Keeping an ear to the ground through mid-semester evaluations

Keeping an ear to the ground through mid-semester evaluations

One of the biggest issues professors face, especially when anything slightly nontraditional, is the risk of negative reactions from students --- "pushback" as it's usually called. And the instrument that causes student pushback to bloom into a full-blown, action-preventing fear is something we all know: The end-of-semester course evaluation. Student pushback can translate into poor course evaluations, and poor evaluations --- thanks to their outsized role in determining personnel decisions in higher ed --- can lead to losing your job.

I'm not writing today about the merits or demerits of course evaluations. Instead, I want to promote a very simple teaching practice that both mitigates the risk of poor course evaluations and helps faculty nip student pushback in the bud: Give an informal evaluation to your students in the middle of the semester, which is right about now for most folks. I've been giving mid-semester evaluations to my students for years and it has never produced anything but great results for me. It's simple, costs nothing, and produces loads of actionable information on your teaching that you can use to catch and address issues before they become irreparable.

The premise behind a mid-semester evaluation is what you think it is: You give some sort of instrument to students that asks them for input on what's working well, what's not working as well, and what changes should be made heading into the second half of the semester. Then you look at the responses and see what the good and bad points are, what issues are arising, and what changes you might make --- and then take action.

Here are a couple of common and simple methods for mid-semester evaluations:

  • Laura McGrath describes in this article at GradHacker the start/stop/continue method. Ask students three questions on a form: (1) What should we STOP doing in the course? (2) What should we START doing in the course? and (3) What should we CONTINUE doing in the course? This is fairly broad, but by casting a wide net, you can often get a good global view of how students are experiencing the course, and plenty of information to make course corrections heading into the second half.
  • In a related article at ProfHacker, George Williams expands this into four questions: (1) What's going well? (2) What needs improvement? (3) What can the students do to improve the class? (4) What can the instructor do to improve the class? This is a nice variation on start/stop/continue in that it stresses that improving the class is a shared responsibility and gets students thinking in concrete terms about both sides of that responsibility.

My own practice for mid-semester evaluations includes a mix of numerical and open-ended items. The numerical items are a selection of items taken verbatim from the actual end-of-semester evaluations. I use some of the more important items, especially any item that has been an issue for me in the past. For example, one of the items on our evaluations is The instructor treats students with respect --- I always include this on my mid-semester evaluations because this is a really important item for me, and I want to know how I am doing. I also include the item The instructor grades and returns assignments and exams promptly because I've had lower scores than I want on this item in past classes, and I want to give it special attention to know if I need to work harder on it before the end of the course.

Along with those numerical items, I include three open-ended questions -- (1) What are some aspects of the course and the instruction that are helping you learn so far? (2) What are some specific aspects in the course or the instruction that could be changed to help you learn better? (3) What other overall comments, concerns, or suggestions do you have about the course so far? This is basically start/stop/continue and you could easily do that instead.

All of these items are packaged up and given to students as a Google Form, with the numerical items phrased as Likert-scale questions and the open-ended questions given as "paragraph" questions. Then just copy the link to the form and send it out or post it on the course LMS, and let students work on it until a preset deadline.

Google Forms helpfully dump student responses into a spreadsheet, so it's very easy to work with the data. Use the numerical results to triangulate the verbal results. Make histograms. Go crazy. And importantly, do something with the results. The whole point of mid-semester evaluations is to get advance warning that there are things to work on before they become unfixable.

To reiterate this last point, if you give an evaluation, then also make a sincere effort to work on areas that clearly need to be worked on. Here we can take an example from Starbucks. Baristas at Starbucks are trained in what's called the LATTE method for dealing with customer complaints[1]. Applied to student evaluations, it would say:

  • Listen to what the students are telling you through their responses.
  • Acknowledge any legitimate problems that are being reported --- and even the ones that are not really legitimate, for example complaints that come from misreading the syllabus or mishearing your expectations for student work.
  • Take problem solving action to address these problems.
  • Thank the students for making you aware of their issues. Remember students are sticking their necks out to make their thoughts heard.
  • Explain why the problem occurred and what you are doing to fix it.

Ever since making mid-semester evaluations a regular part of my teaching, my end-of-semester evaluations have improved significantly. I believe part of that is that by simply giving students a way to be heard, they become more favorably disposed to the class and to me; and if there are real problems to address, I know about them early rather than after the course is over, so issues that arise on mid-semester evaluations rarely recur on end-of-semester evaluations. Mid-semester evaluations also help build a culture of trust and openness that makes your class a more welcoming place for your students.

If you've got ideas or variations on this idea of your own, leave them in the comments.


  1. No, I am not saying that we professors are like baristas, preparing and serving up products on demand to paying customers. I've written about this at length already. This is just an idea --- a very good one, if we have ears to hear --- that can be applied just as well in the client-consultant model of higher education as it can in your local Starbucks. ↩︎

Robert Talbert

Robert Talbert

Mathematics professor who writes and speaks about math, research and practice on teaching and learning, technology, productivity, and higher education.