Exams: Open or Closed Book

Once every five years or so a student in class asks if we can have open-book exams instead of the usual closed-book version. After I explain the types of questions I would ask on an open-book exam to test understanding of the class material, students decide they would much rather stick with the closed-book exam.

Perhaps I’m being unfair and I’m making the exam sound extra-hard for the students. However, it seems to me that if you can look at your notes or the textbook, you could certainly write down a conceptual explanation without understanding it or solve a problem with the help of a worked example not having internalized the actual required learning-to-solve mental models. Now, if the exam was timed it might not make too much of a difference. A student who didn’t know the material will spend time flipping pages or notes and would do poorly anyway. But I thing open-book timed exams don’t make sense. Open-book non-timed exams however make sense. I call these projects. And we do have them from time-to-time. In the jargon, I would place greater weight on higher order Bloom’s taxonomy questions in an open-book exam.

I do provide enough information on a closed-book exam so that students aren’t burdened with sheer memorization. Periodic Table, numerical constants and data values are always provided, along with equations that we have used in class but not extensively. Equations that are used extensively should be second nature to the students (i.e., they should be in long-term memory) because this facilitates learning of more complex material. I have experimented with allowing students to bring in a single index card with any information they choose to write, but I’ve found that on average it advantages the stronger students over the weaker students in my classes, so I’ve discontinued that approach.

In my general chemistry classes this year, I have a mixed approach. Students have three take-home midterm exams which are closed-book and timed. After the timed exam is over, they are allowed to consult their notes, textbooks, classmates, and annotate their exam in a different color. Regardless of how the students actually do, they receive full credit for the exam. The three exams are altogether worth 15-20% of the class grade (instead of 50%) but now the closed-book and timed final exam is worth half of the class grade (instead of a third). So far I think this strategy is working. In the jargon, I have made a sharper distinction between formative and summative assessment in the exam grades.

What prompted my thinking about these issues again is an article published last month (Rummer R, Schweppe J and Schwede A (2019) Front. Psychol. 10:463). The title of the article is “Open-Book Versus Closed-Book Tests in University Classes: A Field Experiment.” The abstract is shown below.

It’s a small study and the main conclusion can be found in the abstract – preparing for closed-book exams fosters long-term memory encoding. You can read the paper for yourself to draw your own conclusions as to how applicable this study might be to your particular situation and subject matter. One useful feature of the paper is that it provides extensive references for anyone interested in what other studies have been performed prior to this study, thus giving you a quick overview on what is known so far.

The authors carefully go through the limitations and interpretations of their study. Aspects of cognitive load theory show up in the article, and I find their arguments plausible. Here’s a snippet: “Thus, the finding that participants learning with a closed-book practice test outperformed those learning with an open-book practice test seems to support the theoretically highly relevant assumption that the testing effect is due to retrieval practice… Another indirect effect of the closed-book test concerns the preparation and repetition of the learning matter at home. Since the content of the learning materials was highly relevant to the students… [this] resulted in more extensive study at home in the closed-book group than in the open-book group.”

For many years I’ve made use of coupling ‘retrieval practice’ to the ‘testing effect’ by having many short 1-5 minute quizzes at the start of my general chemistry classes. These are low stakes enough so that students aren’t overly stressed, but provide enough motivation for students to keep up with the material and strengthen those learning connections they’ve been (hopefully) making. My quizzes are closed-book and feedback is immediate. It’s been helpful to me not just to apply a practice because it seems useful, but also to explore the theoretical underpinnings for its utility – this helps me explain to the students why I employ a particular practice as an instructor to help them learn. By and large I think my students this year ‘grok’* what I’m trying to do with the take-home exams and why it is important to actually do them closed-book and timed. And so I keep reading these articles even if they aren’t about chemistry specifically.

*Not a technical term unlike ‘retrieval practice’ and ‘testing effect’.