What are the best sources on how to be a good teacher?

Charlie Clarke, a Finance PhD student at UConn, and a loyal MR reader, writes to me:

Hi Tyler,

I’m a grad student teaching for the first time, and I was wondering if  you had any recommendations for a book relaying evidence based advice for teaching methods.  I know Cowen’s law, “There is a literature on everything.”  Just hoping there is a good book or two synthesizing that literature so that I can use it to improve my teaching.

Love the blog.

The most important lesson is to use the right textbook.  Beyond that:

1. Give a damn.

2. Get to the point when you speak.

3. Expect something from them.

4. Teach to the students who are interested in learning.

5. At all levels, do not overestimate the attention span of your audience.

6. Do not be afraid to be idiosyncratic, provided you adhere strictly to #2.

Those are my tips.  But to be honest, I do not consider them RCT-tested and I am not sure they maximize social welfare.  They instead start from the premise that the key question is what kind of person do I want to be, and then the method asks the students to conform to that vision.  Some or all of them might prove RCT-neutral, or worse.  Nonetheless, the approach is a good way to motivate me and that is part of the problem.

Doesn’t Bryan Caplan have a post on this?  Here is John Baez on how to teach.  Peoples, what can you recommend from the literature?


No backing data (of course), but anecdotal experience from engineering is competence, confidence, and competition. Master your subject (e.g. maths), teach it as if you are right (this is one reason math is good), then see who can approach your desired answers on a brutal grading curve (lots and lots of math). There is a bit of a free lunch in setting the bar at 'unattainable.' It's a bit of a scam because I never had good engineering teachers per se, but this allows them to teach well without being good teachers.

Think back to which of your teachers were good, which awful. Copy the good ones. The fact that this simple procedure is beyond many academics suggests that rather a lot of them are none too bright.

Isn't the problem that you spend your entire educational life in a system that said "these teachers are so good that if you don't do exactly what they want, you fail!"?

Or people think that this suggestion is so obvious they don't bother to mention it.

Ken Bain's _What the Best College Teachers Do_. Harvard UP, 2004.
Doug Lemov's _Teach Like a Champion_. Jossey-Bass, 2010. This is K-12, but has much to say to post-secondary teachers anyway.

Avoid using Powerpoint, unless you have extremely good reasons to justify it.

Teaching is an art, much like acting or painting and not everyone is gonna be good at it. Still, I would emphasize:

1. lecture preparation: Have a plan, have talking points. Have good short precise prepared sentences for major concepts and don't wing it and ramble on the spot. Like a seminar.
2. examples/stories. Good examples bring life to concepts and theories and make it a lot easier to remember the generalities.
3. ask them questions, let them guide some of the discussion. Ask them what they would do, keep them engaged and awake. Group work also helps because some of them will ask each other stuff they won't ask you.
4. Repeat, repeat and repeat. people's attention will drift in a 75 minute lecture.
5. Have fun. If you're in love with the subject and excited about it then you will be a lot more bearable and will make the class fun even if the subject itself doesn't hold much appeal to your students.
6. Be firm with your policies. Students respect harsh, demanding but fair policies but if they see that you wanna please them then they will exploit that. Nobody likes a pushover.

Considering they charge, what $750 per lecture, isn't the lack of quality control in teaching kind of astonishing? In a more normal universe teaching would require rounds of auditions, mock classes, and peer review before you were sent out there. Anyway what works for me is to constantly while teaching scan my body for tense muscles and relax them.

Perhaps the lack of quality control comes from the fact that the faculty are hired more for research than teaching. Quality control in teaching would compromise research quality, which as opposed to teaching is perhaps what universities are actually reputed for. This need not in every single aspect be anti-student; for instance, a student might get into better research projects at a university that has good faculty.

BTW the $750 per lecture bit is surprising to me. I mean, of a Professor's salary let us say $30,000 per year comes from teaching. If the Professor teaches two 16 week semesters, say 2 + 1 teaching load, something like 3 hours a week. Accounting for holidays etc., may be around 90 hours per year. So roughly $333 the Professor gets per hour (rather, 50 minutes). Where is the extra cost going? If the Professor should get, say $660 per lecture, it means around $60,000 per year he/she gets from teaching. I doubt that. Of course there could be mechanisms I am unaware of.

I was referring to the price the university charges students per lecture, not the price the professor charges the university. It's actually more like $125: $30,000 per year / (8 classes * 30 lectures per class).

To be fair shouldn't you ascribe more to what you get from the $30K than just lectures?

Cynically of course,'no', but in reality your 30K gets you lectures, tests, student life, signalling value, etc.

Bob, I know you were referring to that. My question "Where is the extra cost going" is a real question, not a rhetorical one.

Regarding Rahul's question on middleman below : interesting point. May be to some extent the university channel's the students' funds into the Professors' research compensation? Again, may be the Khan academy sort of thing is a first step towards eliminating the middleman.

I think your math is way off.

On a semester system, students generally take about 15 hours / semester, so 30 credits a year. A standard 3 credit course will meet for 150 minutes per week in 3 50-minute or 2 75-minute sessions, and a typical semester is 15 weeks. I just counted, and Tuesday/Thursday classes at my university meet 28 times this semester. A student taking a full load of twice a week courses would, then, have 140 class meetings.

Also, I'd like to see some evidence that average tuition is $30k / year.

According to The College Board, public 4-year colleges charged an average tuition and fees of $8244 / year for in-state students. Private 4-year institutions charged an average of $28,500. 44% of full-time undergrads attend a 4-year college with tuition and fees < $9,000 per year.

Estimating generously in your favor, let's give an average tuition of $11,000 per year. At 280 class meetings, that's $39.29 per course meeting.

Note the large difference between what the professor charges the university and what the university charges the students.

Isn't there substantial incentive to eliminate the middleman?

Doug Lemov's "Teach Like a Champion"

I second Rahul's point about Powerpoint or Slides. The problem is that in many universities these days you don't have real blackboards or whiteboards and soon run out of space. Or if you change lecture theaters last minute..... Another problem is that if you are a bit tired or so, slides can help. And finally if you teach large groups like I do (350 kids or so) then readability becomes an issue. Plus there is some logic for giving students at least part of your lecture slides/notes. Not all to be sure, then they don't copy it down, which in itself contributes a lot to learning. So my compromise is to use a tablet computer (if you can afford it buy the best, tablets are still in their infancy, so get something like Lenovo Thinkpad Tablet. And by all means get an SSD harddrive) and I use the blackboard/whiteboard in addition if available. In some lecture theaters where you have two projectors, you can use one with your tablet and the other one with a normal computer. Of course non of this applies if you are in a good old lecture theater with six blackboards, chalk, and porters that clean it every break with water. But where do you have these conditions today?

Another great thing is using clickers (personal response systems), see for example this video by Harvard's Eric Mazur:
If you cannot afford clickers you can do it by show of hands or by using colored cards (buy one for each student, even if you have a thousand students you can easily afford that out of pocket). The main advantage of clickers is simply that they are NOT visible to peers, so you have less problems with everyone being too embarrassed to volunteer an answer and/or everyone imitating the best students' answer. Clickers give you a more accurate picture of how little people know, and are not swamped by everyone imitating 5 kids in row 1, as no one but you can see what these kids think. Mazur has a longer great video on this, and the obligatory Science article.

"Students were passing exams without having understood the fundamental concepts he was trying to teach".

Hmm. Sounds like his exams were not meaningful to begin with.

Mazur is also the main story in "Twilight of the Lecture," which is worth reading on its own, and especially if you're the sort of person who prefers to read over watching.

Much of what people have said will make your students' experience better. Whether your students will actually learn more is another question. It's not just that there are very few randomized clinical trials in education. There are almost no studies which use long term learning as the measure of success or failure. But since almost none of your students will need to use or remember what you taught them a year from now, perhaps that is not really so important.

For math, hand-write notes and restrict exams to what is in both syllabus and notes. This does wonders for keeping to the important stuff. The best seminar speaker I saw wrote everything by hand too.

Feedback is important. That means meaningful and rapid grading of assignments and exams. Another feedback is from students to professors via standardized evaluations of professors by their students. While professors shouldn't teach to standardized evaluations that students fill out on their professors, professors should not be afraid of them and should carefully read them to better understand after the fact how they were or were not reaching their students.

I have a data point that might suggest student feedback is less than helpful. At one university where I taught, we had a scheme with 6 or so required questions; you could then add another 20 or so questions from a list of a couple hundred. For most courses I spent a fair amount of time going through the list and selecting 5 to 10 specific questions I wanted feedback on (usually changes in some part of the course that I was trying out).

At that time the surveys had not yet become official, so it was up to the instructors to collect and turn in the surveys. There was always a huge delay in getting back the results. One semester, I was really curious to know how a major course change had turned out, so I opened the envelope to scan the comments. What I found was that all the work I did selecting the questions seemed to be worth nothing.

Looking over the scoring sheets, I realized I could easily sort them into three piles: the largest one was a pile of scoring sheets where I got the top or next-to-top scores in all the questions; a much smaller pile was students giving me the lowest or next to lowest score on all the questions; the smallest pile of all was where the students gave me a significantly different score on different questions. The last pile was maybe 5 students in a class of 40. So I got a good evaluation score, but not much information.

I started looking at all my classes after that; for as long as I saw the scoring, always three piles and always the ones who seemed to bother taking care with it the smallest pile. So I quit asking the optional questions, and I decided that what students really evaluate is whether or not you are likeable.

So, the algorithm for my score average was approximately:
(top score)*(fraction of students who liked me) + (bottom score)*(fraction of students who didn't like me)

As feedback, it's not all that helpful. I have learned that any time I have to say something negative to a student it will harm the score because students are offended by negative feedback.

This is a bit unconventional, but stake out your "competitors" i.e. check out the coursework, lectures, notes, textbooks etc. of other professors teaching similar classes. Would your students be able to solve their problem sets?

Having studying outside the American system, one feature that I appreciated in hindsight was that the teaching and the testing were done by two entirely different professors. You never knew which "textbook(s)" the tester was partial too; forced us to hedge our bets about textbook loyalty.

>>>The most important lesson is to use the right textbook.<<<

As an instructor what's even more important is to never develop dogmatic loyalty to any one textbook. (You obviously cannot ask your students to buy six textbooks, but still) It's exceedingly unlikely that one textbook will outdo all competitors on all topics.

Much of this advice seems to focus on teaching as performance art, while most of the theory and practice of recent decades leans against the storied "sage on a stage" mentality. Try Elizabeth F. Barkley's "Student Engagement Techniques". If nothing else, it offers a sizable toolkit of methods one can draw from.

Unfortunately, much of the research arguing against the "sage on the stage mentality," has, like most education research, been largely unreliable. This study, http://educationnext.org/sage-on-the-stage/ , tries to inject some rigor into the question and finds that the "guide on the side" or "student directed" or "problem solving" or whatever you want to call it method isn't as effective.

1. Knowing is the ability to do something. Decide what they should be able to do on your tests and work backwards from there.

2. Think of tests as a way to assess how well you taught. If you are going to make 70% a C, 70% of your test should be things you want your C level students to be able to do. Curves are for people who can't be bothered to write a thoughtul test.

3. Your student's attention spans for listening to an uninterrupted lecture are about 20-30 minutes. If you plan on talking for 50 minutes, half of what you say might as well be delivered to a wall.

4. Look at books on coaching sports. Coaches, unlike college professors, actually must teach well to stay employed.

Regarding #4 though, coaches also get more leeway and resources to cheery-pick the right sort of raw-material.

I found Anna Goldsworthy's memoir PIANO LESSONS highly worthwhile:


Depends on what you are teaching.

If you are teaching graduate MBAs with some practical business experience, then introductory lectures followed by in class projects and problems, with them doing the presentations and you guiding the discussion. Follow up with a strong summary incorporating their points, adding points they missed, and summarizing. End with praise.

If you are teaching law students, make them struggle to find the point of the case, ask hard questions randomly, start off a summary or outline and have them complete it with your help, and then give them a problem to solve in the class. Draw in other disciplines--law, psych, sociology, medicine, insurance, business and marketing--to make the course more interesting and relevant and to show how and why the law changes.

There's a booklet produced by an MIT professor called The Torch or the Firehose that may be useful.

I would second basically all of Anshu's points. Prepare well, and attempt to engage the class rather than spew information. "Engaging" them can mean something different depending on whether you're teaching a class of 20 or 400, but even in the large classes you can throw out the occasional question to make sure they're still with you. Depending on the topic, some sort of demonstration or other deviation from a standard lecture can help too.

To this I would also add "Be available, and communicate promptly during not-class." Yes, this takes a lot of time. But students will notice if you consistently answer their emails quickly and informatively.

I don't know where bob went to school, but even more expensive universities have undergraduate tuition closer to $100-150 per lecture, assuming a standard courseload of 12 lecture-hours per week....

Regarding slides, it's worth talking about why many people are advising against them. Objectively, slides are neither good nor bad--they're just one of many possible tools. However, they're a tool that is very easy to misuse in several different ways. Since many lecturers do indeed misuse them, you often hear advice to avoid them altogether (either from people who think from experience that they're always bad, or merely recognize that it's easy to go wrong).
Given this, it's perfectly reasonable to avoid slides. The following applies if you decide to use them anyway:
The chief error lecturers use with slides is going too fast. Since you no longer have to write up the content, you're not slowed by your own writing speed--so there's a temptation to go approximately as fast as you can speak rather than as fast as you can write, which doesn't help the students.
A related problem is trying to convey too much content on a slide. Now, more detail per slide is more acceptable for a lecture than for a presentation-style talk (where you only want minimal bullets). But it still shouldn't require 12-point font.
Don't just read your slides. If you're doing that, you might as well just send students a copy and go home. Discuss them. And be perceptive toward your audience--try to encourage questions about points you're raising.
Another thing to remember is that unlike a blackboard, once slides are advanced, the old material isn't visible anymore. Sounds obvious, but it can really help if you repeat an important definition or equation on subsequent slides if those later slides heavily use that information.
Print out your slides beforehand and provide copies to the students at the beginning of class. This allows them to both look back and reference earlier slides during the lecture if they need to and take notes directly on the slides.

Sorry for the run-on paragraph there; there are supposed to be single line breaks. (Any chance of a preview button?)

A "preview button" would be good; being able to edit after posting would be excellent.

After posting the above, I was sent back to the top of the page and lost my place. Inefficient and awkward.

Lemov's book is very good and easy to read. It's aimed at the K-12 level, but probably applies more to higher levels than most people usually think. The APA just released this, too:


I have been teaching MBAs and Executives for five years. Not successfully, initially. But I have got pretty good at it. What changed? Well, I understood that at that level you don´t need to teach theories or details. You need to provide a vision, a perspective that helps them to achieve whatever they want to achieve (get a job, impress others, etc.). In many professional environments, there is knowledge already in the student and the job of the teacher is to make the student aware of this. I teach Marketing, but my wife is a Finance professor and the same applies. Why do you teach option pricing or the Miller Modigliani Theorem if your students will not be finance quants or academics? Because these ideas will change the way they see what they already know. But the first question you should ask yourself is "what is the vision change I want to encourage?" Then find a way to do that, examples, cases, lectures (powerpoints also), activities, games, discussions, stories.

Re textbook choice, let me offer a modest revision. Choose the right textbook and hide it in your office. Assign the second best textbook to your students and dazzle them with your brilliance.

+1 Lemov's Teach like a Champion. I teach middle school but can see how the techniques in the book apply at all levels.

Tyler's #5 (don't overestimate attention spans) is crucial to being engaging as a teacher/lecturer.

Lastly, think about what you want students to do during class and outside of class. Be transparent with those expectations in a syllabus and your introductory lecture. Whenever possible, have students do something besides listen and take notes during class - lasting knowledge comes from engaging more modalities than just auditory & visual.


Be open to feedback, try different things, and try to improve one year to the next.

Two books that were influential on me are "How People Learn" and "Understanding by Design." I would add one critical bullet point to Tyler's list: you are not a good model for your students. I.e., don't just copy what you liked in your teachers, because the sort of people who become college professors are not typical. My best advise, as someone else mentioned above, is think of yourself as a coach. To teach someone to say, shoot a free-throw, you don't give them a lecture then go straight to grading them on their performance. You have them do it, repeat it, and give them feedback. You put them in a situation where they are free to try things and make mistakes. Ultimately, if you adhere to #1, you'll be fine. But that's pretty difficult in the long run. I lasted about five years.

In defense of presentation slides, think about Apple product announcements. Hardly anyone finds them boring, but in essence they are just a Keynote presentation. Having an iPhone to unveil does help, but Apple's competitors did not manage to successfully copy what they do.

Here's a few things I'd focus on:

1) Do not copy lecture notes on a slide. Make a few short (2-5 words) bullet points per slide. The powerpoint should not include all the information you want to convey. It should be more of a supporting structure, a frame.
2) Use graphics, pictures, even short YouTube videos.
3) Buy a remote control and set up animations - each click should make a new bullet point appear. Otherwise your audience reads the whole slide once it appears. It creates two problems - they do not listen to you just after the slide comes up, because they are reading it and they think they know all you are going to be talking about while this slide is up. Think of it as a spoiler.
4) When creating your presentation: a) show/create a problem or a puzzle, b) solve it. Have you ever noticed how Apple presentations begin with an 'adversary' (i.e. phones with keyboards, thick, heavy computers, CDs) and go on to present a solution (i.e. iPhone, MacBook Air, iPod)? It engages the audience. Create a narrative.
5) Creating a good presentation requires some time and preparation. The slides are set in stone, so you will need to think about the structure of the lecture, main points etc. It is both good and bad. In some cases, it works perfectly and you should stick to it. But do not be afraid to improvise if a chance comes up, and remember than not every class/subject requires a presentation.

Best advice I ever received: Make sure that YOU (the teacher) are always having fun. Don't think about how to entertain them, and only secondarily about how to accomplish your teaching goals. Your first priority should be to entertain yourself.

BTW: John Baez is a freaking hero-teacher. His trilogy of Glasgow lectures on his favorite numbers was absolutely masterful. He did not talk down to his audience but he made some rather advanced mathematics comprehensible and vivid. Possibly the best academic lectures on the internet. I'm a guy from the humanities, but I was riveted.


I'm a grad student in philosophy, have taught two classes, and had similar questions as the finance student, so I looked for some research to read. Here is a book by Daniel T. Willingham (a cognitive scientist at UVa(?) that I liked a lot that I don't think anyone has mentioned (he says some things others have mentioned but also more):

"Why Don't Students Like School: A Cognitive Scientist Answers Questions About How the Mind Works and What It Means for the Classroom"

Amazon link:

That got me thinking how far can one really go with generic, subject-agnostic teaching advice. e.g. For a field like philosophy won't the optimal teaching strategy be far different from, say, finance?

Rahul, I know absolutely nothing about finance (let alone teaching finance) so I couldn't say how similar or different the optimal teaching strategy would be. However, there are commonalities among brains with regard to the psychology of learning and many of the lessons seem to be easily be translated among disciplines. Many of them involve things people have mentioned (e.g. attention span, organization, etc.). Willingham does a great job presenting these ideas accessibly.

Depends on how you define success. If your goal is student satisfaction, the advice presented here is pretty good and entirely consistent with the evidence in the literature, On the other hand, if success means that your students can actually do something, the evidence suggests that show, do, teach, repeat (until the behavior becomes habitual), is more effective (this is not to negate the advice above on how to show). That is: show them how to do it, have them do it, repeat until they get it right, have them show someone else how to do it, and repeat this step until they can do it without thinking about it (as opposed to thinking about the context, the people involved, and the application). If the skill isn't important enough to justify this kind of investment, it probably isn't worth your time or theirs.


It seems that we have very weak to no evidence that student satisfaction is correlated with student learning. What if the best way to teach involves repetition that students find boring? What if occasional humiliation is actually productive? What if it is better as a teacher to be hated than to be loved? Would a polarizing teacher do a better job than a popular one? I truly don't know the answers to these questions. But I had teachers who were great and inspirational lecturers who nonetheless were not very good at teaching me how to handle the more tedious homework. And I had others whom I wasn't impressed by but who helped me get through difficult material and master it.

Won't this essentially devolve to a paternalism debate: Does the teacher know what's better for the student than the student himself?

No. Good teachers get good student evaluations. Period.

Some bad teachers get good evaluations by bribing students with good grades or fun activities. They are frauds and showmen.

I commonly hear the excuse from bad teachers that the students are lazy and apathetic, and they can't be expected to teach to students who refuse to learn. This is denial. It usually comes from an arrogant teacher who expects the students where he got a job to be as good as HIM and as good, in general, as students where HR got his BA, MA, or PhD. Ain't so.

I disagree. I went to college before there were student evaluations of teaching, but I vividly remember most of my classes. The correct outcomes would be:

1. Good teacher I would have give a good evaluation.
2. Bad teacher I would have given a bad evaluation.

I probably would have rated at least half of my instructors correctly.

But I would have given good evaluations to teachers who were entertaining (relative to the subject material) even when -- in retrospect -- I realize they were just entertainers.

And I would have given harsh evaluations to teachers who were cold and uncompromising and maybe unfriendly; the ones who hammered home the material that has been the most valuable to me ever since. The ones who made the most lasting contributions to my intellectual development.

Where Dr. Brown and RJ disagree may be the margin between a good teacher (who coldly hammered home material of lasting importance) and an exceptional one (who did all of that while inspiring the class).

2. Get to the point when you speak

That's what I love about Tyler's speeches (the ones online). He goes to the podium and immediately starts conveying ides - none of the usual perfunctory nonsense/ throat clearing.

No one has discussed students ubiquitous smartphones, blackberries, notebooks, netbooks, and assorted gadgets, most of which they drag around with them everywhere they go, like futuristic umbilical cords (but wireless).

What does one do with them? Ban them? That's hard to do/annoying to enforce, and it makes the students irritated and fidgety. Allow them? That makes the teacher annoyed, as he or she sees quite a few tops-of-heads.

Final thought: if you (over)script your lectures, you will be boring. On the other hand, spontaneity is draining and exhausting, like stand-up. Okay, yes, exhilarating but still draining and exhausting. So tread that fine line...

What Hmmmm said. Bain's book has a lot of insights, but his approach is not for everyone.

There are journals out there that do present some evidence-based analysis of what works and what doesn't. (Much as I love the Journal of Economic Education, it could do better on this score.) Much of the evidence suggests that the productivity of lecturing is limited.

If you're teaching economics, sigh up on the tch-econ listserv (http://org.elon.edu/econ/tch-econ/). Lots of really good stuff appears there.

Do this one little thing: Ask questions *before* you call on anyone.

Exhortations to be exciting, get to the point, etc. offer little practical help because they do not offer specific guides to behavior. Good teachers have good habits--habits that can be tested, refined, and adopted regardless of innate dispositions. The best work I've found on that front: http://www.amazon.com/Teach-Like-Champion-Techniques-Students/dp/0470550473 (written for elementary school teachers, yes, but so what?). The single best technique, out of the many offered there: Ask questions before you call on anyone. That gets everyone thinking about the answer. If instead you call on somebody and then start asking questions of him/her, the rest of the class relaxes and tunes out, because they know they are off the hook.

Also: Repeat important points. See, e.g., supra.

When I was ABD, I told one of my professors about a really cool computer simulation I had written for students. He looked at me and said, "If you are putting your energy into teaching, you'll never finish your dissertation."

It's sad, sad, sad, but your institution does not care whether you are doing a good job teaching; they care about whether you finish your degree and get some publications written. (Proof: if they cared about your teaching, they'd put institutional effort into teaching preparation, provide incentives to good teaching or disincentives to bad teaching.)

I'm not saying this is a good system; just saying this IS the system.

Peoples, what can you recommend from the literature?

Why Don't Students Like School: A Cognitive Scientist Answers Questions About How the Mind Works and What It Means for the Classroom; see also the books Amazon recommends in addition to Why Don't Students Like School. Though it might sound strange, see the Atlantic article "What Makes a Great Teacher? From it, I learned to seek regular feedback in the form of hands; see this post for more. One other thing I do: ask students to write short passages, then I go around and read / skim about half the answers, mostly to figure out what, if anything, students seem to be getting, and I try to adjust what's going on based on that.

One major thing I'd add to Tyler's list, though it may already be implied by 1., 4., and 5.: be ready to change. Have a plan, but if you realize you need to do or teach something else, do or teach something else. If you're just mechanically executing a series of steps, regardless of feedback loops, you might as well be a YouTube video.

Most of the books I've read on teaching and learning haven't been very good.

>>>Most of the books I’ve read on teaching and learning haven’t been very good.<<<

That's because most of the research on teaching itself is pretty sketchy. We hardly know much rigorously about what makes a good teacher.

An interesting quote from the Atlantic article was this:

"Meanwhile, a master’s degree in education seems to have no impact on classroom effectiveness."

This makes me wonder fundamentally about the quality of education departments and their research product; is the degree they peddle quackery?

"Research on teaching"

Research is the problem. Professor incentives are aligned heavily toward research with a minimum teaching quality constraint.

Research on teaching is like a modern anthropologist studying aboriginal society. The researcher can't relate and brings too many biases.

Good teaching is an art. It can be learned, but doing research on teaching is largelt just a way for good teachers to meet minimum research requirements or for "specialists" in learning to justify their existence.

This makes me wonder fundamentally about the quality of education departments and their research product; is the degree they peddle quackery?

Quick answer: Yes.

Longer answer: If you talk with teachers about ed school, and they think that you are sympathetic, they will tell you,

1. Student teaching was useful.

2. I got a few useful things from my education courses.

3. Mostly, my education courses were a waste of time, but I needed to take them to get a teaching job.

“Meanwhile, a master’s degree in education seems to have no impact on classroom effectiveness.”

I see a lot of the best teachers in public schools around me spending time away from their craft in order to climb the degree ladder to get better pay. I've sometimes wondered if there is a difference between getting the degree before teaching and getting it while teaching. And, if there is a difference, is it one of selectivity? If there is a distinction between the value of earning it before or during teaching, might that mean that earning it during teaching has a negative impact on the classroom? While I can't answer that question, I suspect that there is a small negative impact. If that is the case, the current incentives for climbing the pay ladder are infuriating (moreso than they already are).

1. Give a damn. Everything else follows from that.

If you're asking how to teach well, you're on the right track. I second much of what the others say, especially emulating good teachers you've had. Ultimately you must become your own best example. Teaching methods and style depends crucially on the number of students, classroom characteristics, subject matter, and student ability. You can't expect to teach Statistics the same way you teach Principles of Economics or Classical Poetry.

You can't teach students at a community college or Tier 3 university the same way you teach at a flagship university. Set your standards NOT based on what you had to meet to prepare for grad school but what THEY need when they graduate. Many professors step DOWN in academic tiers when they get a job - your students could not get admitted into the university you went to. Even your best students will retain only about 40% of what you teach. So set realistic standards or you will frustrate them and disappoint yourself.

Using a blackboard or whiteboard is best. It helps you pace the student's writing. It allows you to add, ad hoc, information motivated by student questions.

You can use PowerPoint. In some cases you must. But you must understand its strengths and limitations. Slides should have one main idea per slide. Don't have distracting animations. In charts, have lines appear as if you were drawing them. Most of all, remember that students are taking notes and you must give them time to do so. If you make the slides available online, attendance will drop. On the other hand it gives them something tangible to use.

Teach to ALL your students. It's your job to motivate them. The 5% who cannot be motivated aren't probably not in class anyway. Your goal should be to lift every student to a threshold level of understanding. You will undoubtedly fail to reach everyone, and no matter how much of it is their fault, you should consider where your efforts came up short.

Motivate them at the beginning of the course by explaining why your course is important to them. The most immediate reason is that they need to pass to graduate, you are their partner, not their enemy, in that goal, and being in class is the most important factor of success. Then talk about why the stuff you're teaching has some practical importance in their lives. Then you must deliver on your promise to keep it important to them with each class.

Pass out your syllabus on Day 1 and continue to make it available on subsequent days for late registrations. Make the key points on your syllabus testable material.

Give very small homework assignments for each class. About ten to twenty minutes of homework. They will do this. Thirty small assignments are better than five big ones. It keeps them engaged between classes.

Give small quizzes every two weeks (small class). Have students grade each other in class and then spot check them. Exams should look similar to the quizzes so they understand your style.

Always remember that your course is only one of four or five courses your students are taking. Your course is not more important to them than the others.

Consider the prerequisites for your course and make sure the other professor did what he was supposed to do. You probably have to refresh student memories on some things. Consider what courses have yours as a prerequisite. Make sure you cover everything you need to.

Prepare! Get your game face on before class. Encourage questions but stay on schedule. Close by summarizing what you covered, tell them what to read for the next class, and assign homework.

Encourage them to study in pairs or sets. Always have an open door, but have them contact each other for help or questions before they come to you.

Don't force or grade attendance. They are adults and know how to budget their time. Motivate them to be there and have them get notes from fellow students if they miss class. Don't reteach a class. Set your standards on Day One.

Don't grade on a curve - it's a tool for bad teachers. Teach to a fixed standard and grade everyone to the same standard. However, if the mean grade on your exams is 47, then YOU or your exams are the problem. Curve only to correct problems caused by your mistakes or bad test writing. Don't give extra credit- you will regret it and they will abuse it.

Writing good exams is HARD, so spend a lot of time on them. Exams should represent what you taught. Good exams and quizzes are teaching tools themselves because they help reinforce what they know. When they get a question right, they should be thinking, "Wow, I know that!" Multiple choice exams are hard to write and easy to grade. Essay/ short answer us the other way around. Your exam questions (and multiple choice answers) must be CRYSTAL CLEAR. Anticipate possible misunderstanding.

Work current events into the class.

Good luck and good teaching.

Some very good points! But if they are undergraduates, they aren't exactly adults, paradoxical as that might sound. If you don't incentivize attendance -- or make it highly important in some manner (final exam) -- many will not attend.

Lectures should be informative enough that skipping them will be its own punishment. (If a particularly bright student wants to try to learn everything from the textbook and sleep in through the lecture, let them try. Most will fail.) If someone can skip most of the lectures and still do well on the papers and exams, the fault is with the lectures.

I've been told that in lower division classes, the lectures are to explain the readings; in upper-division classes, the readings are to explain the lectures. (That was at Berkeley in the 80s. The dividing point may be undergrad/grad these days, depending where you are.) Either way, figure out whether you're explaining the readings, or providing material for which the readings are merely the background, and explain that to the students on the first day, and again throughout the course. Especially if your lectures are upper-division.

I just want to follow up with "good tests are extremely hard to write." You can usually tell if it's a good test by taking it, but not always.

What's interesting is that today you can actually see what good college lectures look like in your field, and bad one's too by going to ITunesU or oercommons.org

You can see the difference.

After reviewing some of the courses in your field in ITunesU: ask: what made one course good and interesting and one course on the same subject bad and boring.

For large lectures, if the lecturer speaks in paragraphs, with lead in sentences and summaries, it is more interesting than if the prof appears unprepared, goes all over the place, and doesn't have discernable plan.


Check for understanding, check for understanding, check for understanding. Don't assume you are doing a good job if only the most brilliant students get your point. You *must* take a random sample of the class when questioning them in order to see whether they are getting the idea. Choose students randomly from your roster and call on them. If a majority don't understand the concept, then you need to re-teach it.

If you don't have time to stop and question during your lecture, have the student answer one key question on a slip of paper and give it to you as they leave. It can be anonymous. The most important thing is for *you* to see whether your lecture enabled them to answer an important question. If it didn't, then you need to re-think your way of teaching.

As a student and starting techer, I always wondered whts others take on who to teach to? I am affraid that teaching to median runs the big risks of boring better students and driving them out of the class, thus leading to market for lemons problem. Also, it is half way down the road of relaxing the standards every year a bit, lowerung the standards overall. For example in my school ids learn not to give a damm and study couple days before the exam very fast, and then the culture of the whole university is set. However, teching to best is not viable strategy either, so I alwasy thought something like teaching to second quartile from the top. And saying "teach to all" is doable partly in some situations but unapplicable in others. Whats your take on this?

"4. Teach to the students who are interested in learning."

This can become a self-fulfilling prophecy propelling students fortunate enough to have inspiring teachers/parents upward while keeping those less fortunate from participating or caring at all.

When I teach, I spend considerable time getting to know each student and trying to deconstruct reasons why they might not be motivated. While I can't always find that button and push it, I often can because...I give a damn (circling back to 1) and have practiced the process.

If we're not selecting for teachers who can do this, then we aren't *really* caring about social mobility to the extent possible.

Refuse to accept anything that's not stapled. Charge 25 cents for you to staple it yourself.

I hate getting unstapled homework.

.......at that price-point I bet he'l get more un-stapled homework turned in, not less.

1. Go read some John Holt books. Now.
2. realize that Teaching is impossible, what you are attempting to do is facilitate your students' Learning

Good summary of the literature here: http://www.csub.edu/tlc/options/resources/handouts/scholarship_teaching/HalpernHakel.pdf

As they say "it would be difficult to design an educational model that is more at odds with the findings of current research about human cognition than the one being used today at most colleges and universities"

Basically, if you want your students to actually remember anything in one year's time, you need to teach more like the military and less like college.

Ignore the advice you'll hear about trivial issues like PowerPoint vs. chalk or stapled papers. Such things have nothing to do with learning. Focus on actual research-based learning principles. There is this field called "Educational Psychology" that most people have never heard of. We've been studying how people learn for over 100 years, and in the process we've also learned that fostering learning is just about the last thing most schools and universities actually care about. Simple logistics tends to override all other concerns. The big exceptions are in areas where outcomes are easily observable AND the stakes are high. Look to military training, nursing, medical, and dental education for the best examples of how to actually educate. It is not a matter of being a good teacher, it is about designing a SYSTEM that results in learning.

1) The single most important variable in promoting long term retention and transfer is “practice at retrieval.” This principle means that learners need to generate responses, with minimal cues, repeatedly over time with varied applications so that recall becomes fluent and is more likely to occur across different contexts and content domains. In other words, you MUST make demands on your students or they will quickly forget the week's lessons. [The simplest method of doing this is to give daily short-answer quizzes with feedback. This method has a ton of empirical support.]

2) Varying the conditions under which learning takes place makes learning harder for learners but results in better learning,

3) Learning is generally enhanced when learners are required to take information that is presented in one format and “re-represent” it in an alternative format.

4) What and how much is learned in any situation depends heavily on prior knowledge and experience.

5) Learning is influenced by both our students’ and our own epistemologies.

6) Experience alone is a poor teacher.

7) Lectures work well for learning assessed with recognition tests, but work badly for understanding.

8) The act of remembering itself influences what learners will and will not remember in the future.

9) Less is more, especially when we think about long-term retention and transfer.

10) What learners do determines what and how much is learned, how well it will be remembered, and the conditions under which it will be recalled.

If you are forced to teach in a system that does not follow these principles, then my advice is just to enjoy yourself and don't take it too seriously. Teaching can be very stressful. Focus on the most interested students, while keeping the rest happy.

One more principle they should have listed: Spaced practice leads to longer retention than massed practice. Dozens of studies have shown that the interval time between practice sessions is directly related to forgetting rates, and this principle has been know for over 70 years.

Spitzer, H. F. (1939). Studies in retention. Journal of Educational Psychology, 30, 641–657.

A familiarity with the VAK learning styles would go a long way to helping most teachers get the most out of many more students:

http://nwlink.com/~donclark/hrd/styles/vakt.html (short article)

Comments for this post are closed