My students and I agree on at least one thing: exams suck.
I haven’t given a test since last spring. Although several concerns factored into this decision, it was mostly inspired by several semesters of observation, in which I saw promising students seriously contemplating dropping my class because they didn’t do as well as expected on the first exam. Truly, no amount of consolation (“the first one is always hard, don’t worry!”) makes much of a difference when the test is worth 20% of the total grade. Sure, I try to use the experience to help students refine their study skills and critically examine how they’ve been approaching the material, but that lesson comes too little too late, and often they are unable to turn around old habits in time for the next test. It was unfair, I reasoned, to expect students to cram so desperately for my exams, knowing that a slip-up could permanently sink their grade — especially during midterms and finals, when they are cramming for several difficult exams all at the same time. Of course, this process also creates some discomfort for me as an instructor; I am tasked with grading all the exams, determining how to deal with questions that many people bombed or how to fairly grade open-ended responses, and breaking the news to my students.
So, faced with these clear problems, I decided it was time for an overhaul.
For the past couple semesters, I have instead been using a quiz, writing, and project-based format. My quizzes are smaller, administered weekly online, and both open-note and open-book. Because quizzes are more numerous and shorter, they are perceived as being “low-stakes” compared to quizzes (altogether, though, they still take up a sizable chunk of the course grade). My aim with quizzes is not just about testing learned material; they also give students a glimpse of how well they understood the previous chapter and might help them understand that they need to read more closely or pay more attention for the next one. Thus, they (hopefully) provide a sort of automated, incremental feedback system. Writing assignments can be used to shake up a traditional lecture and encourage students to think critically about the topic of the day, which helps keep them invested in class and invites discussion. Projects are focused more on application, my Bloom’s taxonomy level of choice, along with some opportunities for analysis and evaluation of existing ideas. I want to see my students relate to the topics we cover, create something new, and get excited about their newfound knowledge, a new tool with which to confront their world.
Testing is a controversial topic in modern pedagogy. David Perry points out that in-class exams force differently abled students to out themselves on order to get accommodations, and they create a context in which a student’s ability to memorize answers and quickly regurgitate them is sometimes more directly assessed than the student’s actual learning. David Gooblar argues that testing has its place, but he agrees with my low-stakes approach, favoring shorter, low pressure tests over traditional multi-chapter exams. Gooblar also argues that using a combination of assignments (i.e., tests and writing) can help bridge the gap between STEM and the humanities and more adequately evaluate students’ diverse abilities than an approach that exclusively utilizes one method.
I’m excited to discuss this topic with the academy next week, and I can’t wait to hear from all of you: what methods do you use, and why? Are you pro-tests, anti-tests, or in the middle? How does your discipline shape these choices? Which levels of Bloom’s taxonomy do you currently tick off, and which would you like to incorporate in future course plans?