peterr@utcsrgv.UUCP (Peter Rowley) (01/04/84)
Teaching someone something well is difficult; a science of teaching would need a comprehensive theory of psychology, which doesn't exist. We are left with anecdotes and some tried and true practice. All of which fill many books. Thus to substantiate something would take more space than available here, even if it could be done at all. Still, there's an important point to be made: grades are the single biggest negative factor in education. If you want to teach rote knowledge, grades are good motivators. You can easily test rote knowledge and give appropriate feedback. If a student knows the facts, s/he will get a good mark (setting aside the problem of "exam nerves" which can be lessened by a clear statement of goals-- easy when one is only teaching facts). If they don't, they will get a bad mark. One has the rather odious task of deciding how many facts a student should know to pass, however, and the preposterousness of using a single number to express how much a student knows is hard to brush away. Still, one cannot be satisfied with teaching rote knowledge; facts without understanding. In a world that changes significantly in a lifetime, one cannot learn set rules in youth and expect to apply them from then on. One can see numerous examples of people trying to apply the old rules to new situations, in industry, government, and day-to-day life. They just don't work that well--- and when they don't, all they can say is "well, that's what I was taught would work...". Note, though, there was a time not long ago when rules learned in youth COULD last a lifetime. Teaching facts as building blocks, ways of getting new facts (from established sources or by experimentation), and, most important, ways of putting facts together to apply to new situations, has always been important for fostering the creative thought needed for progress. Einstein said "Imagination is more important than knowledge". But now, we need such thought not just for the part of the population responsible for invention, but so that people at large can adapt to changing situations. Worse, as we can build machines to apply the simple, unchanging, rules for us, day-to-day employment will require more and more creative thought-- the mundane will be done by machines. This is particularly true of programming, which requires much creativity. This is a *serious* change that the people being educated now are not being adequately prepared for. And if the back-to-basics nonsense dominates, the situation will be even worse. (Seymour Papert's book "Mindstorms" echoes educational theorist Jean Piaget on the importance of such learning) Back to grades. I contend that the introduction of grades into an educational setting increases the emphasis on rote knowledge, at the expense of learning how to combine facts. As grade pressure increases, students turn their attention to the specific behaviours needed to perform well on exams and assignments. On assignments, cheating increases (and in large classes this is very tough to catch). As for exams, it is difficult, and even inappropriate, to test the ability to combine facts on an exam. To do so, novel situations must be presented, and these are likely to cause panic in a good number of students. Further, creative thought requires time for reflection, time not available in an exam. Thus, due to lack of time and skill for overworked CS profs (who were never taught how to teach), and the desire/need not to panic students, exams invariably wind up primarily testing rote knowledge, sometimes relegating creative work to bonus questions. Given the above, I feel this is clearly the wrong proportion-- we need to foster creative thought a great deal more. So, rote knowledge is not enough. We can't test creative thought well on an exam. Grade pressure causes cheating on assignments. And grades add to another social ill: they foster competition over cooperation by making grade comparisons inevitable; after all, some fail and some pass, even if you do not mark on a curve. Grades currently play a very negative role, even if they are calculated with care and precision, such as by Larry W. I am tempted to adopt some of the guerilla tactics of Robert G., but, as a grad student without much authority, am hesitant to do so, even if I were convinced the benefits would outweigh the feelings of distrust, suspicion, and even hate that the students would, I am sure, develop. These feelings could significantly interfere with learning. If I had a free hand, I would adopt a two-part approach. Throw away the old lecture format. Teach the facts using CAI and tutorial sessions, with the TA's and the professors. Teach creative thought relating to the topics of the course by solving problems in class, with as much student participation as the class size permits. Assign a letter grade on the basis of CAI performance (maybe 30% of the mark) and performance on assignments, done individually and in small (2-3) groups. Use automated methods to vary the assignments from student to student, within the constraints of marking resources available to the course. To reduce the load on those resources, and on the students, do NOT assign make-work assignments (such as many programming assignments are). Make each bit of writing the student does require them to think as much as possible. This can be done by providing skeleton programs for them to fill in, or, better, by giving written assignments where possible. And, finally, attempt to lessen the impact of grades as much as possible. Make it clear that good grades won't help the students once they are hired. Adopt a marking scheme which is perceived as comforting (e.g. 3 sub-schemes and the best mark of the 3 is the final mark). Make the course pass/fail if you can. All these comments are based on the teaching of 2 CS courses, and the taking of dozens of courses over the years; I'm hardly a veteran, or a properly trained teacher (not that many of the certified teachers I've had were any good). But, above all, grades are the enemy. We need something else, but I have no alternative to offer, just the perspective that the influence of grades needs to be reduced as much as possible. p. rowley, U. Toronto