Several years ago, William Bowen, a former president of Princeton, and two other researchers found that top colleges gave no admissions advantages to low-income students, despite claims to the contrary. Children of alumni received an advantage. Minorities (except Asians) and athletes received an even bigger advantage. But all else equal, a low-income applicant was no more likely to get in than a high-income applicant with the same SAT score. It’s pretty hard to call that meritocracy.In response, Greg Mankiw writes:
Maybe David is right, but to convince me, here is what I would like to see. Regress some measure of college success (such as GPA) on SAT scores and the student's family income. If David is right, then the coefficient on family income should be negative. That is, a lower-income student should do better in college, holding reported SAT score constant, because he managed to get that SAT score without all those extra benefits. This is a regression that some enterprising college admissions committee could easily do.Mankiw goes on to post an email from an economist at The University of Western Ontario who ran just this kind of regression and found—unsurprisingly—that the coefficient on family income is positive, not negative. Thus, it seems top colleges should not expect students from lower-income backgrounds to receive higher grades, on average, than their middle-income peers with identical SAT scores.
These findings call into question at least one version of Leonhardt’s hypothesis. But I think the implications here are a bit more complicated, depending on how you view the concept of meritocracy.
If some “enterprising college admissions committee” did run Mankiw’s regression and found that the coefficient on family income is negative, what would this really mean? Putting aside the question of significance, it would tell us that a unit increase in annual family income (let’s say $1000) is generally associated with a certain percentage point decrease in GPA, holding SAT scores constant. In other words, among students with identical admissions test scores, the kids with higher family incomes tend to have lower college GPAs.
It seems obvious to me that this is not the case, as family income likely provides advantages beyond those captured by higher SAT scores.
There are lots of reasons why low-income students with good SAT scores would have lower college GPAs, on average, than their middle- or high-income peers. For one thing, higher family income is firmly linked to higher levels of parental education, which is a strong indicator of focus on education and education-based achievements. This matters in terms of performance, and it matters in terms of our expectations for low-income students who apply to top-tier colleges.
Family income is also likely to affect students after the admissions process. Kids whose parents make less money face all sorts of obstacles in college, like having to work longer hours to support their studies.
If you believe that a true meritocratic system should account for these disadvantages, it makes sense to argue that college admissions boards should give stronger consideration to an applicant’s household income. But this isn’t the same thing as saying students from lower-income backgrounds will perform better than other students with equivalent admissions test scores.
To be fair, I don’t think Mankiw is actually arguing that college admissions boards should completely discount income disadvantage. He’s simply pointing out a fundamental conceit in Leonhardt’s argument. Mankiw is correct that if an admission committee’s only metric is performance, it makes little sense to consider family income, except to the extent that this aids the committee in its selection of higher-achieving students. (One possible problem with Mankiw’s equation is that it assumes the relationship between GPA and SAT scores is identical at all income levels. Even for a performance-focused admissions committee, that assumption is pretty problematic.)
The bigger challenge in all of this is figuring out what we actually mean when we talk about meritocracy. Certainly, income provides unearned advantages, but so does innate ability. In a country that values both exceptional performance and equality of opportunity, college admissions boards need to strike some sort of balance.
While I agree with Leonhardt that income is one factor college admissions boards should take into account, I can’t seem to justify why it’s necessarily more important than other kinds of advantages.
-Jeremy
No comments:
Post a Comment