One problem with the traditional ‘2 exams’ format of many polisci classes is that we are never offering students an opportunity to learn the material. We test to see if they’ve learned the material, but fail to offer students low-stakes opportunities to practice and confront the material. This article by the New York Times cites a study that indicates that students more from taking a test than they do by studying for it. Aha! We might say–test-taking helps students to learn! My job is done! Indeed–but the test is supposed to capture evidence of learning for the purpose of evaluation. It is not supposed to work as a formative tool that aids students learning. The student may be better equipped to show us evidence of learning from taking the test, but the point of data collection has passed, and they are instead evaluated on the basis of their performance prior to this potentially transformative experience. Sadly, taking an exam is often the first and only experience students get with active learning in a classroom.
Exams can work as a chance to practice and learn the material, but that is a positive externality of their usual purpose, and thus both succeed at an unintentional consequence (learning) and fail at their usual function (evaluation of learning).
How to correct this? There are multiple approaches. The article suggests multiple quizzes and tests but acknowledges that these often engineer student resentment. My solution is one I use in my methods class, which I grant is an easier venue than a traditional polisci course. But it may be useful as a model nevertheless.
I design the course to give students multiple chances of working with each piece of the new material. They confront it first in their reading, followed by lecture in class. Each idea in the lecture is followed by an example that I walk the students through. Then they work in small groups in class on another example, which we then review as a class. This is followed by an individual homework assignment, each of which is worth a small number of points. By this time, they have encountered the concept, examples, and problems associated with it five times. I usually post further examples with the answers on our class website as practice for students that need more work; sometimes I have allowed students to correct their homeworks, fixing mistakes they made the first time around. By the time students take the exam, they are very familiar with the kinds of questions and problems they need to know. This both increases their chances for success and lowers the stress of taking exams. The two largest components of their grade–their project and their final exam–come at the end, after the students have actively engaged with these concepts on multiple occasions. They know at this point what they truly understand and what is still confusing, allowing us to have productive office hour meetings where we target their areas of concern.
While overlapping projects, class activities, exams, and homeworks in this manner leads to a loss of content, I believe this to be a fair trade. We can never cover all the content in a course–every subject in US politics can be its own course or sequence of courses–so this simply requires making somewhat deeper cuts. Better in my mind that students truly learn the core of the material than vaguely familiarize themselves with its periphery.