Do frequent tests in a course encourage the right sort of learning and student outcomes- perhaps we need to rethink the way we construct our courses and ways of assessing student performance, and if we use tests (quizzes/exams) maybe they can be adapted to provide a better way for students to learn.
As another semester starts and I remind myself of the principles of reverse design of my classes I started revisiting some of my thoughts on tests and exams. I first started seriously thinking about this over 20 years ago (1) and some of my thoughts at the time were captured in “The Quantum Theory of Education” cartoon that appeared in a review I wrote for Nature (2). In the last few years there have been a number of articles in the popular literature (3) and even Scientific American (4) with titles implying that frequent tests can boost learning, and it started me thinking again about this issue.
Let me start by exposing my own biases: in keeping with Vision and Change (5) I think we need to shift the focus of science education to concepts rather than content, and we need to make it more interdisciplinary rather than siloed, and we need to emphasize student skills at least as much as content. Finally, we need to encourage and develop long term understanding of scientific principles and concepts and ways of approaching the many problems that science presents to us and, perhaps more importantly, will, in the future, present to us. As much as we might like to believe that students need to “know” the pillars that future science is built on, that may not still be quite as true.
So, back to the title of this post - do we test too much? Over the 40 years I have been teaching the structure of the average course has stayed fairly constant in respect to assessment of student ability and assignment of grades- every so often during the semester there is “ test” on the material covered in the preceding weeks and a final exam (sometimes comprehensive, sometimes not) that account for a large fraction of the total grade - the remainder of the grade is made up of things such as homework and other assignments etc. While a future blog will address the potential balance between tests and other ways of assigning grades, this blog will focus on tests. How do students study for such a scenario? The few days before each “test” they might try to revisit their lecture notes and assigned readings and memorize as much content as they can which they regurgitate on the exam. We often give them trial questions to think about etc to focus their study, which certainly helps but can still encourage them to operate in the memorize mode of learning. How effective is this approach? The answer to that question is beyond the scope of this post, but thinking about this reminded me of three excellent articles in the “Chronicle” I’d seen over the years that made what I think are still very salient points. After a while I tracked them down in my files. Two, by James Lang, “Metacognition and Student Learning” emphasizing the need for students to have opportunities to reflect on their mistakes ( 6), and “Small Changes in Teaching: Space it out” pointing out that we don’t always give students the time they need for meaningful learning (after-all there is all that content to cover in the course!) (7), and one by David Jaffe “Stop Telling Students to Study for Exams” (8 ) are well worth taking a look at.
In the past several years I have tried a “new” - at least for me - approach to exams. After the exam I give the students an overall numerical grade (on a 0-10 scale) on each question and give them a week to find out what they should have put down and rework their answer – I then give them upto 50% of the missed score on the question. The students like it of course because they end up with a higher score on the exam that they would otherwise have received, and I like it, not just because of the ease of grading but also because it gives students an incentive to reflect on the material and time to think. Anecdotally this seems to help students build a better realization of concepts and to get out of the silos we tend to put them in. What I need to do now is construct a controlled study using a concept focused comprehensive final exam – I remember telling students many years ago at First Year orientation one year that the ideal exam is the one they take a year after I finish teaching the course! – maybe this is the use of a well- constructed, validated national exam such as the one ASBMB is developing for Biochemistry and Molecular Biology (9) so students can receive a professional society certified degree when they graduate.
Articles and resources referred to:
2. Future of Education in the Molecular Life Sciences, Nature Reviews: Molecular Cell Biology, Vol2, 221-225 (2001)
3. Jessica Lahey in The Atlantic, Jan 21, 2014
4. Annie Murphy Paul in Scientific American, August 1st issue, 2015