5 ms·

The best study that I know of on student evaluation of teaching was done at the Air Force Academy, where they assign students to teachers at random. Bottom line

The best study that I know of on student evaluation of teaching was done at the Air Force Academy, where they assign students to teachers at random. Bottom line is that student performance in later classes is negatively correlated with student evaluation of teaching. http://faculty.econ.ucdavis.edu/faculty/scarrell/profqual2.pdf http://faculty.econ.ucdavis.edu/faculty/scarrell/profqual2.p...

analog31 6y ago
There might be a confounding factor here. For a variety of reasons, I know a lot of academics. It's pretty widespread for teachers to prefer an early schedule for a variety of reasons. One is simply to help them organize their time better. Another is a rumor that earlier classes get better students. In any event it many mean that the teachers with better seniority get first crack at the better schedules. I've known teachers who bring cookies on evaluation day. Another rule of thumb is that evaluations tend to favor teachers who give out easy grades. The evaluations that my spouse received contained threats of violence.
- jimhefferon 6y ago
  There are thousands of potential confounding factors. But this study seems to me to address many of the obvious ones, and provides reasonable data suggesting that, while there may be good ways to assess teaching, student evaluations are not one of them. > I've known ... That is common. I had a friend who left a picture of their family on the desk. I am sorry to hear about your spouse's evals. That stinks.
- karatinversion 6y ago
  > Bottom line is that student performance in later classes is negatively correlated with student evaluation of teaching. I at least read “later” here as “subsequent, following” classes, not classes which take place later in the day.
  - HarryHirsch 6y ago
    The later the class is in the day the lousier the students and the less inclined to learn. Can't expect a half-literate teenager to give you good ratings when the class is about writing coherent English. That's an observable fact. But the abstract in the paper is talking about follow-on classes, not a section of the same course that's given on a different hour of the day.
    - karatinversion 6y ago
      Other than abusing teenagers, I don't really see what the point of this comment is.
  - analog31 6y ago
    Well, I'm beyond the point of retracting the comment, but thanks for correcting that. I guess a few well deserved downvotes won't kill me.
- whimsicalism 6y ago
  I think you misunderstood what was meant by "later" :)
hyperpallium2 6y ago
Wait, you have an above average teacher and do well, then have an average one and do average (worse).
- srtjstjsj 6y ago
  What are you trying to say?
mjburgess 6y ago
ie., teachers which were more highly evaluated on 101 classes worsened the performance on 201s of their students. I work in education and this has been my long held belief; it is interesting to see good evidence for this effect. Student evals measure their happiness; it is trivial to increase them by limiting their frustration (ie., the difficulty & depth of content). The goal of education isn't short-term comfort, it is long-term competence. For example, in my profession, I typically only have a few days with a group to rapidly onboard them in complex areas. It was my approach to end my session by considering all the issues the learners did not understand; and to review them again. However, this causes them to be aware of the limitations of the learning, and to frustrate them again with whatever was too complex. And thereby evaluate me lower. Now I am forced into a 'review' which convinces them they have learnt something (true), only because that feeling of accomplishment is better for the eval. It isnt better for them.
hyperpallium2 6y ago
Their concluding speculations (on why less experienced/qualified teachers' students perform better now and worse later) omitted the two I've seen: (1) You're better able to teach if you've learnt it recently (say, within 5 years). This is because you remember what it was like not knowing it, so can better communicate with someone who doesn't know it. e.g. "demonstrators" (often last year's students) often communicate better than lecturers. (2) The paradox of excellent teaching is that students get less practice in struggling, working it out for themselves, finding other resources (peers, texts). They learn the material, but not how to learn. That is, spoon feeding is a form of "disabling help", where doing something for someone that they could do for themselves deprives them of developing that capability. Perhaps... extremely knowledgable but hopeless teachers are best.

analog31 6y ago

There might be a confounding factor here. For a variety of reasons, I know a lot of academics. It's pretty widespread for teachers to prefer an early schedule for a variety of reasons. One is simply to help them organize their time better. Another is a rumor that earlier classes get better students. In any event it many mean that the teachers with better seniority get first crack at the better schedules. I've known teachers who bring cookies on evaluation day. Another rule of thumb is that evaluations tend to favor teachers who give out easy grades. The evaluations that my spouse received contained threats of violence.

jimhefferon 6y ago
There are thousands of potential confounding factors. But this study seems to me to address many of the obvious ones, and provides reasonable data suggesting that, while there may be good ways to assess teaching, student evaluations are not one of them. > I've known ... That is common. I had a friend who left a picture of their family on the desk. I am sorry to hear about your spouse's evals. That stinks.
karatinversion 6y ago
> Bottom line is that student performance in later classes is negatively correlated with student evaluation of teaching. I at least read “later” here as “subsequent, following” classes, not classes which take place later in the day.
- HarryHirsch 6y ago
  The later the class is in the day the lousier the students and the less inclined to learn. Can't expect a half-literate teenager to give you good ratings when the class is about writing coherent English. That's an observable fact. But the abstract in the paper is talking about follow-on classes, not a section of the same course that's given on a different hour of the day.
  - karatinversion 6y ago
    Other than abusing teenagers, I don't really see what the point of this comment is.
- analog31 6y ago
  Well, I'm beyond the point of retracting the comment, but thanks for correcting that. I guess a few well deserved downvotes won't kill me.
whimsicalism 6y ago
I think you misunderstood what was meant by "later" :)

hyperpallium2 6y ago

Wait, you have an above average teacher and do well, then have an average one and do average (worse).

srtjstjsj 6y ago
What are you trying to say?

mjburgess 6y ago

ie., teachers which were more highly evaluated on 101 classes worsened the performance on 201s of their students. I work in education and this has been my long held belief; it is interesting to see good evidence for this effect. Student evals measure their happiness; it is trivial to increase them by limiting their frustration (ie., the difficulty & depth of content). The goal of education isn't short-term comfort, it is long-term competence. For example, in my profession, I typically only have a few days with a group to rapidly onboard them in complex areas. It was my approach to end my session by considering all the issues the learners did not understand; and to review them again. However, this causes them to be aware of the limitations of the learning, and to frustrate them again with whatever was too complex. And thereby evaluate me lower. Now I am forced into a 'review' which convinces them they have learnt something (true), only because that feeling of accomplishment is better for the eval. It isnt better for them.

hyperpallium2 6y ago

Their concluding speculations (on why less experienced/qualified teachers' students perform better now and worse later) omitted the two I've seen: (1) You're better able to teach if you've learnt it recently (say, within 5 years). This is because you remember what it was like not knowing it, so can better communicate with someone who doesn't know it. e.g. "demonstrators" (often last year's students) often communicate better than lecturers. (2) The paradox of excellent teaching is that students get less practice in struggling, working it out for themselves, finding other resources (peers, texts). They learn the material, but not how to learn. That is, spoon feeding is a form of "disabling help", where doing something for someone that they could do for themselves deprives them of developing that capability. Perhaps... extremely knowledgable but hopeless teachers are best.