winter book forum 2018, part 2: what do people actually get out of college?

This Winter, we are discussing Bryan Caplan’s The Case Against Education. The main issue: We invest a ton in education and it seems to do good. But is that because schooling acts as a filter or because schooling gives your concrete skills or better ways of thinking? If education is mostly a filter (the signalling model), we should probably cut back on education a lot.

In this post, I’ll discuss the types of evidence that Caplan reviews. His book is empirical in that the strength of the argument relies on what other researchers have found. A short blog post does not do justice to this work. For example, he asks – how much do people learn in college? How much do people use specific skills (like algebra) in the workplace? Is there any evidence that learning is transferable – that people acquire “critical thinking?” Each of these topics commands one’s full attention, but we can only skim through the best here.

As you can expect from the title of the book, the direct benefits of education are pretty sparse. Probably the most damning evidence are studies that show that people don’t learn that much in college to start with. Another important fact is that few people ever use the skills – the few they may remember  – in work. Thus, it is very hard to argue for the simple human capital argument – educations makes you better because you learn valuable things. This can’t be right because people don’t learn or retain much in college.

Two related points: In response to those who argue that education imparts critical thinking, he points to evidence that learning is actually domain specific. Learning one area doesn’t seem to help in most others. This is called “transfer learning” in psychology and it’s been rejected for a long, long time. Another fascinating point – if education improves you via human capital development, we’d expect your income to increase for every year of education you get. Instead, Caplan reports that studies of income show no increase in income until you hit 4 years of college – a classic sign of signalling – which economists call sheepskin effects.

Of course, no single study seals the deal and it may be that Caplan has misread some, or even a lot, of the studies. But is is unlikely he misread it all and it is consistent with the everyday view that formal education is not a particularly good way to impart skills. Thus, we should be very skeptical of claims that education is a great way to train people for the labor market. Next week: So what?

50+ chapters of grad skool advice goodness: Grad Skool Rulz ($4.44 – cheap!!!!)/Theory for the Working Sociologist (discount code: ROJAS – 30% off!!)/From Black Power/Party in the Street / Read Contexts Magazine– It’s Awesome!


Written by fabiorojas

February 19, 2018 at 5:01 am

5 Responses

Subscribe to comments with RSS.

  1. This perspective suggests that income is actually an accurate measurement of skills/knowledge. There are lots of reasons to be suspicious of such a claim, not least of which are gender and racial disparities in income.



    February 20, 2018 at 3:19 am

  2. Income may not be an accurate measure of skills or knowledge but educational researchers, psychometricians and industrial psychologists have put a lot of effort into measuring skills and knowledge. Colllege students learn remarkably little. That’s the point of Acaremically Adrift, which Dr Rojas referred to above. Regarding skills students do learn some, but they’re relatively narrow, deteriorate over time unless they get into a job where those skills are used and are specialised. Transfer learning doesn’t exist. Philosophy students get better at arguing, engineering students get better at math but they do t get better at what they don’t practice.

    Even vocational schools aren’t that good at what they’re supposed to be doing. Law school is done teaching what is professionally useful by the end of the first year of three, and that’s generous. A psychiatrist of my acquaintance once said he could teach the average person capable of doing a medicine degree enough to be better than the average psychiatrist in two weeks. Business school is certainly no preparation for founding, running or managing a business.


    Barry Cotter

    February 20, 2018 at 12:23 pm

  3. Academically Adrift ONLY measures transfer knowledge.

    Liked by 1 person


    February 20, 2018 at 1:36 pm

  4. Academically Adrift may be a good book (I would have to read it to be sure), but it has one very important limitation: it only deals with higher education in a U.S. context. In order to make a claim such as “college students learn remarkably little” it would have to use data from other national contexts.

    Liked by 1 person


    February 21, 2018 at 12:49 am

  5. Ironic that this is coming out in the wake of a well-publicized study in NBER that uses a quasi-experimental design to show that people really do learn something in college.

    Academically Adrift also had more nuanced findings than “college students learn little in college.” There was a lot of variation across majors: business and marketing majors showed no improvement in their scores on that particular test, but others did. Business students also constitute about 20-25% of undergrad degrees.

    Trump, arguably the most well-known undergrad BA in business (although he likes to pretend it’s an MBA), exemplifies the lack of critical thinking skills learned by the modal business and marketing student.



    February 25, 2018 at 9:13 pm

Comments are closed.

%d bloggers like this: