-Tutor system, a combined ITS and simulation of the human circulatory system, has been modeled after the patterns of hinting behaviors exhibited by humans tutoring medical students about cardiovascular physiology (16). However, obtaining sufficient information to draw meaningful comparisons is often difficult because the processes underlying expert human teaching or tutoring are not well understood (17,18). For example, the extent to which expert human tutors do extensive student diagnosis has been widely debated (5,16,19,20). Partly because of this, and partly because of the limited bandwidth in human-computer interaction (21), student diagnoses by successful ILEs also vary in their degree of detail. Nevertheless, the clear benefits of human tutorial guidance have led many system developers to try to emulate characteristics of human tutors. System Components The majority of ILEs we surveyed are complex systems, made up of a number of different components or modules. Most ITSs are comprised of four components: a domain knowledge or expert module, a student model, a tutoring or pedagogical module, and a user interface (22). ILEs may contain other components as well, including a simulation environment (e.g., 23,24); a learning component, for systems that employ machine-learning techniques to improve themselves (2,18); and a control component, to coordinate all the other components (18,23). Although often regarded as discrete units, the various system components are usually interrelated in function and often in features (25). We describe briefly each of the four most common ILE components, those of an ITS. The expert module of an ITS contains the target knowledge of the domain that the system is designed to teach. This

knowledge may be declarative (concepts, system models, etc.) or procedural, or both (20). Ideally, this knowledge has been verified by human domain experts, who either examine the knowledge base or interact with prototypes of the learning system (18). In most cases the expert module consists of a working model capable of solving target problems in the domain (3). Such expert models may differ in the degree to which their problem solving processes correspond to those of human problem solvers (3,26). For instructional purposes, a model that can explain what it is doing is much more useful than a model with optimal, but unexplainable behavior. So, for example, an expert model consisting of a probability network with no conceptual underpinnings would not be very helpful for instruction, since humans cannot readily assimilate complex systems of probabilities. Student modeling components, which diagnose students’ emerging competence, appear in several forms. An overlay model represents a student’s knowledge as a subset of the expert model’s knowledge. Overlays can reveal those pieces of expert knowledge that are either missing or misapplied in a student’s mind, but they cannot capture student knowledge that is qualitatively different from an expert’s (10,21). So, for example, an overlay model for mechanics could represent that a student does not know that objects maintain their velocities absent outside forces, but it would not be able to capture that a student uses the term acceleration in ways that subsume parts of acceleration and parts of velocity. Another type of model matches student errors to a bug library of common misconceptions in the domain. Such models can detect not only the correct knowledge students do not possess but also much of the incorrect knowledge they do possess (10,21). Some bug libraries are comprised of prespecified bugs, while others generate models of student bugs from a set of underlying principles (3,10,21). Some other models try to simulate the student’s learning processes, relying on the difficult assumption that human and computer learning processes can be equated (2). However, Reusser (20) notes that ‘‘machine tutoring based on cognitive simulation of the student is still not possible across a full range of tasks and in open-ended domains.’’ Student models diagnose in different ways, including knowledge tracing [keeping track of which skills or pieces of knowledge a student has mastered (27)], model tracing [matching a student’s solution steps to those of an expert problem solving model (5)], plan recognition (inferring a student’s plan based on subgoals accomplished so far), and decision trees (21). Generally, student diagnosis is more difficult in student-controlled systems than in systems that maintain some degree of control over the interaction (26); but even in the latter systems, student models should diagnose conservatively. While expert knowledge is fully in tune with real-world systems and hence constrained by them, students sometimes begin with very divergent and idiosyncratic beliefs and backgrounds. Consequently, it is easy to misdiagnose a student’s misunderstandings, making a conservative approach to student diagnosis appropriate (6). Developers must consider the dangers of misdiagnosis (28), as well as the validity and reliability of computer-generated diagnoses (18). A variety of measures are available for evaluating student models’ diagnostic success (29). It is also possible for a model simply to query the student to resolve ambiguities in diagnosis (10).

INTELLIGENT TUTORING SYSTEMS

The pedagogical module structures the interaction between the system and the user, deciding what task material to present and what kind of feedback to provide, if any (20). Its behavior depends upon the domain knowledge and student modeling components (18). Pedagogical styles can differ along such nonorthogonal dimensions as prescriptive versus discovery learning, tutoring versus coaching, and student-directed versus system-directed (19,20). Some pedagogical approaches are directive (7), some are noninterventionist (19), and some are in between, such as the cognitive apprenticeship teaching methods of scaffolding and fading (9,28). Thus, feedback from a system can play any number of roles, from corrective to regulative to informative (3,30), and it may differ in relative amount and timing (e.g., immediate vs. delayed vs. on demand (1,5,28,31)). The pedagogical module is best evaluated against standards of its underlying instructional theory or of expert human teachers or tutors in the respective domain (5,18). Although often de-emphasized relative to the other modules, the user interface component of an ILE is critical to its success or failure (3,6,10,18). In addition to being the means by which the user and the system communicate, the interface can also function as an external memory for the user, reducing his or her cognitive load (6). Common screen interfaces may contain text, hypertext, graphics, or even animation. While graphical interfaces are generally more engaging than text-based ones (26), in some domains they have been found to be disadvantageous (32). Some interfaces are capable of adapting to individual users, although resultant changes in screen displays may cause confusion (6). In some systems, the interface may be rich enough to also play a pedagogical role, in which case the separate effects of the interface and the system’s pedagogy may be difficult to ascertain (5,6). Evaluation Why Evaluate?. While the specifications of a declarative knowledge or CAI system can be validated via careful inspection by experts in relevant field(s), this type of validation is insufficient for complex systems such as ILEs (18). Generally there is no way to guarantee that such a complex system does what it purports to do unless its effects can be demonstrated in the behavior of actual users in the target population (33). One radical view of evaluation in the educational technology field is ‘‘to move towards a situation where we have reliable, precise theories from which systems may be formally derived which will produce the learning benefits predicted by the theories’’ (34), thereby eliminating the need for evaluation. However, we have not yet reached that utopian position. There are other, more specific reasons for evaluating an ILE. A system under development should be tested in the field to assess ‘‘its actual impact on a broad array of teacher and student behaviors’’ (35). Developers must assess not only the effectiveness of a system but also the likelihood that it will be fully accepted into the work or school culture of the target audience (36). Formative versus Summative. The ultimate validation of a learning tool involves formal, controlled experiments that follow a scientific methodology (6,37). However, the literature shows that there are many fewer controlled evaluations than systems (3,15,38). In most cases, a summative type of evalua-

531

tion is reserved for completed systems. Some have argued that such evaluations are inappropriate because formal summative techniques for systems as complex as ILEs do not yet exist (18,23). While global effects can readily be measured, we have insufficient tools for validating specific mechanisms within a system or for verifying exactly what is learned from a system. Thus, many developers prefer to conduct formative, internal evaluations. Although much more informal and less rigorous than summative evaluations, formative evaluations can produce results of much greater detail (18). Such detail is especially important during system development. Many developers use formative evaluation studies for rapid prototyping and incremental improvement of system parts. In early stages of system development, a ‘‘Wizard of Oz’’ (6) approach is used, in which humans simulate missing system components (18). Techniques used in formative evaluations include additive design comparisons, to assess the impact of various components on the overall effectiveness of the ILE; and lag sequential analysis, to measure the impact of system feedback on the user (29). Even pilot evaluation of completed systems is usually formative in nature (38), often progressing from laboratory sessions to field testing (18). Whereas the purpose of a summative evaluation is to validate a system’s advantages, a formative evaluation should emphasize weaknesses and other negative aspects, so that they can be corrected (6). Assessing Educational Impact. An ILE evaluation may address a variety of issues. Developers and researchers may ask how much users gain by interacting with the system, while teachers, administrators, and other outside parties may ask whether users learn the same things as students of more traditional instruction (37). Corporate trainers may focus on transfer, the extent to which the system not only trains workers for a specific job but also prepares them to quickly learn other jobs that are closely (near transfer) or more distantly (far transfer) related. In addition to transfer, an evaluation may measure attitude change about computers (39). Mark and Greer (18) list several criteria for evaluating an ILE’s effects on achievement and affect. Achievement measures include transfer, retention, time to mastery, and dropout rates. Affective measures include student motivation (including the intrinsic motivation of computer use), self-esteem measures, and time on task. While most achievement measures are objective, many affective ones are subjective and usually gathered by questionnaire.

SYSTEMS CATEGORIZED BY DOMAIN/SKILL In the sections that follow, we present a number of examples of intelligent learning tools for various subject matters, commenting on any evaluation data that are available. Programming Many of the ILEs we reviewed were designed to teach computer programming. This was an attractive domain for early intelligent tutoring system efforts. The subject matter was familiar to developers (2), the procedures to be taught were well defined, and programming novices were readily available. Most of these systems teach only the basic, introductory ele-

532

INTELLIGENT TUTORING SYSTEMS

ments of a programming language, with some covering the content of a one-semester course. The LISP (LIST Processing) Tutor. One of the most wellknown and thoroughly evaluated ILEs is the CMU (Carnegie Mellon University) LISP Tutor, also known in various incarnations as GREATERP (4), LISPITS (27), and the GRAPES LISP Tutor (31). The system is an ITS that has been used to teach introductory LISP programming, both in laboratory studies and in one-semester college courses. Its design principles are based on Anderson’s ACT* theory of cognitive skill acquisition (7,11,27,40). It contains a problem-solving expert component, a bug catalog, and a tutoring module for assessing student knowledge, assigning appropriate problems, and providing feedback (4). Problem-solving rules are represented as productions (IF-THEN statements) in GRAPES (Goal Restricted Production System) (41), and the system uses model tracing to diagnose student solution plans (7) as well as knowledge tracing to select appropriate problems for students to solve (27). In essence, the student tries to write programs prescribed by the tutor, and the tutor intervenes with advice whenever the student’s activity deviates from what the expert model would do. The interface includes a structured editor using LISP templates, so that the student does not have to concentrate on checking syntax. Whenever the student makes an erroneous step, the system intervenes with immediate feedback. Two early evaluations of the LISP Tutor (7) clearly demonstrated its educational effectiveness. One study compared groups of students learning LISP from a human tutor, from the ITS, or on their own. Although post-test scores were equivalent across groups, the human-tutored and ITS-tutored groups took significantly less time to cover the material. The second study found that ITS-tutored students took less time and scored better on a final exam than control students working on their own (7,27). Students in the studies liked the tutor and rated it as superior to traditional programming courses. Results showed that while a human tutor was still best, the ITS was a close second, far ahead of classroom instruction (4). The performance and mastery time data were consistent with the 1.0 effect size found in evaluations of some other extensive ITSs (42). Corbett and Anderson (27) found that students who had received more explanatory feedback from the ITS made fewer errors per goal than those who had received less explanatory feedback, but did no better on post-tests. They also found that students had equal post-test performance but longer solution times when they controlled the timing of feedback presentation than when the system did, suggesting that students either were more careful about making errors or were spending more time detecting and correcting them (27). Students rarely requested immediate feedback from the ITS; most of them wanted feedback only when they were finished coding a problem (43,44). GIL. Another LISP tutor, GIL [Graphical Instruction in LISP (45)], was built using many of the same principles as the CMU tutor. One difference is a graphical interface. This allows students to learn programming concepts without having to deal with syntax concurrently, and it imposes a lower cognitive load than a text-based system. The basic idea is to describe a program as a graph of processes that connect in-

puts to outputs. With this format, students can construct program graphs both forward from a problem’s input(s) and backward from its output, thus eliminating the top-down, leftto-right constraint found in the LISP Tutor. Another improvement is that GIL’s rule base can be used not only to solve LISP problems but also to explain to the student why a particular step is appropriate in a given situation. In a study comparing four versions of GIL with different degrees of feedback (46), undergraduate LISP novices wrote program graphs with GIL and then completed a post-test based on elements of similar problems. Students who had received greater degrees of feedback tended to commit fewer errors, to immediately fix them more often, and to request system help less often than those in the minimal and delayed feedback conditions. They also scored significantly higher on the post-test. Thus, unlike in Corbett and Anderson’s (27) feedback study, GIL students who received the most informative feedback both solved the LISP problems better and apparently learned the material better (46). A visually explicit interface alone can serve pedagogical functions, even in a system with little or no tutoring. Another study (cited in Ref. 5) compared students using the standard version of GIL to those using an exploratory version without model tracing. Although the exploratory students took twice as long as the standard GIL students to complete the training problems, they scored as well as them on post-tests. ADAPT. Another programming system, ADAPT (ADA Packages Tool), was designed to teach a second language to programmers already experienced in Pascal or C (47). Since syntax from these prior languages shows positive transfer to ADA but solution planning shows negative transfer, the focus with ADAPT is on planning rather than syntax. ADAPT’s user interface includes plan menus, from which plan components are chosen. Some of the plans are buggy; immediate feedback is delivered when one of these is chosen. ADAPT is more flexible than the LISP Tutor, generally allowing planning of steps in any order and enforcing top-down, left-toright order only at the coding level. In a formative evaluation study (47), six undergraduates who knew both Pascal and C solved problems using ADAPT. As in the coding-order manipulation studies with the LISP Tutor (27,44), students did not exercise control over the order of step planning; for the most part, they developed plans in the sequence in which they appeared in the interface. Positive transfer of prior syntax to ADA was found, as expected; and on the few ADA syntax errors that students did commit, the system’s error-location feedback was usually sufficient for them to be corrected immediately (47). EGO. Ego is an ITS that teaches Gries’ methodology for developing programs and proofs in parallel (48). Although the system teaches program writing, it helps focus students on the overall methodology by helping them with algebra, logic, and code syntax. The system also contains a context-sensitive advisor, available on demand. Ego allows students to make and correct their own errors, although the system can intervene to prevent excessive drift down a bad solution path. The system maintains an overlay student model and a goal library, and each goal module incorporates both tutorial knowledge and a bug catalog specific to that goal. Ego thus can utilize different teaching strategies as appropriate in the con-

INTELLIGENT TUTORING SYSTEMS

text of particular goals. Unlike many other programming tutors, its interface allows a student to undo a bad path back to the origin of the error. PROUST. PROUST is an ITS for beginning Pascal students that identifies and provides feedback on program bugs. Unlike the other programming tutors we reviewed, PROUST accepts only complete, syntactically correct programs. The system is noninteractive, and does not tutor on writing correct code per se, but rather compares a student’s submitted code to its library of plans, identifying bugs (3,15). While noninteractiveness may seem like a limitation of the system, recall that students in one of Corbett and Anderson’s (44) studies requested feedback only when they had finished coding their programs. This form of tutoring is consistent with the revised ACT-R theory’s claim that students can learn from complete problem solving products, not just from correcting erroneous steps en route to complete solutions (43). In addition, structured editors are available to facilitate the writing of syntactically correct program code; therefore, PROUST’s input constraints are not extraordinarily difficult to satisfy. In general, tutoring based upon completed problems will likely work so long as the student is afforded enough help to assure problem completion. Summary. The programming ILEs we reviewed differ along several dimensions. Some teach or enforce correct syntax, while others focus more on solution planning than coding syntax. Some impose a strict order on program development, while others are more flexible and allow users to choose the order in which they work (even if they usually opt not to exercise their choice). Some utilize model tracing to provide immediate feedback, keeping students on-path during solution coding; one provides delayed feedback, allowing students to commit and correct errors; and one delivers feedback only on completed programs. Of those ILEs that provide immediate feedback, most provide information on at least the location of coding errors, if not more explanatory feedback. Generally, in each of the studies comparing degrees of feedback, more was found to be better than less.

533

them to proceed with any legal inference, even if it is not on a solution path included in its expert model (7). In a formal evaluation study (49), GPTutor students averaged a letter grade higher on exams than the control class. In another study (11), using a version modified to treat legal but nonoptimal inferences as illegal moves, high-school students who received only minimal feedback on their moves made significantly more errors per proof on a post-test than those receiving any combination of goal and condition-violation feedback, and also tended to immediately fix fewer of their errors. These results are consistent with those found in the aforementioned feedback study with the GIL programming tutor (46). ANGLE. ANGLE (A New Geometry Learning Environment) is a more recent tutor based on principles similar to the GPTutor, but further emphasizing the diagrammatic reasoning inherent in expert geometry planning (33). Diagrams also facilitate novice problem solving by making subgoals explicit and reducing cognitive load, in domains including geometry (11), propositional calculus (6), programming (45), argumentation (50,51), and certain procedural task simulations (39). A formative study of the ITS in geometry classes showed that ANGLE students tended to solve more proofs on a posttest than control group students, although the results varied by teacher. However, in another study, ANGLE students made more execution errors than GPT students on a posttest. This may reflect ANGLE’s higher emphasis on tutoring proof planning over execution, or possibly ANGLE’s more flexible interface, which does not enforce a planning approach (52). Summary. Both of the geometry ILEs share many features with the LISP Tutor and GIL, including diagnosis by model tracing and the graphical nature of the latter’s interface. The tutors differed from each other in the extent of their emphasis on planning, as with the programming ILEs. As in the programming domain, minimal feedback on errors was less beneficial than more informative feedback, with goal-related feedback being the most valuable.

Geometry/Diagrammatic Reasoning After programming, the ITS community began to focus on mathematics instruction and then on a number of other areas. Geometry received considerable attention, partly because it relies more heavily on reasoning with diagrams and partly because it was found that the proof process could be understood more readily when presented graphically as the task of finding a path from premises to conclusions. Conveniently, geometry proofs start with a conclusion to prove, and involve the relatively simple operators of forward and backward inference (10). The Geometry Tutor. The Geometry Tutor (also known as GPTutor or just GPT), also inspired by the ACT* theory, teaches students how to construct geometric proofs (7,11,35,49). Much like GIL’s program-graph interface, the Geometry Tutor’s interface uses graph-like displays (proof trees) to represent both directions of inference and to reduce students’ cognitive loads (40). The tutor intervenes immediately if students attempt illegal inferences, but will allow

Algebra PAT. PAT (Practical Algebra Tutor) is another ITS based on the ACT theories of skill acquisition (53). It was designed specifically to support a new algebra curriculum, produced by the Pittsburgh Urban Mathematics Project (PUMP), emphasizing real-world problems. PAT users work on word problems using tables, graphs, and symbolic equations, while the system does model tracing using correct and buggy rules. The system also does knowledge tracing, displaying cumulative skills acquired by the user in a ‘‘Skillometer’’ window. In an admittedly confounded formative evaluation of PAT (53), algebra classes used PAT plus the new curriculum, in comparison to control classes using the traditional curriculum without the tutor. The PAT students scored 100% better than controls on tests of the new curriculum, but also 15% better on standardized tests of the traditional curriculum. The tutor has been integrated into many of the ninth-grade algebra classes that implement the new curriculum and is being adopted by districts elsewhere in the country.

534

INTELLIGENT TUTORING SYSTEMS

RAND Algebra Tutor. Another algebra tutor, developed by the RAND Corporation (37), has a less sophisticated student model, based on the number of problems tried and solved. Although it can sort problems based on student ability (via a form of knowledge tracing), and provide hints and answers to student questions, it cannot tailor its feedback to individual student characteristics. Field evaluations of multiple versions of the tutor with different pedagogical goals for teaching highlevel skills have met with mixed success (37). Pixie. Different instructional strategies were also compared in the Pixie algebra tutor (cited in Ref. 2). Developers exposed students to two pedagogical strategies, Model Based Remediation (akin to human or ITS tutoring) and Reteaching (akin to CAI). Although each strategy led to superior performance over control students, who received error notification only, performance with both strategies was equivalent. Although the developers concluded that ITS effects were similar to those obtained with traditional CAI approaches, their instructional manipulation was on too small a sample and for too short a duration for meaningful comparisons to be drawn (42). When the amount of content conveyed and the criterion for evaluation are both minimal, there is no reason to expect intelligent systems to excel over older approaches. Summary. The ACT-based algebra ILE (which used both model and knowledge tracing) demonstrated tangible benefits for its users. Studies of the other two systems, however, had equivocal results. Possible reasons for this include division of focus between different pedagogical goals and strategies, less effective diagnosis (via knowledge tracing alone in one case), and problems with experimental manipulations and with cultural and other field factors. Other Mathematical Skills GIDE. GIDE (54) is a goal-based diagnostic system for problem solving in algebra and statistics. It assumes that problem-solving errors are systematic, and that they must be considered in the context of a student’s solution plan in order to permit ‘‘intention-based’’ diagnosis (54), similar to PROUST. GIDE uses buggy plans and rules as necessary for diagnosis, and attributes missing or skipped steps in student solutions to inferred prerequisite knowledge or other such conceptual dependencies. Evaluations were conducted of two implementations of the system, one for statistics and one for algebra word problems (54). GIDE-Stat was able to recognize almost all of the goals, including implicit ones, in problem solutions from students in an introductory statistics course. It also identified most of the missing goals in students’ errors, but the implicit inference engine was too powerful, often attributing known or acquired concepts to students where an expert instructor would not. GIDE-Algebra’s evaluation was more extensive, involving thousands of problem solutions. The system made interpretable diagnoses for all of the solutions, and its diagnoses for a random subset of the solutions corresponded highly with those of human raters. POSIT. POSIT (Process-Oriented Subtraction Interface for Tutoring) teaches subtraction by presenting declarative knowledge from a bug library to correct student errors. The statements it presents are based on the system’s error diagnoses. In a formative evaluation, POSIT correctly diagnosed

most student errors, and misdiagnosed fewer errors than it failed to diagnose at all. In a summative evaluation, time to mastery was shortest with a human tutor and longest with classroom instruction, with the ILE group in between. The effect size for POSIT relative to classroom instruction was 0.74 (29). Summary. The ILEs in this section cover the domains of subtraction and diagnosis in algebra and statistics. Each system’s diagnoses were extraordinarily successful. One reason for this could be the relative simplicity and formality of both the problem solving operators and the inference procedures in these domains, in comparison to programming and to natural domains such as physics and medical diagnosis (10). Electrical and Economic Law Induction MHO. The MHO system teaches basic principles of electricity, using an overlay student model based on mastery of a collection of curriculum issues or ‘‘bites,’’ each of which was a simple concept or a combination of concepts taught earlier (3). It does knowledge tracing to select issues to address in problems it presents to students, based on prerequisite knowledge mastered thus far (3). A summative evaluation study (39) compared two versions of the tutor, one that provided ruleapplication feedback (i.e., it told students directly which principles to apply) and one with rule-induction feedback (i.e., students had to induce principles based on the relevant variables). More exploratory students did better if they received rule-induction feedback, and less exploratory students did better with the rule-application version. Voltaville. Voltaville is a microworld for learning about electric circuits via self-directed experimentation in a computer-based circuit laboratory (55). There is no direct instruction by the system. Rather, students try to discover as many electric laws as possible, which they submit to the system for feedback. Voltaville returns feedback both on the hypothesized laws and on the sufficiency of the evidence gathered in support of them. In an evaluation study, undergraduate students’ electrical knowledge improved from pre-test to posttest, and they discovered most of the laws included in the microworld. Students who showed the most improvement with the system were better at algebra and on learning indicators such as data management, devising correct hypotheses, controlling variables, and interpreting evidence in their simulated experiments (55). Smithtown. Smithtown (12) is a similar microworld, using essentially the same interface as Voltaville, for law induction in the domain of microeconomics. It also has no fixed curriculum, and seeks only to coach scientific inquiry skills via a form of knowledge tracing (3). While interacting with the system, users can alternate between simply observing the effects of manipulated variables (exploratory mode) and devising and testing hypotheses about them (experiment mode). Smithtown’s coach intervenes when a student exhibits buggy behaviors while in experiment mode, but remains silent while the student is in exploratory mode. The coach’s intervention threshold can be modified to present anything from immediate feedback to complete silence. In one experiment, Shute and Glaser (12) compared undergraduates in an introductory economics course using Smith-

INTELLIGENT TUTORING SYSTEMS

town, receiving only classroom instruction, or receiving no economics instruction at all (control). Both treatment groups outperformed the control group; and although Smithtown did not teach economics directly, students who used it did as well as classroom students on post-tests, after having spent less than half of the latter group’s time on task. The better Smithtown students were differentiated by learning and performance indicators similar to those in the Voltaville study (55). Shute and Glaser (12) ran a second experiment on a larger sample (over 500), of military recruits. Better learning correlated with more hypothesis-driven learning indicators; results from less-able learners showed them to be limited to datadriven indicators. Summary. The ILEs reviewed in this section were much less directive than many of the aforementioned systems, precluding model tracing in the purest sense (3). However, these systems were able to get a lot of mileage out of knowledge tracing; evaluations of each system showed at least some learning benefits for its users. The extent of benefits varied with user ability or practice. Given the less powerful forms of student modeling employed in exploratory environments, this is not surprising. Operative Skill GT-VITA. GT-VITA [Georgia Tech Visual and Inspectable Tutor and Assistant (23)] is a tutoring architecture for training NASA satellite ground controllers. It teaches operative skill, or ‘‘how to use declarative and procedural knowledge to manage complex systems in real time’’ (23), using a cognitive apprenticeship approach. The system includes an operator function model (OFM) to teach and evaluate operative skill, and an associated expert system (OFMspert) for presenting context-sensitive advice. The system also has a pedagogy module to provide immediate feedback (early in an interaction), coaching at critical checkpoints in a task (later in an interaction), and other ‘‘lesson objects’’ (23). The system also combines overlay and buggy student models for diagnosis. In a field evaluation (23), the tutoring architecture was used in the context of a payload-operations control center by novice satellite ground controllers. Students had difficulty at first with some of the operational skill demands during real-time satellite pass simulations; however, they did well on all declarative and most procedural prerequisites, and eventually on the essential operational skill measures. Students rated the system highly, and based on the system’s effectiveness, NASA has adopted a newer version of the architecture to be used in required ground control training (23). ALM. The ALM (Advanced Learning for Mobile subscriber Equipment) tutor (29) is a system for training operative skill for Army communications equipment. It uses an overlay student model and provides advice when students commit procedural errors. ALM’s advice templates cover anywhere from five to seven procedural steps, which may impose an excessive cognitive load. As a potential remedy, developers created MALM, a modified version of the tutor, in which each advice template covers only one procedural step. In an informal evaluation, initial error rates were comparable with both tutors, as were error and correction rates immediately following advice. However, MALM led to better performance that persisted, whereas the benefits of ALM’s advice lasted through

535

only a brief interval (29). Thus, MALM’s simplified feedback worked better than ALM’s more complicated feedback, suggesting that the cognitive load imposed by the latter was nontrivial. Summary. These systems differ from those reviewed previously in that they add real-time complexities to already complex tasks. Thus, cognitive load is even more a concern than with the other domains discussed above. ALM’s developers were able to achieve improved performance by simply reducing the amount of feedback presented by the system at any given time. However, because GT-VITA employed different types of feedback at different times, as well as different forms of student modeling, it is difficult to determine which of its complex features are most effective or which could be best improved. Miscellaneous Systems Andes. Andes is a tutoring system for teaching university physics (56). Students use the system to solve physics problems and study example problem solutions. Andes uses quantitative as well as qualitative physics problems, which are less likely to perpetuate students’ misconceptions than traditional quantitative problems because their focus is not on algebraic manipulation [56; see also (6)]. Based on results of a pilot study in which the majority of students’ requests for help were made when they were lost (56), Andes’ student model has been extended to do diagnosis by plan recognition. Its coaching component has been billed as ‘‘the first computer tutor aiming to improve learning by guiding self-explanation’’ (57), the extent of which is gauged by using a ‘‘poor man’s eyetracker’’ (56) to measure students’ reading times for various elements of example problems. Andes provides immediate error-flagging feedback on a student’s problem-solving steps, except when such feedback could lead to error correction via simple guessing; in such cases, Andes instead presents questions focusing on the student’s reasoning. VCR Tutor. Mark and Greer (8) developed a device tutor to teach VCR programming. They created four versions of their VCR Tutor with different pedagogical approaches, to examine the role of knowledgeable feedback on a task as predominantly procedural as programming a VCR. Tutorial interactions ranged from simply forcing the user through a predetermined programming procedure, to giving error notification feedback, to giving informative feedback drawn from a conceptual model of the task. Only the most informative version employed student modeling and error diagnosis, using a bug catalog. Students who had used the most informative version had fewer steps, errors, and error types, and did marginally better on all other post-test measures than students who had used any of the other three versions. Thus, although each of the four versions was sufficient for teaching VCR programming, knowledgeable feedback led to performance advantages at no additional cost in training time. This is consistent with McKendree’s (11) finding that ‘‘tasks that are quite constrained may not require maximally informative feedback, but even these tasks may be learned at least as effectively given the more informative feedback.’’ CATO. CATO (50) is an ILE designed to teach beginning law students to argue with cases. Despite the limited feed-

536

INTELLIGENT TUTORING SYSTEMS

back it generates, the system fosters argumentation skill via an interface that reifies argument structure and helps to manage the complexity that a solely text-based system would present. It provides a set of core argument moves that students can use, and a hierarchy of factors that represent similarities and differences between legal cases. A controlled evaluation study (50) using first-year law students found no differences between groups on either a pre-test or a post-test of basic argument skills, but the control group did better on a more advanced, memo-writing assignment. The evaluators concluded that CATO was able to improve basic argument skills as much as the traditional instructor, but because its method of teaching was not holistic like the instructor’s, it was unable to prepare students for more integrative tasks such as memo writing (50). Turbinia-Vyasa. Turbinia-Vyasa is an instructional system for training operators to troubleshoot failures in marine steam power plants (24). It includes an intelligent tutor and a domain simulator with high degrees of dynamic, structural, and temporal fidelity to power plants on naval vessels. As with many engineering applications, such plants are quite complex and contain many interrelated subsystems, making component failures difficult to troubleshoot. To help minimize the operator’s cognitive load during training, as well as the computational requirements for representing such a complex system, the simulator employs qualitative approximations of system states rather than numerical values. The tutor includes highly organized system and troubleshooting knowledge, including limited case-based diagnostic knowledge linking symptoms to components. A student model keeps track of students’ failure hypotheses, and the tutor also uses a record of students’ actions to infer their misconceptions about the plant system. The tutor responds immediately to student queries and also provides feedback when it infers a student misconception, either immediately or at the end of the training session depending on context. At session’s end students can also review correct problem solutions with explanations. An experiment compared groups of Naval ROTC cadets trained with the simulator and active (system-initiated), passive (student-initiated), or no tutoring. Both tutored groups learned to formulate and test failure hypotheses well, while the untutored group mainly used guessing. Some students became overly dependent on the active tutor’s feedback, using it to evaluate their hypotheses instead of theorizing and seeking evidence themselves. In another experiment, groups of cadets solved troubleshooting problems with some combination of diagnostic cost and time limits, to better reflect fidelity of interaction in real-world diagnostic tasks. While the unlimited group was most successful, cadets subjected to both limits were second best. The imposed limits induced them to abandon bottom-up, experimenter strategies in favor of more efficient, top-down, theorist strategies (24). Sherlock. Sherlock is an extensively evaluated ILE for training a technical job in avionics troubleshooting. Specifically, it is a ‘‘computer-coached practice environment’’ (17) that combines intelligent coaching facilities with a realistic work-environment simulation, emphasizing the latter over the former and over student modeling precision (58). Sherlock II, the most recent incarnation of the ILE, utilizes hierarchical, fuzzy student modeling variables that approximate an

overlay onto both its expert and curriculum models (59). Sherlock I provided both conceptual and procedural hints on demand, on an ascending scale of explicitness (58). Hint levels were matched to the current student model (59), and were faded as the student became more proficient (17), in order to keep students ‘‘in the position of almost knowing what to do but having to stretch their knowledge just a little in order to keep going’’ (36). Sherlock II added facilities to support reflective follow-up after problem solving, including goal-related presentations such as intelligent replays of problem solving steps, critiques of those steps, and information about what an expert might have done (60). These capabilities were added to help compensate for the learning opportunities that are precluded by the high cognitive effort expended during problem solving (36,61), as well as to coach situations in which students were able to solve the problems but did so in a nonoptimal way (62). Sherlock follows a cognitive apprenticeship approach of holistic instruction, rather than pacing students through a series of separate lessons (17,58). Because traditional on-the-job training often spans many years, Sherlock accelerates the skill-acquisition process (62). A benchmark study showed that trainees with 20–25 hours of experience with Sherlock I performed at a level equivalent to that of technicians with four years of on-the-job experience, with 90% retention of performance gains after six months (36,59). Thus, even relatively brief interactions with Sherlock produced troubleshooting skills that were durable (36). In one controlled evaluation (17), using groups of airmen matched for ability on pre-tests, Sherlock students solved significantly more problems, used more expertlike problem solving steps, and executed fewer bad steps than a control group. Use of Sherlock I led to more expert troubleshooting solutions in fewer steps, for both low-ability and high-ability students (17). An evaluation of Sherlock II was conducted with Air Force master and apprentice technicians (62). Again, experimental and control groups were matched for ability, based on verbal troubleshooting tests. The Sherlock students scored higher on post-tests both of standard avionics troubleshooting tasks and of tasks involving another, fictitious troubleshooting system, thus showing transfer to novel troubleshooting tasks. Control students performed many more nonoptimal solution steps, such as swapping of electronic components, in both troubleshooting environments than tutored students and master technicians. Sherlock I effect sizes for some post-test measures were greater than 1.0. Effect sizes as great as 2.0 were obtained in the evaluations of Sherlock II (60). Clearly, both generations of Sherlock have been successful. However, because Sherlock incorporates multiple elements and instructional strategies, it is difficult to attribute its success to any one of them (17,36). Its developers have proposed ways of partialing out its effectiveness, such as applying elements of Sherlock to other domains or gauging Sherlock’s effectiveness with certain elements or pedagogical strategies removed. However, they concede that ‘‘it may be inevitable that successful training requires confounding of approaches’’ (17). Collaborative Systems Beginning with the advent of serious interest in computersupported collaborative work (CSCW) in the late 1980s, a recent trend in the design of ILEs is to support collaboration

INTELLIGENT TUTORING SYSTEMS

(19). Collaborative systems offer many potential benefits over single-user systems. One benefit is cost-effectiveness; more students may be able to use fewer computers at the same time (25,39). Students working together in groups may be able to diagnose and teach each other, relieving the computer of such burdens as diagnosis and natural language parsing (39) as well as exposing all students in a group to alternative hypotheses and multiple perspectives (63,64). Some collaborative systems involve ‘‘pseudo-social’’ interactions between a human learner and computerized agents (65). Developers may envision the human and the computer as a single system (65) and evaluate the various elements of their software in terms of how they foster activity in the entire human-computer system (63). HERON. HERON (20), named for a Greek mathematician, is a graphics-based tool for helping students in grades three through nine solve mathematics story problems. It supports the use of graphical solution-trees for problem representation and planning, in which students can link problem conceptnodes using arithmetic operators to map out a solution plan, working forward or backward. Erroneous input will cause HERON to intervene, based not on student diagnosis but on solution-tree content. A field evaluation (20) compared pairs of fifth-graders solving story problems with HERON to pairs working without HERON. The system improved story problem comprehension and solutions on a post-test, promoted useful dialogue among the pairs, and was liked by both students and teachers. GDSS. Alavi (66) conducted a study using a group decision support system (GDSS) with classes of MBA students. The software included tools for brainstorming; categorizing and ranking ideas; and scoring, rating, and voting on alternatives. Effectiveness of collaborative learning was measured by students’ self-perceptions of learning and evaluative ratings of their classroom experiences. Results showed a significant effect of GDSS use over the traditional group; it positively affected students’ perceived learning and skill development, interest in the subject matter, and appraisal of the group learning exercises and overall classroom experience. GDSS students had significantly higher final exam scores than control students. It was unclear which tools or features of the GDSS had the greatest impact on group learning, and neither experimenter bias nor novelty effects could be ruled out (66). CLARE. CLARE (Collaborative Learning and Research Environment) is a system that supports collaborative knowledge construction from published research papers (64). The elements comprising CLARE are a knowledge representation language, a process model for collaborative learning, and a hypertext-based interface that integrates them. The representation language includes node primitives to denote epistemological concepts (e.g., claim, theory, and question) as well as relationships between them. Students are led, in two phases, ‘‘from an external, isolated and individual position inward toward an internal, integrated and collaborative perspective’’ (64). Most students agreed that the two-phase collaborative process model helped to promote the formation of individual views. Students also found CLARE to be useful for collaboration, for understanding research papers in a novel way, and

537

for understanding other perspectives. Although students were not pleased with CLARE’s interface, most students said that the representation language helped to expose different points of view, and that the primitives were among CLARE’s most useful features. However, many students used the primitives incorrectly (e.g., stating evidence as claims, listing problems as disagreements). The evaluators note that even incorrect use of the node primitives can be useful for collaborative learning because, as indicated in their study, it stimulates discussion among students about the roles of the different concepts in understanding the research papers (64). Belvedere. Formative evaluation studies of Belvedere (51) have shown similar patterns. Belvedere is a networked graphical environment designed to foster scientific argumentation skills in middle-schoolers. Students use node and link primitives similar to CLARE’s to construct graphical argument representations of scientific problems. Problems can come from any source, but Belvedere’s developers have created specialized databases about several scientific debates, which are accessible via a World Wide Web browser. Belvedere’s interface was designed to resemble that of familiar drawing programs, so that students could learn to create argument diagrams with only minimal training. With both diagram sharing and chat facilities, Belvedere enables students to discuss and reflect upon their argumentation processes and products. A computerized coach is available on demand to provide guidance in developing argument diagrams (67). Several formative evaluation studies of Belvedere were conducted with middle- and high-school students (51). Students used Belvedere’s node and link primitives in ways that were inconsistent both with their intended usage and with their own and other students’ usage, much like students in the CLARE evaluation. Although Belvedere’s developers agreed with CLARE’s that such unintended usage actually served to stimulate collaborative discussions (63), such usage can cause problems for the automated diagnostic coach. Although the coach was still under development during the aforementioned formative evaluation studies, it has since received some empirical validation (67). In addition to syntactic node patterns, the coach is now able to respond to consistency relations between any nodes in a diagram that were copied from one of Belvedere’s semantically annotated knowledge bases. Developers applied the coach to a subset of one of the knowledge bases used in the formative evaluations and found that, in most cases, its consistency judgments agreed with their own. The authors then semantically annotated that entire knowledge base, and then had the coach evaluate a diagram produced by students from one of the earlier studies. The coach agreed with all but one of the students’ links; and on that particular link, the developers agreed with the coach. Thus, with only minimal knowledge engineering via semantic annotations to an existing knowledge base, developers were able to extend the capabilities of the coach to include consistency checking. Such capabilities could be useful for coaching collaboration by pointing out inconsistent relations between or within students’ diagrams (67). Adapting Existing Systems. Note that none of the previously discussed collaborative systems involves a student-modeling component. Partly this is due to the difficulties inherent in trying to maintain separate models for each student in a col-

538

INTELLIGENT TUTORING SYSTEMS

laborative setting, especially when only one computerized agent is involved (26). Another reason is the viewpoint of many developers that ‘‘collaboration should be concerned with what is happening on the screen’’ (65), as opposed to knowledge hidden inside the learners’ heads. However, diagnosisbased tutoring or coaching, as employed in many of the systems we reviewed, appears to provide substantial benefits for single users. It seems a shame to have to sacrifice such system intelligence in order to support collaboration among groups of students. One system that has attempted to bridge this gap is Sherlock, whose developers argue that ‘‘affording students significant opportunities for collaborative learning is not going to be any harder than developing high quality computer-based systems for solo learning’’ (61). They somewhat circumvent the student-modeling problem by defining different roles for Sherlock’s coach in a collaborative setting (9). In one scenario, Sherlock can work with a student as a problem-solving collaborator; in another, students work in groups as a single student, with Sherlock as coach. The developers also envision situations in which Sherlock can help ease a student’s transition from self-critique to peercritique. In one, Sherlock acts as a peer during review of the student’s problem solving trace, constructing explanations jointly with the student. In another, a group of students constructs such explanations, with Sherlock available on demand for assistance. In these scenarios, Sherlock is concerned not with modeling individual student knowledge but with fostering peer interaction and critique involving single or joint problem solving activity; however, it can still apply its diagnostic expertise to promote such activities (17,25,61).

DISCUSSION In focusing on systems for which evaluation results were accessible, this article does not completely reflect the diversity of domains for which ILEs have been, and continue to be, developed. The majority of systems we reviewed support welldefined tasks and domains, many of which contain largely procedural components. This is due partially to the course of history in the field of ILE development, which began almost exclusively with such constrained domains (2) and only more recently began to steer toward more ill-defined, ambiguous tasks. Nevertheless, across the domains and systems we did review, a number of similarities and differences emerged. With few exceptions, users liked the various systems and were motivated to use them, regardless of domain. This was borne out not only by attitude measures but also by observations of their use of the systems. Although only a subset of the systems have been judged favorably enough to be adopted by their target audiences, at least most of them have hurdled the initial barrier of capturing user interest. Among the other similarities between systems were the beneficial aspects of their user interfaces. Across many domains, interfaces functioned as external memories for their users. This was true for both single-user and collaborative systems. Many interfaces also served some pedagogical functions, usually by design but sometimes by accident (6). Interfaces helped to reify forward and backward problem solving or inferencing not only in geometry but also in programming (GIL) and mathematics (HERON). Evaluation studies re-

vealed unanticipated shortcomings in some interfaces, one of which (ADAPT’s) precluded observation of the effects of other system components. Such findings further underscore the importance of formative evaluation to system development. Approaches to student modeling were also similar across domains. No clear within-domain preferences for particular student modeling techniques emerged from our review. For example, it was not the case that ILEs for one domain used bug libraries exclusively while those for another domain used only overlay models. Some ILEs for programming and operative skills (Ego and GT-VITA, respectively) even used both types of model. Similarly, variants of both model tracing and knowledge tracing were used for diagnosis in programming and mathematics domains. Only in the discovery-world environments were knowledge tracing variants used exclusively, because model tracing was not feasible with such unguided instruction (3). In general, the relative effects of systems’ pedagogical modules tended to vary with the specifications of their underlying instructional approaches. Among the primary characteristics that differentiate these approaches is the type of system intervention they advocate. Immediate Feedback A major issue in the design of interactive learning environments is the role of system-generated immediate feedback. Although used to varying extents in the operative skill tutors and in limited ways in the economics microworld, immediate feedback approaches dominated many of the tutors for programming, geometry, and mathematics, including all of the ACT* tutors. Indeed, the ACT* commitment to providing immediate feedback in its tutors is one of the theory’s most controversial features (27). One reason for preferring it is to ensure that feedback is delivered in the context in which it is needed, that of the student’s current goal and working memory states (27). Another reason to provide corrective feedback immediately is to prevent students from floundering while trying to recover from lengthy incorrect solution paths (11,27,40). Although the revised ACT-R theory and its newer tutorial instantiations permit off-path problem solving (43), they still focus students toward correct solution paths, and immediate feedback still plays a major role in the interaction. However, research has shown immediate feedback to be disadvantageous in certain situations and with particular tasks (1). In one experiment using a modified LISP Tutor, students who received immediate feedback solved training problems faster than students who received delayed feedback, but when solving test problems took more time and made more errors than delayed-feedback students (31). In addition, delayed-feedback students seemed to be better at planning problem solutions than immediate-feedback students. The experimenters argued that the absence of immediate feedback in the delayed condition allowed students to redeploy their cognitive resources toward developing secondary skills such as error detection and correction. A study comparing versions of the GIL tutor (5) provides further evidence of this: Students who did not receive GIL’s immediate model-tracing feedback scored better on a transfer test of program debugging skills than those who did. Thus, the value of immediate feedback seems to vary with not only the task but also the desired learning outcomes of the intervention.

INTELLIGENT TUTORING SYSTEMS

Cognitive Load Cognitive capacity must be considered when assessing the actual or potential benefits of system-generated feedback. Because information processing limitations arise more often in the student than in the computer, the cognitive load issue pervades nearly all domains and design rationales to at least some degree. Tutorial feedback is but one of many things that must compete for a user’s cognitive resources while he or she is interacting with an ILE. With more complex tasks, feedback may be best left for post-problem reflection, when cognitive resources are no longer being taxed by immediate problem-solving demands (36). Perhaps, with further research, McKendree’s (11) claims (see previous section) regarding the relationship between task complexity and feedback content could be extended to account for feedback timing as well. Feedback Content In addition to McKendree, several other researchers have conducted studies, using ILEs included in our review, which manipulated feedback content. In many of these studies, performance of users given varying amounts of feedback was compared. In some cases, minimal feedback (e.g., error notification) was sufficient to allow users to eventually solve problems. However, in nearly all cases, the maximal amount of feedback resulted in the greatest learning outcome, whether measured by test scores, time to mastery, or both. In some cases, the highest level of feedback involved goal-related information (e.g., GPTutor, GIL). Given the cognitive capacity limitations delineated earlier, this is not surprising. One can easily lose track of goal-related information while engaged in complex problem solving. Sherlock tries to compensate for this by presenting goal-related information during a reflection period after the problem has been solved. Developers have described methods for varying the nature of system feedback content based on problem characteristics (Andes) or on the chronological point in the interaction during which one type may be more appropriate than another (GTVITA). Researchers have discussed the prospects of fading out feedback content as users become better able to proceed without it (e.g., Andes, Sherlock). Other researchers have investigated the effects of removing tutorial feedback entirely (e.g., comparing versions of an ILE with and without feedback). A study using GIL (see previous section) showed that its model tracing feedback did have an effect beyond the pedagogical effects of its interface. Concluding Remarks Aside from the general similarities and differences already discussed, it is difficult to identify any domain-specific effects of, or any clear preferences between, the various approaches to providing feedback. Again, this may be due partially to the preponderance of well-defined, primarily procedural domains represented in the literature on evaluated ILEs. However, it could also be the case that a simple classification of approaches based solely on the domains or skills they involve is insufficient. Indeed, although such a classification is a useful starting point, our review suggests that perhaps other crossdomain factors, such as task complexity and priorities of learning outcomes, must be considered as well. We are inclined to agree with Shute and Glaser’s (12) characterization

539

of learning from an ILE as an interaction of many factors, including domain subject matter, targeted learning outcomes, and the type of instructional strategies employed by the system. BIBLIOGRAPHY 1. J. A. Kulik and C.-L. C. Kulik, Timing of feedback and verbal learning, Rev. Educ. Res., 58 (1): 79–97, 1988. 2. R. J. Siedel and O. C. Park, An historical perspective and a model for evaluation of intelligent tutoring systems, J. Educ. Comput. Res., 10 (2): 103–128, 1994. 3. E. Wenger, Artificial Intelligence and Tutoring Systems: Computational and Cognitive Approaches to the Communication of Knowledge, Los Altos, CA: Morgan Kaufmann, 1987. 4. J. R. Anderson and B. J. Reiser, The LISP tutor, Byte, 10: 159– 175, 1985. 5. D. C. Merrill et al., Effective tutoring techniques: A comparison of human tutors and intelligent tutoring systems, J. Learn. Sci., 2 (3): 277–306, 1992. 6. M. Twidale, Redressing the balance: The advantages of informal evaluation techniques for intelligent learning environments, J. Artif. Intell. Educ., 4 (2/3): 155–178, 1993. 7. J. R. Anderson, C. F. Boyle, and B. J. Reiser, Intelligent tutoring systems, Science, 228: 456–462, 1985. 8. M. A. Mark and J. E. Greer, The VCR Tutor: Evaluating instructional effectiveness, Proc. 13th Annu. Conf. Cogni. Sci. Soc., 1991, pp. 564–569. 9. J. S. Brown, A. Collins, and P. Duguid, Situated cognition and the culture of learning, Educ. Researcher, 18 (1): 32–41, 1989. 10. W. J. Clancey, Qualitative student models, Annu. Rev. Comput. Sci., 1: 381–450, 1986. 11. J. McKendree, Effective feedback content for tutoring complex skills, Hum. Comput. Interact., 5 (4): 381–413, 1990. 12. V. Shute and R. Glaser, A large scale evaluation of an intelligent discovery world: Smithtown, Interact. Learn. Environ., 1: 51–77, 1990. 13. P. A. Cohen, J. A. Kulik, and C.-L. C. Kulik, Educational outcomes of tutoring: A meta-analysis of findings, Amer. Educ. Res. J., 19: 237–248, 1982. 14. B. S. Bloom, The 2 sigma problem: The search for methods of group instruction as effective as one-to-one tutoring, Educ. Researcher, 13 (6): 4–16, 1984. 15. D. C. Littman and E. Soloway, Evaluating ITSs: The cognitive science perspective, in M. C. Polson and J. J. Richardson (eds.), Foundations of Intelligent Tutoring Systems, Hillsdale, NJ: Erlbaum, 1988, pp. 209–242. 16. G. D. Hume et al., The use of hints as a tutorial tactic, Proc. 15th Annu. Conf. Cogni. Sci. Soc., 1993, pp. 563–568. 17. S. P. Lajoie and A. Lesgold, Apprenticeship training in the workplace: Computer coached practice environment as a new form of apprenticeship, Mach.-Mediated Learn., 3: 7–28, 1989. 18. M. A. Mark and J. E. Greer, Evaluation methodologies for intelligent tutoring systems, J. Artif. Intell. Educ., 4 (2/3): 129–153, 1993. 19. E. De Corte, Changing views of computer-supported learning environments for the acquisition of knowledge and thinking skills, in S. Vosniadou et al. (eds.), International Perspectives on the Design of Technology-Supported Learning Environments, Mahwah, NJ: Erlbaum, 1996, pp. 129–145. 20. K. Reusser, From cognitive modeling to the design of pedagogical tools, in S. Vosniadou et al. (eds.), International Perspectives on

540

21.

22. 23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

INTELLIGENT TUTORING SYSTEMS the Design of Technology-Supported Learning Environments, Mahwah, NJ: Erlbaum, 1996, pp. 81–103. K. VanLehn, Student modeling, in M. C. Polson and J. J. Richardson (eds.), Foundations of Intelligent Tutoring Systems, Hillsdale, NJ: Erlbaum, 1988, pp. 55–78. M. C. Polson and J. J. Richardson (eds.), Foundations of Intelligent Tutoring Systems, Hillsdale, NJ: Erlbaum, 1988. R. W. Chu, C. M. Mitchell, and P. M. Jones, Using the operator function model and OFMspert as the basis for an intelligent tutoring system: Towards a tutor/aid paradigm for operators of supervisory control systems, IEEE Trans. Syst. Man Cybern., 25: 1054–1075, 1995. T. Govindaraj et al., Training for diagnostic problem solving in complex engineered systems: Modeling, simulation, intelligent tutors, in W. B. Rouse (ed.), Human/Technology Interaction in Complex Systems, vol. 8, Greenwich, CT: JAI Press, 1995, pp. 1–66. S. Katz and A. Lesgold, The role of the tutor in computer-based collaborative learning situations, in S. P. Lajoie and S. J. Derry (eds.), Computers as Cognitive Tools, Hillsdale, NJ: Erlbaum, 1993, pp. 289–317. R. R. Burton and J. S. Brown, An investigation of computer coaching for informal learning activities, in D. Sleeman and J. S. Brown (eds.), Intelligent Tutoring Systems, New York: Academic Press, 1982, pp. 79–98. A. T. Corbett and J. R. Anderson, LISP Intelligent Tutoring System: Research in skill acquisition, in J. H. Larkin and R. W. Chabay (eds.), Computer-Assisted Instruction and Intelligent Tutoring Systems: Shared Goals and Complementary Approaches, Hillsdale, NJ: Erlbaum, 1992, pp. 73–109. A. Collins, Design issues for learning environments, in S. Vosniadou et al. (eds.), International Perspectives on the Design of Technology-Supported Learning Environments, Mahwah, NJ: Erlbaum, 1996, pp. 347–361. P. J. Legree, P. D. Gillis, and M. A. Orey, The quantitative evaluation of intelligent tutoring system applications: Product and process criteria, J. Artif. Intell. Educ., 4 (2/3): 209–226, 1993. P. M. Fischer and H. Mandl, Improvement of the acquisition of knowledge by informing feedback, in H. Mandl and A. Lesgold (eds.), Learning Issues for Intelligent Tutoring Systems, New York: Springer, 1988, pp. 187–241. L. J. Schooler and J. R. Anderson, The disruptive potential of immediate feedback, Proc. 12th Annu. Conf. Cogni. Sci. Soc., 1990, pp. 702–708. M. Oliver, T. O’Shea, and P. Fung, An empirical study into which style of representation works best for learners of modal logic, in B. du Boulay and R. Mizoguchi (eds.), Proc. AI-ED 97 World Conf. Artifi. Intell. Educ., Tokyo: IOS Press, 1997, pp. 371–379. K. R. Koedinger and J. R. Anderson, Effective use of intelligent software in high school math classrooms, in S. Brna, S. Ohlsson, and H. Pain (eds.), AI-ED 93: Proc. 6th World Conf. Artifi. Intell. Educ., Charlottesville, VA: AACE, 1993, pp. 241–248. J. Self, Computational mathetics: Towards a science of learning systems design [online], 1995. Available www: http:// www.cbl.leeds.ac.uk/~jas/cm.html J. W. Schofield, D. Evans-Rhodes, and B. R. Huber, Artificial intelligence in the classroom: The impact of a computer-based tutor on teachers and students, Social Sci. Comput. Rev., 8 (1): 24– 41, 1990. A. Lesgold, Assessment of intelligent training technology, in E. L. Baker and H. F. O’Neil, Jr. (eds.), Technology Assessment in Education and Training, Hillsdale, NJ: Erlbaum, 1994, pp. 97–116. C. Stasz et al., An intelligent tutor for basic algebra: Perspectives on evaluation, Instructional Views of Intelligent Computer-As-

sisted Instruction: Data and Issues, Symp. Annu. Meet. Amer. Educ. Res. Association, San Francisco, March 1989. 38. V. J. Shute and J. W. Regian, Principles for evaluating intelligent tutoring systems, J. Artif. Intell. Educ., 4 (2/3): 245–271, 1993. 39. J. W. Regian and V. J. Shute, Evaluating intelligent tutoring systems, in E. L. Baker and H. F. O’Neil Jr. (eds.), Technology Assessment in Education and Training, Hillsdale, NJ: Erlbaum, 1994, pp. 79–96. 40. J. R. Anderson et al., Cognitive principles in the design of computer tutors, Proc. 6th Annu. Conf. Cogni. Sci. Soc., 1984, pp. 2–9. 41. J. R. Anderson, R. Farrell, and R. Sauers, Learning to program in LISP, Cognit. Sci., 8: 87–129, 1984. 42. P. J. Legree and P. D. Gillis, Product effectiveness evaluation criteria for intelligent tutoring systems, J. Comput. Based Instruct., 18 (2): 57–62, 1991. 43. J. R. Anderson et al., Cognitive tutors: Lessons learned, J. Learn. Sci., 4 (2): 167–207, 1995. 44. A. T. Corbett and J. R. Anderson, The effect of feedback control on learning to program with the Lisp tutor, Proc. 12th Annu. Conf. Cogni. Sci. Soc., 1990, pp. 796–803. 45. B. J. Reiser et al., A graphical programming language interface for an intelligent LISP tutor, Proc. CHI’88, Conf. Hum. Factors Comput. Syst., 1988, pp. 39–44. 46. J. W. Connelly, An empirical investigation of the effective degrees of feedback content in GIL, an intelligent tutor for programming, unpublished manuscript, 1989. 47. V. Fix and S. Wiedenbeck, An intelligent tool to aid students in learning second and subsequent programming languages, Comput. Educ., 27 (2): 71–83, 1996. 48. F. Ng, G. Butler, and J. Kay, An intelligent tutoring system for the Dijkstra Gries methodology, IEEE Trans. Softw. Eng., 21: 415–428, 1995. 49. R. Wertheimer, The geometry proof tutor: An ‘‘intelligent’’ computer-based tutor in the classroom, Math. Teacher, 84 (4): 308– 317, 1990. 50. V. Aleven and K. D. Ashley, Teaching case-based argumentation through a model and examples: Empirical evaluation of an intelligent learning environment, in B. du Boulay and R. Mizoguchi (eds.), Proc. AI-ED 97 World Conf. Artifi. Intell. Educ., Tokyo: IOS Press, 1997, pp. 87–94. 51. D. Suthers et al., Belvedere: Engaging students in critical discussion of science and public policy issues, in J. Greer (ed.), AI-ED 95: Proc. 7th World Conf. Artifi. Intell. Educ., Washington, DC: Association Advancement Comput. Educ., 1995, pp. 266–273. 52. K. R. Koedinger and J. R. Anderson, Reifying implicit planning in geometry: Guidelines for model-based intelligent tutoring system design, in S. P. Lajoie and S. J. Derry (eds.), Computers as Cognitive Tools, Hillsdale, NJ: Erlbaum, 1993, pp. 15–45. 53. K. R. Koedinger et al., Intelligent tutoring goes to school in the big city, in J. Greer (ed.), AI-ED 95: Proc. 7th World Conf. Artifi. Intell. Educ., Washington, DC: Association Advancement Comput. Educ., 1995, pp. 421–428. 54. M. M. Sebrechts, From testing to training: Evaluating automated diagnosis in statistics and algebra, in C. Frasson, G. Gauthier, and G. I. McCalla (eds.), Intelligent Tutoring Systems, Berlin: Springer-Verlag, 1992, pp. 559–566. 55. L. Schauble et al., Causal models and experimentation strategies in scientific reasoning, J. Learn. Sci., 1 (2): 201–238, 1991. 56. K. VanLehn, Conceptual and meta learning during coached problem solving, in C. Frasson, G. Gauthier, and A. Lesgold (eds.), ITS96: Proc. 3rd Int. Conf. Intell. Tutoring Systems, New York: Springer, 1996, pp. 29–47. 57. C. Conati, J. H. Larkin, and K. VanLehn, A computer framework to support self-explanation, in B. du Boulay and R. Mizoguchi

INTERACTIVE VIDEO (eds.), Proc. AI-ED 97 World Conf. Artifi. Intell. Educ., Tokyo: IOS Press, 1997, pp. 279–286. 58. A. Lesgold et al., A coached practice environment for an electronics troubleshooting job, in J. Larkin and R. Chabay (eds.), Computer-Assisted Instruction and Intelligent Tutoring Systems: Shared Issues and Complementary Approaches, Hillsdale, NJ: Erlbaum, 1992, pp. 201–238. 59. A. Lesgold et al., Possibilities for assesment using computerbased apprenticeship environments, in J. W. Regian and V. J. Shute (eds.), Cognitive Approaches to Automated Instruction, Hillsdale, NJ: Erlbaum, 1992, pp. 49–80. 60. A. Lesgold, Ideas about feedback and their implications for intelligent coached apprenticeship, Mach.-Mediated Learn., 4: 67–80, 1994. 61. A. Lesgold et al., Extensions of intelligent tutoring paradigms to support collaborative learning, in S. Dijkstra, H. P. M. Krammer, and J. J. G. van Merrie¨nboer (eds.), Instructional Models in Computer-Based Learning Environments, Berlin: Springer-Verlag, 1992, pp. 291–311. 62. S. P. Gott, A. Lesgold, and R. S. Kane, Tutoring for transfer of technical competence, in S. Dijkstra et al. (eds.), Instructional Design: Vol II: Solving Instructional Design Problems, Mahwah, NJ: Erlbaum, 1997, pp. 221–250. 63. D. Suthers and A. Weiner, Groupware for developing critical discussion skills, in J. L. Schnase and E. L. Cunnius (eds.), Proc. CSCL ’95: 1st Int. Conf. Comput. Support Collaborative Learning, Mahwah, NJ: Erlbaum, 1995, pp. 341–348. 64. D. Wan and P. M. Johnson, Experiences with CLARE: A computer-supported collaborative learning environment, Int. J. Hum.-Comput. Stud., 41 (6): 851–879, 1994. 65. P. Dillenbourg, Distributing cognition over humans and machines, in S. Vosniadou et al. (eds.), International Perspectives on the Design of Technology-Supported Learning Environments, Mahwah, NJ: Erlbaum, 1996, pp. 165–183. 66. M. Alavi, Computer mediated collaborative learning: An empirical evaluation, MIS Quart., 18 (2): 159–174, 1994. 67. M. Paolucci, D. Suthers, and A. Weiner, Automated advice-giving strategies for scientific inquiry, in C. Frasson, G. Gauthier, and A. Lesgold (eds.), ITS96: Proc. 3rd Int. Conf. Intell. Tutoring Syst., New York: Springer, 1996, pp. 372–381.

JOHN CONNELLY ALAN LESGOLD University of Pittsburgh

INTELLIGENT VEHICLE HIGHWAY SYSTEMS (IVHS). See INTELLIGENT TRANSPORTATION SYSTEMS. INTERACTIVE TELEVISION. See SET-TOP BOXES.

541

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2908.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Management Education Standard Article Laurence D. Richards1 1Indiana University East, Richmond, IN Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2908 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (68K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2908.htm (1 of 2)17.06.2008 21:43:46

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2908.htm

Abstract The sections in this article are A Brief History Management Philosophy Educational Models Issues in Management Education | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2908.htm (2 of 2)17.06.2008 21:43:46

MANAGEMENT EDUCATION ENGINEERING MANAGEMENT TECHNOLOGY MANAGEMENT Engineers have been candidates for managerial positions in industrial and government organizations for most of the twentieth century. During the latter half of the century, the percentage of engineers moving into managerial positions increased dramatically, and that trend has continued into the twenty-first century. About two-thirds of all individuals with engineering degrees now pursue a management career track, as opposed to design, research, or other career tracks, and about 85% of all engineers will, at some point in their careers, have managerial responsibility. With the current emphasis on integrating sophisticated information, communication, and automation technologies into engineered systems as well as into the processes for designing, manufacturing, and marketing those systems, adequate technological knowledge to understand the intricacies of these systems and processes is recognized as an important attribute of the managers whose task is to coordinate the processes. Hence, the percentage of individuals being promoted into managerial positions who have some form of technical background (engineers and applied scientists being prominent among that group) has been increasing over the past two decades. With the trend toward wider distribution of management functions, and in particular the emergence of the project form of organization in the high-tech industries, these managerial positions tend to have titles like project manager (or leader), program manager, product manager, systems manager, operations (or production) manager, and field office manager or assistant manager. Both the redistribution of management functions and their redefinition within a framework of integration has led to special needs for the management education of engineers.

A BRIEF HISTORY During the first half of the twentieth century, most organizations relied heavily on on-the-job training and internal management training programs to prepare engineers and other employees for positions of management. While large organizations continue to staff training departments and organizations of all types contract for internal management training services, there emerged after World War II an awareness of the value of more formal management education. For those with engineering degrees, a master’s degree in management, most notably the master of business administration (MBA), became the program of choice. Many employers selected certain individuals to enroll in these programs full time and set up reimbursement schemes to encourage others to enroll part time. By the late 1950s and early 1960s, the approaches and techniques being developed to manage new and complex military and space technologies, in particular, sug-

gested a form of management education different from what was being offered through MBA programs. The new form needed to be designed specifically for those who were to become program and systems managers. For engineers and applied scientists, some early responses to this need included master’s degree programs in engineering administration at George Washington University and in engineering management at the University of Missouri–Rolla, each of which evolved into its own department. Other programs began sprouting within existing departments (e.g., a master’s degree in engineering management within the University of Pittsburgh’s department of industrial engineering), and some departments changed their name when an engineering management program was added (e.g., the department of industrial engineering and engineering management at Stanford University and the department of operations research and engineering management at Southern Methodist University). There are now, in the United States, more than one hundred master’s degree programs in management that cater exclusively to engineers and applied scientists, as well as many technology and systems management programs for both engineers and non-engineers. Each of these programs has its own flavor, representing a variety of curricula, forms of staffing, and delivery modes. The importance of employing engineers with a managerial perspective has also led to the development of undergraduate programs, both majors and minors, and undergraduate management courses (required of engineers at some institutions) to provide that perspective. Non-degree certificate programs have recently become popular alternatives to master’s degrees as a way to deliver management education in a shorter timeframe and in nontraditional formats. The history that has created the current variety of management education programs presents, for the engineer, the problem of discerning the differences between the options. For the providers of management education, the absence of standardized curricula presents a problem of identity, which is important in developing employer credibility. The flavor of each type of program is influenced by the clientele to be served within the geographical region of the institution, and by the nature of that institution and its other academic programs. Some management education programs have a manufacturing orientation, others a project management orientation; some cater to the private sector, others to the public sector; some reflect a traditional business administration curriculum, others an industrial engineering curriculum, and still others a mix of the two. The American Assembly of Collegiate Schools of Business (AACSB) has, over the years, provided curricular guidance for management programs offered through schools of business. The Accreditation Board for Engineering and Technology (ABET) focuses its attention on undergraduate programs and has not served to provide curricular guidance for graduate management programs in schools of engineering. The American Society for Engineering Management (ASEM) runs a certification program for graduate programs in engineering and technology management. ASEM recognizes that these programs must reflect the rapidly changing needs for management education in a high-tech world and that a diversity of perspectives and

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering. Copyright © 2007 John Wiley & Sons, Inc.

2

Technology Management

program types can be desirable. There is general agreement that management education has been and needs to continue to be multidisciplinary and interdisciplinary. This, by itself, presents special problems in staffing and delivering nontraditional management programs within the traditional university. There is also agreement regarding the high priority that should be placed in management education on continuing to develop communication skills, thinking and problem-solving skills, and interpersonal and team-building skills. This presents special problems when using distance learning technology, a common and increasingly used mode for delivering management education. There is disagreement on the extent to which the philosophy of management appropriate for the type of managerial position into which engineers are likely to move differs from the philosophy of management predominant in traditional education programs. Or, to put the same issue into the form of a question, Does the emergence of the project form of organization imply a need for a form of management education quite different from the traditional forms in order to reflect the new philosophy? And, if so, how can it be programmed, staffed, and delivered so that it creates an identity readily recognized and valued by students, teachers, educational administrators, and employers? MANAGEMENT PHILOSOPHY The term management philosophy is used here to refer to a system of motivating concepts, principles, and values, and hence a viewpoint on or way of thinking about the practice of management. Over time, discernible shifts in philosophies of management tend to be aligned with shifts in socioeconomic factors, organizational structures and processes, and/or technological breakthroughs. The currently emerging, although not yet completely defined, philosophy of management can be linked to the role of information and communication technologies in facilitating new, streamlined organizational structures, and of automation and simulation technologies in coordinating and customizing the processes of design, production, and delivery of an organization’s goods and services, which themselves are increasingly incorporating those technologies. It is in this context that engineers are attractive prospects for managerial positions due to their familiarity and comfort level with the technology. The diversity of management education programs for engineers is a consequence of differences in perception of or approach to the emerging philosophy of management. Organizational Trends Although ad hoc project teams (work groups, task forces, program offices, etc.) have been a part of organizations for a long time, the matrix form of organization was the first formalization of a project-oriented structure. Companies in the military, aerospace, electronics, chemical, and other high-tech industries needed a way to free project managers from the usual chain of command, which involved reporting along functional lines of authority (e.g., engineering, manufacturing, marketing, and finance). The complexity of

the technologies being employed required that all functions be considered simultaneously and without bias. By creating a line of reporting separate from these functional lines, this integration could occur. While functional management serves to maintain stability in organizations, project management serves as a change agent. The matrix organization is not without its problems. Members of project teams typically have homes in functional departments, creating opportunities for conflict between functional and project managers. Also, projects are temporary, so when a project is completed there may or may not be a new project to which the project manager can move. Many organizations have lost some of their top management talent as a result of an uneven flow of projects. The multiple chains of command in the matrix organization also present a special problem of coordination. Computerized information systems, even in the early years, provided a means to facilitate the coordination of projects. With further advances in information and communication technology, a project form of organization that is significantly different from the matrix organization is becoming possible. The trend involves a redefinition and redistribution of traditional functions of management—both in the sense of functions like production, marketing, and finance and in the sense of functions like planning, organizing, and controlling. For example, management control is being redefined in terms like dynamic coordination of teams rather than in terms like chain of command and span of control. The redistribution of functions suggests a greater need for individuals with some management education, rather than less, as every member of a project team must develop an appreciation for all the functions and take responsibility for the success of the team. This is demanding a shift in thinking about management. Management Paradigms The popularity of the term paradigm can be traced to Thomas Kuhn and his book The Structure of Scientific Revolutions (1). As the use of the term spread, it was picked up by writers and thinkers in nonscientific disciplines, including management consultants and educators, who began using it to address the dramatic changes occurring in the marketplace and in society and the need for managers to embrace new ways of thinking in order for their organizations to survive. The word has now taken on the status of a buzzword, devaluing some of the original impact intended by Kuhn. The word paradigm can denote a model, a pattern, or an example. Kuhn preferred to use the word exemplar for example and saved paradigm for talking about a predominant pattern of thought or worldview within a profession, discipline, or sector of society. He specified two conditions for qualifying a change in thinking as a paradigm shift: First, the change has to be linked to an accomplishment of sufficient magnitude to attract a group of adherents away from competing patterns, and second, the new pattern must possess an open-endedness sufficient to give these adherents something to do. Paradigm shifts are not, and cannot, be planned in the usual sense. They occur when they are needed, and they become needed when desirable concepts, tools, systems,

Technology Management

and processes cannot be implemented without them. Implementation of modern information, communication, and automation technologies, competition from international and multinational corporations (particularly in Pacific Rim countries), and socioeconomic changes in North America are cited as the factors that provide the impetus for recent shifts in management thinking and practice. Whether or not these changes qualify as a transformation of management paradigm remains for history to record. The number of new management concepts, tools, and systems cited by educators, writers, and consultants over the past few decades is so great that it would not be possible or fruitful to list them here, and more are being created daily. The typical management section of one of the large chain bookstores typically contains 40 to 50 titles on management tools and philosophy, leadership style, and organizational change, with each claiming new ways of thinking. There are, however, a few writers whose ideas and language have become so widely known and used in both the rhetoric of everyday management life and actual management practice, that they deserve mention. The first three items in the following list offer a societal context in which new thinking is being stimulated; the remainder focus specifically on management thinking. 1. The Third Wave With the publication of his books Future Shock (2) and The Third Wave (3), Alvin Toffler has acquired the status of social visionary. Focusing on the impact of information, communication, and automation technologies, Toffler chronicles structural shifts in economic, political, and social systems worldwide that are having consequences on and creating new possibilities for national priorities, organizational transformation, individual work, and everyday life. The first wave was agricultural, the second industrial, and the third informational. 2. The Knowledge Society As Toffler is widely regarded as America’s social visionary, Peter Drucker is widely regarded as its management visionary. Through a long history of books, Drucker has popularized certain ideas on management. In Post-Capitalist Society (4), he introduces the idea of the knowledge society and prophesizes that information and knowledge will soon replace labor, land, and capital as the most important (maybe the only important) resource (and product) of economic organizations. 3. The Age of Unreason/Paradox The British author Charles Handy has gone a step further in his books The Age of Unreason (5) and The Age of Paradox (6), examining the contradictions created by sudden and dramatic change and the implications of that on organizations of all types. As does Toffler and Drucker, Handy addresses educational institutions as well as economic and political ones, and the changes needed to manage these for the social good. The reader is led to conclude a need for new logic, a logic of change. 4. Systems Thinking The early history of systems thinking in management begins at the Tavistock Institute in London in the 1960s. It was here that experiments with a variety of corporations were conducted, em-

3

ploying ideas in democratic management and participative decision making. Emery and Trist (7) are credited with developing the sociotechnical systems approach to organizational change, stimulating their colleagues and sponsors to follow with books like Towards a New Philosophy of Management (8) and Alternatives to Hierarchies (9). In the United States, Russell Ackoff became a dominant advocate of systems thinking in management, building on his work with Emery, On Purposeful Systems (10), and popularizing it in books like Redesigning the Future (11), Creating the Corporate Future (12), and The Democratic Corporation (13). He developed idealized design as a tool for participative and consensus decision making and applied it to many types of organization. The concepts of the circular organization, the internal market economy, and the multidimensional approach to organizational design represent ways to implement democratic management. He offers a challenge to total quality management (see list item 6). 5. Change Management The first management book to reach the status of a national bestseller in the modern era was Peters and Waterman’s In Search of Excellence (14). Based on case studies of successful and unsuccessful firms, the authors identify eight attributes of successful organizations that can serve as guidelines for change. Peters followed this with some sequels, including A Passion for Excellence (15) and Thriving on Chaos (16). Bradford and Cohen in Managing for Excellence (17) offered a practical approach to thinking about these changes, including the transformation of the role of the manager from that of technician and conductor to that of developer. They suggest that the role of developer requires a team approach to work and its organization. The idea of self-directed work teams has received substantial attention in recent years. (See Ref. 18.) Rosabeth Moss Kantor, also relying on a set of case studies, declared in The Change Masters (19) the importance of flat, team-based, entrepreneurial-style organizations for success with innovation and change. Additional case studies of highly successful companies can be found in Collins and Porras, Built to Last (20), and Collins, Good to Great (21). 6. Total Quality Management The apparent success of Japanese companies in the 1970s and 1980s has been attributed in part to the implementation of ideas developed by W. Edwards Deming of the United States. Deming was hired by the Japanese in the 1950s to apply statistical concepts of quality to design, production, and other processes in Japanese firms. The design quality and reliability of many of the products of those firms made them competitive in the global marketplace. As a result of his experiences in Japan, Deming extended his ideas to general management thinking. His 14 points of management presented in his Out of the Crisis (22) and in Scherkenbach’s The Deming Route to Quality and Productivity (23) form the core concepts of total qual-

4

Technology Management

ity management (TQM). TQM has received attention in many corporate and government organizations in the United States and Europe. Implementation of TQM involves both the introduction of new tools (e.g., statistical process control, robust design, quality function deployment) and the development of a new culture oriented toward the customer and continuous improvement of products and processes. U.S. companies have attempted to use Deming’s 14 points of management as a way to accomplish the latter, but few have been successful in implementing all of them. The primary obstacle appears to be the degree to which the reliance on numbers-based management systems inhibits a customer orientation and continuous improvement. These systems— namely, systems for work standards, training, purchasing, performance appraisal, production control, and financial management—are deeply embedded in the thinking about accepted ways to do business. Irrespective of the difficulties, Deming and TQM have influenced management thinking by raising awareness of the importance of satisfying customers, encouraging employee creativity, maintaining the flexibility to change quickly, building quality into the processes of design and production (rather than inspecting for quality in the end products), and focusing attention on variation in those processes. 7. Process Reengineering With Hammer and Champy’s Reengineering the Corporation (24) and Champy’s Reengineering Management (25), the idea of continuous improvement has met its greatest challenge. While the focus is still on building quality into the fundamental processes of a business and its organization, the thesis is that creating quality in a rapidly changing and turbulent world requires fundamental transformation of these processes and discontinuous thinking. These radical and dramatic transformations rely on breakthroughs, not on ideas for incremental improvements (or Kaizen). Considered one of the major process innovations in recent years is that of concurrent or simultaneous engineering—an approach to product development that emphasizes integration of the design, production, marketing, distribution, and other aspects of a product by considering them concurrently rather than sequentially. The idea is not only that interrelationships among these aspects are better addressed, but also that the time to market is less and the responsiveness to the customer greater. 8. The Learning Organization Argyris and Schon introduced the concept of organizational learning in their 1978 book by that title (26). It was then picked up by a number of authors, most notably Peter Senge in his book The Fifth Discipline (27). The learning organization is one in which learning holds a position of the highest priority in the strategic mission, as well as the daily operations, of the organization. The five disciplines of effective learning organizations are shared vision, mental models, team learning, personal mastery, and systems thinking. A re-

lated concept is that of the virtual organization, one that transcends the boundaries of any single organization by taking advantage of the collaborative possibilities offered by multiple organizations. The resulting organizational arrangement is not a physically defined entity, but a virtual one with the ability to respond more quickly, more frequently, and more innovatively than any of its component organizations could individually. This attribute has been given the name agility. (See Ref. 28.) 9. Nonlinear Dynamics and Chaos The theoretical foundations for the ideas just presented have not been well articulated. Systems theory is a recurring theme, but only in its system dynamics version does it reflect a rigor deserving of scientific status. The science of nonlinear dynamics and chaos has provided an additional foundation for management theory. Notable books that have explicated this theory include Wheatley’s Leadership and the New Science (29), Priesmeyer’s Organizations and Chaos (30), and Goldstein’s The Unshackled Organization (31). Common themes include advocacy of self-organization (as opposed to planned change), flexibility as a criterion for decision making (as opposed to optimality), and variety generation (as opposed to efficiency and predictability). The shift in thinking from productivity to agility is consistent with a shift from a goal orientation to a process orientation, the latter requiring special attention to the dynamics of the organization and its operations. Strategic planning becomes an ongoing process of reevaluating and resetting goals, rather than an occasional exercise in evaluating strategies for achieving fixed goals. Paradoxes no longer need to be treated as aberrations; they become sources of creativity and transformation. 10. The Knowledge-Creating Company In their book by this title, Nonaka and Takeuchi (32) build on the ideas in nonlinear dynamics and make a case for an alternative view of the paradigm shift taking place. Rather than treating middle management as superfluous and advocating significant reductions in middle manager ranks, they regard middle management as acquiring a central role in the management and operations of the firm. They contend that the primary product of modern organizations is knowledge, and to think otherwise is to put the firm at a competitive disadvantage. Middle managers are the knowledge engineers of the organization. What is particularly noteworthy is that these authors are both professors in a Japanese university who have worked in industry and did their graduate education in the United States. Their contention that the success of Japanese companies in the 1970s and 1980s was based on skills and expertise in organizational knowledge creation is in contrast to the commonly held belief that success was attributable to the implementation of quality control, quality circles (bottom-up decision making), and lean management. While not in contradiction with these principles, knowledge creation does represent a perspective that perhaps has more po-

Technology Management

tential as a way of thinking about change in Western organizations than do ideas in books like Ouchi’s Theory Z (33) and Pascale and Athos’s The Art of Japanese Management (34). While many of the ideas discussed in the preceding list are contradictory, there are also some common themes. These include a trend toward semiautonomous (self-directed) project teams or work groups; a focus on flexibility/agility in planning, design, production, and marketing; treatment of the entire life cycle of a product/service concurrently throughout its development; and recognition of the importance of information sharing, knowledge building, and learning in everyday activities. Organizations that embrace these notions tend to exhibit distributed, dynamic, networked, and multidimensional structures, and parallel, integrated, nonlinear, and circular processes. Where these organizations exist, they rely heavily on information, communication, and automation technologies to help maintain coordination. Intelligent systems are often used and will be increasingly used in the future. Implications for Management Education A case has been made by many (the aforementioned authors among them) that if shifts in management paradigms are occurring in industrial and government organizations, then shifts in the philosophy and delivery of management education should follow. These shifts, it is argued, should not be limited to changes in what is taught, but should include changes in how it is taught as well. What can be said at this time is that there are multiple strands of change occurring simultaneously, some of which could be viewed as contradictory. For example, one shift is being driven by information and communication technologies (particularly Internet-based technologies) and is directed at the ability to deliver courses to a greater number of geographically distributed students. While this mode of delivery has been heavily content oriented, the pressures from industry are for a more process-oriented form of education, directed at developing the new thinking, interpersonal, and communication skills suggested by the paradigm shift(s) discussed previously. Master’s degrees and certificates in engineering management and the management of technology have been prime candidates for experiments with both content-oriented, online delivery and process-oriented, weekend/evening programs that make extensive use of case studies and group projects. While these divergent philosophies of management education continue to specify two quite different categories of program, they are beginning to merge into an integrated concept and approach. EDUCATIONAL MODELS There are at least three ways to distinguish different approaches to management education for engineers. There are different program types corresponding to educational level; different curricula depending on the student and employer market targeted; and different pedagogical approaches reflecting teaching styles and modes of delivery

5

supported by different institutional types.

Program Types Management education programs designed specifically for engineers and applied scientists began with master’s degrees, including MBAs with concentrations in areas like industrial management and technology strategy, master’s degrees in engineering management or engineering administration, management concentrations within industrial engineering master’s programs, and master’s degrees in the management of technology. MBAs are offered by business schools; master’s degrees in engineering management/administration tend to be offered by engineering schools either as stand-alone programs or as programs within an engineering department, and may or may not involve participation of a business school; industrial engineering concentrations in management tend to favor a more quantitative approach than do the other programs; and management of technology programs tend to be offered as collaborative efforts between business and engineering schools. In all of these categories, there are some programs that are designed for full-time students and others designed for part-time, working professionals. While some of these programs do not require work experience for admission, others require two years or more of full-time work experience in an engineering environment since receiving the bachelor’s degree, and virtually all regard such work experience as highly desirable. MBAs emphasize functional areas of business—particularly production, marketing, human resources, and finance—and integrate these through a policy perspective. Engineering management/administration programs emphasize the management of technical projects, programs, and operations, integrating functional areas of business throughout the curriculum. Management of technology programs tend to emphasize the strategic management of technology and technical resources. Universities with ABET accredited undergraduate degrees in engineering management in the United States University of Missouri–Rolla, Stevens Institute of Technology, and U.S. Military Academy. These programs emphasize a curriculum in engineering fundamentals, a concentration in an engineering discipline, and a broad set of courses in management topics. Another trend in undergraduate programs is to offer a minor in engineering management. The undergraduate minor at Old Dominion University is a four-course sequence emphasizing decision making, quality control, economic analysis, project management, and team building. In contrast, the undergraduate minor in technology and management at the University of Illinois at Urbana–Champaign is offered jointly by the colleges of commerce/business and engineering and is open to students from both colleges. The curriculum consists of six courses, three of which combine the students of both colleges, and an industrially sponsored capstone project. The use of industrially sponsored projects in undergraduate engineering curricula is quite popular; such projects offer opportunities to expose students to the management and business aspects of the engineering profession.

6

Technology Management

Certificate programs as alternatives to degree programs are being introduced at both the post-bachelor’s and postmaster’s levels. These programs allow greater flexibility in curriculum and format than do many degree programs, and employers are becoming increasingly supportive of programs with educational content that meets their needs, even if such programs cannot be accomplished within an official degree-granting structure. Case Western Reserve University’s certificate in the management of technology has received strong support from industry. It can be anticipated that as the demand for continuing education in management beyond the master’s degree continues to grow, professional doctoral programs (as opposed to researchoriented doctoral programs) that can be delivered in parttime and nontraditional formats will begin to emerge. Such programs will not require a research dissertation. Curricular Content The focus of the curricula of management programs varies substantially from one institution to another. There is general agreement that an introduction to the functional areas of finance, marketing, and production is important; however, in some programs this is accomplished in separate courses, while in others it is accomplished by integrating these subjects across the curriculum. Programs at institutions that reside in regions with a high concentration of federal government employers and government contractors tend to focus on program and contract management. Programs at institutions that reside in regions with high concentrations of industrial employers tend to focus more on manufacturing and/or project management. Programs associated with industrial engineering departments tend to favor either a manufacturing or an operations research focus. Programs associated with business schools tend to favor a strategic management focus. Programs within schools of engineering tend to be more technical and quantitative, although some programs have maintained a strong behavioral component in their curricula. Ongoing debate about the degree to which management curricula for engineers should be standardized is likely to continue. Pedagogical Approaches Just as there is a range of curricular content in management programs that target engineers and applied scientists, there is also a range of pedagogical approaches to the teaching and delivery of that curriculum. While there has been a move toward more project-based education in undergraduate engineering programs, the demand for master’s programs in management that are accessible to the parttime student has mitigated the degree to which this occurs at the graduate level. Many master’s programs do require a capstone project or thesis; and some on-campus programs, including those with weekend or other nontraditional formats, may permit a greater concentration of project work throughout the curriculum. Capstone projects conducted as individual study have a history in some programs of taking multiple semesters to complete. At many institutions, master’s degrees in engineering management have been targeted for delivery via distance

1.

learning technology. The latest trend is toward Internetbased delivery. This approach is primarily asynchronous, requiring a high level of motivation on the part of the student for successful completion to be realized. Asynchronous learning has been around for many years in the form of correspondence courses. Curricula developed for the World Wide Web offer many advantages over the traditional correspondence course. It remains to be seen if this form of delivery can be integrated with other modes of delivery to circumvent its drawbacks. ISSUES IN MANAGEMENT EDUCATION The changes that are occurring in the structure and management of economic organizations, and that will certainly continue to occur, are forcing institutions of higher education to reconsider the way they deliver educational products in general. The demand for management education, in particular, is such that a variety of innovative programs have been forthcoming in recent years. The growing political pressure to hold public institutions accountable, through demonstrations of the value of their services to the public, along with the tightening of budgets, has raised questions about the future role of the university in our society. The rise of the “virtual university” certainly suggests some profound changes. The predominant opinion at this time is that there is much of management education that can be delivered through a virtual university, much of it better (particularly when the visual capabilities of the various media are utilized) than what could be delivered through the traditional university; there are other needs of management education (like face-to-face group exercises and team-based projects) that have not been met as well in this way. The following trends in the content of management education for engineers and applied scientists are regarded as significant: 1. The team-based project form of organization 2. The management of parallel and distributed (as opposed to sequential) processes 3. Global and multicultural perspectives on business 4. Entrepreneurship/intrapreneurship skills The following trends in the process of management education for engineers and applied scientists are regarded as significant: 1. 2. 3. 4.

Industrial involvement in courses/projects Interdisciplinary approaches to teaching The use of educational technology A mix of scheduled and asynchronous learning environments

The extent to which management education continues to embrace these trends will depend in great part on how institutions of higher education respond to their changing roles.

Technology Management

BIBLIOGRAPHY

2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16.

T. S. Kuhn, The Structure of Scientific Revolutions, 2nd ed., Chicago: The University of Chicago Press, 1970. A. Toffler, Future Shock, New York: Bantam Books, 1970. A. Toffler, The Third Wave, New York: Bantam Books, 1980. P. F. Drucker, Post-Capitalist Society, New York: HarperBusiness, 1993. C. Handy, The Age of Unreason, Boston: Harvard Business School Press, 1989. C. Handy, The Age of Paradox, Boston: Harvard Business School Press, 1994. F. E. Emery, E. L. Trist, Socio-technical systems. In C. W. Churchman and M. Verhulst (eds.), Management Sciences: Models and Techniques, Oxford: Pergamon, 1960. P. Hill, Towards a New Philosophy of Management, New York: Barnes & Noble, 1971. Ph. G. Herbst, Alternatives to Hierarchies, Leiden, The Netherlands: Martinus Nijoff Social Sciences Division, 1976. R. L. Ackoff, F. E. Emery, On Purposeful Systems, Chicago: Aldine-Atherton, 1972. R. L. Ackoff, Redesigning the Future, New York: Wiley, 1974. R. L. Ackoff, Creating the Corporate Future, New York: Wiley, 1981. R. L. Ackoff, The Democratic Corporation, New York: Oxford University Press, 1993. T. J. Peters, R. H. Waterman, Jr., In Search of Excellence, New York: Harper & Row, 1982. T. Peters, N. Austin, A Passion for Excellence, New York: Random House, 1985. T. Peters, Thriving on Chaos, New York: Alfred A. Knopf, 1987.

17. D. L. Bradford, A. R. Cohen, Managing for Excellence, New York: Wiley, 1984. 18. J. D. Orsburn, L. Moran, E. Musselwhite, J. H. Zenger, SelfDirected Work Teams, Burr Ridge, IL: Irwin, 1990. 19. R. M. Kanter, The Change Masters, New York: Simon & Schuster, 1983. 20. J. C. Collins, J. I. Porras, Built to Last, New York: HarperBusiness, 2002. 21. J. Collins, Good to Great, New York: HarperBusiness, 2001. 22. W. E. Deming, Out of the Crisis, Cambridge, MA: Massachusetts Institute of Technology, Center for Advanced Engineering Study, 1982. 23. W. W. Scherkenbach, The Deming Route to Quality and Productivity, Washington, DC: CEEPress Books, 1992. 24. M. Hammer, J. Champy, Reengineering the Corporation, New York: HarperBusiness, 1993. 25. J. Champy, Reengineering Management, New York: HarperBusiness, 1995. 26. C. Argyris, D. A. Schon, Organizational Learning: A Theory of Action Perspective, Reading, MA: Addison-Wesley, 1978. 27. P. M. Senge, The Fifth Discipline: The Art and Practice of the Learning Organization, New York: Doubleday Currency, 1990. 28. S. L. Goldman, R. N. Nagel, K. Preiss, Agile Competitors and Virtual Organizations, New York: Van Nostrand Reinhold, 1995. 29. M. J. Wheatley, Leadership and the New Science, San Francisco: Berrett-Koehler, 1991.

7

30. H. R. Priesmeyer, Organizations and Chaos: Defining the Methods of Nonlinear Management, Westport, CT: Quorum Books, 1992. 31. J. Goldstein, The Unshackled Organization, Portland, OR: Productivity Press, 1994. 32. I. Nonaka, H. Takeuchi, The Knowledge-Creating Company, New York: Oxford University Press, 1995. 33. W. G. Ouchi, Theory Z, New York: Avon Books, 1981. 34. R. T. Pascale, A. G. Athos, The Art of Japanese Management, New York: Warner Books, 1981.

LAURENCE D. RICHARDS Indiana University East, Richmond, IN

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2909.htm

●

HOME ●

ABOUT US //

●

CONTACT US ●

HELP

Wiley Encyclopedia of Electrical and Electronics Engineering Professional Preparation Standard Article Leighton E. Sissom1 1Sissom & Associates, Inc., Cookeville, TN, Copyright © 1999 by John Wiley & Sons, Inc. All rights reserved. : 10.1002/047134608X.W2909 Article Online Posting Date: December 27, 1999 Abstract | Full Text: HTML PDF (155K)

Browse this title ●

Search this title Enter words or phrases ❍

Advanced Product Search

❍ ❍

Acronym Finder

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2909.htm (1 of 2)17.06.2008 21:44:01

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%20ELECTRIC...D%20ELECTRONICS%20ENGINEERING/18.%20Education/W2909.htm

Abstract The sections in this article are Baccalaureate Degree Professionalism Graduate Study Career Development Summary Keywords: accreditation; career development; construction; design; development; operation; practice; production; professional registration; research | | | Copyright © 1999-2008 All Rights Reserved.

file:///N|/000000/0WILEY%20ENCYCLOPEDIA%20OF%2...RONICS%20ENGINEERING/18.%20Education/W2909.htm (2 of 2)17.06.2008 21:44:01

J. Webster (ed.), Wiley Encyclopedia of Electrical and Electronics Engineering c 1999 John Wiley & Sons, Inc. Copyright


PROFESSIONAL PREPARATION Today’s engineer is in the most exciting arena of life. Engineering is challenging, rewarding, and limited only by one’s ingenuity. It is a creative profession which continues innovation with the materials and forces of nature. It extends the fundamentals of science to useable products and processes for society’s benefit. Most people were skeptical in May 1961 when President John F. Kennedy announced that the United States would go to the moon and return before 1970. But skepticism disappeared when it was accomplished in July 1969. Jules Verne (1828–1905) conceptualized a journey to the moon in 1873 in his From the Earth to the Moon—and he wasn’t even a scientist or engineer. He was a French science fiction writer, the forerunner of modern science fiction writers, who had ideas ahead of his time but couched them in the basics of science. Present-day engineers with vivid imaginations overshadow the accomplishments of the past. The connotation of “professional preparation” suggests logical methodology. That notion is real. What, then, may one do to prepare for this innovative profession? The discussion centers on undergraduate education, graduate study, and engineering practice (career development). The reader will hopefully recognize that there are many overlapping areas, such as professionalism and ethics, which are both “taught” and “caught.”

Baccalaureate Degree The quickest way for employment in a profession is by earning a baccalaureate degree in one’s discipline of choice. It is becoming more important to also pursue graduate degrees in order to better cope with current technological developments. Accreditation. The baccalaureate degree which is earned should be in a program accredited by the Accreditation Board for Engineering and Technology (ABET), the national engineering accreditation agency. The Engineering Accreditation Commission of ABET prescribes the minimum requirements for basic- or advancedlevel accreditation (1). The programs pertinent to this treatise include electrical engineering, electronics engineering, and other engineering programs with similar modifiers in their titles. For simplification, we shall hereafter refer to all such programs as electrical engineering. The basic ideas behind accreditation of engineering programs are: (a) quality control—that any student who graduates from an ABET-accredited program meets defined minimum requirements; and (b) truth in packaging—employers and graduate schools know the backgrounds of their personnel. It goes without saying that the curriculum must include the technical fundamentals of electrical engineering. Every electrical engineering professional must be intimately familiar with Ohm’s law, Kirchhoff ’s laws, and how electrons flow in both direct and alternating current, for example. They must have a working knowledge of systems and components, circuits and terminal devices, and controls and measurements. They must understand electromagnetic fields and wave phenomena. They must be able to mathematically model physical phenomena and to measure pertinent electrical quantities. Equally important are the other components of the curriculum that tie together the technical elements. It is highly probable that these other components will determine the ultimate success of the electrical 1

2

PROFESSIONAL PREPARATION

engineering practitioner. The other components include, but are not limited to, basic education, social sciences, communication skills (verbal and written), interpersonal relationships, and ethics and teaming. Curriculum. The accredited basic degree must include the components shown in Table 1. It summarizes the progression history of the engineering curriculum from the beginning of ABET [formerly called Engineers’ Council for Professional Development (ECPD)] accreditation in 1934. Definitions of the components and details of the curriculum are included in the integral article entitled Curricula. Interpersonal Skills. Student interactions with other students and with their mentors set the tenor for practitioner interactions. We learn in school how to behave in life. Since most behavior is learned, it is important that engineering students progress through interfacing on laboratory teams, working together on special projects, and interacting with their teachers. They may also learn from other professionals who may be invited to participate in the educational programs via guest lectures, development programs, and research projects. Teaming. One of the most educational aspects of the formal engineering education program is participating as a team member in laboratory projects. Providing critical data on one’s individual part of the project may make or break the entire project, just as in real life. Being on time with the data is crucial. Taking into account other team members’ contributions is a necessity. Continuing inquiries with other team members keeps the project on schedule and endears the individual to others. Cooperation is the key word. These are characteristics of behavior which may be learned and honed to perfection by interacting with others under the direction of a professional mentor. Teaming becomes more important as the student approaches graduation, notably in senior-level capstone design projects. Nothing develops the undergraduate’s ability to interface with others more than regional and national competitive events where teams of students vie for the best robot or the most accurate control system. Examples include the IEEE’s southeast Education Conference’s annual competition and the national/international solarenergy-powered vehicle races. These competitive academic models are extended into professional practice via conferences, competitive paper events, and multilevel grades of membership in professional and technical organizations on a personal basis.

PROFESSIONAL PREPARATION

3

Fig. 1. Hierarchy of needs. From lower-level basics of life to upper-level individualized ambitions. Higher-level needs become important when lower-level needs are satisfied.

Communication Skills. Nothing defines professionals’ progress as much as their ability to communicate, both verbally and in writing. An excellent idea may be lost if left noncommunicated to colleagues. An innovative product may languish in the laboratory if venture capitalists cannot be sold on its merits. A contract may be lost to a competitor who drafts a better proposal or even a letter of first contact. Good ideas must be conveyed to others in order to enjoy success in our society. In school, speech courses encourage us to practice our presentations before a mirror. Better yet, the videotape recorder provides an ideal way for us “to see ourselves.” We would not attend a play where the actors did not rehearse in order to hone their performances. Why should we expect fellow students or practicing professionals to be more receptive to our presentations? We should recognize both the impact and the limitations of audio–visual aids. We live in an era of the 45 s film clip. For the most effective presentation, we must adhere to the attract-and-hold-attention concepts that are extant today. For example, a slide should not normally be left before an audience for more than a minute. And it should only be used when pertinent to the comments being made and when germane to the central subject. Personal Esteem. Perhaps just as important as the message that is conveyed through communication is the building of personal esteem. A well-done paper or presentation gives the author or presenter a satisfying feeling, develops confidence, and makes the next project easier. Personal esteem includes self-respect, achievement, and recognition. Attracting and maintaining the respect of peers is crucial to professional success. Self-esteem is critical to becoming a successful practitioner. It centers on personal worth and completeness. It affects the individual’s ability to produce results. Success breeds success. Motivation. Maslow (2) defined a hierarchy of needs, shown in Fig. 1, which relate to motivation. Since needs must be satisfied from the bottom up, with the most fundamental needs forming a broad base of support for further development, it is easy to see that higher-level needs become important when lower-level needs are satisfied. As one works toward the upper level, ambitions and abilities are realized. It is the need to be the best that drives the ultimate professional. Internship. Recognized for centuries as being one of the most important ways to develop professionally, internship has been somewhat neglected at the undergraduate level in engineering education. Formal

4

PROFESSIONAL PREPARATION

cooperative education programs constitute the most common internship program in engineering education. They consist of alternate work and study programs, ranging from seven weeks to one year, usually quarter-byquarter or semester-by-semester. As an added benefit, the participant learns and earns simultaneously. Ideally, either formal or informal study are continued while the individual is on a Co-Op industrial or governmental assignment. Very few institutions now require engineering students to participate in a Co-Op program. While everyone recognizes the benefits of professional interactions at key junctures within the educational process, both students and faculty members usually want the students to progress as rapidly as possible to graduation. The argument is a financial one since delayed graduation is rarely compensated by higher earnings of the graduate who participated in the Co-Op program. Participation in a Co-Op program should not be based on finances but on motivation. It is also worth mentioning that Co-Op students are highly sought in the employment market. Co-Op students who have been on an industrial assignment always come back to school with a fresh outlook. In isolated cases while on assignment, some students may decide that engineering is not the career for them. In a majority of cases, however, returning students are motivated to excel in their engineering classrooms and laboratories. They are often motivated to accelerate their graduation. In many instances, CoOp assignments provide laboratory projects; and, in some cases, returning students are permitted to work on company-sponsored projects which provide modest stipends for the students while in school. Ethics. Ethics deals with what is right versus what is wrong. Engineering ethics is simply the adherence to prescribed rules of conduct which pervade all society while in school and in practice. Ethics is obviously not limited to engineering. Inculcating good ethical behavior begins with the family, extends through engineering education, and continues throughout the life of the practitioner. Cheating and plagiarism are the first elements of ethics encountered in school. There is sometimes a fine line between working together in order to learn more in the academy, or to produce more in the workplace, and in submitting one’s own work for recognition and credit. Clearly, the best way to learn is by studying or working with someone more intelligent or more innovative than ourselves. The approach to problem solving and to creative concepts “rub off” from others. Getting an answer to a question at the time it arises is the most efficient learning mode. Sometimes a word or two from a colleague (student or practitioner) or mentor (teacher or supervisor) can be more beneficial than a week of self-study. Whether in school or the workplace, credit must be given where it is due. As young children, we learn that we should not take credit for the work of someone else. That principle extends to adulthood in the academy and in the workplace. We don’t “copy” someone else’s work for credit on an exam or for recognition in a treatise submitted under our name. We don’t forge another person’s name on a check. We don’t alter grades in the teacher’s grade book or hack the university computer to alter records. These are examples of violating ethical behavior. As an example of ethics, consider the case of an undergraduate who gained access to a university’s computerized records to get names and addresses of students who belonged to a certain religious organization— in the days when such data could be a part of official records. He wanted the names and addresses in order to invite the students to participate in functions affiliated with their religious body. But he gained access to them by unethical means. His motives were pure, but his actions were not. This illustrates the fine line that is often faced by both students and practitioners. Some people falsify their own records, pad their resumes, and exaggerate credentials in an effort to further their careers. In the science and technology spectrum, one of the most common cases is that of graduates of engineering technology programs or engineering-related programs claiming that they are engineers. In some instances, their alma maters may even be guilty of promulgating this impropriety by publications which are inexact or omit key descriptive material. Perhaps the most widely prevalent breach of ethics in engineering—and probably other professions—is the copying of computer software. Unless in the public domain, a computer program is the work of its author—

PROFESSIONAL PREPARATION

5

just as much as a book. Both are intellectual property, analogous to real property that should not be stolen. Moral dishonesty is a violation of ethics.

Professionalism A profession—such as medicine, law, ministry, engineering, architecture, and teaching—is generally defined as a vocation which has the following common characteristics. • • • • • • •

Specialized knowledge Education and/or training required Standards promulgated by force of organization or concerted opinion Code of ethics Responsibilities to the public exceed those of clients Public recognition Police, reward, and punish practitioners

Professionalism embraces the adherence to these characteristics while engaging in practice. Engineering qualifies as a profession in all of these. A professional engineer is one who practices while adhering to these characteristics. Attaining specialized knowledge is a multifaceted process. The beginning path in obtaining specialized knowledge is that defined above—that is, beginning with an ABET-accredited baccalaureate degree in engineering. After graduation, the path may differ, depending upon individual goals. About one-third of today’s graduates opt to continue their education into graduate school. (See the section entitled “Graduate Study.”) Some go through formal industrial indoctrination programs, including specialized education and/or training. Professional Registration. A graduate engineer cannot legally practice in fields which involve the safety and welfare of our citizenry without being professionally registered. Currently, all of the states in the United States require that a professional engineer, entitled to bear P.E. after his/her name, go through a registration process prescribed by the laws of the individual states, most of which comply with the “Model Law” promulgated by the National Council of Examiners for Engineering and Surveying (NCEES). Most states require: (1) Graduation from an ABET-accredited engineering program. (2) Passing an eight-hour, limited-reference Fundamentals of Engineering (FE) examination near the end of the degree program or after graduation, covering the subjects of mathematics, physics, chemistry, and basic engineering courses. [Some states call their exams Engineer-in-Training (EIT) or Intern Engineer (IE).] The only reference that may be used on the exam is the NCEES’ Fundamentals of Engineering Reference Handbook. (3) Practicing in a responsible engineering position under the direction of a registered P.E. for a period of four years following the baccalaureate degree. (4) Passing an eight-hour, open-book Professional Engineering (PE) examination, which covers only subjects from ones area of specialty. [In some states, the examine is called Principles and Practices, P&P.] (5) Satisfying the cognizant state Registration Board that the candidate’s character is acceptable. This involves references and usually a personal interview by the Board. The FE and PE examinations are offered simultaneously in all states. The morning session of the FE examination covers the subjects of

6

PROFESSIONAL PREPARATION Chemistry Engineering economy Fluid mechanics Mathematics Statics

Dynamics Electrical circuits Material science and structure of matter Mechanics of materials Thermodynamics

The afternoon session covers the subjects of

(1) Applied mathematics (2) Electrical circuits (3) Thermodynamics and fluid mechanics

(1) Engineering mechanics

Engineering economy Engineering mechanics includes statics, kinetics, kinematics, and mechanics of materials. Individual state Registration Boards purchase the tests from NCEES. The Boards are responsible for distributing the applications for the exams, administering them, grading them, and notifying the examinees of the results. The Boards also establish the passing cutoff scores, causing some critics to charge that the exams are not uniform, since the boards in neighboring states may not choose identical cutoff scores. Minor deviations in the path to professional registration is permitted by some Registration Boards. For example, 50% of a student’s time spent in graduate school is usually credited toward practicing under a registered P.E. After an individual is registered in one state, most other states accept his/her credentials, via reciprocity or comity, for parallel registration in their states. The fundamental premise of the registration process is to protect the public by preventing unqualified people from offering engineering services. Engineers who work for companies that design and manufacture products are exempt from the licensing process—known as the industrial exemption. Unfortunately, the industrial exemption has kept some engineers from becoming registered P.E.s. In the view of many, it has also held back the development of professionalism while keeping the number of registered professionals at a low percentage. That this view is correct is evidenced by the necessity of successful examinees to meet minimal pass requirements. While this does not mean that an individual who fails the exam is deficient in his/her education, it does mean that he/she is deficient in that aspect of the education. Despite the arguments for the industrial exemption, most people agree that being a registered professional engineer is an important step in the development of professionalism. It is, of course, not the only step.

PROFESSIONAL PREPARATION

7

Graduate Study One may pursue a graduate degree after obtaining a baccalaureate degree—such as masters and/or doctorates within engineering or master of business administration outside of engineering. Some graduates pursue other professional degrees, such as medicine and law. Most professional schools outside of engineering see an undergraduate engineering degree to be excellent preparation for their post-baccalaureate programs. Most graduate degrees that are obtained by students with electrical engineering baccalaureate degrees are within their discipline of electrical engineering. They involve such subspecialties as (2) Antennas Circuits Computers Digital systems Power Electronics Lasers Plasmas Signal processing Telecommunications Microwaves Networks Instrumentation Microprocessors Some engineering graduates pursue advanced degrees without getting a baccalaureate degree. This is especially true when an advanced degree is considered to be the first professional degree. For example, an institution might offer a master of science degree in biomedical engineering, which requires courses in medicine and electrical engineering. But an engineering baccalaureate degree is required for all engineering graduate programs. Advanced-level criteria promulgated by ABET (1) provide (a) additional depth in the primary engineering discipline, (b) additional breadth in engineering areas related to the primary discipline, (c) deeper immersion in cultural, social, and/or business studies related to engineering practice, (d) emphasis in the broad study of manufacturing, construction, engineering management, and/or engineering entrepreneurship, and (e) programs that are offered jointly by the engineering unit and another academic unit. Very few graduate engineering programs are ABET-accredited because funding bodies, particularly federal agencies, reward the depth of “research” rather than the breadth of “practice.” Graduate study in ones own discipline permits him/her to develop depth, integrate peripheral fields of study, and become an authority in a specialized area. Graduate degrees may be research-oriented or designoriented. The largest percentage of advanced degree programs are currently research-oriented because support for the academic work is provided through research grants/contracts. Research projects then become the vehicle for the students’ specialized area, their areas of depth. Research-oriented advanced degrees are normally entitled Master of Science, Master of Science in Electrical Engineering, Doctor of Philosophy, and Doctor of Science.

8

PROFESSIONAL PREPARATION

Design-oriented graduate programs are often based on industrial-sponsored projects, sometimes centered on a product and/or process development and/or refinement. Such projects often have a shorter time frame since industry has to provide quarterly reports to the stockholders. They are frequently more consumer-oriented. Degrees granted are normally called Master of Engineering, Master of Electrical Engineering, and Doctor of Engineering. It is not possible to say whether the research- or design-oriented program is better. The answer is dependent upon the sponsoring agency, the academic institution, the graduate-study mentor(s), and the individual student—with all of them centered on the goals of the program. Currently, the research-oriented programs more nearly meet governmental agenda, while design-oriented programs more nearly meet industrial agenda. The confusion grew out of the early pattern of science funding from the federal government by agencies such as the National Science Foundation, which is largely based in science rather than engineering. Graduate programs normally require a research thesis/dissertation or a design report. Abridged research theses are generally published in international journals and presented at national forums. But the publication outlet for design reports is often limited to specific industries. Or the design report may not be published if it involves a company-proprietary project. The results of a design report may simply be locked up and kept from the world if deemed proprietary by the industrial sponsor. This restriction keeps the percentage of design-oriented programs low.

Career Development The disciplines of engineering (chemical, civil, electrical, industrial, mechanical, etc.) generally relate to the interest of an engineer. The functions of engineering—which are used as the subtitles of this section (research, development, design, production, etc.)—relate more closely with aptitudes, training, and career assignments of the individual practitioner. Functions of engineering characterize the type of work an engineer does after graduation (3,4,5,6). For example, one may be intimately involved in the design and development of a product which straddles disciplines, say electrical and mechanical engineering. Career development generally parallels the functions of engineering—within, but most often across, engineering disciplines. Practicing engineers are called upon to do so many things that do not fall within their disciplines. The degree of specialization is defined by the job assignment. For example, research in microelectronics is obviously more specialized than the management of a manufacturing plant. The former requires intimate attention to details and technological nuances. The latter demands a breadth of expertise, including interfacing with a myriad of people and functioning with budget constraints. Most modern-day products and processes involve the discipline of electrical engineering, as a minimum in measurement and control. But it is most often impossible to bring a product or process to fruition without straddling disciples. One only needs to be out of engineering school a short time to realize that real-world problems are not naturally compartmentalized by engineering discipline. It is imperative that career development follow the functions of engineering if the practitioners are to advance in their professional careers. Professional preparation does not end with graduation. It is an ongoing process which normally leads to advancement in one’s career. Without it, the half-life of an electrical engineer is less than five years. Creativity and innovation are central to the more esoteric aspects of engineering. But what are they? Creativity is artistic or intellectual inventiveness which arises from imagination. Children often conjure up entire armies or invaders from outer space with cardboard boxes and blocks—that is, until adults sever that imagination by forcing them into their molds. Creativity is a very personal and individual thing. It is the spark, the light bulb, the new idea. It strikes at uncommon places and at uncommon times—for example, when one is sitting in church or sleeping.

PROFESSIONAL PREPARATION

9

Innovation is a new concept—method, custom, device, process, or product. It is a change in the way of doing things. It is the entire scenario from creating a new idea through development, production, and use of the product or process. Creativity is usually required in the early stages of this scenario, and often new sparks occur along the way when creative people are involved. But ideas are not enough; they must be good ideas. Benjamin Disraeli said about a colleague, “He had only one idea, and that was wrong.” We do not think in a vacuum. Even the most abstract ideas are influenced by one’s environment. Research. A pundit has characterized research as “looking in a dark room for a black cat which is probably not there.” Research demands detail; concentration on current technological developments; and being able to extrapolate beyond current products, processes, analytical techniques, and common approaches. The researcher must be well grounded in the basic sciences and mathematics. Research is often perceived as the glamorous function in engineering since the practitioner is always “on the edge, pushing on the boundaries.” Most researchers enter their positions after obtaining a graduate degree, or degrees, which prepare them for specialization. They often delve into the nature of matter and investigate the behavior of materials when subjected to a variety of forces and environments. Their work frequently overlaps with that of scientists. Stated another way, their function is close to the science end of the technological spectrum. An engineer conducting research is expected to be acquainted with the most advanced analytical tools— mathematical techniques and computer codes, for example. The objective of research is to discover truths and find a practical use for them. Oftentimes there are surprises, such as the unexpected making of nylon, super-glue, and a variety of other products which “popped up” while the researcher was working on other matters. Research engineers, more than in any other function, must remain abreast with the developments in their areas. That requires reading (a) the current research journals, (b) key dissertations and theses produced in graduate schools, (c) papers presented in national and international meetings, and (d) specialized monographs. It requires attending forums, workshops, seminars, and short courses within the research area. Contact with key colleagues, within one’s own organization and without, is imperative. Except in proprietary cases, it is usually to the practitioner’s advantage to share expertise with colleagues, taking advantage of synergism. Reviewing papers prior to publication and proposals before awarding helps keep one up-to-date. Innovations may need to be patented, which requires extra detailed efforts but which are worthwhile in building a personal resume and in furthering one’s organization’s interest. To stay at the forefront, it is often necessary for a research engineer to take graduate courses which have developed since getting his/her degree(s). Such specialized courses are often offered at a time and place convenient to the practitioner since graduate schools and employers cooperate in the interest of both parties. Employers often pay for such specialized education, but the serious practitioner will not let reimbursement policies stand in the way of getting the needed updating. Professional preparation for research includes developing patience and learning to cope with failure since much of the work involves trial and error and painstaking details. The researcher must be very perceptive, latching onto details that have previously been ignored or unseen. Oftentimes, the final result is far different from that envisioned from the outset. Development. The development engineer takes the natural phenomena and concepts discovered or formulated by the researcher and “develops” processes and devices that utilize those phenomena and concepts. Development often overlaps with research so much so that it is common to see Research and Development (R & D) departments. Engineering development involves the generation of models that may be used in pilot processes or techniques. Innovation and creativity are central to the attributes needed by the development engineer. Development engineers meld a clear understanding of the fundaments of basic science and cleverness in making things work. They are ideally tinkerers who enjoys “bread-boarding things together.”

10

PROFESSIONAL PREPARATION

Development engineers’ efforts involves layout, building, and testing of the experimental models which will lead to the ultimate product or process. They are always interested in producing a process or product that will function commercially. The product or process must work, be economically feasible, and meet specified objectives. While the researcher deals most often with unknown factors, the development engineer frequently uses existing products and processes to perform new and different functions. The interlinking of products and processes is important in development. Patents are more common in development than in research. In professional preparation, the development engineer must amass a mental library of techniques, processes, mechanisms, and components whose functions are well known. They must be such that they can be brought together into a new functional relationship. A control circuit may trigger a laser in intergalactic mapping as easily as it switches on a water pump in a fossil fuel power plant. Knowing the capabilities and limitations of the library of elementary units is, therefore, an absolute necessity if they are to benefit the development engineers in their quests for a “better mousetrap.” Since we never want to “reinvent the wheel,” searching the literature for the right component in a new device is one of the steps in development. The practitioner must then be aware of the key journals and monographs which might provide clues to the new product or process. Manufacturers’ literature is also a major source of information to meet the objective(s). Nowadays, many data are available on the Internet or World Wide Web, which must be readily accessible to the development engineer. A systematic search of the literature may reveal existing units which may be integrated to meet the objective without substantial modification. Ideally, the integration should be done with cognizance of the economic benefits. The savvy developer is discriminating in drawing a line between modifying existing products and starting over. Sometimes mathematical analyses or computer models will facilitate making a decision in the development process. After dividing the anticipated new device or process into several integrated segments, the development engineer may then piece together a variety of solutions to reach the desired result. Modifications may be made in an individual segment without interfering with the other segments. When the system is in a workable state by engineers who understand its intricacy, it may be refined to be operable by less knowledgeable people in a safe manner. Development engineers often begin professional practice after getting an advanced degree(s). Their professional preparation closely parallels that of the research engineer. In many respects, the developer stands between the researcher and the design engineer but overlaps with both. Design. The definition of engineering design by the Engineering Accreditation Commission of ABET is an appropriate introduction to this engineering function (1): Engineering design is the process of devising a system, component, or process to meet desired needs. It is a decision-making process (often iterative), in which the basic sciences, mathematics, and engineering sciences are applied to convert resources optimally to meet a stated objective. Among the fundamental elements of the design process are the establishment of objectives and criteria, synthesis, analysis, construction, testing, and evaluation. The engineering design component of a curriculum must include at least some of the following features: development of student creativity, use of open-ended problems, development and use of design methodology, formulation of design problem statements and specifications, consideration of alternative solutions, feasibility considerations, and detailed system descriptions. Further, it is necessary to include a variety of realistic constraints such as economic factors, safety, reliability, aesthetics, ethics, and social impact. In general, design is characterized by efficiently bringing together components and current technology to result in a device or process which will benefit society. It normally entails using up-to-date analytical techniques, materials, and manufacturing processes. Form and function are often equally important.

PROFESSIONAL PREPARATION

11

The design engineer bridges the gap between the laboratory and the production process or construction. Design engineers probably need an even greater repertoire of procedures, processes, mechanisms, components, and subsystems than do development engineers. Their professional preparation necessitates cataloguing such entities so that they may be readily brought to bear on the solution of a new configuration or objective. In the most general sense, design engineers take the “bread-board innovation” from the developer and give form and function to it. They devise the shape, strength, and operational configuration of the machine or process, specifying dimensions, materials, and the intricacies necessary for manufacturing or construction. The design engineer must anticipate misuses of the product, failure mechanisms, and maintenance requirements. Professional preparation to become a design engineer requires the following: solid grounding in mathematics and basic sciences; an understanding of machine capabilities; the ability to “see” things that have never existed; an understanding of various materials; an appreciation of the environment; knowledge of codes and standards; application of safety principles within our litigious society; and familiarity with manufacturing and construction techniques. Most of these requirements necessitate (a) continued learning via reading design magazines, manuals, and trade journals, (b) keeping track of changing codes and regulations, (c) being sensitive to costs of producing a device or structure and for the maintenance of them, (d) knowing life cycles, and (e) environmental changes. One of the important attributes of a designer—the ability to see things that have never existed—appears to depend upon an innate ability, which must exist before being enhanced by education and training. Some people just visualize things better than others. The most esoteric researcher frequently has trouble seeing an innovative application. Engineers sometimes have trouble visualizing the flow of electrons without likening it to the flow of a fluid. In the arena of mass production, the cost per piece is very important, often making or breaking a product whose cost is within two mils of another one. Style, appearance, and expected life of the product are very important. A consumer product must appeal to a customer in order to be marketable and in order to survive the competition. In construction, the resulting design is for one building, bridge, or control system. While total cost is still an important factor, maintenance becomes even more important. The replacement cost and component life cycle are, therefore, crucial to the long-term acceptance of a design. Another design parameter, which is not yet spelled out in the ABET criteria for accreditation but which is becoming increasingly important, is disposability, or recycling. Its importance is very paramount where health factors are involved. There are some rational reasons for recycling: to reduce pollution (water and air), to save landfill space, to cut down depletion of natural resources, and to reduce cost. The benefits are both environmental and economic. Clearly, recycling requires judicious design of the products being manufactured so they will be inexpensive to recycle. In general, if a product is not biodegradable, it should be recyclable. The safety design parameter merits special attention in our litigious society. The basic tenet in engineering is that a danger must be “designed out” of machines, processes, and structures wherever possible. If the danger cannot be eliminated by design, such as at the point of operation in a machine, we must guard the danger region. If unable to totally guard, then we must warn users by signs and instructions. For example, a table saw will not function if the blade is totally inaccessible. The blade, therefore, must be guarded, but the material must come in contact with the blade in order to be cut, requiring exposure to the danger element. Warnings are then placed on the unit about the danger and its consequences. In engineering, we must always design systems that will “fail safe.” For example, if a tractor-trailer loses its air pressure on the highway, the trailer brakes automatically lock. Engineers design interlocks to prevent inadvertent injury. For example, a clothes dryer quits rotating when the door is opened. If an elevator cable snaps, letting the elevator begin to fall freely, a set of brakes grabs the ways in which the elevator runs, thereby stopping it quickly and gently. Automobiles have redundancy built into their brakes and accelerator linkage systems. These design schemes are a part of the engineer’s repertoire to prevent death, injury, and destruction

12

PROFESSIONAL PREPARATION

of property. In devising a fail-safe system, the design engineer must always keep in mind that only gravity can be depended upon in the event of a failure. Codes and standards play a major role in protecting our citizenry. No pressure vessel can be sold in the United States, and many foreign countries, without having the American Society of Mechanical Engineer’s (ASME) code stamped on it, signifying that it complies with rigorous safety standards. Water heaters must be equipped with code-approved pressure relief valves. Portable electric heaters must be equipped with switches which deactivate them if they are overturned. Lawn mowers are equipped with kill switches to reduce their propensity to cause injury. Electrical design engineers must be at the forefront in seeing that machines and structures are equipped with interlocks which prevent injury to users of them. As an important part of professional preparation, the design engineer must remain abreast of the pertinent regulations and standards, which are promulgated by hundreds of agencies and organizations. Regulations are requirements by governmental agencies. Examples include: (1) the Federal Motor Vehicle Safety Standards (FMVSS) issued by the US Department of Transportation and (2) Occupational Safety and Health Administration (OSHA) regulations by the US Department of Labor. Voluntary standards, promulgated and/or coordinated by the American National Standards Institute (ANSI), call for that which should be done by a prudent manufacturer or builder, and less frequently by the employer or user. Oftentimes, standards are adopted by governmental agencies, which gives them the force of law, making them regulations. For example, pertinent sections of the National Electrical Code, by the National Fire Protection Association (NFPA), have been adopted by all states to govern the safe installation of electrical wiring and components within building. The National Electrical Safety Code (ANSI C2), promulgated by The Institute of Electrical and Electronics Engineers (IEEE), governs transmission lines and associated equipment (outside of buildings) throughout the United States. Multiple designs exist for any objective. Optimization then becomes important to the design engineer. The optimization process must include not only quantifiable factors—such as current, voltage, and cost—but form factors—such as shape, color, and appeal. Evaluation methods are continually being upgraded to account for all of the factors of optimization. The up-to-date design engineer must remain current in these matters via participation in seminars and other specialized forums. Judgment eventually determines the final design. And judgment is the sum of one’s total experiences. Design engineers must be more attuned to people problems than either the research engineer or the development engineer. They must interface with those who manufacture the products or build the structure. And those who manufacture and build never hesitate to tell the designer all of the reasons why a thing cannot be done. The design engineers must then be persuasive, being well acquainted with their arguments before trying to “foist them on the old pros.” Since every design involves a departure from that which has been done before, it is imperative that the design engineer be creative and innovative. Production (Manufacturing). The conversion of raw materials into finished products is the objective of the production engineer. Cost of production is usually paramount. It is dependent upon the manufacturing process and the cost and flow of raw materials. The production engineer should interface directly with the design engineer while products are being designed, making suggestions for ease of production—selection of materials, processes and procedures. Such collaborative efforts are sometimes referred to as concurrent engineering and take advantage of total quality management principles. Frequently a small change in design will significantly facilitate manufacturing and assembly, thereby reducing cost. For example, loosening a tolerance from 0.01 mm to 0.03 mm may reduce production costs by 20%. Production connotes manufacturing, which may be quite varied depending upon the materials being used and the processes selected. For example, the manufacturing of plastics components is significantly different from making parts of metals. A plastic blow molding operation has very little similarity to a sand casting

PROFESSIONAL PREPARATION

13

operation. The production engineer must be thoroughly acquainted with the capabilities and limitations of the machines which are utilized in manufacturing. Electrical engineers are often needed in planning for energy requirements in manufacturing, which may involve short-term bursts of power. Cycling and timing are generally intimate to processing. The emphasis of the production engineer is on how to manufacture the product after the design engineer has defined what to produce. The technical aspects of manufacturing include: flow of the product; distribution of energy (electric, gas, hydraulic, pneumatic); forming of components; specialized treatment of components; timing of subassemblies in the flow; in-line testing of components and subassemblies; inspection and rework; packaging; and shipping. The production engineer is responsible for selecting tools and processes to be used in the manufacturing process. The production team consists of engineers, technicians, craftsmen and assembly line workers. Their expertise ranges from highly specialized to very limited, requiring the production engineer to have a special knack for placing team members in the best niche for optimized results. The more difficult part of manufacturing may be scheduling and staffing. Startup and shutdown of assembly lines become important in shift scheduling. The most difficult of all is the relationship with people, but it may be the most rewarding. Assembly lines are so often staffed by lower paid workers that training become a very crucial aspect of manufacturing. Training is normally a joint responsibility of the production engineer and the manager of human resources. Many of the problems associated with production are not covered in engineering schools. Most manufacturing operations are not organized under engineering but rather under management, which requires the production engineer to interface with sometimes competing forces. Professional preparation must, therefore, include attending specialized seminars and in some instances proprietary workshops conducted by the providers of assembly line equipment. Specialized training in production processes, economic analysis, and quality control is often a necessity. Ideally, the production engineer should work closely with the sales staff. This is especially important since the production engineer operates under constant pressure and deals with machines that are often pushed to the limits of their capability. Sales wants more output; maintenance wants more time; some materials are not available and substitutes are necessary; and the design engineer may want to make a last-minute change in the product. Life is not easy for the production engineer! In every decision, the production engineer must make the product for the lowest possible cost. This necessitates professional preparation with a strong background in engineering design, economics, business, and psychology—some of which must come through continuing education and professional development activities. Construction. The primary difference between production and construction is in the quantity of the end product. The production engineer may turn out millions of components per month, whereas the construction engineer may spend many months on a single product. As used here, the term construction engineering means the work of building structures that have been designed previously and will be operated subsequently. The structure may be a building, a dam, a bridge, or an elaborate control system for a nuclear power plant. Scheduling of tasks and utilization of people are key aspects of construction. In most projects, the construction engineers will have been associated with the bidding on the contract—time requirement and people needed—making them cognizant of detailed costs. The flow of construction materials is very important since every delivery may put them in the way, and late delivery may require layoffs of people and missed target dates. The construction engineer must be able to relate to construction craftsmen, deal with labor unions, and interface with subcontractors. Professional preparation for the construction engineer includes: a strong background in design (structures and components), economics, business matters, labor relations, and interactions with people. Interactions with subcontractors and craftsmen—carpenters, pipe fitters, electricians, plumbers, and so on—are often crucial to completing a project on schedule and within cost.

14

PROFESSIONAL PREPARATION

The most difficult problems of construction engineers require a strong technical background, but their ability to work with people generally decides success or failure of a project. The ultimate capability of the construction engineer requires experience and judgment which can only be obtained by actual practice. Operations and Maintenance. After a structure—building, dam, bridge, or system—has been completed, it must be maintained and operated by someone with unique capabilities. Such a person may be called plant engineer or operations engineer. Operations include: maintenance of the structure, grounds, equipment and utilities; setting up and regulating facility equipment and automated machinery; optimizing operating costs; keeping the environment acceptable; and phasing in new equipment when needed. Professional preparation of the plant engineers, or operations engineers, includes a broad background in the functions of engineering, ideally in most of the engineering disciplines. They must deal with operating personnel, vendors, community leaders, and other professionals (lawyers and doctors, for example). Often regulations imposed by governmental agencies or an internal board of directors may require significant time and effort. Plant engineers plan new facilities; supervise construction if done in-house; select and install equipment; allocate space; and are responsible for the overall safety in the facility. They usually have crews of craftsmen and technicians to carry out the needed functions. Operating cost containment is generally a crucial role of the plant engineer. This routinely entails monitoring utility costs and energy losses from the structure; maintaining the environment (lighting, heating, ventilation, air-conditioning) within acceptable standards; and protecting the facility and operating personnel. Successful operations engineers are not so much dependent upon esoteric mathematical concepts and scientific principles as they are on being able to optimize the output of machines and the input of operating personnel. The operations engineer is analogous to the conductor of a symphony, making sure that each person carries out the assigned tasks. On-the job training should be supplemented in professional preparation by knowledge of labor relations, psychology, economics, and interactions with people. In general, the plant engineer is responsible for the smooth operation of the facility, including the safety of personnel. Testing and Instrumentation. Some would argue that testing is not a separate engineering function since it is done at all levels of research, development, design, and production, sometimes even in operations. It is so important, however, that it deserves separate note. During research, the testing that is done falls into the area of experiments to examine the validity of hypotheses or wild ideas. At the development stage, it is usually necessary to test prototypes of components and subsystems as well as new processes. Clearly, a product must be tested periodically during its manufacture in order to assure its quality. The responsibility for conducting some of the tests in the engineering spectrum rests with test engineers. They must be able to establish the test protocol, setting the test parameters as well as the quantities to be measured. Statistical standards are important to the test engineer, requiring that an adequate number of tests be conducted in each sequence of data collected. The data must then be analyzed with appropriate conclusions drawn. Quality assurance (QA) engineers are responsible for the quality control (QC) in a manufacturing or construction operation. The requirements involve periodic inspections in addition to maintaining adequate control procedures for assuring that the quality sensed at periodic junctures is held fast. Professional preparation for the test engineer includes: a sound background in the basic sciences and the engineering sciences; keeping current in measurement techniques; a strong base in probability and statistics; knowledge of calibration methods; and familiarity with pertinent codes and standards which govern the products and/or structures and the test procedures. Test engineers are often called upon to specify the instrumentation that is required in manufacturing operations, not just in the testing and sampling processes. They are normally better able to provide instrumentation details than the research, development, design, or production engineers whose responsibilities are much broader but less attuned to the intricacies of measurements and control.

PROFESSIONAL PREPARATION

15

Sales and Marketing. At the outset, it should be noted that all other engineers usually claim that sales engineers promise more than they can deliver. For this reason, it is important that those who are in marketing keep in touch with their colleagues that are producing the things they sell. In order to maintain personal and organizational integrity, the sales engineer should not sell anything that is not useful to the customer. The sales engineer is expected to have a broad engineering knowledge, be intimately acquainted with the product that is being sold, and understand the needs of the customers. Contrary to some pundits, being a responsible sales engineer is not always easy. It requires being alert to anticipated changes in models, government regulations, and new materials which might be used to advantage. It requires knowing the competitors’ products about as well as one’s own. The responsible sales engineer might even sell a competitor’s product on occasion. This brings up the concept of being a representative of several manufacturers, complementary or otherwise. This is often done in the case of after-market parts and components where replacement units may be built by a number of manufacturers, say in motor vehicles and farm machinery. The manufacturers’ representative needs an even broader background than a sales engineer, who deals only with proprietary products. Professional preparation for the sales engineer is first and foremost being equipped to work well with people. This requires a background in the social sciences, psychology, and business as well as a clear understanding of the fundamentals of engineering. Sales and marketing seminars and workshops are a must if one is to keep up with the competition. Anything that can be done to improve one’s personality and confidence is a plus. Speech courses and leadership conferences usually help. But most of all, the sales engineer must enjoy people and believe in the product that is being sold in order to succeed. Management. The management function of engineering may be viewed with respect to the other functions by considering their roles (5). Research determines if a thing can be done; development determines what can be done; design and production determine how it will be done. But management must decide whether to do it at all. The best overall attributes of all functions of engineering are brought to bear in management. Management is always crucial to the success of research, development, design, and production or construction. Since the managers at this level must have strong technical abilities, it is logical then that managers arise from the functions of their expertise. As they advance in management activities, they usually become less involved in details of technical matters. Their attention is diverted to the overall operation of the organization, as well as to matters that involve all of the engineering functions. In the past, many managers—including those who may be executives—were selected with backgrounds in business, law, or the social sciences. Today there is an increasing trend in industry and government to choose managers who have an engineering background, perhaps because of the increased complexity of our products and processes. This is especially true in industries which are involved in advanced technology and large volume manufacturing. The principal advantage that engineers have over professionals from other backgrounds is their methodical approach to problem solving. No one knows better than an engineer that a problem cannot be solved until it is well-defined, even if nebulous assumptions must be made at the outset. The basic principles of management are the same without regard to the operation, whether the company is building microprocessors or drilling for oil. The objective is to use the resources of the organization to the best advantage in a competitive economy. The company resources will depend upon the nature of the business. The engineer in management must continually make decisions in the optimal use of facilities, including equipment, capital, and personnel. The decisions may be touchy on occasion—for example, when a decision is made to down-size staff in order to buy and utilize a sophisticated automated machine. Such decisions are often made on the basis of return to the stockholders or taxpayers. Most managers that come from engineering need additional training in the business aspects of the operation. Development of the needed skills may be facilitated via Saturday management programs at local universities or by specialized short courses, say in cost accounting. It is generally agreed, however, that such deficiencies can always be augmented more easily than a person with a nontechnical background can learn enough about the technology of the operation.

16

PROFESSIONAL PREPARATION

Clearly, the manager must have a variety of blended skills from technology and business. This is especially true in the aerospace and electronics industries which are highly automated or computer intensive. One bright spot for the electrical engineer is that microprocessors drive the world! The professional preparation of an engineer in management parallels that of other engineers in the fundamentals of engineering. Having an aptitude for working with people is the attribute that commonly leads to an engineer being selected for management. Being able to sell oneself and the organization’s product or service is central to success. The manager must be able to lead teams of multitalented personnel, accepting many of their nuances and channeling them into a cohesive unit. Successful managers must care for people, be sensitive to their personal welfare, and share their successes without placing themselves in an intellectually superior position. Any engineer who has the ability to apply engineering principles in the supervision of large numbers of people while they expend huge sums of money can qualify to be a manager of an enterprise. This is without regard to one’s individual discipline or engineering function background. In addition to being responsible to the stockholders and organizational personnel, managers are also responsible to the general public. It is not acceptable to make more money for the stockholders if the organization pollutes adjacent streams or the air in the community. Since managers normally speak for the organization, they must present a good appearance and appeal to a variety of publics. Their written and verbal communications skills must be excellent in order to succeed. Nowadays, the way one projects on television is very important in some organizations. Managers have the ultimate responsibility for (a) recruiting and training personnel, (b) leading in the formulation of the objectives of the organization, (c) the acquisition of materials and equipment, and (d) optimizing the organization’s capital. They will be dependent upon key personnel for input required in the decision making process. This dependency determines how the business is organized, who reports to whom, and so on. Systems. The concept of systems engineering has come to the fore in recent years. Notably in the space program, all engineering disciplines and functions had to be brought together systematically in order to achieve the results of an extensive project—reaching the moon and beyond. The concept of systems engineering can be thought of as a super management tool. There are no new principles—only methodical checks and balances, schemes for bringing experts together, and optimization via tradeoffs. Teaching. Teaching is a unique function of engineering. In a way it subsumes all of the other functions of engineering. More than in any other engineering function, teaching offers the practitioner the best opportunity to continue learning and influence newcomers to the profession. In order to be successful, engineering teachers must remain at the forefront, keep up with all of the newest developments, and be able to translate them to a variant audience. Today’s university faculty members are more than transmitters of knowledge. They are intimately involved in the acquisition of knowledge via research and making it applicable by development in the laboratory. Faculty members’ research leads to the publications which are generally the most esoteric and universally applicable throughout the technological spectrum. That this is true derives from academic freedom and not being bound by proprietary interests. Fewer patents are granted to faculty members, however, than to industrial practitioners since patents normally arise from proprietary interests. Engineers who teach have a greater freedom to select their areas of concentration than other engineers. In general, engineering teachers select their own research areas for investigation, establish their own standards of performance, and enjoy greater flexibility than other practitioners. These freedoms require an unusual measure of initiative, personal esteem, enthusiasm, and perseverance. Teachers commonly rub shoulders with other professionals who are experts within their own areas, which provides fodder for further learning and motivation to emulate the best.

PROFESSIONAL PREPARATION

17

To be a successful teacher requires mastery of one’s subject and the ability to communicate. Engineering teachers usually come from the top quartile of their classes. They can do anything. The pundit who says that “those who can’t do anything else, teach” is just wrong! Professional preparation for teaching requires an advanced degree(s). The doctorate—Doctor of Philosophy, Doctor of Science, Doctor of Engineering—is now considered the required degree for one entering teaching. This is based upon the special competence that comes from original research done at an advanced level, which only comes after developing an extraordinary measure of personal discipline and perseverance. Ideally, one who goes into teaching should be acquainted with educational methods and have a mastery of communication skills. Engineering teachers routinely participate in two other functions of engineering: research and consulting. Research is their vehicle for enlarging the borders of their areas of interest and expertise. It also provides an avenue for publication, which is now expected of those who teach in engineering schools. And it provides a way for those with common interests to come together in symposia and forums to share their findings with others. The expertise developed in their push to excel then makes them attractive as consultants to industry and commerce as solvers of unique and intractable problems. Consulting. Consulting engineers are recognized as such by their ability to solve difficult problems. They are often called when others have failed to solve existing problems. They generally act as a source of information for a variety of industrial and governmental organizations. The connotation of the title consultant suggests specific skills or expertise brought about by extensive education and experience. A new graduate is rarely asked to be a consultant! Industries seek consultants to design and develop new and unique products and processes and to solve problems that their own personnel have not been able to handle. Some governmental agencies seek consultants to ply their expertise to specialized tasks and to bring an outside view to projects. The legal profession is increasingly using engineering consultants to investigate failures of machines and processes and to testify in the courts in resolving conflicts that have resulted in litigation—commonly in products liability matters, personal injury, patent infringement, and business practices. In every consulting assignment, the consultant is expected to quickly evaluate the matter, estimate the effort required to remedy the problem, and offer solutions which may solve the problems or resolve the issues. Recommendations are often sought which provide alternatives for managers. In some cases, financial implications have to be factored into the proposed solutions. Engineering consulting usually begins individually, although there are a number of companies which specialize in offering consulting services. Some consulting firms offer a wide variety of services—including design and construction of buildings and public projects; water treatment plants; airports, chemical facilities; oil refineries; and the operation of private ventures or public services. Profession preparation to be an engineering consultant requires development of an extraordinary expertise. The best preparation is being well-grounded in the fundamentals of engineering and having the ability to extrapolate those basics to a variety of circumstances. Consultants are rarely called upon to solve problems that they have seen before. The problems are unique or the circumstances differ, requiring that fundamental principles be brought to bear in reaching a solution. A most important attribute of the successful consultant is communication skills, written and verbal. Most consulting projects require a written report in addition to verbally reporting to those who have engaged the consultant. Clear thinking, while applying engineering basics, and verbal skills are especially important in consulting in the legal arena. The choice of words and the intonation of the presentation may influence the judge or jury in a key point of a trial. Successful consultants must be adept at business. Their livelihood depends upon selling their services and assuring the financial success of their work. The best preparation for being a consultant is knowledge and experience, knowledge and experience, knowledge and experience!

18

PROFESSIONAL PREPARATION

Fig. 2. Functions of engineering. A qualitative representation of how the functions of engineering relate to the Principles of Science and Engineering (top abscissa, heavy on technical aspects) and to Economics and Interactions with People (bottom abscissa, heavy on societal interactions).

Other. Graduates of engineering schools sometimes go into other professions. Business gets the largest percentage of those who step across the border from engineering with a number pursuing Master of Business Administration (MBA) degrees. The MBA is sometimes viewed as a shortcut in preparing those engineers who want to go into management. The next larger number of engineering graduates go into law. Their engineering backgrounds serve them well in the logic that is necessary in the preparation of cases. Engineering fundamentals are a tremendous benefit to attorneys who practice in patents, personal injury, and products liability. Engineering graduates are increasingly going into medicine. This is a boon for the area of biomedical engineering. In this area, engineers are better equipped to make major contributions to medical instrumentation, prosthetic devices, and biofeedback systems. Their knowledge of materials, motion, and structures augments that learned in medical school, facilitating their treatment of special maladies. Engineers are also going into banking and commerce more than in the past. Their backgrounds make them well-suited to evaluate business ventures, especially those entailing technical matters. In short, engineering education and experience equip individuals for any endeavor in life. The ordered approach to identifying problems and solving them in a methodical fashion contributes to success in any career.

Summary The engineer is a problem identifier and solver, whether technical, socioeconomic or otherwise. The problems range from highly technical research, dependent upon theoretical principles and creativity, to the less technical, more people-oriented management challenges which are centered on people and capital. Figure 2 illustrates the functions of engineering, in a qualitative fashion, as they relate to engineering principles and business

PROFESSIONAL PREPARATION

19

aspects of the engineering spectrum. Note that the figure depicts the bridging functions of systems, testing, teaching, and consulting as spanning the other functions. Professional preparation for engineers in all functions of engineering have a common beginning to career development—the basic engineering degree. But that is only the beginning, as suggested by the term commencement applied to the graduation ceremony. The engineer must continue to learn via professional development activities—additional formal education, short courses, membership in professional/technical societies, workshops, professional conferences, and other means of education and training. But the best preparation is personal study and development via interfacing with others who know more than ourselves. The spectrum of activities in the engineering profession is very broad. Engineering touches every other aspect of life. Electrical engineering is crucial to all engineering disciplines and to every engineering function.

BIBLIOGRAPHY 1. Accreditation Board for Engineering and Technology, Criteria for Accrediting Programs in Engineering in the United States, Baltimore, MD, 1996–1997. 2. G. Maslow, Motivation and Personality, New York: Harper and Row, 1970. 3. G. C. Beakley, D. L. Evans, J. B. Keats, Engineering: An Introduction to a Creative Profession, 5th ed., New York: Macmillan, 1986. 4. S. G. Walesch, Engineering Your Future: Launching a Successful Entry-Level Technical Career in Today’s Business Environment, Englewood Cliffs, NJ: Prentice-Hall, 1995. 5. J. C. Martin, The Successful Engineer, New York: McGraw-Hill, 1993. 6. R. B. Landis, Studying Engineering: A Road Map to a Rewarding Career, Burbank, CA: Discovery Press, 1995.

LEIGHTON E. SISSOM Sissom & Associates, Inc.