Intelligence vs. Priorities

Steven Pinker has a piece at the New Republic arguing that Ivy League schools ought to weight standardized test scores more heavily in admissions. this has prompted a bunch of tongue-clucking about the failures of the Ivy League from the usual suspects, and a rather heated concurrence from Scott Aaronson. That last finally got me to read the piece, because I had figured I would be happier not reading it, but I wanted to see what got Scott so worked up.

Sadly, my first instinct was correct. It starts off well enough, taking down an earlier anti-Ivy League piece by William Deresiewicz for being full of Proof by Blatant Assertion and empty verbiage. Then it takes off into the land of Proof by Blatant Assertion and meaningless anecdotes. With some bonus impossible statistics, which Ephblog analyzes in some detail (also: Ephblog is apparently running again; I had dropped them from my feeds a couple of years ago). Something about this whole issue just degrades every argument.

There are a bunch of things to not like about Pinker's article, but to my mind the biggest is that there are a whole lot of unexamined assumptions underlying the assertion that there is a problem here that demands fixing. This is based largely on anecdotes about Kids These Days, and the interpretation of these strikes me as kind of dubious. There's also a weird confusion between an argument about intelligence and priorities.

The chief evidence that the Pinker offers for there being a problem is that students don't go to his classes:

Knowing how our students are selected, I should not have been surprised when I discovered how they treat their educational windfall once they get here. A few weeks into every semester, I face a lecture hall that is half-empty, despite the fact that I am repeatedly voted a Harvard Yearbook Favorite Professor, that the lectures are not video-recorded, and that they are the only source of certain material that will be on the exam. I don’t take it personally; it’s common knowledge that Harvard students stay away from lectures in droves, burning a fifty-dollar bill from their parents’ wallets every time they do. Obviously they’re not slackers; the reason is that they are crazy-busy. Since they’re not punching a clock at Safeway or picking up kids at day-care, what could they be doing that is more important than learning in class? The answer is that they are consumed by the same kinds of extracurricular activities that got them here in the first place.

Some of these activities, like writing for the campus newspaper, are clearly educational, but most would be classified in any other setting as recreation: sports, dance, improv comedy, and music, music, music (many students perform in more than one ensemble). The commitments can be draconian: a member of the crew might pull an oar four hours a day, seven days a week, and musical ensembles can be just as demanding. Many students have told me that the camaraderie, teamwork, and sense of accomplishment made these activities their most important experiences at Harvard. But it’s not clear why they could not have had the same experiences at Tailgate State, or, for that matter, the local YMCA, opening up places for less “well-rounded” students who could take better advantage of the libraries, labs, and lectures.

This is, at bottom, nothing but a complaint that a bunch of 18-to-22-year-olds don't share the priorities of a prominent professional academic. Which is something that I, for one, am utterly shocked to learn for the first time ever when reading this today.

But the really problematic part of this is the jump where he starts attributing this mismatch in priorities to the personal deficiencies of the students admitted to Harvard, and their families. For one thing, as the Ephblog post shows, it's not all that mathematically plausible to claim that huge numbers of substandard students are getting into elite schools on the basis of extracurriculars alone. To be fair, this may just be a matter of poor word choice on Pinker's part, or his editor's-- while the claim he appears to be making in the article is pretty clearly nonsense, it might be plausible to claim that less than 10% of Harvard students were admitted on the basis of academics and nothing else. That's a conveniently unanswerable question, without access to the inner workings of the Harvard admissions office.

But there are a bunch of assumptions here that don't sit all that well with me. One is, ironically, a problem that Pinker attributes to the "admissocrats": "perpetuating the destructive stereotype that smart people are one-dimensional dweebs." Three paragraphs after rhetorically hanging that on people who favor "holistic" admissions, he does exactly the same thing, by assuming that admitting more one-dimensional dweebs would provide a student body who "could take better advantage of the libraries, labs, and lectures."

There's also the assumption that students are only doing these extracurricular activities to get into college; that if students didn't need to play sports, or music, or do community service, or be active in student government in order to get into college, they wouldn't do those things, and would instead focus on academics. That's kind of difficult to square with the class attendance anecdote, though-- if students were actually engaging in these activities cynically as an admissions strategy, they would drop them immediately on getting to college. And, indeed, a large fraction of them do. But the ones who continue on in their extracurriculars are presumably doing so because they actually enjoy doing these things, and evidently more than they enjoy Pinker's award-winning lectures.

But the most problematic of the assumptions is that admitting a different category of students selected more on the basis of test scores would produce a better match between the priorities of young adults and middle-aged professors. I'll give Pinker credit for being relatively forthright about the sort of shifts you would get from a more-test-centered admissions process, which is to produce a student body that is even more drawn from the socioeconomic upper classes. His contention that this indicates a correlation with inherited intelligence, not a causal relationship between money and high scores is, let's say "not uncontroversial," but I think a more fundamental problem with this is the assumption that this will produce students more to the liking of university faculty.

I mean, let's imagine that Pinker's contention is correct, and the relationship between economic status and test scores is mostly caused by innate abilities-- that is, smart people tend to get rich, and smart people have smart kids, who score well on tests. (Yes, this is lightly brushing aside hundreds of years of institutional racism, etc.-- it's a hypothetical to show that even in the best imaginable case where the problematic assertion of meritocracy is true, his argument is dubious.) Even if that's true, it's not clear to me that this necessarily gets you a set of students who will be more interested in going to Pinker's class, or even the sort of liberal education he lauds earlier in the article.

In the end, most of what he talks up is stuff that matters very deeply to people who go on to become university professors. And while I can't speak to the contents of Steven Pinker's bank account, in general, one does not become extremely wealthy by becoming a university professor. Most of the career tracks that lead into the upper socioeconomic strata benefit far more from the "soft skills" and relationships that students build through extracurrciular activities than formal classroom activities. If students inherit personality traits as well as raw intelligence from their parents, it's entirely possible that admitting a higher-scoring cohort of children of intelligent parents who got rich because of their intelligence will produce a class whose priorities align even less well with those of the faculty. Children of corporate lawyers and venture capitalists may very well choose to pursue careers in those same fields for themselves, and those paths may not require regular attendance at classes on linguistic theory.

This isn't a sure thing, of course-- you could also imagine that, freed of mundane concerns about needing to make a living, the children of the rich would be better able to pursue the life of the mind, and thus end up more like their faculty. But that's the point: the supposed problem is a matter of student priorities, and the connection between those priorities and student intelligence is an assumption that's being made without being stated outright.

So, while in general I find Pinker less of a pompous ass than Deresiewicz, ultimately, I don't find him any more convincing. His piece is the same sort of hopeless muddle of unstated and unquestioned assumptions about Kids These Days, just with a thin veneer of scientific objectivity.

-----

Since there appears to be an absolute obligation to include anecdotes in writing about this stuff, I will tack on a note about my own experience. Like most of the people writing about this stuff, I am not without a personal-historical stake in the matter.

On the one side, I would almost certainly have gotten into college under the more test-based and less "holistic" scheme Pinker and others are pushing. I had excellent (though not perfect) SAT scores back in the day, excellent high school grades, got high scores on all the AP tests I was able to take, etc. And I'm doing my bit to uphold the correlation between high test scores and success in life-- I graduated college, got a Ph.D., hold a tenured faculty position at an elite school, and have published multiple books. I don't think anyone at Williams would have cause to regret having admitted me back in the day, or having given me an extremely generous scholarship.

In between, though, I was the sort of student Professor Pinker grumbles about. I spent a good chunk of my college career more interested in partying and playing rugby than studying and going to class. Had you asked my professors about me during my sophomore year, say, I'm sure a number of them would've shaken their heads and questioned the standards being employed by the admissions office. I wasn't in any particular danger of being kicked out, or anything, but I'm not sure many of my professors in those years would say I was making a positive contribution to the intellectual life of the campus that would justify the institution's investment in me.

That wasn't because I lacked ability or intelligence, but because I was kind of an asshole when I was nineteen, and my priorities then were not my priorities now. This is not an uncommon state of affairs, and like a lot of nineteen-year-old assholes, I got better.

But I suspect that in that interim period, I had professors who assumed, incorrectly, that I must've gotten in on some basis other than innate ability. Because, of course, high test scores are not an obvious and visible trait.

(Dumb story: once, my sophomore year, I was playing a drinking game with a couple of other guys who were noted as hard-drinking idiots, and a couple of freshmen. One of whom was getting trounced in the game, and busted out "Oh, yeah, well who has higher SAT scores?" Turns out he had the lowest scores of anybody in the game, which didn't go well for him after that...)

So, I have personal reasons to doubt the link between innate ability as measured by standardized test scores and student priorities that align well with those of the faculty. And I've known a fair number of bright students in my years as a faculty member who were the same kind of asshole I was at nineteen, and got better by the time they graduated. Even very good students by whatever objective measure you favor aren't always going to value the things faculty think they ought to, because students aren't faculty.

I also bristle a bit at the assumption that people do extracurricular activities solely in a cynical sort of way, to game the admissions process. In addition to my very good grades, I played three sports and was in the band. Not because I felt I needed to do those things to get into a good college, but because I enjoyed doing them (and probably would've been bored silly had I not been doing them, but that's another issue). I still block out time in my schedule to play basketball as often as I can manage it, because I really love the game.

So, you know, there' my personal story about why I don't give a lot of credence to Pinker's arguments. Which has exactly the same evidentiary value (a small number) as any of the other anecdotes trotted out in these various discussions, but fulfills my contractual obligation as a person writing about college admissions.