A different model for assessing students

January 23, 2008, 7:34 pm

Coming fresh off my two-week stint doing promotion and tenure portfolio evaluations, I’m in the middle of a three-day blitz to design and prepare all three of my spring semester courses — Calculus, Linear Algebra, and Differential Equations. Why go so fast, you ask, since classes don’t start for another week? Simple. I am wanting, badly, to go see Cloverfield and the only hope I have of doing so with two kids under 4 is to go during the day when they’re in school. So I need to get stuff done!

That bit of professionalism aside, I was going to say, course design has been on my mind a lot in the last few days. Especially assessment. I wanted to throw out a major change in the way I approach assessment in my classes that I started to use last semester and am building in prominently in my spring courses. The model is as follows:

  • Assessments are to be clearly delineated into two types: Formative and summative. Formative assessments are intended to gather information on short-range, micro-scale student performance on small amounts of material, with my feedback mainly being used to flag areas of concern and give students a chance to improve. Summative assessments are intended to see, basically, if the formative assessments have done any good — whether the students have attained the course goals on the material.
  • There are to be no more than three summative assessments in the course (including the final exam) and collectively they should be worth at least 50% of the grade. So, a small number of high-point-value assessments to gauge whether students actually know what they are supposed to know in something like a final analysis.
  • On the other hand, formative assessments are to be short, graded quickly, and handed back no more than two class meetings after collection. And there should be a lot of them. So, a large number of low-point-value, quick-turnaround assessments that help correct student’s conceptions more or less “on the fly”.

The old model was to have three tests, a final, and several assignments that involve lengthy problem solving and/or writing. I had two problems with that model. First, everything took a long time to grade; students always complain about work not getting handed back quickly, but in my case they were justified — and how could I not have a problem with being timely, when every assessment was such that it required 5-10 hours of grading? The second problem was cheating. All those lengthy, high-cost, out-of-class, writing-intensive assessments fed naturally into problems with academic dishonesty. The potential repercussions of getting caught were small when compared with the amount of work to be done and the ease with which plagiarizing could be accomplished.

So I realized that my problems with students cheating and plagiarizing on everything, and bad course evaluations having to do with handing back work quickly, were rooted in a common problem: having a concept of assessment that featured assignments that took a long time to grade, didn’t provide immediate feedback, and lent themselves to cheating. And rather than try to crack down harder on cheating on the one hand and grade faster or spend more time grading on the other, I decided the way to go was to reinvent the way I assess.

This is a simple, perhaps obvious, way to do it. But I’m pretty good at missing the obvious.

Educators, does this sound like something you do? Or would like to do? Or tried and it didn’t work for you? Or what?

This entry was posted in Education, Life in academia, Teaching and tagged , , , . Bookmark the permalink.