Teaching Evaluations

by Derek Smith

With the Winter quarter ending in March, student teaching evaluations became available last week. I’m always excited to see the results of the comment section, but I’m beginning to realize that this excitement may just be an issue of ego. I just don’t get very many useful ideas out of the surveys and I’d like to change that. I’ll explain a portion of the survey system used at UCSB to measure TA effectiveness. But really, I’d like to hear from others about systems in place at other institutions; or any other techniques you have discovered for evaluating your performance in the classroom.

There are numerous reasons for attempting to measure such a nebulous quantity as “effectiveness”–determining allocation of resources, feedback for future students–but I’m approaching the topic from the point of personal improvement. In this regard, student performance and end-of-course evaluations do not provide the level of detail necessary to gauge the success of a particular lesson, demonstration or interaction. To improve weaknesses highlighted by results of a macroscopic measurement requires collecting information at a higher resolution. The situation with end-of-course evaluations is directly analogous to annual employee evaluations. Unless a manager has kept good notes throughout the couse of the evaluation period, the review is of little relavance to the person being evaluated. I believe it’s the case that most managers don’t keep detailed enough records unless something is going wrong, so why would we expect a student to do any better?

On the last day of lecture and discussion section, the students at UCSB are given a few minutes to rate the instructor and TA (separately). Here are the questions used on the TA evaluation (rate the following an a 1-5 scale):

  • willingness to answer questions
  • clarity of explanations
  • coordination of the discussion with lecture
  • availability for help
  • how often you attend your discussion section

This is followed by a free-form comment section; the majority of those I received are either blank or contain positive comments. A few express frustration over time constraints: section too short, office hour conflict, etc. However, a troubling few used the word “confusing”. Due to the time delay between the misunderstanding and the evaluation, any specific critique that may have been gained is long gone. Worse, I don’t know the true extent of the issue because most people left the comments blank!

In thinking about improved techniques to gauge my performance, I asked myself, what exactly is the point of discussion section? If you look at syllabi of courses which include a section meeting you’ll find that the is time set aside to reinforce and apply material from lecture. In practice this translates to reviewing and solving problems. The real intention, however, is to uncover and identifying student weaknesses.

An ideal measurement of performance would take into account the instructor’s model of the student’s learning, which sounds expensive. For instance this paper describes “structured interviews” conducted mutiple times per semester with individual students and two teachers (one acting as a recorder). A related and slightly less invovled technique for getting feedback is classroom observation by a more experienced instructor. But then I realized that I previously mentioned an idea I came across for collecting student feedback which is designed to be “cheap” enough to be performed at every meeting. I picked up this combination interview/evaluation from this video presentation on interactive teaching methods.

To use the method allow students to hand in, at any time, notecards detailing the most confusing portion of class, the most useful/useless example or demonstration and any other comments and questions. Use the first few minutes of the next class to address the issues. You can find another explanation of the approach in this letter.

I have not yet utilized this idea, but will do so this quarter. I will let you know how it goes. But I’d also like to hear about other universities. What sort of TA evaluation system do you use? What do you get out of system personally. What would you like to improve about the system? Do you have any other in-class techniques to get real-time feedback? Let me know!


About Derek Smith

Former weather dude and scientific software developer. In the upcoming 2015-16 year I will complete my PhD at UCSB in nonlinear dispersive equations. I enjoy spending time with my two young daughters and running.
This entry was posted in General. Bookmark the permalink.