Bryan Caplan  

The Iron Laws of Pedagogy

PRINT
Reinhardt on Doctors' Monopoly... Henderson on the Three Nobels...
Everyone who's ever been a student can vouch for what I call the Iron Laws of Pedagogy:

First Iron Law: Students learn only a small fraction of what they're taught.

Second Iron Law: Students remember only a small fraction of what they learn.

Third Iron Law: Most of the lessons students remember lack practical applications.

Research on Transfer of Learning strongly confirms a fourth, less obvious conclusion:

Fourth Iron Law: Even when students remember something with practical applications, they still usually fail to apply what they know... unless you explicitly tell them to do so.

If you're tempted to yawn at these truisms, reread the Fourth Iron Law.  Anyone who jumps through the hoops of formal education witnesses its multiplicative inadequacies.  Yet when former students argue about education policy, most fail to apply their first-hand knowledge - precisely as the Fourth Law predicts!  What do they do instead?  Give in to wishful thinking and Social Desirability Bias - and hail education as the the key to the universe.



COMMENTS (12 to date)
Kevin writes:

I actually think your class was the most valuable one I took in college. Not because I remember anything about public choice, but realized that I had no skills other than showing up somewhere 70% of the time! So as a sobered senior I turned to free online tech education and arduous professional exams to supplement my degree, and it seems to have payed off. So thank you for being honest with students!

Daniel Fountain writes:

Possible reason for the bias:

Logical complements of your laws can be summarized as

! 1+2: Students have high learning retention rates (Sharpness)
! 3: Students seek out or remember the lessons they view to have practical relevance (Forethought)
! 4: Students are creative and can apply their knowledge to different, seemingly unrelated fields. (Critical thinking ability hereinafter Criticality)

Accepting that people use Sharpness, Forethought, and Criticality as qualitative metrics for gauging peoples intelligence (which I think is a safe bet) its important to note that education is irrelevant to these qualities unless !4 is true.

If !4 is true then students can theoretically learn !1+2 and !3; if 4 is true then !1+2 and !3 are inherent qualities (i.e. genetic qualities). Of course talking of genetic intelligence is very closely tied with racism and as such is very unlikely to be accepted by the public at large. This also blows a huge hole in the ideas that you can "learn to learn", parenting being able to change the general intelligence level of a child, and of course teachers being able to increase the grades of students in unrelated coursework.

Obviously there will be a distribution of Criticality among the population; however, all individuals have strong personal incentives to put forth the idea that everyone has high Criticality regardless of if his true belief. It insulates them from being labeled as racist, multiplies their personal ability to impact people in lasting ways, justifies paternalism as a cost-effective behavior changer, etc.

I also don't think its a stretch to say that if !4 were true for the average individual then Overlearning would be much more apparent in the general population in a much wider array of thought processes than it actually is.

Tracy W writes:

First law of human reproduction: most sperm cells are unviable.
Second law of human reproduction: most eggs will be discarded before the mother reaches menarche.
Third law: most times a couple has sex they won't fertilise an egg.
Fourth law: even when they do, there's a very high rate of spontaneous miscarriage.

The odds are definitely against getting a human baby out the other end. But Caplan writes a book called Selfish Reasons to Have More Kids". Why? Because the rare times the whole system works the outcomes are wonderful!

Education is not reproduction of course. But even if someone only rarely transfers learning to new situations those few times can be immensely valuable, as other people can then copy that application of knowledge. Or benefit by buying the product, one doesn't need to know how vaccines work to have your life saved by them.

Ricardo V writes:

I completely agree with Tracy W.. Anyway, I write what I thought before reading her comment. I had thought about sowing and seeds, and also financial investments and the principle of diversification. In my view, formal education is pretty much like financial investments: by applying the principle of diversification, one both increases return and reduces risk.

About the first law:

Even if students learn only a small fraction of what they are thaught, that fraction differs from individual to individual. And not only schools, but more importantly, EACH INDIVIDUAL student does not and cannot know beforehand which fraction of what he studies he will eventually learn. Hence, he has to try and learn many different subjects: some will be learnt, and might become professionally relevant in the much later future, others won't.

Hence, if formal education was reduced to a very small fraction, that could be the right fraction for some individuals, but not for most others. And there is simply no way to tell.

And if education was reduced to a set of "small fractions", which were available for students to freely choose one, they would not be able to tell which one was right for them BEFORE trying! They would better pick many of them - as they end up doing when they go to school.

Hence, if one individual decided to freely choose what to study, still the same "inefficiency" would arise: after studying one little topic, the individual would understand it was not for him, and pick another. At the end, "too many" topics would have been studied, and maybe only the most relevant really learnt. But this is in the assumption that people do learn only a small fraction of what they study (I disagree on that infra).

Furthermore, if they learn only a small fraction, the absolute size of it would be even smaller if the total amount of education was to be reduced. Again, the argument of the seeds: if I want at least a few trees to grow, I need to sow many seeds but I CANNOT know beforehand which specific seeds will actually develop in what specific square-centimeters of land.

Further still, if the first law is true, that would be an argument for increasing the quality of teaching and the quality of studying methods, not just to reduce the total amount of education. If the place where I eat provides low quality food, should I just eat less, or pick a better school, ops, I mean, pick a better restaurant?

Two extra remarks: I doubt very much the very first "iron law". I doubt that students learn only a small fraction of what they have been taught.

What I have experienced as a student and as a teacher is that, quite often, people have learnt more than they think they did, and quite often, they learn things in ways that are not captured by intermediate tests, and exams.

Second, this article is based on the reductionist view that the only objective of education is to give practical skills for people to employ in the job they actually have. This view reduces the whole idea of education to an absurd in which it might be true that the costs of education are much above the benefits. But that is only the case because most sources of benefit are, by assumption, excluded: all those benefits which do not relate directly and immediately to the skills to be used in the job you have.

Brian writes:

None of the 4 "laws" you mention is an iron law at all. They are often true, yes, but not inescapably true, which is what an iron law ia all about.

Using the right pedagogical methods (active learning), students can learn and retain a high percentage of what they're taught. It depends crucially on re-wiring their brains in a coherent way.

Law 3, while not iron, is also frequently true, but somewhat irrelevant. It depends on how one defines practical, but education plays an important role in a number of impractical but fundamental areas. As Ricardo V points out, academic specialization and trade drives productivity gains through intellectual comparative advantage. Students become more comfortable doing a wide range of things, from writing and speaking and learning on their own, that expands their later range of options.

Finally, transfer of knowledge is hard but greatly increased when the right pedagogical techniques are used.

Glen S. McGhee writes:

This raises the question of HOW we have to "hail education as the key to the universe," and WHERE do we go from here? Both are important questions.

Ricardo V writes:

@Brian:

This Caplan article on education is based on the same false assumption that can be found implicit on other education articles of his.

The assumption is that there is only one motive behind the demand for education: to acquire skills in order to work in the job you already have.

Any skills or knowledge that you do not use in the job you already have, are considered a pure waste.


Under that assumption, the aggregate private demand for education should be much smaller than thought: if there is only one motive for it, education must be much less demanded.

As demand is small, so it is easier to make the case that the costs are much higher than the benefits. Hence, one concludes that education is a "big waste of time and money".

The problem with all this is that the assumption is false: there are many and many, lots of motives behind the private demand for education. People educate themselves for many reasons.

The reasons are so numerous, that people keep spending their own money (not taxpayers') and time on education, even when it is not compulsory, even when it is not cheap, even when they already have a job!!


Of course, if people benefit so much from education, to the point of paying from their own pockets, one can ask: why should any education be funded by tax-payers money?

That is a good question, but whatever your favorite answer to the question is, the question or the answer does not change the reality that people do demand huge amounts of education even if it is expensive, even if they are already employed.

As the motivations for education are many, the case that "education is a big waste of time and money" just crumbles: with so many motives for education, demand must be high, at which case *private* benefits only (i.e., even without considering any externalities) should be in fact much higher than costs, and education is efficient.


With all these in mind, these article is not much surprising. Take the "third iron law", for example: even if it was true, why would it matter? Well, it would only matter in the strange world in which people would care to educate themselves exclusively for the sake of applying practical knowledge in their jobs.

In the real world we live in, however, even if the third iron law was true, it wouldn't matter much: people demand education for way more motives than "practical application", and so, if only a fraction of it was practical, people would still value it highly.

MingoV writes:

The Iron Laws arise due to educational failures since childhood. Once children are past the basics, they are expected to acquire knowledge, learn practical applications of that knowledge, and develop judgment about when to apply the knowledge. Unfortunately, most educators: commonly test memorization instead of knowledge, rarely test application of knowledge, and almost never test judgment. Students quickly acquire their own knowledge: they don't have to learn anything to succeed. Students learn that cramming before tests yields outcomes just as good as spending ten times longer mastering the subject. This pattern occurs at all levels of education. I saw it in medical school. My school provided copies of previous exams and guaranteed that 70% of new exam questions (all multiple choice) would be from old exams. Some students read almost no textbook chapters or class handouts. They memorized old test questions and passed their exams. Aren't you glad that thousands of physicians never learned biochemistry, pathology, and pharmacology?

JKB writes:

Well, the one thing the "education" system refuses to teach, at least overtly, is how to study. Instead, they continue to promote the 1st and 2nd grade methods of memorization and repetition. Useful when learning your numbers and the alphabet, but distracting when a student needs to process more complex material.

So, the student pushes forward, using more and more brute mental strength until they reach their limit and then "fail" or slip and fall behind. Some students drop out, some reach the "graduation" only to stop, some continue in mediocrity, a select few either figure it our for themselves or more likely encounter a professor who breaks their "school helplessness" opening up the world to them.

It doesn't have to be this way. But the multiple choice test reinforces regurgitation rather than reflection and thought. And, our "educators" are generally those who did well in the zombie system so see it only as a need for poor students to apply themselves more.

A few years ago, I did a casual google survey hoping to find some school, private, public, university, etc., that had posted the ultimate study guide for their students. Many had decent advice, the old quiet place, free of distractions, etc., but, all left the student staring at the book as if the knowledge would just beam into their eyes like some sci-fi movie. You'd think how to extract the knowledge from the book would be lesson one? We teach fry cooks how to actually fry, should we not teach students how to actually study. With the goal of their study skills development that any book, no matter how ponderous, could be distilled by the student with aplomb?

Daniel Fountain writes:

@Ricardo V

If you've read/heard of Bryan's work on signaling you would realize that

The assumption is that there is only one motive behind the demand for education: to acquire skills in order to work in the job you already have.

is not his position. What Bryan is clearly pointing with these ILOP out is that the first hand experience for the vast majority of the population demonstrates that demand for education is not rooted in skill development. He makes the case elsewhere it is in fact rooted in signaling. I think his position is something like a 20:80 ratio for skills:signals.

Kendall Ponder writes:

What is the time element in #1 and what is "small"? Students demonstrate on ACT/SAT tests every year that they remember significant portions of what they have been taught in school. Ten years later they may have forgotten most of it except for the part that had direct application and/or interest for them.

Ricardo V writes:

@Daniel Fountain:

I believe I know very well Caplan's position, and my interpretation of it actually seems to coincide with yours!


I am not saying Caplan acknowledges the assumption I wrote, and I strongly believe he wouldn't acknowledge it, because there's a dilemma here: without the assumption, the whole argument that "education is a big waste of time and money crumbles"; and by acknowledging it, the absurdity and lack of realism of the whole story becomes much more visible.


Argument: people educate themselves *believing* they will get skills, and believing it is those skills that will allow them to earn higher incomes than people with less education. If they knew that the mechanism between more education and higher wages is not skill development, but signalling, they would consider spending less time and less money on education IF there was some cheaper mechanism to signal their ability. If that mechanism existed, it would then be clear that "education is a big waste of time and money".

@Daniel: do you agree that is a fair description of Caplan's position?

Now, that argument crumbles without the following assumption:

"The assumption is that there is only one motive behind the demand for education: to acquire skills in order to work in the job you already have."


How? Simple: let's drop the assumption and think: what if people in the first place are NOT educating themselves in order to acquire skills, and they are not even educating themselves to signal ability? What if they are educating themselves for a variety of other reasons?


If that is true, and I think it is - that's why I have mentioned people educating themselves even when education is not compulsory, and it is expensive, and they have to pay it from their pockets, and even when they already have a job - well, if people value education for so many reasons OTHER than skills and signalling, their private demand (excluding externalities, etc.) must be big, much bigger than *under the assumption*, at which point it may well cover all the costs. At that point, education is definitely NOT "a big waste of time and money".


Of course, the assumption itself is pretty absurd, unrealistic, etc.: that's why it is not a good strategy to acknowledge it if one wants to push for less government spending on education. Nonetheless, it is an enabling assumption for the whole signalling story.


The assumption is pretty much there all over the place in such articles in this blog as this one and that one:

http://econlog.econlib.org/archives/2011/11/the_magic_of_ed.html

http://econlog.econlib.org/archives/2012/08/the_marginal_pr.html


And these "iron laws" are best understood under the same hidden assumption. I just repeat myself here: what is the relevance of, say, the "third iron law": "Most of the lessons students remember lack practical applications."?

If people did not educate themselves in the first place because they cared for "practical applications", what would be the relevance of such "law"? None.

If the hidden assumption does not hold, and people educated themselves for many reasons, the "third iron law" is pretty much irrelevant whether it is true or false! But it would be relevant if people, as the assumption says, had "only one motive behind the demand for education: to acquire skills".

Comments for this entry have been closed
Return to top