Jerome Karabel, who has written widely and well about admission to elite colleges (I’ve discussed his work here, here, and here), has an interesting review of the new book by Daniel Golden on “how the ruling class buys its way into elite colleges” (from the subtitle). Golden’s book, you will recall, is the basis of the arguments I discussed in this recent post below.
The entire review is worth reading, but (for personal as well as blog-related reasons) I was particularly interested in the following observation:
After spending most of the book roundly criticizing the admissions practices of many of the nation’s most prestigious colleges, Golden turns to what he considers a model institution: The California Institute of Technology. Unlike other leading colleges, Caltech does not allow the prerogatives of privilege — whether wealth, fame or legacy status — to affect who gets in. In stark contrast to other top institutions, Caltech believes that it is possible to raise the funds necessary to maintain a great university without using admission as a bribe, and its own distinguished history supports that belief.
But the Caltech admissions policy, though exemplary in its integrity, is not without problems. In no small part because of its narrowly conventional definition of merit (primarily scores on standardized tests, grades and rank in class), it has been notoriously unsuccessful in enrolling African Americans; in 2004, just one out of 207 Caltech freshmen was black (for purposes of comparison, the black proportions of the undergraduate student body at MIT, Stanford and Harvard — all of which use a more flexible definition of merit — were 6, 10 and 8 percent, respectively).
What I would really like to know (and as both an undergraduate and graduate alum of Stanford I would especially like to know about it), but what I am sure MIT, Stanford, and Harvard will never tell, is whether they apply their “flexible definition of merit” only to minority applicants. I assume all three august institutions would say, in a huff, “No, absolutely not!” But I also assume none of the three would be willing to say how many non-minorities were admitted under this “flexible” standard, or to make public the average test scores of students broken down by race, gender, and ethnicity.
One way to apply the “relaxed” standards would be to allow the same or equivalent legacy and sports “relaxations” to selected candidates from low income and disadvantaged populations without regard to race.
Because racial minorities are overrepresented in those populations, that might help accomplish diversity without overt racial discrimination.
But the Caltech admissions policy, though exemplary in its integrity, is not without problems. In no small part because of its narrowly conventional definition of merit (primarily scores on standardized tests, grades and rank in class), it has been notoriously unsuccessful in enrolling African Americans;
A few reactions:
1) Apparently being “exemplary in integrity” is insufficient.
2) It’s still a “problem” if the outcome of the policy doesn’t match the preferred racial mix.
3) And when I say “preferred racial mix,” I mean “black people.” Note that he doesn’t identify what percentage of Caltech’s class is minority, perhaps because we all know what it would show: that minorities are not “underrepresented” at all. Except, of course, by redefining Asians as notminorities.
4) To say that the school has been “unsuccessful in enrolling African Americans” implies that the school’s goal is to “enroll African Americans.” (As opposed to enrolling the most qualified students without regard to race.)
5) The phrase “narrowly conventional definition of merit” is so Orwellian, isn’t it? Normally, redefining words “unconventionally” is considered a bad thing, isn’t it? (Jim Crow was only racist by “narrowly conventional definitions of racism,” right?)
Actually I think Karabel _is_ implying that Harvard, MIT, etc use a more “flexible definition of merit” for everybody and not just for blacks. One of the main findings of Karabel’s book is that college admissions discrimination is rarely so blunt as the pre-Gratz University of Michigan system. Rather the typical solution to an undesirable racial balance — whether that be too many Jews before 1960 or too few blacks since 1960 — is to go fuzzy and emphasize holistic review, character, etc. Note that this week UCLA switched to a very holistic means of admission and they as much as admitted that they were doing so as an end run around 209.
Actually I think Karabel _is_ implying that Harvard, MIT, etc use a more “flexible definition of merit” for everybody and not just for blacks.
That may well be what he is implying, and it may even be true that they do. But if it is true, there should not be a significant difference between the average grades and test scores (what Karabel calls the “narrowly conventional definition of merit”) of minority and non-minority admits. I don’t believe that is the case, and I also believe that these schools refuse to make the relevant data public.
Now that UCLA has moved to “holistic” criteria, it will be interesting to see if the racial and ethnic gaps in the “narrowly conventional” measures of merit that largely disappeared after Proposition 209 will reappear.
Cal Tech has received the message, at least for graduate student admissions:
http://www.gradpreview.caltech.edu/