When thinking about how to adjust Intro Bio for the realities of Fall 2020 teaching, we made a few changes to the course. One major change is that the class is now fully flipped — the students watch asynchronous lectures on their own (we made these using lecture recordings from previous semesters, which had the advantage of having responses to questions previous students asked when they first learned the material) and the synchronous (but recorded for asynchronous viewing if needed) sessions are used for “active learning sessions” where we work through old exam problems and activities to help students organize information (e.g., a table comparing and contrasting gene flow and genetic drift). I think the active learning sessions have been really helpful for student learning, and all of the student feedback about them has been positive. They love the active learning sessions! But that’s not the focus of this post.
The focus of this post is another change we made: moving to more frequent exams. In previous semesters, students had two low stakes quizzes a week, and four high stakes exams across the semester. (The four exams together were worth ~75% of the final grade.) This year, students had one low stakes quiz a week, and six exams. The six exams together are worth about the same amount as last year’s exams — ~76% of the final grade — so each exam on its own is lower stakes (while admittedly still not trivial!), which is part of inclusive course design. We had several reasons for wanting to give six exams vs. four, including:
- By lowering the stakes on each exam, we hoped that a student illness or any other life event or disruption would have less of an impact on how the student did in the course overall.
- By lowering the stakes on each exam, we hoped it would reduce exam anxiety and the temptation to cheat (we did other things to try to reduce these two things, too).
- Having six exams meant we could down-weight the two lowest exams and up-weight the two highest (as opposed to what we did with four exams, which was to down-weight just the one lowest and up-weight just the one highest).*
- Each exam could be shorter and, unless a student had accommodations that gave them extra time, meant the exams could fit in the normal class period, rather than being in the evening.
- Students would get feedback on their study strategies and preparation earlier in the semester and more often.
At the beginning of the semester, I had the impression that students mostly seemed to like that we were going to have six exams, but I didn’t have a great sample. Our first exam was in week 3, and exams 2, 3, 4, and 5 were in weeks 5, 7, 9, and 11, respectively. It turns out that fitting in six exams in the semester makes for a schedule that is kind of intense.
This intensity was definitely felt by the instructional team. The point that seemed the most ridiculous to me was when, on the same day, I was giving exam 3, entering exam 4 questions into Canvas so that exam 4 could be vetted, and writing exam 5 questions. It was a lot.
As my co-instructor, Cindee Giffen, and I discussed it, we realized we didn’t really have a good sense of how students felt about the number of exams. Did they like having 6? Did it feel like too much? What would they do if they were in charge and we were just starting the semester?
So, when we released exam 5 grades, we also included a survey (with students getting an extra credit point if they filled out the survey, to incentivize participation). I was surprised by the responses, and figured they might be of broader interest, so I’m sharing the results here. (414 students responded to the survey.)
Here are the key results:
At the beginning of the semester, students mostly thought 6 exams would be a good thing (41%) or weren’t sure whether it would be good or bad (41%). Only 19% thought 6 exams would be a bad thing.
How did those views change over time?
A) 68% of students who thought they would like six exams at the beginning of the semester felt the same after five exams (green part of left bar in figure below); equal numbers liked six exams more (14%; orange part of left bar in figure below) and less (14%; gray part of left bar in figure below).
B) Students who thought they would not like six exams at the beginning of the semester were more mixed after five exams. 39% said they were now even less in favor of six exams (gray part of middle bar in figure below). 31% said they were now more in favor of six exams (middle bar, orange), and 24% felt the same (middle bar, green) as they did at the start.
C) Of students who weren’t sure at the beginning of the semester (right bar in figure below), 43% still weren’t sure after five exams (green) plus 6% were not sure if their feelings changed (blue); 30% were now less in favor of six exams (gray), and 21% were now more in favor of six exams (orange).
After 5 exams, about half of students felt 4 exams would be much worse (19%) or a little worse (32%) than 6 exams. 9% said it wouldn’t make a different and 5% weren’t sure. A bit over a third of the class said 4 exams would be much better (11%) or a little better (24%) than 6 exams.
When asked how many exams they would prefer to have if they were to retake the course, the most common response was 6 (44%), followed by 5 (28%), then 4 (20%), with 8% saying they weren’t sure.
So, overall, they seemed to prefer having more than 4 exams, even without the benefit of distance from the intensity of the frequent exams. I was definitely surprised that the most common answer was preferring 6 exams! I was not surprised that there was variation in the opinions (sometimes strong variation!) One thing that is abundantly clear when teaching a class this size is that there is no strategy that is best for all students.
What are some of the things they liked about 6 exams?
We included some options of things we thought were upsides of 6 exams, and also gave the option of them entering their own answers. I haven’t had a chance to go through the free responses yet, but here are the percentages that chose the options we gave them:
- Ability to upweight & downweight two exams (93%)
- Lower stakes on each exam (81%)
- Exams during class time rather than in the evening (75%)
- 80 minute exam vs. 120 minute exam (67%)
- 6 exams helped me to stay on top of the material (51%)
- Feedback on study strategies earlier in the semester (36%)
- 6 exams is less stressful. (34%)
An option that I forgot to include but should have is that there was less material to focus on for each exam. The exams were semi-cumulative (exam 2 built on exam 1, exam 3 built on exams 1 & 2, but then we reset and exam 4 focused on just unit 4 material). In conversations with students, they have said that being able to focus in on just four lectures was helpful for studying. I also think that was helpful for them.
What are some of the things they disliked about 6 exams?
Here are the percentages that chose the options we gave them:
- If I got a little behind, I didn’t have as much time to catch up (73%)
- Less time to learn and synthesize information (68%)
- 6 exams is more stressful (47%)
- I’d prefer fewer exam days, even if that means each exam is higher stakes (28%)
- I’d prefer evening exams (6%)
Something I didn’t think to include but that has come up in discussions with students is that, to do six exams, we need the exams to be done via Canvas, rather than as printed exams done in person. Not being able to easily mark up the exam is definitely a drawback of exams taken digitally (and I wonder about inequities that result from this — having a tablet or easy access to a printer probably helps a lot).
Based on all this, what would I do if I was starting the semester over again?
Probably 6 exams, but 5 would be tempting, to try to have it feel like there was a little more time to catch up on things. I think that would have especially helped students around exam 5, but some of the problem there was that there were major covid-related changes around then (including a shift to mostly remote classes, with many students moving around the same time). So, a little more time would have helped some students then….or, alternatively, it might have meant that the exam that came at a particularly hard time was worth more of their final grade.
I also still haven’t decided what I think would be best once we’re able to do things in person again. There are some things I really like about online exams (including the ability to offer them during class time and to give more exams), but there were also definitely downsides (including that most students were taking them on screens, which made things like phylogenies and pedigrees much harder). So, I’m not sure what would make the most sense. But it definitely is useful to know that, overall, students prefer having more than four exams!**
Have you tried different numbers of exams? Did you also find there were tradeoffs? How are you balancing them? I’d love to hear thoughts from others in the comments!
*This semester, each exam is out of 80 points. I will take the two lowest exam grades and rescale them to be out of 65 points, and take the two highest exam grades and rescale them to be out of 95 points. There was one year when we dropped the lowest exam, but that had the unfortunate effect of having students who were happy with their grade heading in to the last unit check out for that unit, so, since then, we’ve down-weighted rather than dropping.
**Someone asked me if I have considered not giving any exams. I have, but can’t really figure out how to make a class like this work with no exams.
Dynamic Ecology 
I switched from a two midterms and a final to 5 exams, one every 3 weeks. It was a huge improvement in everyway. All exams count equally except the last, which is a comprehensive final and is worth 2x as much. The benefit if that students don’t have to cram for so much material and learn none of it well. Concepts that are built upon over the semester can be assessed as we go and adjusted as we go. I also learned to never set an objective in the lecture material that must be reached before the next exam. That way, if things are going rough, we can slow down and get it right before going forward, and if things are going very well, we can cruise, but taking away the need to reach strategic endpoints before each exam was obviously having negative repercussions in the classroom.
I really like the idea of having more exams, although the amount of time for the instructor to put in is scary! I already do weekly assignments and quizzes, and 3 tests so it is a lot.
I wonder if the results you have are biased in that students might think that since you as the authority figure and expert teacher and designing this course to have 6 exams, then that must be the best way to do it and thus they are more likely to state that they like or prefer this structure?
Emily, the extra exams are little less daunting for me because they are somewhat shorter. This makes them easier for me to face when grading, which is by far, my least enjoyed academic activity. I am fortunate in that my class is small, and thus, the exams are more problem solving and critical thinking oriented and can be hand graded, not multiple choice exams. I might think differently were this a big intro lecture class.
First: welcome back Meghan! We’ve missed you and your posts!
Surprising and interesting poll results.
Here at Calgary, back when all our classes were in person there was concern that every instructor was giving a midterm and a final (or maybe two midterms and a final), with the result that students were getting stressed by having high-stakes exams in all their courses at the same times. Once our classes moved online, most every instructor switched to frequent low-stakes quizzes (often weekly) with no midterms, because that’s the advice we were given. Frequent low stakes quizzes are better for keeping the students on track, and reduce pressure on students. Well, except it turns out it doesn’t reduce pressure, because our students reported that their total workload had jumped massively. Without intending to, our switch to frequent low-stakes quizzes in every class pushed all students to spend much more time on studying than they would otherwise have done in a normal year (and it’s not as if they were slackers in the Before Times). The students reported feeling under constant unrelenting pressure.
Jeremy, this was my immediate thought. Six exams might work for students for one course, but it doesn’t scale up to a full course load. With thirty exams per semester, you’d cease to be a Biology major and become an Exam Triage major!
Similarly, it doesn’t scale up for many faculty members, particularly those at smaller institutions where teaching loads are high. If one is fortunate to teach two or three courses per year, or have ample TA/grading support, it might be feasible. But I don’t see how this would work otherwise (without resorting to multiple choice questions). Exam practices need to be sustainable for instructors as well as students.
PS. 30 exams per semester is an exam every other weekday, all semester long, on average 🙂
The other issue (that may be specific to my course) is that I mandate that the weekly quizzes be completed within a specific 1 hour window (10-11 am on Fridays). That’s to reduce the opportunity students have to communicate with one another about the quiz, which they’re not allowed to do. I also made the quizzes open book, and adjusted the sorts of questions I ask accordingly (e.g., I don’t ask them for definitions that they could just look up). And I wrote the quizzes so that they could definitely be completed in 15-20 min. max, so that the 1 hour window allows *plenty* of time, even for students entitled to extra time on exams. But what’s ended up happening is that students take *far* longer than I anticipated–most use the full hour! Because they’re spending a *lot* of time looking at their notes and rewatching bits of the lecture videos. They’re also triple-checking their answers.
The students haven’t complained, and they certainly would complain if I cut the time available to, say, 20 min. (i.e. the amount of time I’d allow if I were giving the same quiz in person in class, with open notes). But even so, I don’t feel like this is optimal. I don’t want them spending an hour every week on a low stakes quiz!
So I dunno. I’m going to have to change how I do things next term anyway, because I’m teaching a different, much larger class, so I won’t have time to mark weekly quizzes and neither will my TAs. But I’m still not sure what the sweet spot here is. What I may do is go back to fewer, longer, higher-stakes exams, and find other ways to oblige them to keep up with the course material (as opposed to slacking off, then doing panic cramming the day before the exam). For instance, ask them a bunch of quick multiple choice questions each week, but don’t mark them for correctness. Just give them the mark if they answer the questions, and a zero if they don’t answer them. That’s not a perfect system–a student could just pick “A” for every question without bothering to even read the questions. But I did this last winter in the same class after we were forced to move online partway through the term, and only one student in the whole class just lazily picked “A” for every multiple choice question.
Any suggestions?
If the goal is for students to keep up with the course, I like your final idea of questions graded only (or mostly) for participation. It’s heartening that only one of your students was cynical enough to put “A” for everything under this scenario, and I think most students prefer to put in a good-faith effort.
If you’re not going to grade things, multiple choice becomes less necessary. I’ve had good luck with posting 10-20 open-ended, pre-class questions per class to help direct reading and preparation. I never take them up (at least not directly) or grade them, but a lot of students seem to like to have a goal in mind as they prepare for class. Of course, I’m sure a lot of other students never download them at all, but I’m OK with that.
I think sometimes we are so focused on finding processes that can’t be gamed that we dismiss solutions that are broadly effective. The odd student who puts “A” for everything, or the students who never once complete the pre-class questions, might just be the cost of doing business for implementing a system that helps many other students keep up and follow the course material.
More generally, I think that a creeping focus on evaluation can be damaging because it validates the notion that the whole point of a course is to get a grade rather than to learn something interesting and useful. Sloggification can’t compensate for interesting material, thoughtful presentation, and genuine engagement.
Julia’s tweet approximates my approach. I too want to drop students’ lowest exam — partly because anyone can have a bad day, but partly because arranging makeups is such a pain, and if someone misses one I can say that that will be the one that’s dropped. But dropping one out of four exams is not great; if the first 3 exams go well, students can blow off everything after that. So I’ve gone to six exams overall, despite not being thrilled with sacrificing that much instructional time (we have 10-week quarters plus a finals week).
If we assume an instructor has a fixed amount of time dedicated to the course, 6 exams means less time for lecture/activity prep, refreshing content, and other things that might be good for students. I wonder if the benefit of 6 exams outweighs the cost to the students in terms of getting a teacher who has spent less time on other things. You could still heavily up weight/down weight with 4 or 5 exams, it may feel less natural, but there is nothing preventing it. Alternatively, you could also make quizzes worth more. 4-5 exams is still quite a bit and would still promote gradual learning.
I did something like this in my Animal Behavior course not long ago. In addition to a lot of exams, they built up in points. Can’t recall exactly, but something like 20 points, 30 points, 50, 70, etc. Each one was fully cumulative. Some students loved it. Some students said they didn’t like it because they were “constantly studying” (which was sort of the point!). Not surprisingly, they did incredibly well on the final exam, which was really just the last, fully cumulative and highest point-value exam. Some said what they liked about the structure was that by the end of the semester, they knew the older material so well that studying for the final exam was almost unnecessary and this was nice given how many other classes were piling on. I’ve also had success with a similar approach in classes where a few larger, higher-stakes exams have “mini-exams” in between them. For example, they might have “mini-exam 1” and then a couple weeks later the “regular” exam 1, which would include all the material on mini-exam 1 as well as any new material. After regular exam 1, though, we re-set and mini-exam 2 only covers new material since regular exam 1. This gets them used to the exam format and keeps them studying, but isn’t quite as intense as 6 or more full exams. All that said, the change I made for some classes that I prefer over all of these different exam structures as a “specifications grading” approach that largely eliminated or reduced the need for exams at all. That approach is probably not practical in larger classes though.