« How to sound smart around the water cooler | Main | The Australian Labo(u)r Market »
In the shower with Robert Frank
I tend to listen to NPR while showering, and really enjoyed this morning's interview with Robert Frank. The interview draws heavily from his book, The Economic Naturalist - previous blogged about by Tyler, here and here.
Robert Frank's observations on economics teaching will fundamentally change what I do in the classroom. What he has to say is important. Read it. Here.
Yes, this was previously covered on MR (here and here). But I am intrigued by Frank's ambition in arguing that we need to emphasize the "deep" concepts of economics in a way that transforms how our students see the world. We econ profs probably fail, and it is hard to see how to do better. But it is worth doing.
Perhaps blogs like Marginal Revolution help one better see the world through an economists lens. But most econ profs teach in the classroom, not the blogosphere, and so I want to ask: How can we do a better job teaching what is important, true and beautiful in economics? Comments open. But a request: Please only comment if you have taken the time to read the Frank piece (this one).
Posted by Justin Wolfers on October 3, 2007 at 08:29 PM in Economics | Permalink
Comments
My current Econ prof teaches like this too. I was quite surprised when I walked into his class and he said "I don't want you guys to have many referencing for your essays. It's not important. It's more important you see it from the economic POV".
No Naturalist essays though (I'd kill to do some).
But this kind of method is kinda fuzzy (coz he'd teach us the Mainstream view and then 5 mins later talk about the heterodox POV, then switch back to mainstream 10 mins later). But of all the econs courses I've taken (4 so far, not inclusive of introductory), this is by far the most interesting. At some point I wished my introductory econs lecturer taught like that, but ah.
I'd say teaching econs with an anchor in everyday life is extremely useful.
Posted by: Chewxy at Oct 3, 2007 8:59:07 PM
This sounds like an excellent method for teaching. I love the idea of 2 page term papers.
I took macro my first semester at a little liberal arts college (Carleton) almost 2 decades ago from an instructor
I can now recognize was burned out. It was nothing but diagrams and graphs with little explanation and few examples
to tie it to the real world. I zoned out and did almost no work and precious little listening after the first few weeks.
I took the final in about 45 minutes, hopelessly lost and firing bullshit from a firehose.
The next semester I ran into someone who told me I started quite a bit of chatter with the early departure.
No one else had left for the next half hour or so, so they wondered if I was a genius or had flunked.
The answer was neither.
The instructor had done such a poor job no one could answer much on the test and a huge curve was required.
I got a B in a class where I learned nothing.
I've really taken an interest in economics here during the recent "pop" boom, and I wonder if I might not have
stuck with it had I had better instruction.
Frank definitely sold a copy of Naturalist with that interview. Hell, maybe a copy of his principles book too.
Posted by: Different Jeff at Oct 3, 2007 8:59:43 PM
Hello Dr. Wolfers,
I have a technical question regarding economics. One of the concepts central to modern economic theory is that of rationality. By standard definition for decision makers to be rational the consumer theory requires the decision to be both 'complete' (in a binary sense) and 'transitive.'
However, as you would know most humans are bad at satisfying both the conditions to be rational and people like Dan Kahneman etc have shown that to be the case.
Now to get to the point, I once asked a very famous econometrist the question of rationality and he responded by saying that he did not believe in the concept of rationality, but believed in incentives. From your work, I am not so sure how you go about doing things, but do you also have skeptical views about rationality assumed by economic theory and just take incentives to be the real starting point of your work? If this is so do you have a formal definition of what you mean by "incentive." Is it just a general self-interested tendencies of most organisms or do you have something else which is more specific?
Thank you.
~ Mickey Mouse
Note: I believe this skepticism about rationality is widespread in economics and I have confirmed it with a few other economists. Also somebody once asked James Tobin to define economics in one word and I believe his reply was incentives!
Posted by: mickey mouse at Oct 3, 2007 9:01:54 PM
I'm not a teacher of economics, but I'm a sociologist who teaches economic sociology and I also used to be a math teacher, so I've been around the block a few times. I also took some econ courses as a grad student and, as you asked, I read the article you linked. Here are my thoughts:
1. Don't teach as if students are going to graduate school. 95% aren't going to have a career in the academy. Heck, most Americans don't even read books. Drop all math and proofs in the intro course. Math is not intuitive for most folks, even at elite colleges. Trust me, I used to be a math teacher! People will resist any message that comes with math. Instead, boil down the intro econ course into a handful of super important ideas that can easily be taught. And then hammer home a bunch of examples, endlessly.
2. Create something interactive and fun about the course. If you make a course about memorizing stuff for tests and solving math problems, then people will spend most of their time memorizing stuff for tests. And then they will promptly forget it! Instead, make some sort of hands on exercise that illustrates the idea. For example, could people somehow run an electronic market within in a class? Most people will remember that more than any supply or demand curve. Doing it is better than memorizing it.
3. If your department doesn't have it already, restructure the rest of the econ courses into a math intensive track for grad school/math literate students and courses for the rest of the undergrad population. You will broaden the appeal of these courses and get more people, which means better transmission of economic ideas. Right now, people think econ is the sure fire path to riches, and Freakonomics has helped increase enrollments. That means that you can teach anything and fill the classes. But if you want better transmission of ideas, it may not be a bad idea to reach more people with a simpler batch of ideas.
Posted by: Fabio Rojas at Oct 3, 2007 9:18:38 PM
I have a technical question regarding listening to stuff in the shower.
How do you do it? Do you just have a portable stereo on the counter, and tune it to what you want before hopping in? Or do you have some way of controlling, while you're in the shower, what you hear and how loudly?
And, finally, doesn't it have to be pretty loud to drown out the sound of the water?
Posted by: zlguocius at Oct 3, 2007 9:28:48 PM
mickey mouse...the more relevant question in much of economics is whether individual irrationality implies aggregate irrationality; the answer is not necessarily, and there's a good literature on this. Further, rational preferences are fairly not restrictive, and if you're unhappy with completeness, we can always start with choice and revealed preference. Transitivity also tends to fade away quickly due to the "Dutch Book argument" in market situations.
The more interesting behavioral economics has been about the shape of the utility curve, not the rationality of preferences.
I think you'll find a definition of incentive unsatisfying - in general, economists mean, by an incentive, that if given two choices, where the utility from A is more than the utility from B, people will choose A. The reasons for this stem from the rationality of preferences, so there's a bit of a link here.
Posted by: cure at Oct 3, 2007 9:37:29 PM
Fabio, I disagree with chucking out the math. My previous courses were quite mathematical, and even though they are boring (I slept in a lot of the classes), I find them quite useful (especially in understanding some weird arse game theory approach papers).
Math is important. Just make it relevant.
Posted by: Chewxy at Oct 3, 2007 9:47:00 PM
I will say that he even applies this to his macro book with bernanke. That book does a great job of taking supply and demand and marginalism to every chapter. So all the longrun chapters are held together well by that unifying framework, and the shortrun chapters do a good job of integrating keynesian cross with AD/AS. He's very good at taking a few simple ideas and using them repeatedly to enhance your understanding of complex things, which is what I think his principles of macro textbook does.
I've been thinking alot about Alex's blogpost about Ayres' book and his discussion of "Direct Instruction." Maybe I should be aiming for more scripted lectures. I have had the msot success when my lectures are logical and scripted with an emphasis on problems to be worked out. I don't think anyone learns economics except through problem sets - and usually only after some weeping and gnashing of teeth has taken place.
Posted by: jason voorhees at Oct 3, 2007 10:10:06 PM
I think it is not exactly clear that the objective of the course IS to teach them material they will remember forever by simplifying to the few basic concepts. Consider most courses you take in college, and they will have two things in common: 1) You will forget 99% of the material; 2) You will never use what you learn again in the real world. Consider your history classes, philosophy classes, geology, etc. Most of the material is useless beyond exam day, so why do we set up a system like this?
The answer is because it comes as a bundle that confers two major benefits upon the student. For starters, many of us take some time to decide what we want to do with our lives, and getting a sampling of what each field entails through the introductory courses is useful. That I am not interested in geology is almost as useful as my learning I was interested in Economics. Secondly, taking these courses sends useful signals to future interested parties. Perhaps I cannot remember how I calculated the area of a sphere in geometry, but the fact that I was capable of doing it at one time is a useful signal. The same is true with economics. So let's not get carried away with ourselves and throw out teaching the concepts of matters like deadweight loss. After all, they still may remember the point, like that taxes cost more than they take, and that just because a party writes the check for the tax bill doesn't mean they're the ones who are paying it.
Posted by: Justin Ross at Oct 3, 2007 10:16:46 PM
Chewxy - I wish I had you in my classes! It would be nice if most people absorbed math into their intellectual repertoire, but the truth is that most people are fairly math resistant. They seem to memorize the minimum to pass a test and immediately forget it. To really learn math, not only do you need exposure in a classroom setting, but you also need to hang out with people who are constantly doing math, so you don't forget it.
While serious economics should employ math, the *average* college student should be exposed to a math minimal economics course. If you read Frank's comments, he discusses a study indicating that people retain very little from econ courses, and I bet it has to do with the fact that people quickly get hung up on math. I agree math is wonderful (I have a degree in math), but the truth is that math is a very hard language for most folks. So for the masses of college students, dump math, teach ideas.
Posted by: Fabio Rojas at Oct 3, 2007 10:18:51 PM
I'm a 2nd year graduate student, and I can't tell you how many times I've heard, "we are giving you the tools, but it is your job to come up with interesting questions." I think tools are important, but the professors act like coming up with good questions is an exogenous process, something you are just born with. It isn't. Prof. Frank shows convencingly that coming up with puzzles and thinking in economic terms is something that takes practice and can be learned. Once you learn it, it informs your whole life.
I thought the quote, "they’ll have been gone 10 or 15 years — many of them, the first thing they want to do, is tell me about extra questions they’ve been thinking about, and what do I think of the answer to this question?" Isn't thought how we all do research? We come up with an interesting question and then we try to answer it. Often, we can come up with lots of different answers, so we use data and models to see which answer is the best. Maybe several answers are correct, and all working in concert, some affecting behavior more than others. Frank's metho seems a lot more useful for people that want to go onto graduate school. They teach the undergraduate material totally differently in graduate school, anyway. I don't remember the little graph models from intermediate courses anymore.
I became interested in econ, because of being introduced to Paul Krugman in my first college econ course. He really explained what a model was and why it should be simple, and then showed how the power of a little economic lifting could destroy an argument. In contrast, many of my friends who took the intro econ courses (I took in a much more relaxed way in high school) got stuck in big classes learning definitions and thinking economics made too many assumptions to be useful to the "real world." When there are just tons of examples (especially now with the freakonomics trend), where very simple concepts can yield big dividends. I also felt my undergraduate intermediate courses came very easy. So many people were trying to memorize how models shifted, whereas if you have economic intuition, you just say, "ok, what's that line mean, ok, and that one, oh ok, well then if you shock it like that this moves right, this stays the same, oh neat result."
Posted by: Charlie at Oct 3, 2007 10:23:27 PM
Sorry, but the teacher Prof. Frank is naive. It seems that since he's learned enough that he's achieved mastery so that he can "thinking economically", he now has forgotten just how important those quantitative problems were in developing "thinking economically". You know how one author here (Alex, I think?) says "it's taken me 35 years to read this book", because the ability of one's human mind to read has continued to improve with practice? Well, it's the same for thinking "economically", or thinking "physically" or thinking "mathematically".
Yes, the narrative is helpful. But without actually training, no student will REALLY understand how to do it. The ATM example is limited, but let's pick something similar. Frozen juice containers are STILL made out of paper with that annoying strip on the top side. Why? If you weren't used to economic thinking, you might say "must not be any financial benefit in changing the packaging, it's not costing market share so there's no reason to improve it." But if that's the case, how do you explain when packagding does improve? why have milk cartons changed? ARe you SURE there's no market share improvement to make? How would you know without drawing the demand curve? Why have other frozen food containers changed? What about automation? Are you sure that every other process in the company hasn't changed--if the other packaging processes have changed, is it REALLY a financial benefit to maintain one old one?
How would the unskilled economist who only has "economic thinking" to use judge the difference between the above instances?
They wouldn't. Now, does narration have a place? Sure--as a way to make the subject tangible and fun. Most O-chem courses have used one day a week as a "better living through chemistry" course for decades, as have other kinds of courses found "real applications" to showcase important ideas. But those examples are CHERRY PICKED FOR A REASON: to teach A SPECIFIC IDEA that relates to actual course material--relates in the scientific principle at play or in techniques. To act like you can withhold practice in those principles and techniques and actually
teach something, you will fail your students.
A good example of this are the Feynman Lectures on Physics. History has heralded these books for teaching physics. But actually, they were a COMPLETE FAILURE for the Caltech freshmen who were taking the course. They learned nothing. They failed. By the 2nd trimester, the Caltech physics profs were the only attendees, and they were heralding the material for teaching insights and narration. But only they were capable of getting anything out of it. Freshmen were dropping the course, or flunking. Great genius and insight is nothing without enough underlying bedrock of working problems, doing the mechanical practice.
Finally, I think the notion that you are teaching all of your college students the same way in the first place is silly. Why? Certainly there are at least half a dozen different primary motives for students to take intro econ: those who want to major in it, those who need the credit to pass some other college breadth requirement, those who are trying to please their parents, those who needed 4 units at the time your course is offered, those trying to sleep with the TA, those who have no idea what to do with their lives but took the advice of someone else who told them to. Their different motivations will drive them to study differently, and drive them to listen differently. The idea that they all want the same thing out of the class--an A, for example--is also silly. So why teach to one style anyway, unless you've already decided to only teach to those who have one specific motivation. Fine, specify that from the beginning. But most undergrads won't have enough maturity to understand that specification and align themselves with it.
It's always good to teach the big picture. But the big picture without data, evidence, and working ACTUAL QUANTITATIVE PROBLEMS doesn't stick.
Posted by: Allison at Oct 4, 2007 1:46:30 AM
It seems that since he's learned enough that he's achieved mastery so that he can "thinking economically", he now has forgotten just how important those quantitative problems were in developing "thinking economically".
This is completely, utterly wrong. Please read some of Von Mises, Hayek or even Adam Smith's works - then you can comment on whether "thinking economically" depends on quantitative skills.
Posted by: anonymous at Oct 4, 2007 5:25:52 AM
1. Don't teach as if students are going to graduate school.
I do have to admit that Prof. Frank's ideas sound ideal for a survey course (i.e. a course where students are not expected to take further economic courses). However for a mainstream economics course, it may be more appropriate to teach in a fashion that is more in keeping with how the field is taught in general. i.e. teach as if students are going to graduate school.
Posted by: Tom West at Oct 4, 2007 7:33:28 AM
I had Professor Frank while I was a graduate student at Cornell - and I really enjoyed it. However, I was in class of graduate students, so it was hard to tell whether his techniques were improving learning or whether we already knew the stuff (and it will surpise folks to know how poorly we grad students did on intermediate level exams). The most important thing he did for us was make us write, and write and write (clearly) about the basics.
That aside, the most important thing that we as teachers can do is to integrate the actual DOING of economics into the classroom and outside of the classroom. You want students to learn about how exchange can create value, have them read about the red paper clip - and then have them DO it over the course of a semester - I had some students trade up to some pretty significant items, and they were pretty proud of it. You want students to appreciate how markets work, run some simple experiments. You want students to pay attention in your class, well make an effort to learn what they are doing in OTHER classes, and address these issues in your class. And, having students read outside of the textbook (I nominally used Mankiw's book, but my primary readings came from all over the place) will intrigue them. Have them read a few chapters of Landsburg's "Fair Play" and selected chapters from Tim Harford's book, etc; have them read some stuff at Liberty Fund, have them read some small pieces of original texts, etc.
Most of all, I think, is to introduce the ethical foundations of economics early on, and to also address the common claims made against economics early on. In some classes, I'd start with three or four common claims made against the study of economics, get people all heated up about them, and then spend the rest of the semester returning to them (e.g. trade and exchange is zero sum, things are getting worse, we are running out of resources, we should protect jobs, economists are all corporate lackeys, etc.)
Admitting there is no one best way to teach economics is probably a useful thing to do. If anyone ought to recognize that, it is an economics teacher. I happened to learn well from traditional texts and graphs, and that helped me get my PhD - but man do I wish I were more creative in my economic thinking.
Posted by: wintercow20 at Oct 4, 2007 8:50:18 AM
---This is completely, utterly wrong. Please read some of Von Mises, Hayek or even Adam Smith's works - then you can comment on whether "thinking economically" depends on quantitative skills.
So now we compare the thinking of those attending undergraduate education to that of brilliant geniuses? Of course Mises, Hayek, and Smith had mastery, but just reading them isn't enough most undergraduates to actually learn how to think economically. See my above example. Students will apply the supposed "method" incorrectly. Without actual quantitative data they won't know how to assess if they are right or wrong.
Posted by: Allison at Oct 4, 2007 10:03:27 AM
See my above example. Students will apply the supposed "method" incorrectly.
Your example was not entirely clear, but my hunch is that even a quantitative economist cannot answer your question without enlisting a subject matter expert e.g. a packaging engineer providing the appropriate data. If so, this is all the more reason to de-emphasize quantitative skills in favor of general concepts.
Posted by: anonymous at Oct 4, 2007 10:26:36 AM
Somehow I made it to middle age with a graduate degree and 25 years of practice in a medical profession without taking an econ course. I remedied that with undergraduate intro to micro and macro courses at age 50, and felt like all the lights were turned out to illuminate and confirm what I believed intuitively about the world I'd been perceiving. I had the good fortune of 1)decades of life experience to provide a context for this new way of looking at the world, and 2) instructors who described economics as a particular way of looking at the world, rather than merely another body of information to absorb and repeat at exam time. In addition, at least one of the instructors encouraged us to subscribe to and read one of the major publications that reports the news cast in an economist's worldview--I chose to subscribe to the Wall St Journal--and to read other sources, including MR. That was my intro to this fine blog. I'll never be the same.
But when I enrolled in an intermediate econ class I was stopped cold by the required calculus that I hadn't touched in 30 years. The demands of full-time professional practice kept me from self-tutoring calculus so I could keep up with the class. The take-away for would-be excellent econ teachers would be to offer a track for the economically-minded student who does not intend to pursue an economics degree. I manage by reading MR, Megan McArdle, Ezra Klein and others, with my cursor hovering on Wikipedia to look up the terms which are new to me.
Posted by: JohnS at Oct 4, 2007 11:48:58 AM
Lol at any quantitative tools being taught to intro students.
Posted by: Charlie at Oct 4, 2007 1:28:42 PM
My experience was just like Charlie at Oct 3rd so I endorse his comments. I add only that one should a) be nonideological - the coolest moment I had in undergrad social sciences was when my communist political systems teacher spent the Tuesday class brilliantly expounding and justifying the Marxist critique of capitalist society and then came in on Thursday and did exactly the reverse - it lingers in my head 30 years later; and b) give people tools to see how to detect bulls***.
Posted by: MT57 at Oct 4, 2007 1:56:38 PM
Definition
Assumptions
Lemma
Theorem
Proof
Definition
Assumptions
Lemma
Theorem
Proof
Intermission 1: Joke about economists.
Definition
Assumptions
Lemma
Theorem
Proof
Definition
Assumptions
Lemma
Theorem
Proof
Intermission 2: Story about some weird mathematician.
Definition
Assumptions
Lemma
Theorem
Proof
...
That's all you need. It's not fun when it's happening to you but it's great when you finally understand it and it does help you understand it.
Ok, I'm only a little bit serious. That's how I like(d) my classes. No goofy word problems, no cutsy distracting examples, no stretching to tie everything however weakly to some current event or the latest pop culture fad among the youth, no pathetic trying to be "hip" and "with it". Just the meat and potatoes.
I realize other people have more gaudy tastes.
Posted by: notsneaky at Oct 4, 2007 3:01:07 PM
Has anyone in Frank's classes asked the question about why so few college professors are good teachers in any subject ... economic incentives right? ... you don't get tenure by being a good teacher.
Posted by: RTS at Oct 4, 2007 3:54:35 PM
Allison, economics needs math like a fish needs a bicycle. If you understand economics, you don't need math. If you don't understand economics, all the math in the world won't help you.
Example: you don't need a calculator, or calculus, to understand the economics calculation problem.
Posted by: Russell Nelson at Oct 5, 2007 1:46:14 AM
"How can we do a better job teaching what is important, true and beautiful in economics?"
I submit that undergraduate economics, even beyond introductory classes, could benefit greatly from a limited case method, popular in business schools. In particular, cases could work very well for the target Mr. Wolfers and Mr. Frank are discussing: non-majors who are taking micro 101 as a distribution requirement (see Joel Spolsky). It is better to teach them lasting, useful principles than to give them a foretaste of the economics profession.
I don't recommend teaching only with cases - the Harvard Business School method - but mixing cases and lectures. A well-taught case reinforces learning in at least three ways: (1) it usually reviews material previously taught in a lecture, (2) students get the experience of unfolding its lessons themselves, and (3) its storyline aids recall.
Clearly the case method is only as good as its execution: how well written is the case, how well the teacher leads discussion, and how well the students prepare (writing up few hundred words before discussion is the best way). A good case is rich and multifaceted; it continues to unfold throughout the course.
Before you snort, go to a top business school, ask students who is the best case teacher, and sit in on a couple of classes.
Posted by: Timothy at Oct 5, 2007 2:17:33 AM
I should add that (properly-led) case discussion works very well at least up to ~60 students and could probably go well beyond that. Some sort of grading on participation and preparation helps.
Posted by: Timothy at Oct 5, 2007 2:37:23 AM
