I am a physics lab TA and I was recently informed that my engineering students will receive a lab final at the end of the semester worth 50 percent of their grade. The final exam is not personally created by me, but is decided amongst all of the lab TA's teaching the same course in the same semester. The meeting amongst the TA's will occur next week, however I am attempting to prepare for whatever will come, for better or worse for my students. I want my engineering students to succeed, and I want to give them all of my resources to give them the preparation to pass the final exam. I assume that I will not be allowed to tell my students what the lab final will be, or what types of questions will be asked. My students are required to complete 9 weekly lab worksheets corresponding to their specific lab of the week. All sum total of all lab worksheets are worth the other 50 percent of the lab grade.

One issue I am currently weighing in my head is how to grade the lab worksheets. Should I go easy on the grading to soften the blow of the final exam, or should I grade harshly to force them to correct every mistake they make in prepration for the final exam?


While I don't TA physics, but Geology, our labs tend to have the same structure. Luckily the labs are small enough that I can get to know the groups ect, So I can determine how hard they are working on the labs ect, so I generally grade 70% effort and 30% correctness. The difference is though, that you are teaching engineering majors, where I tend to teach non geology or STEM majors. Getting something in the lab 100% correct for my students isn't as important and following the scientific process, but, for a class of engineers, I would have higher expectations since the content they are learning is helpful for their careers later. IE, they should have a vested interest beyond the grade.

My experience with the Lab exams is that the students either do very well or very poorly. I would be somewhat lax on the labs, especially on first time mistakes. But there isn't much you can do if they aren't working hard and aren't completing the labs correctly and they will likely fail the test if thats the case anyway.


People learn physics (and many other things) by making mistakes, and then trying to correct those mistakes. With that in mind, I would always grade activities in a way that was supposed to motivate my students to fix their mistakes. However, it's also useful not to penalize them too much for making mistakes when they are first learning the material. Usually what I wound up doing was grading in a way such that students had to basically know what they were doing to get a 90% or higher, and they would have to really get everything right to get a 100% on an activity, but even if they didn't understand some concepts they would still get around 80%.

Of course, this grading scheme was specified by the administrators I was working for, so I don't really have any data comparing this with alternate grading schemes. I can offer the anecdotal evidence that when I was taking physics classes as an undergraduate, it was understood among the physics students that the grading would be harsh and thus grades as low as 60% or so could still be taken as an indicator that you were basically getting it.


One of my lab instructors would force us to correct the mistakes until the report was of sufficient quality. Sometimes, the problem would be forgotten units on a table, or a numerical error in a formula (fixable in a few minutes); others would be conceptual mistakes (requiring more work).

On another similar lab, the instructor would clearly indicate any errors and discuss them with us to make sure we understood them. There was then no need to rewrite the report, and the grade was only affected mildly. But, on the following reports, you better not make the same mistake, because it will then be penalised.

With the first option you can give them an opportunity to correct the mistakes without harming their grades, albeit at the cost of more work for all parties. On the second, you can make sure they learn from their mistakes, while being nice when they do it for the first time.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.