Whenever I teach an intro class, I say to my students, “Remember that movie with the teacher from L.A. who has all those students who think they can’t learn math, then he teaches them how to do it?” They respond, yeah. “Well,” I continue, “as you probably know, Columbia scours the country looking for these top teachers, then hires them. [pause] Not.” Actually, I tell them, “Columbia scours the country looking for the best students–that’s you–and then finds the world’s best researchers–that’s us. We can teach you stuff because we know everything. But we’re not such good teachers, so you have to use your student skills to learn. My goal as a teacher is to teach you how to teach yourself statistics.” Etc. etc.
Anyway, Kipp Herreid, the Director of the National Center for Case Study Teaching in Science, writes,
One of our case study workshop “alumni” wrote in to say that he is a member of a faculty team that is trying to develop a working model for the university to use in judging whether one has attained “excellence in teaching.” Their challenge is to develop a set of metrics similar to those commonly used to judge research performance in their Promotion and Tenure process, such as the number and quality of peer-reviewed publication, the type and number of grants obtained, the number of invited lectures given, etc.).
As he noted in his email to us, when you ask someone at a typical top-40 research university what constitutes “good” research, they seem to have little trouble defining what that means, how to quantify it, or how to form an opinion on one’s performance. However, when you ask those same people what constitutes an “exceptional” teacher or perhaps a great outreach program, they usually stumble and have few equivalent and easily measurable metrics.
We’d like to hear from you if you have any thoughts or suggestions on where to find such criteria that might be in use today at your school or other institutions. Our correspondent notes that while he has been able to find numerous descriptors of what a “good” teacher is (in qualitative terms), he has found nothing on how to quantitatively measure that in meaningful ways.
Please direct any answers to [email protected]
I don't know the movie in question, but it's a common formula – though I think my favorite instance was "Cool Runnings".
I've come to believe that the truly exceptional teachers are also judged as absolutely lousy teachers by those students who don't think on the same level, but there are, hopefully, several students in a big class that it all just comes together with – and those are the students who speak out the most.
Of course, the other bunch could come away with little to nothing.
I would start (and probably end) with reading "What the best college teachers do" and building a metric based on the findings there. It may boil down (partly) to what Andrew said above.
I think the best teachers remember what it was like to lack understanding, to be completely confused by a topic. Then they explain it in terms that speak to the as-yet unsophisticated intuition of the students.
A simple criteria is to determine how well the students do in preceeding classes and how many drop classes and school. Finally, how well do the students do in their post school employment? Would the students write a letter about the prof?
While this would spread the evaluation out with a 3-4 year time lag, It is better than believing the instant gratification of student teacher ratings is the best evaluation criteria!
I think that it is exceptionally difficult to be a "good teacher" to a broad range of students. The teacher at McGill that taught me the most is also famous for confusing students.
I think that it is an unsolved problem to take a group of people with widely varying backgrounds, aptitudes and interest levels and be effective for the whole population. This seems to be especially true when the class size is large.
In a small journal club I can scan the faces of the students and adapt the presentation accordingly. For example, when I was teaching the basics of multiple imputation I ended up having to make a long diversion into types of missing data.
But if there are more than 10-12 people in the room this strategy isn't feasible.
I wonder if anyone is good in front of 100 students? If they are, I would be grateful for tips on how to emulate them!
I'll start this comment by saying that I may be six sigma.
I look at the problem of teaching as similar to presenting during a meeting or conference.
I learn very little outside of solving problems, unless I have a question related to solving a problem. Seminars are useful because its discussion, and I can ask a question about a problem that I attempted to solve. Lectures are generally not an efficient use of my time, but the best lecturers assign reading before the class which requires thinking/problem solving and then present students with a problem or two to solve in class. Then they walk them through "non-obvious" aspects of the solution.
I actually measure the performance of my teachers by doing the following: (1) write down a list of things I want to learn before entering the class (based on syllabus); (2) after class, measure whether I think I improved my knowledge of the topic, using a thermometer survey question; (3) compute statistics for the semester. If I can't tell what I should be learning before I show up for class, I give the instructor the equivalent of 0 for the day.
Personally, I think if you sampled 4 classes from any teacher and gave computerized pre and post survey polls to the students, you would learn a great deal about who was a good teacher and who wasn't. Given that a lot of students have laptops now, they could even instrument the process in real-time.
When I worked for a large, Fortune 10 company, detailed post-conference speech surveys were normal. I had amazing feedback about nearly every aspect of my performance and all of the program managers in the company/division/group were ranked by their statistics. So there is a great model to build upon in the private sector. Consultants were available to help you "reach" your audience.
I have been struggling with what it means to be a 'good teacher' and how I can be one too, for many years. I really appreciate this opportunity to engage with others who might be interested in this topic too. I agree that one of the most important attributes is an awareness of what is difficult in the material – and why – and I also agree with other posts that it is very difficult to be 'the best teacher' (or even a good one) for every student in a class.
The university, as an institution, does very little to help faculty actually improve their teaching. The student evaluations are designed to be summative (summarizing performance), not formative (providing feedback that can help to improve performance). I believe that a good way to document and/or quantify good teaching, or when excellence in teaching has been met (for a given class) would be to compare the results of formative evaluations, not the results of summative ones, across terms. Currently, the standard university evaluation (-possibly an outlandish generalization since I’ve only been a student or instructor at five places in D.C., California and Maryland) asks questions like, "compared to the other instructors at this university, how would you rate this instructor?" This is a normative evaluation, comparing instructors at a particular point in time. I feel that the single numeric summary based on evaluations like this – typically, the answer to one ‘summarizing’ question – is not an appropriate representation of teaching performance. Instead, I would like to see/develop/employ a rubric where the dimensions of teaching are outlined and performance levels are explicitly articulated. Asking students to rate an instructor using this rubric would provide a very rich, single-value summary. (I have created what I call a “mastery rubric” like this for clinical research and am working on one for physicianship – anyone who is interested in this or who has done something similar should please email me!! [email protected]) While I see the practical utility, a quantitative value is over-rated, in my opinion, when qualitative statements like “86% of students rated this instructor as “excellent” using the instructional excellence rubric” give a great deal more – and richer – information.
In a setting where formative input is obtained from students, then a single summary value can still be obtained as a function of how the formative input changes over time. That is, it would be reasonable to declare that the teaching performance has reached some kind of high mark
when the formative input ‘levels off’ for a given course over time. Formative queries like, "what can this instructor do to improve his/her teaching" will usually elicit litanies of characteristics that students believe they like about teaching; a reflective and responsive instructor might observe a change from reasonable (Less reading! More examples!) to unreasonable litanies (give out chocolates! Fewer exams! Grade on a curve!). I would argue that when formative input reaches a consistent level – you ask, “what can I do to teach better/make this course better?” and students answer, "I really don't know" or, "nothing, the course was great/as good as it could be", then you'd have a criterion-referenced indicator that the course (or teaching – it’s important to separate these as much as possible) is as good as it could be.
Formative input to teaching is difficult to elicit, and sometimes even harder to respond to – particularly, as Dr. Gelman notes – if we were hired for our research skills and histories, and not so much for our teaching abilities. I work exclusively with graduate and post-graduate students, and would be interested in how reasonable these suggestions appear for individuals teaching undergrads.
I have various thoughts here, but just briefly:
1. Yes, you can be a great teacher for some students and a poor teacher for others. But there are people like Frank Morgan who are great teachers for everyone. Such people also get good evaluations, I imagine.
2. I suspect that incentives matter. One thing I like about student evaluations is that they motivate the teacher to be a little bit prepared in class.
3. I agree with Seabreeze that some of the best teachers are those who have had to struggle with the material. (But Frank M. doesn't fall into this category, I think!)
4. I agree with Jhgeorge and Stephen that some pre-test and post-test would help a lot. One difficulty is that with college classes there aren't always easily-available standardized tests. I suspect that the common practice (which I do too) of writing up a new exam each semester, is actually a terrible way to to.
5. I agree with Rochelle that it would help if teaching were taken more seriously. It's probably done better in professional schools than in Arts & Sciences places like my own departments.
Your first paragraph reminds me of Richard Felder, who asked whether good researchers make good teachers. "Any public challenge to the dominance of research in the academic incentive and reward system inevitably generates a spate of rebuttals from professors and academic administrators, who argue that teaching can only be done well by researchers. Surprisingly for research advocates, they offer no evidence to support this argument: the best most of them can do is point to some professors who are good at both activities, which is like claiming that you can only be a world-class organist if you practice medicine in Africa and pointing to Albert Schweitzer to prove it." from Who Should Teach in College
I am currently a first-year grad student attending a predominantly research-oriented universtiy. I find myself often frustrated with the quality of teaching. I am near the top of my class, but sometimes it seems that this is only because of long hours outside of class studying the textbook or searching the internet for a better explanation of the material than presented in class.
I am not sure what to make of the fact that my teachers, while presenting the material, refer to some of the concepts as "basic" or "easy." Granted these have been 1st year introductory probability and statistics courses and I am sure they are "basic" and "easy" for them, yet I still find the notion to be a bit insulting. I wonder if perhaps the material was easy for them the first time, or if perhaps they have forgotten how difficult it was and how much time they spent studying the material to achieve their current level of proficiency.
Another one of my impressions, and please correct me if I am wrong, is that some of these programs desire to weed out the less intelligent or serious students. Unfortunately, it seems that in so doing, the classes seem harder than they should have to be. In other words, I thought the goal was to teach us the material, not to make every lecture and homework assignment feel like a test of your aptitude.
My feeling, unfortunately, is that the professors don't really desire to become better teachers, the good, intelligent students with an aptitude for the subject will simply find their way to understanding.
Too busy with their own research to teach…. and altogether seemingly uninterested in teaching the material. I actually went to an office hour and had a teacher tell me that he didn't want to talk about that homework problem because it wasn't intersting to him.
Other times when I have gone to office hours, instead of trying to help fill in gaps in my understanding of the material, the professor makes me feel like an idiot for not knowing how to do the problem.
I know that with much material, students need to struggle with it for a while and only then will you truly learn it, but I thought that professors were supposed to be there to make the journey to understanding easier…not to make you feel like you are stupid or haven't tried or studied hard enough.
I don't know. Perhaps this is just part of the process of preparing us for research on our own someday when no one is going to be there to help you understand the material. I wonder if someday I will look back on this experience as a positive upon realizing how much I learned during these years. Perhaps even realize they were just pushing me to excel. Whether or not my professors are performing their duty in the way most conducive to learning, I'm not sure… but sadly, it sometimes doesn't even feel like one of their goals.
Perhaps that is what is really going on here, research professors just aren't concerned with improving their skills as TEACHERS. The impression I get from the teachers is " I am smart, I understand the material, if you are smart like me, you too will figure out the material and start publishing just like me. Until then, you better hurry and catch up if you can."
Who makes the best teachers you ask? I say it sometimes comes down to whether or not the teacher cares about presenting the material in a way which is most conducive to learning. Granted this will not be the same method for everyone, I am sure that even a modest effort put forth will be noticed and appreciated.
I don't know. Maybe I just don't "think on the same level," but in the end I am understanding the material and ending up at the top of my class… the only question is, could it have been easier for me to attain my current level of understanding with a better teacher, and perhaps more students would have learned the material as well. I think that before we sit down and think about what makes a good teacher, we need to start with some specific, well-defined goals… then good teachers are simply those who achieve these goals. Too often it seems that professors don't have well-defined objectives for the course except for to get through the book and touch on everything that is expected of them to cover.
The modeling concept is universal for every learning topic.
I am a strong believer in guided reading programs.
My wife is a special ed. teacher and recently impressed me with this program called, Read Naturally.
I did a write-up on it here:
http://productreviewratings.com/2008/12/read-natu…
I hope you find it interesting.
Thanks,
Rob
I see that this discussion is several years old, but I'm very interested in the outcome. Is there a metric for identifying great or even just good teachers?
I'd rather ask a different question: What is the potential for candidate A, B or C to be a good teacher?
Maybe "teaching" is too broad and we need to focus on specific skills that good teachers possess. The excellent teachers probably score highly in most skills. Poor teachers may not score high in any skill.
I believe that we are born with a fairly unique set of strengths. As we discover those strengths and move towards a career that utilizes those strengths we will find greater satisfaction and success. (Google "strengthsfinder" to learn more)
I think the earlier comment from Joseph Delaney hinted on one of the key skills I think teachers need to possess; that is the ability to perceive and respond to the questions of the students that are yet unasked. This skill varies from teacher to teacher in how many people they are addressing. Some do it well with only a single student. Others do it well with a huge audience. The metric might be to determine how many people are in your sweet spot.
Back to my question. Let's say I have 25 people that work in my company and I want to invest time and money in the one with the greatest potential to be a good teacher to help with training. Who should I choose?
I don't really want to know if they think they are a good fit for the job (a limitation of self assessment tools), I would prefer to use something less subjective, if possible.
Any suggestions?