Search the Lounge


« Court Affirms Access to NY Transit Hearings | Main | A Day at Bar-Ilan, and Reflections on Workshop Culture »

December 29, 2009


Feed You can follow this conversation by subscribing to the comment feed for this post.

Larry Rosenthal

The heart of the problem, of course, is that there is little agreement on what constitutes good teaching. The most common measure of the quality of teaching is through student evaluations, but students attend law school in order to obtain a marketable skill, and until they have experienced the practice of law, they are in a poor position to evaluate whether their teachers are preparing them effectively for practice. Indeed, complaints that the standard case method fails to prepare students adequately for the rigors of practice date back nearly a century, and they are likely to become more urgent. In the current job market, employers will be less willing than ever to absorb the training costs that law schools effectively externalize to employers so that law faculties may concentrate on the pursuit of scholarship that may advance the interests of individual scholars, and perhaps even institutional interests, but which will offer diminishingly value to students in an increasingly competitive job market in which employers become ever less willing to except the prestige of a graduate's law school as a proxy for the likelihood that the graduate will prove to be a profitable investment for the employer. Until law schools can develop an understanding of good teaching that is geared to serving the interests of students, rather than the interests of law faculties, there is little reason to believe that incentive systems will promote more effective teaching.

Larry Rosenthal
Chapman University School of Law


I agree with Mr. Rosenthal. Good teaching is very, very subjective. Even the best teacher won't be able to really teach some students.


I think Brian's comment to your last post hits the nail on the head. I was told by one important senior faculty member, in my second semester of teaching, that my teaching should be "ok, but no better." And that is, I suspect from what I hear from friends, the prevailing view just about everywhere. We pay lip service only to teaching because it can't be publicized and measured the way scholarship is. (In fact, I think teaching awards are really more for scholarly distinction -- for Deans to dress up their prolific ones in good teacher garb.) As a result, there is a structural problem that will cause teaching to continue to be slighted. Who has ever gotten a lateral offer for good teaching? Until this incentive structure changes, I think professors who like and excel at teaching are acting altruistically but ultimately against our own best interests. Those who won't spend the time to make their teaching more than ok are the ones who will be celebrated. Which is a shame.

Greg McNeal

I suppose this recommendation will fall into the encourage/reward category. I have always thought that it would be worth exploring whether a small portion of a faculty member's salary should be contingent on their teaching evaluations (say 5-10%).

Recognizing up front the fallibilty of teaching evaluations I think evals can serve as a good measure of certain teaching skills. Larry is right that students don't really know what it is they should be learning, but they do know some things which are measured or could be measured.

For example, our teaching evaluations at my current institution use an open ended narrative form and a scantron form. The scantron uses a scale of 1-7 to measure skills which I believe students are in a position to gauge. The university also allows the instructor and law school to customize the forms using 177 additional questions (

Here are the questions currently used on our form which I think students are capable of assessing:

-Rate the instructor's skill in handling student's questions effectively.

-Rate the instructor's skill in using examples and illustrations.

-Rate the instructor's skill in maintaining a positive atmosphere for learning.

-Rate the instructor's availability for individual help and consultation.

-Rate the instructor in terms of his/her preparation for class.

-Rate the instructor's skill in using class time effectively.

-Rate the instructor's skill in encouraging class participation and discussion.

-Rate the instructor's skill in improving critical thinking skills.

Now, I'm not arguing that any of these particular measures are the right measures for any institution, even my own, but they do seem within the ability of students to gauge. So, if a school wanted to value and reward teaching they could make some portion of the professor's salary contingent on performance on each of these measures (perhaps benchmarked against the rest of the faculty, more on that challenge in a bit).

Making a portion of pay contingent on performance would serve two ends. First, if the instructor was consistently performing poorly on any measure, they would have a strong incentive to improve. I know a few tenured faculty members at other institutions who tell me they don't even read their evals anymore. Rest assured if 5-10% of their salary were contingent on measures which included: "maintaining a positive atmosphere for learning" or "availability for consultation" they would strive to improve on those measures. I imagine such improvement would also increase student satisfaction and perhaps even student learning. There are broader benefits too, for example if in the aggregate faculty members found that they were all rated poorly on "availabilty for consultation" the faculty may have a strong argument that scholarship is interfering with the student experience and teaching loads need to be adjusted, or perhaps to improve "podium time" more time is necessary between classes.

Whatever the case may be, connecting dollars to evaluations causes everyone to pay more attention to what is being measured. A school may value "integrating theory and contemporary legal scholarship into the classroom" or "integrating examples from law practice" these could be encouraged by measuring them on the evaluations.

This brings me to my second point, there seems to be a consensus amongst law professors that teaching evaluations are a poor measure of instructor quality. If evaluations are a poor measure, change is more likely when pay is tied to them. If most of the institution has a financial stake in the measures of evaluation, much more time will be spent on properly calibrating those measures (and not just once, but repeatedly over time and in light of new data/issues/institutional focus).

There are some challenges associated with this proposal. If the standards and evaluation criteria are developed by faculty they have an incentive to not include those things which will dramatically change old habits ("Incorporate practice examples/theory/visual aids/etc. heck no!"). Perhaps student involvement or outside consultants could counteract this effect. ( To overcome institutional inertia (and employment contracts) it may be necessary to implement the system just for new faculty, grandfathering the more seasoned faculty.

Also, there are challenges with the evalation forms themselves. For example, my current institution uses two forms, the first is a scantron with questions like those I included above. The second is the open ended narrative form (e.g. What did you like/dislike about the course/book/instructor; How does instructor compare to others, etc.). The contingency system I've outlined wouldn't be practical with the open ended narrative form alone, it lends itself best to the scantron type of evaluation. However the open ended form and scantron can work together to provide narrative richness which can aid the instructor in improving their performance.

Let me provide a personal example: after my first semester teaching my evaluations for one of my classes indicated that on a scale of 1-7 with 7 as outstanding and 1 below average, I was rated a 4.1 on "Availabilty to Answer Questions." I was surprised by this because I was in my office everyday with the door wide open, and no one ever came to my scheduled office hours or any other time. Upon reading the narratives what I found was that students did not like the fact that I showed up approximately 1 minute before class started, and left immediately after class ending (rookie mistake, I thought I was punctual). The students wanted their "podium question time" so I made it a point to arrive a few minutes earlier and to stay a few minutes later and the next semester this rating jumped up and has remained high ever since. The narrative portion helped me improve on the scantron evaluation, I was also motivated to improve because I cared about my students and I knew this would be evaluated by appointments committees and others. Of course if I were more concerned with those extra minutes before and after class (think of all the writing that can happen in 5 minutes a day, three times a week, over a 14 week semester!) I might not have made the change...unless of course it were tied to money.

A final problem I see with this contingency model relates to developing the measure or standard the instructor is measured against. Is it an objective scale (e.g. 1-7, average, above average, etc.) or is the scale measured against the performance of other faculty members (e.g. above the mean/median on a given question) or some combination (you must have at least a 4 rating in every category and be in the top 2/3 of faculty)? I'm not sure what the exact model would look like, but I'm certain an effective one could be developed (with faculty motivation of course).

Just some thoughts, more for discussion than anything else as I'm not totally convinced that this is a good idea without further reflection and input from more seasoned professors.


The funny thing is that, while it is true that good teaching is extremely hard to measure, we all know who the good and the bad teachers were when we were in law school. Is there any doubt that Prof X was inspiring and effective, and that Prof. Y was a waste of time. Perhaps good teaching is a "know it when I see it" kind of thing.


I want to go Vladimir one better: I am not convinced that good teaching is any harder to measure than good scholarship. When I was in law school, I (and most fellow students) knew who the good teachers were, and now several years later, and having been on the other side of the podium, those judgments hold up pretty well. At the school where I currently teach, it's pretty clear who the good teachers are, from both teaching evaluations and student word of mouth--and the teachers with the best reputations aren't necessarily those who are easiest. For whatever reason, schools have over the past decade decided to reward scholarship rather than teaching. If more schools chose to focus their attention on good teaching, I have little doubt that they would be able to develop a reasonably good (if imperfect) system for identifying superior teachers and directing rewards their way.


Hi anotheranonprof,

Very apt comments. But I wonder, aren't schools already taking teaching into account in rewarding faculty? Doesn't your dean take teaching evaluations into account when deciding on raises -- in addition to a professor's scholarly output and contributions to the law school in terms of service? In setting raises (at least back in the day when we got raises), doesn't your dean look to see how many students each professor teaches (which is a proxy for quality -- not perfect, of course, but a proxy)? In every yearly evaluation I can remember, the dean has had at least the past year's teaching evaluations and course enrollment, in addition to information on what committees I was on, what pro bono work I did, and what scholarship I was working on.

Then there's hiring -- those who want to move laterally better be pretty good teachers (that is, have good teaching evaluations and have colleagues who'll speak well of your teaching). I think teaching's important and schools recognize it.

The comments to this entry are closed.


  • StatCounter
Blog powered by Typepad