Always solve your own exam first

I recently had a learning experience that I hope I can help others¹ avoid: I gave an exam with a significant error. I try very hard to ask thoughtful, interesting, and creative questions that are clearly stated and difficult to misinterpret. In this particular exam question, I presented a mathematical equation as well as a plot of the function (the problem can be solved mathematically or graphically). The issue? The equation was wrong (slightly) and didn’t match the given plot. I found the glitch immediately when I later worked out the solution. I couldn’t believe I made such a stupid mistake.

I felt really horrible over this – professors out there know what a quagmire this situation is. First of all, I understand that exams are very stressful for undergraduates and genuinely don’t want to make it worse. Second, whether they realized it or not, I think the cognitive dissonance from my mistake would throw off any student because their result wouldn’t correlate with the physical system they see on the paper. Third, some students spent the majority of their time trying to resolve this one question, resulting in lower performance on other problems.

In the end, here was my response. First, I clearly admitted the mistake, apologized to my class, and pledged to be more careful next time. I could have covered it up because only one student caught my mistake, but I’m held to the same standard of academic integrity they are and I don’t want to violate their trust. Second, I graded the exam very generously, resulting in a substantial curve. I felt this was the only reasonable amends I could make. And the final resolution? To never to give an exam that I haven’t first solved.