The proper goal of education is utility, not learning

[T]he correlation between student evaluations and student learning is at best close to zero and at worst negative. Student evaluations measure how well liked the teacher is. Students like to be entertained. Thus, to the extent that they rely on student evaluations, universities are incentivizing teachers to teach in ways that the students like rather than in ways that promote learning.

“Democratic education, says Aristotle, ought to mean, not the education which democrats like, but the education that will preserve democracy….” C.S. Lewis, from “Notes on the Way,” published in the English literary magazine Time and Tide, vol. XXIV, at 717 (September 4, 1943).

It’s remarkable that student evaluations haven’t already been lawsuited into oblivion given that student evaluations are both useless and biased.

The final sentence is a good example of what’s wrong with our legal system. To an increasing extent, the courts are outlawing voluntary behavior between consenting adults.

Was Tabarrok perhaps exaggerating for effect? Last I checked, the US legal system no longer criminalized gay sex, is rapidly decriminalizing marijuana, is removing most legal sanctions for gender-nonconformity, and thanks to the internet, barely regulates prostitution. In 2017, Kentucky became the 27th state to adopt (and retain) a right-to-work law. Can anyone think of an era when voluntary behavior between consenting adults was LESS regulated than today?

We’d like to believe that education is about learning, but in fact it’s ultimately about utility. Learning is just a means to an end, and not the most important role of our educational system.

Bryan Caplan has persuasively shown that much of what we teach our students has little practical value. Students are forced to learn an enormous amount of irrelevant trivia, and quickly forget what they cram into their minds before exams.

College can boost utility in two ways. First, students can learn things that make them more productive. That increased productivity can then lead to higher incomes, which boosts utility. Second, colleges can directly provide utility through entertaining classes, sports, parties, etc.

Adam Smith opposed putting teachers on a payroll, arguing that doing so would create a principle/agent problem: The teacher would not have proper incentive to educate the student.

Yet Smith recognized that education provided a social benefit (specifically, against the evil of religiously derived superstition). So, what remedy? The famously laissez-faire Smith advocated … licensure! That is, recommended that many public roles/offices require people to pass a relevant test. Students would then have an incentive to do whatever it took to pass the test—including hiring people who had a reputation for helping people in that endeavor.

Sounds like the bar exam.

[T]here’s a danger that our educational system will begin to resemble some of the more dysfunctional regimes in East Asia, where students lose the best years of their lives endlessly cramming for exams….

Nobel Laureate Richard Feynman realized that his students could perform well on physics tests—cranking through problems and calculating answers—but didn’t internalize that the answers they gave were supposed to correlate with the world of their own experience. They could tell you how much a given fluid would deflect light, but didn’t associate this with the fact that a pencil would appear to bend when it stuck out of a glass of water.

Harvard prof Eric Mazur reached the same conclusion. But he also learned that students would better internalize material through a process of peer instruction—basically, small-group discussion. The downside? This process takes longer than simple lecturing, so the class covers less material. And in the final analysis, teachers are evaluated not based on much education occurs, but how much teaching occurs—that is, how much material the professor could claim to have “covered,” regardless of whether any students actually benefits from the coverage.

In short, the system rewards the credential of education—going through the motions—rather than getting more people to understand more stuff. And because high-performing students (just as the professors themselves) perform well on tests regardless of the uselessness of lectures, the current system of going-through-the-motions perpetuates itself.

Scott,

My impression is that Tabarrok meant that he’s surprised the use of student evaluations of professors for the purpose of determining career advancement hasn’t been stopped by lawsuits. Employers who use irrelevant metrics to make hiring, firing, and promotion decisions leave themselves more vulnerable to claims of illegal discrimination, for example.

It’s tempting to assume colleges and universities have strong HR oversight to prevent obvious vulnerabilities, but they get sued too.

Scott Sumner wrote:

College can boost utility in two ways.  First, students can learn things that make them more productive.  That increased productivity can then lead to higher incomes, which boosts utility.  Second, colleges can directly provide utility through entertaining classes, sports, parties, etc.  In my view, this second type of utility augmentation is much more effective.  I wish it were not so, but that’s my impression after 35 years of teaching.

(Bold added by me.)

Scott, why do you wish it were not so? The only way I can make sense of that phrase is if we both agree that the proper goal of education is learning, not utility–thus contradicting the title of your post.