As I was grading my final exams last week, I wondered about ‘quantity’ of answers to written questions as opposed to ‘quality’ of the answer: in other words, some students write a lot of stuff for an answer, but could have received full points on a question without filling a page with tiny handwriting. Here’s what I tweeted about this.
The students that fill the page certainly take longer to complete an examination, and this reminded me of a little project I did many years ago* about the speed at which students write their exams relative to the grade they received on that exam. For one of my larger undergraduate classes the final exam is meant to be about a two hour exam, but some students finish in just over an hour, and some students wait until we take the exam from them at the end of the three hour exam period.
One year I tracked the order that students forwarded me their exams and after the course was over I plotted their grade on that exam relative to the time it took them to complete the exam**. I’ve always wondered whether or not students who finish quickly are the ones that really know the material, or whether the ones who take the longest are so careful to check and re-check everything that they tend to do better than their peers. Here are the results:
A grade on an exam relative to the order in which that exam was handed in.
Bottom line: there is NO relationship.
Some students write an exam quickly and do very, very well. Some students simply do not know the material, and write the exam quickly and hand it in early. Some students stay to the bitter end and do very well. For some students, staying to the end doesn’t help their grade. And, of course, a lot of students are in the middle.
This was a good little project to do, and actually helps me as an instructor. It helps me to NOT judge a student by the speed at which they write the exam. It provides a reminder that everyone works at their own speed, and will do poorly, or very well, even if they write quickly or slowly.
The other lesson this teaches me is that it’s important to have an exam length that aims for the middle ground. A three hour examination that takes the ‘fastest’ students three hours just isn’t fair. Every student is different and need different amounts of time to read, digest, think and answer questions. Assessments are always tricky business, but one overarching goal of assessments is to test about how well a student may be able to recall content, integrate that content, perhaps do calculations, and think about the material that was discussed in the classroom. Assessments shouldn’t be used as a means to weed-out poor students, especially when the ‘tool’ for this is the length and size of an examination.
Good students write exams quickly or slowly, and should be allowed the time to do so.
—–
*To preserve anonymity around this project, I will not comment on what class it was, or where I was working at the time, and the data are presented in relative terms (i.e., the actual scale is not presented)
**The order was tracked by writing a small number in the top corner of the front page of the exam as students handed in the exam: I did not look at or see this number when I graded the exam, to ensure I would not be biased or influenced by that number when I was grading. The scatter plot was done weeks after the final grades were submitted.
© C. M. Buddle (2015)
Arthropod Ecology 
Chris, you eloquently wrote,
“Assessments are always tricky business, but one overarching goal of assessments is to test about how well a student may be able to recall content, integrate that content, perhaps do calculations, and think about the material that was discussed in the classroom.”
To that list I’d add assessments should allow students to demonstrate their incomplete knowledge. That is, so you can assess the level of their understanding. A single multiple-choice questions are “all or nothing”, “you’ve got it or you don’t”. That don’t leave any room for students who understand some things but not others. Maybe that’s why we typically see a bunch of multiple-choice questions on an exam.
Questions that require multi-step calculations should not be marked “correct/incorrect” based on the number the students finally write at the bottom of the page. (I know, we always give partial credit for mathy questions – 1 pt for setting up the equations, 3 pts for solving, 1 pt for sense-making, etc.) I think that’s why we create those grading scemes / marking guides, to reward students the knowledge and skills they possess, however complete or incomplete that knowledge.
Peter
@polarisdotca
Peter: THANKS for the comment. Excellent points. I agree – I generally avoid MC questions for exactly that reason – incomplete knowledge is important and the idea of multi-step (& “show your work”) Qs are really important.
Pingback: Do students do better when they write exams faster (or slower)? | Teaching for Learning @ McGill University
As someone who takes exams very quickly, thanks for looking into this. I had a professor tell me that that they looked down on students who took exams ‘too fast’ and it always annoyed me to think that she might judge my exam differently because I finished quickly. I am the kind of person who if I sit there and wait longer I will end up second guessing myself and then I do poorly.
Thanks Auriel – for the comment. That is an important insight you provided. I think “we” (profs) judge students all the time, whether or not we know it. Sigh.
Pingback: Recommended Reads #52 | Small Pond Science