Slashdot Mirror


Harvard Ditching Final Exams?

itwbennett writes "According to Harvard magazine, Harvard's Faculty of Arts and Sciences voted at its meeting on May 11 to require instructors to officially inform the Registrar 'at the first week of the term' of the intention to end a course with a formal, seated exam, 'the assumption shall be that the instructor will not be giving a three-hour final examination.' Dean of undergraduate education Jay M. Harris 'told the faculty that of 1,137 undergraduate-level courses this spring term, 259 scheduled finals — the lowest number since 2002, when 200 fewer courses were offered. For the more than 500 graduate-level courses offered, just 14 had finals, he reported.'"

1 of 371 comments (clear)

  1. Re:one step closer to drive thru degrees by onionman · · Score: 5, Insightful

    i mean if you can trust the professor without testing the student, why not trust the student directly? why make the student get out of their car?

    Well, I am a math professor (although at a much lowlier school than Harvard) and I've never had a great opinion of in-class testing. The simple fact is that in the short duration of an in-class test you can't give the students substantive problems to work on. Thus, in-class tests (or any other short-duration timed test) is really an exercise in "how quickly can you work lots of relatively shallow problems".

    I far prefer to give my students lots and lots of really hard take-home problems. I call on them randomly in class to present their solutions at the board and explain their work. This is virtually cheat-proof... if you copy from someone, then it is obvious when I'm quizzing you at the board to prove your assertions. The only draw back of this method is that it takes a lot of effort on the professor's part, and it's only feasible on reasonably-sized classes. I can't do this when I'm teaching a 30-student class of freshman calculus.

    My guess is that Harvard is the type of place where class size isn't an issue. When you've got really small classes (under 10 students) then you can really gauge the knowledge level of each student because you are engaging each one individually in every class meeting. That's the ideal learning environment, but it's expensive.