Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!purdue!decwrl!labrea!rutgers!paul.rutgers.edu!byerly From: byerly@paul.rutgers.edu (Boyce Byerly ) Newsgroups: comp.edu Subject: Re: Exam Files Message-ID: Date: 4 Feb 89 17:45:41 GMT References: <1461@trantor.harris-atd.com> <19554@agate.BERKELEY.EDU> <27541@bu-cs.BU.EDU> <15993@joyce.istc.sri.com> <429@laic.UUCP> <19810@agate.BERKELEY.EDU> <9388@pasteur.Berkeley.EDU> <135@jarthur.Claremont.EDU> Distribution: usa Organization: Rutgers Univ., New Brunswick, N.J. Lines: 25 Finding old exams is certainly a sport practiced by any serious student, but I don't really think it's entirely fair: 1. There is a penalty for introverted students, freshmen, and parttimers; they probably don't nearly as good sources as many. 2. This practice tends to encourage memorization over understanding, which is never a good thing. I like to make up a "sample test" which is far harder than the real thing, and distribute copies to everyone. The problems should be difficult and require heavy thinking. For any creative answers, like an essay question or writing a program, no answer key should be given. This should make students work together, discussing and defending their answers. This is where "learning" takes place. I usually pull applicable questions from old exams. This eliminates my tendency to get lazy when writing the real exam, and keeps the memorizers from gaining much of an advantage. It also gives all the students an equal starting line for studying. A final advantage is that when the students encounter the real exam, they find it's much easier than the sample, which usually relaxes them and allows them to express what they know. Boyce