PARTICIPATORY TEACHING METHODS IN COMPUTER SCIENCE James S. Jones Computer Science Department Graceland College Lamoni, IA 50140 jsjones@graceland.edu ABSTRACT This paper discusses the role of participatory teaching methods in the computer science classroom. The list of these methods includes brainstorming, directed dialogues, small discussion groups, role playing, games, debates, panel discussions, and Socratic dialogues. The author has used such methods in Computers and Society classes and to a limited degree in Compiler Design, Computer Architecture and Operating Systems classes and believes that such techniques have a place in the computer science classroom. It is the aim of this paper to provide an overview of participatory teaching methods and their use in computer science education. INTRODUCTION Participatory teaching methods are those which draw the student into the classroom learning process. The student becomes a participant who articulates in some way what is learned. It is process-oriented teaching. Didactic teaching methods are those which impart content or information to the student as is done in a lecture. It is content-oriented teaching. The former is interactive, the later is batch. Computer science is a process oriented discipline. This is evident to the student writing that first computer program as well as the practitioner planning and adjusting for the next software/hardware release. We are fortunate to be educators in such a discipline because process learning, which is experiential in nature, is so much more satisfying and memorable than content learning. The very appeal of Tracy Kidder's Pulitzer Prize winning Soul of a New Machine was that it transported the reader right into the process of a computer design, complete with human relationships, struggles and decisions. As our discipline matures and the body of content to be taught grows there is a danger that our teaching will similarly become more content oriented with student assignments taking on a narrow focus in order to illustrate specific principles. The process-oriented learning experience could shift out of the academic environment and into the workplace altogether. Participatory teaching methods may help by developing students with greater depth and breadth of thought, willing and able to articulate a position and make compromises in the face of other viewpoints. Interaction increases teacher awareness of what is actually being taught. Classes become more interesting and motivating. The whole educational process becomes more appealing. Our challenge is to find points within computer science courses where such methods will be appropriate and effective. The difficulty will be in sacrificing some content in order to do this. BACKGROUND Participatory techniques are certainly not new to educators. Socrates was perhaps the original teacher employing participatory methods and we sometimes refer to his style of teaching as Socratic. John Dewey, American philosopher and educational theorist, carefully laid out a theory of experiential learning in education early in the twentieth century. Research in the 70's, first by Carroll and later by Newstrom, on the relative effectiveness of training methods showed that case studies, business games, and role playing techniques were far superior to lectures and films when the aim is teaching problem solving skills. And yet, didactic teaching seems to be most prominent, perhaps because it is easier for which to prepare, to write textbooks, and to do. Although participatory techniques are more likely to be found in use by those teaching humanities, there has been some interest expressed recently for using such techniques in the sciences. In teaching the Computers and Society course at Graceland College it became clear to me that topics such as freedom of information vs. privacy, computer ethics, and the impact of computers on society could best be taught in such a manner so I turned to the psychology, history, speech, and education faculty at the college who were successfully using various participatory methods. I owe much to them for getting me started. I began experimenting in the Computers and Society classes and later in the upper division Compiler Design, Computer Architecture and Operating Systems classes. After several years of introducing participatory methods in my classes I have observed that the students are more interested and engaged in the material, give higher ratings to the classes on evaluations, and write more meaningful responses to essay questions on tests. Although most of my teaching is still in a lecture format I will continue to move toward more participatory classes. PARTICIPATORY METHODS What follows is a list of participatory teaching methods and the way they have or could be used in the computer science classroom. It is not meant to be exhaustive and there is some overlapping of concepts. Brainstorming In a brainstorming session a problem or concern is presented to the class and then all are invited to freely think about solutions or possible answers and call them out so that they can be listed for all to see. It is a fast paced, spontaneous session in which a mass of ideas are collected without the merit of any being considered when first given. Students are encouraged to piggyback on other ideas and not to think of any idea as their own. Afterwards the ideas can be organized and evaluated in class. I have used brainstorming in Computers and Society to have the class list the possible effects, good and bad, that computers will have on crime, transportation, cities, family life, etc. I have had the class try to list all the purposes that they can think of for a computer and then organize those ideas into central themes such as speed, memory, and so on, prior to a discussion of what makes the computer such an important tool. Brainstorming has been useful prior to assigning a class paper or prior to preparation by students for class debates and panel discussions. Having the students list possible future inventions using embedded microprocessors can provide a foundation for a summary lecture on microprocessors, their limits and their strengths. In Operating Systems I have had the class list the possible objectives that an operating systems designer could have when designing an operating system [see the lists in exercises 10.4 & 10.5 of Deitel] and then organize them according to what audience such an objective appeals to (users, programmers, systems engineers, administrators, etc). A discussion of which objectives are in conflict with each other followed to underscore trade-off issues. In Computer Architecture I have students list all possible categories for classifying a computer (word length, bus widths, hardware technology, speed, etc). Following that they are presented with Flynn's, Shore's and other taxonomies, their strengths and weaknesses. In an A.I. class students could list possible expert system projects with instruction to follow on problems suitable for expert systems. Numerous upper division possibilities exist for brainstorming sessions. Directed Dialogues A directed dialogue is a goal directed session where the instructor directs the class toward a particular solution to a problem. A problem is presented and class members respond with what are usually partially correct answers. The instructor continues by presenting the overlooked dimensions of the problem and the students further refine their answers. Such a class dialogue continues until the goal is reached. It is as if the instructor is constantly placing roadblocks before the students in order to keep them on a single path. The students will know they are being manipulated so only certain topics will work otherwise they will quietly wait out the instructor to give the answer. Similarly, overuse of the technique will render it ineffective. Perhaps the best subjects are those in which there was an actual historical trend in solving some problem from poor to better to best solution, a trend that can be recaptured in class. The students must feel that the solutions that pop into their heads are just what the instructor is looking for or they will not share it. The atmosphere created is similar to one where a puzzle or brain teaser is given to a group to consider. The most effective demonstration of this technique was one that I witnessed in an Algorithms course in which the instructor preceded a lecture on sorting algorithms with a directed dialogue that began with the question "if you had a million papers to sort how would you do it?" and finished with the class discovering on its own some divide and conquer approaches, not unlike Quick Sort. The instructor's "roadblocks" kept the class on the path of common sense and away from programmer logic ("would you really do it that way?"). I have often felt that Deitel's development of the various algorithms for mutual exclusion primitives leading up to Dekker's algorithm is a promising topic for this technique [see Deitel, pp. 80-88]. In Computers and Society I lead a directed dialogue having students construct and refine a definition for computer. I do a similar session for the definition of operating system in my Operating Systems class. In both cases the students give responses which are usually incomplete and in need of refinement. The lectures I enjoy giving most seem more like directed dialogues that I have with myself where I present a problem, give a possible solution, identify problems with that, and so on until I give the final solution. I am reconsidering those lectures to see if they can be more effective with student interaction. Small Group Discussions The dividing of a class into small discussion groups seems almost too obvious to list. However, it is perhaps the easiest of the methods to use with success and it requires very little planning or preparation. I have frequently used the method in a spontaneous way when I wanted the students to consider some question that had several possible answers. I simply tell the class to divide in groups of 4 or 5 and to choose a spokesperson who will record the group's response to a particular question and then report back to the class. Normally, when such questions are given to the class as a whole only a few assertive students participate in the discussion with the rest taking notes (or naps). Small groups force more students to engage themselves in the subject and to take on leadership roles in the discussion. Since each group feels a certain ownership for it's answer due to the time invested in coming up with it, there are stronger feelings and a greater willingness to debate the merits of the various answers. In a Computer Architecture course teams could be given pieces of information about a machine (for instance, several assembly language instructions and their respective execution time formulas) and then called on to infer all they can about the underlying machine's design, making logical assumptions if necessary. Several teams could then be joined to negotiate a single response to the question by discarding the least likely ideas. Role Playing/Drama Role playing is simulating someone else's position. My use of role playing in class is coupled with debates, panel discussions, and Socratic dialogues which are discussed later. In Computers and Society I assign various roles to students for a panel discussion on such topics as freedom of information vs. privacy or professional ethics and the students must research those viewpoints enough to play out that role. On essay tests I frequently ask students to put themselves into the role of the designer, user, or administrator when answering various questions. Perhaps such questions could be turned into classroom role playing exercises. Students could simulate a development team doing a walk-thru of each other's designs or code in class. One possibility for an introductory computer science course is to have students simulate actual machine components such as AND, OR and NOT gates with white and black marbles for data bits and arrange them in such a way as to build a 2-bit adder. During such a demonstration the instructor could easily point out the importance of the clock and it's speed, the concept of levels in a circuit design, the presence of a power source, and the virtual nature of computation from switching circuits. Role playing is an extemporaneous form of drama. A more structured form would be to have students write and perform some dramatic piece. A history professor at our college has had students select one of a variety of things as a term project, one of which is a dramatic interpretation of an interview with a famous person such as Ben Franklin or Abe Lincoln following extensive research and a basic theme underlying the interview questions. Having students create and perform a dramatization of an interview with the Countess Lovelace or an imaginary 1940 meeting between Howard Aiken and Konrad Zuse might be interesting and educational but I have not as yet been willing to try formal drama in the class. I am certain that more learning takes place with the students who do the research and prepare the dramatic presentation than with the other students who make up the audience. If only a few students are involved then one might ask how this is any more effective than more didactic methods such as student oral presentations or lectures. In any case, creative- minded faculty may wish to give more attention to drama as a possible teaching method. Games Games arouse interest and are highly interactive. Some games are appropriate for use in class as an instructional method. I have used a game called QUERIES 'N THEORIES in my Compiler Design class prior to a lecture on formal grammars. Some compiler texts refer to a "syntactic dominoes" game which could be constructed and played out in class as an introduction to parsing. Artificial Intelligence courses provide a natural environment for game playing exercises. If symbolic logic is being taught somewhere in a computer curriculum then WFF 'N PROOF is an excellent game that can be used in small classes or groups. The instructor could create some game-like exercises from LOGO or Karol the Robot languages to introduce students to programming logic. Programming contests could be implemented as a class activity rather than just an extracurricular event. A trivial pursuit type of game could be constructed for content-intensive material (as in a computer literacy course). Such a game could even be played by teams repetitively with team members being changed each time to remove single-player influence and then the game scores could be used as one measure of a student's performance. Panel Discussions At professional conferences it is often the panel discussion sessions that draw in the larger audiences. Panels are interesting because they are used for airing different sides of an issue -- and that means controversy. Conflict attracts interest and participation. This is especially true in panels that give ample opportunity for audience participation. Panels will become didactic in nature if the panelists each give lengthy statements of position leaving little time for interaction between panelists or with the audience. Panelists should give a brief statement of their response to some problem or issue. Clarifying questions could follow. Finally time is allowed for interactive discussion and debate between panelists and perhaps with the audience. Questions should be directed to specific individuals unless a moderator distributes the questions amongst the panelists. The instructor could be the moderator, however, if the panelists are to be graded then the instructor may need to tape record the session or be an outside observer. Panelists could be required to turn in a paper or outline showing their research and anticipation of opposing points. Panel discussions can also be fairly structured like a formal debate. My use of panel discussions in class have been more like a debate or like a role-playing panel engaged in a Socratic dialogue (both discussed later). Any controversial issue that is appropriate for debate can be handled less formally in a panel discussion. Controversial issues which clearly have more than two sides may be more appropriate for a panel discussion than a debate. Case studies can also be reviewed in a panel discussion as an alternative to using Socratic dialogues. Debates In a debate, debate teams test opposing sides of an issue. A conclusive proposition statement must be formulated giving rise to two teams, one for the proposition and one against it. Students in class may have had debate experience elsewhere and be logical choices as team captains. I am always amazed by the amount of effort and research that the teams put into a debate. In Computers and Society I have had student teams debate the proposition that "electronic democracy, even if taken to the extreme, would be beneficial to our nation". A lively debate could be had on any of the following propositions: "total freedom of information would be good for humanity", "software should be patentable", "firmware (microcode) is software, not hardware, and the courts will see it that way too", "hackers are not criminals", "companies should allow employees reasonable personal use of desktop telephones and computers", "companies have a right to monitor telephone traffic over company leased lines to the fullest extent", "violent crime will decline as EFT expands", "the computer age will cause people to interact more and draw closer together", etc. Since trade offs are inescapable in considering hardware or software design decisions, there may be opportunities for debates in upper division courses. How about a simulated debate between hardware design team members split into "two warring factions, one advocating making the 8-bit byte the basic unit of memory, and the other advocating the 32-bit word as the basic unit of memory" [see Tanenbaum, 2nd ed., p. 202]. Socratic Dialogues The Socratic dialogue is a dialogue between instructor and students in which the instructor plays the role of the inquiring teacher who asks leading questions and probes the answers given by asking more questions of the student. The instructor thus brings out points of concern or conflict to further test the student's thoughtful consideration of the subject. The instructor does not lead the student to a particular answer as in the directed dialogue nor does the instructor give answers. The purpose is to insure that the complexities of an issue and the variety of possible solutions are understood. It is a process of exploration into the subject by both instructor, who acts as moderator, and students alike. I developed an interest in using this approach after observing several Public Broadcasting Service programs on television in which this technique was used to explore a variety of controversial issues with the likes of Harvard Law School professors Arthur Miller, Charles Nesson, and others as moderators of such a dialogue. The Media and Society Seminars of the Columbia School of Journalism in conjunction with several affiliate PBS stations have produced these programs including "The Constitution: That Delicate Balance" and "The Military and the News Media". These forums are really a mixture of panel discussion, role playing and debate with a moderator facilitating the Socratic nature of the seminar. I used such an approach in my Computers and Society class to explore the complexities of a professional ethics case. I assigned to participating students various panelist roles in a hypothetical case taken from Deborah Johnson's Professional Ethics text (programmer, immediate supervisor, top management, defense secretary, reporter, editor, etc). They were to research the likely positions they would take prior to the session. I was very pleased with the outcome. Prior to this approach, student responses to essay questions on professional ethics were much more naive and single minded than when this was introduced in my classes. I hope to do more of this in Computers and Society and perhaps in other courses if I see a feasible topic for doing that. LIMITATIONS AND CONCERNS There are a variety of limitations or problems to consider before implementing various participatory teaching methods. First, participatory techniques require more class time than do didactic techniques. One must be willing to sacrifice some of the time devoted to in- class content teaching. Additional responsibility will be placed on the students to read the text and dig out the content on their own and in their research. More class time will be devoted to helping students understand the complex interrelationships in which that content exists. In general, more preparation time is needed with participatory techniques because instructors must be prepared for the unexpected question or turn of events in such a class. The possibility for this dynamic is almost eliminated in a lecture. As with any technique, the more practice one gets using it the less preparation time it requires. Student evaluation becomes more subjective and difficult. Content- oriented classes lead to objective tests for evaluating students and are easier to grade whereas evaluation of panels and debates are more difficult. One could decide not to evaluate them at all, no more than one would evaluate student notetaking, but then the question of student motivation for participation in debates and panels needs to be addressed. I found it helpful to seek the advice of liberal arts faculty on how they made subjective evaluations of essays, papers, oral presentations, panels and debates. I have also found subjective grading less of a hand wringing experience as I have done it more. It does help if I remind myself that grading techniques never are perfect, that it's the overall GPA average that counts, and that life isn't always fair. Appropriate texts are hard to find. Most are content oriented. For Computers and Society I prefer texts that present varied points of views or ideas without lengthy content-oriented computer literacy sections. Computers in Society, by Kathryn Schellenberg (Ed.), The Information Technology Revolution, by Forester (Ed.), Future Mind, by Ed Lias, and Computer Ethics, by Deborah Johnson, are examples of either primary or reference texts that I have found useful for that course. On the other hand, upper division texts must be highly content specific. I have found that a careful reading of these texts and the exercises in them have led to creative in-class participatory opportunities. Class size and arrangement may be a problem. If the class is too large then not everyone can be a part of a debate team or panel discussion. Perhaps those students seeking bonus credit or an "A" in class will prove to be a manageable subset and the rest of the class will benefit as members of the participating audience. Not everyone in my classes of 40-50 Computers and Society students participate in a debate or panel because there are alternative projects or events to choose from. Small group discussions can be done in large classes but they may not all be able to report back to the class. Ideally classes are not arranged in rows of immovable chairs. I often work in such environments and would certainly not discontinue participatory teaching because of it. Are all students given equal opportunity to participate? When using participatory techniques it is too easy to engage in a dialogue with the more assertive or the more knowledgable students. One must develop strategies to see that all are participating. I have heard of studies suggesting that instructors often ask the male student more abstract or evaluative questions continuing with follow-up questions whereas they will ask the female student more content-oriented questions moving more quickly to the next student. Most instructors lack training in the use of these techniques. There are plenty of resources other than formal training. One could do as I did-- read about the different teaching methods, talk to other faculty who have used them or could list useful resources, day dream about how a particular technique can be tried in class (always a useful learning method), and then get some hands- on trial-and-error experience. I also found myself being a keen observer of the PBS program mentioned earlier regarding Socratic dialogues. Perhaps the greatest limitation that must be overcome is the fear and anxiety for venturing into the unknown. No one likes to fail or look like a fool and lectures are a nice security blanket because they create distance between instructor and students. The instructor is comfortably insulated from the student's views, not only about the subject but also about the teaching. Often one likes to feel well prepared and in control of the situation when trying something new. Unfortunately, one may never take that first step to try something new even if the desire is there because one may never feel perfectly ready. The solution to this is simply to decide to try these techniques, mention them as upcoming assignments to the class in the syllabus, make student assignments for panels and debates as if they are really going to occur, and then let it happen and learn from the experience. SUMMARY Although participatory teaching methods such as panel, debate, and Socratic discussions are not new to the field of education, they may be unknown or untried by computer science educators. Courses such as Computers and Society are particularly suited for such techniques and other upper division courses such as Operating Systems and Computer Architecture may also be suited for these techniques when complex analysis and design issues are raised. Other participatory techniques, such as role playing, brainstorming and games may be applicable as well. It is my view that such techniques increase student interest and motivation. The students enter into the "process" of thinking about the subject and articulate what is being learned with their peers. The disadvantage is that less time is devoted to the learning of facts in class but this is perhaps offset by the advantages gained by requiring students to read, understand and articulate the material on their own. The difficulties in choosing a participatory approach to teaching includes seeking appropriate texts, increasing preparation for classes, moving toward more subjective evaluation of students, and overcoming the apprehension one has to try the untried. These difficulties diminish with time and practice. I believe that the rewards are worth it. REFERENCES M. Adler (Ed.), "The Conduct of Seminars" (Ch. 1), The Paideia Program: Pointers and Prospects, Macmillan Publishing Co., 1984 L. Allen, P. Kugel and J. Ross, "QUERIES 'N THEORIES: The Science & Language Game", WFF 'N PROOF Publishers, 1111 Maple Ave., Turtle Creek, PA 15145, Copyright 1970 E. P. Bell, Jr., "Debating Controversial Issues", The History Teacher, Vol. 15, No. 2, pp. 207-223 (Feb. 1982) M. Buchmann, "Improving Education by Talking: Argument or Conversation?", Teachers College Record, Vol. 86, pp. 441-453 (Spring 1985) S. Carroll, et al, "The Relative Effectiveness of Training Methods-- Expert Opinion and Research", Personnel Psychology, Vol. 25, pp. 495-509, 1972 J. Dewey, Experience and Education, Collier Books (a division of Macmillan Publishing Co.), 1938 D. Gallagher, et al, Using TAKING SIDES in the Classroom, (a general instructor's manual for the entire Taking Sides series), The Dushkin Publishing Group, Inc., Sluice Dock, Guilford, CT 06437 (phone 800/243-6532), 1986 P. Hounshell and G. Madrazo, Jr., "Debate: Verbal Encounters In the Science Classroom", School Science and Mathematics, Vol. 79, pp. 690-694 (Dec 79) J. Newstrom, "Evaluating the Effectiveness of Training Methods", Personnel Administrator, January 1980, pp. 55-60 H. Schroeder and D. G. Ebert, "Debate as a Business and Society Teaching Technique", Journal of Business Education, Vol. 58, pp. 266-269 (April 1983) G. Stanford and A. E. Roark, Human Interaction in Education, Allyn and Bacon, Inc. 1974 "The Constitution: That Delicate Balance" and "The Military and the Media", each is a 13-part PBS series produced by The Media and Society Seminars, Columbia University School of Journalism REFERENCED CLASSROOM TEXTS H. M. Deitel, An Introduction to Operating Systems, rev 1st ed., Addison Wesley, 1984 T. Forester (Ed.), The Information Technology Revolution, The MIT Press, 1985 D. G. Johnson, Computer Ethics, Prentice Hall (Series in Occupational Ethics), 1985 E. J. Lias, Future Mind, Little, Brown and Company, 1982 K. Schellenberg (Ed.), Computers in Society, (an Annual Editions publication), The Dushkin Publishing Group, 1986 A. S. Tanenbaum, Structured Computer Organization, 2nd ed., Prentice Hall, 1984