“Massey Resistance”: Wrong In Theory And Practice, Part II

The first part of my discussion of Princeton Prof. Douglas Massey’s most recent defense of affirmative action appears immediately below, or, if you somehow stumbled on this continuation first, here. Part I dealt with the first of what are claimed to be only “three basic charges” we critics level against affirmative action, that it “constitutes reverse discrimination, lowering the chance of admission for better-qualified white students” and, further, that “affirmative action generally has had only small and insignificant effects on the admission prospects of white students.” As I argued at length — and, I hope and believe, in depth — in that first, long post, baloney!

Here is what they say is our second charge:

They also contend that it creates a mismatch between the skills of minority students and the abilities required for success at selective institutions, setting those students up for academic problems.

Their attempted rebuttal here, however, like their attempted rebuttal to the first “charge,” is mere assertion, with little argument and less evidence. Indeed, all they offer here is a nod to Bowen and Bok’s The Shape of the River, as though it settled all questions. This cavalier disregard of both argument and evidence borders on the bizarre, given the prominence of UCLA law professor Richard Sander’s work on this topic (and the work of his critics, and his replies to those critics) since the appearance of Bowen and Bok’s thinly supported book. As Roger Clegg noted in his own criticism of the Massey article (linked in an UPDATE to Part I), “to refute the mismatch hypothesis by citing a 1998 book and ignoring the work by, for instance, Richard Sander since then is breathtaking.”

I assume Massey et al. respond to Sander in their book. They should have done so in this “adaptation” as well, or not mentioned “mismatch” at all.

Moving on, most of the Massey et al. article deals with what they regard as the third “charge” against affirmative action: “that it stigmatizes minority students as less than fully qualified, which results in demoralization and substandard performance, when in fact those students may be well qualified.”

This third “charge” is the only one they seem to take seriously, and the bulk of their article is devoted to an (ultimately unpersuasive) attempt to rebut it. Here’s the launch:

If white students believe that many of their black peers would not be at a college were it not for affirmative action and, more important, if black students perceive whites to believe that, then affirmative action may indeed undermine minority-group members’ academic performance by heightening the social stigma they already experience because of race or ethnicity. In addition, we have uncovered a fourth possibility: the idea that affirmative action exacerbates the psychological burdens that minority students must carry on campuses. Those who feel threatened because they have internalized negative beliefs about their group will find that they feel even more so if they themselves fall below the institutional norm for SAT performance. Likewise, those who feel they are representing their race every time they are called on to perform academically will have a heightened sense of responsibility, or what we call a “subjective performance burden,” when their group’s average SAT score is known to be well below that of other students at the institution.

First, note that by saying that the third (of only three!) “charges” against affirmative action is that it “stigmatizes minority students,” resulting in “demoralization and substandard performance, when in fact those students may be well qualified,” Massey et al. assume that the only serious (in their view) criticism of affirmative action is in essence psychological and hence that the relevant evidence will turn on such matters as “what white students believe” and “what black students perceive whites to believe” and “the psychological burdens” weighing down black students “who feel they are representing their race” every time they open their mouths and who “feel threatened” because they have “internalized” the negative beliefs that they perceive whites to believe, etc., etc.

Predictably, all this feeling and perceiving and believing and internalizing, etc., leads to an extended riff on “stereotype threat,” which Massey et al. posit as the real villain, not affirmative action. “Indeed,” they write,

our research suggests that the extensive use of race-sensitive criteria under institutional affirmative action, when it produces a large test-score gap between minority and other students on campuses, appears to lower minority achievement in two ways: Directly, it creates a stigmatizing social context within which black and Latino students find it more difficult to perform. Indirectly, it heightens the subjective performance burden experienced by individual minority students.

They do not, of course, advocate eliminating the “large test score gap” that is at the core of affirmative action, even though “lower minority achievement” flows, even in their analysis, from that large gap. Far from it:

the results of our research do not mean that affirmative action is necessarily detrimental to the academic interests of minority students and should be abandoned. Rather, the results imply that as currently administered by selective institutions, the application of race-sensitive admissions criteria appears to create a stigmatizing setting and should be reconsidered. Indeed, if the way affirmative action is administered and framed can be changed so as to mitigate the stigma now being created, its negative academic effects might disappear….

In the end, our finding that affirmative-action programs can undermine grade performance by stigmatizing students and increasing the pressure they feel to perform tells us less about the inherent weakness of affirmative action than about the poor fashion in which programs are carried out.

They would thus leave affirmative action itself, with its large test score gaps, etc., alone and instead try to install some stigma filters and a new “frame.” If I were their PR person I’d suggest they push “Frame It, Don’t Blame It!” as a substitute for the now dated Clintonian “Mend It, Don’t End It!” (For criticisms of “framing” in other, but very similar, contexts, see here, here, and here, as well as here and here.)

What might these stigma filters look like? How, that is, do Massey et al. suggest that racial preferences be “administered and framed” so that their stigmatizing effects are eliminated? As one example, they mention that Stanford psychology professor Claude Steele, the father of the “stereotype threat” theory, and several colleagues

initiated a special program for African-American students at the University of Michigan at Ann Arbor, in which students weren’t stigmatized by labeling the program as remedial or compensatory. Rather, they were told that, as Michigan students, they had survived a competitive selection process, and that their assignment to the program was intended to maximize their strong potential….

Let us not pause to question the legality of a special program at a public university presumably limited to African-Americans and instead marvel at what can be accomplished by not calling a remedial program a remedial program and lavishly praising its members. (As I noted here, the former dean of black affairs at the University of Virginia once wrote that he “let mothers know that there’s an office … that’s going to shower love on their children.”) Maybe the new slogan should be “Rename It, Don’t Blame It!”

To Massey et al., however,

[t]hose findings suggest that if minority students were welcomed and supported at selective institutions in the same way that star athletes and legacy students routinely are, the grade performance of black and Latino students might improve markedly. But, if anything, elite colleges and universities now seem to be doing the opposite of wisely intervening in support of minority students. When they arrive on campus, black and Latino students are often singled out for special treatment in ways that typically imply a need for remediation. They have been far more likely than white or Asian students to report the use of a tutor or the receipt of special instruction in reading, writing, mathematics, test taking, and study skills. Over all, black students have been more than twice as likely as white students to receive institutional help.

The worst possible situation is for minority students to be typically selected for remedial programs on the basis of race and ethnicity rather than ability or level of preparation, thus communicating the tacit assumption that all black and Latino students are of suspect intellectual quality, no matter what their class or educational background….

There are a number of glaring flaws in, or at least obvious questions about, this framing-not-blaming approach to affirmative action’s stigmatizing effects. Here are a few:

• Even if it is legitimate to single out one racial group of students to receive lavish large doses of praise, care, and attention, why would such race-based attention not simply reinforce “stereotype threat”? As Roger Clegg noted in the post linked above,

I have no particular problem with putting a happy face on these intervention programs, but I am skeptical that doing so will solve the problem of less academically competitive students not performing as well as more academically competitive students…. I would also note that limiting these intervention programs on the basis of race is not only illegal (the Supreme Court made clear in its 2003 University of Michigan decisions that “individualized consideration” of students rather than the mechanical application of racial categories is required by schools), but is more likely to aggravate “stereotype threat” … than allowing students of all racial and ethnic backgrounds to participate in them.

• If selecting students “for remedial programs on the basis of race and ethnicity rather than ability or level of preparation” is the “worst possible situation” because it communicates “the tacit assumption that all black and Latino students are of suspect intellectual quality,” why does not substantially lowering the admissions requirements “on the basis of race and ethnicity” reinforce exactly the same tacit assumption?

• Since university officials know very well that a substantial proportion of the minority students they admit are less prepared for college work than their peers and more in need of remedial (by whatever name) “instruction in reading, writing, mathematics, test taking, and study skills,” wouldn’t it be cruel of them not to provide tutors, special instruction, etc.?

The larger problem here is the authors’ overarching (or is it undergirding?) assumption that what they acknowledge as the “demoralization and substandard performance” of minority students is caused by the “stigmatizing social context” created by the way affirmative action is framed rather than by the fact, which they not only freely acknowledge but actually document, that minority students are admitted in large numbers with substandard qualifications. They are admitted because of the “difference” that the religion of “diversity” proclaims they provide to those admitted without preferences; they are admitted under different standards from the other students; and they are thus not surprisingly regarded as different by their non-diverse peers and faculty.

Finally, I would be remiss if I did not confess that there always seems to be something about the arguments of Prof. Massey and his colleagues that I simply fail to understand. Other examples of my difficulty with what strike me as inconsistencies in their work can be found here and here. In the current article, I haven’t yet figured out how the following two paragraphs fit together. After noting the substantial SAT score gap that exists between minorities and other students, Massey et al. write:

Based on those findings, our research has revealed that black and Latino students with relatively low SAT scores do no better or worse than their counterparts who scored at or above the average for their institutions. Affirmative action does not appear to set individual students up for failure by creating a mismatch between cognitive skills and academic demands at competitive colleges and universities. Other things being equal, individual affirmative-action beneficiaries earn the same grades as other students.

But at the same time, we have found a significant effect of institutional affirmative action on the grade performance of black and Latino students. A sizable minority-majority test-score gap within any given institution appears to create a social context that makes it more difficult for minority students to perform academically. The greater the discrepancy in SAT scores between minority students and others on a particular campus, the lower the grades earned by black and Latino students as a group on that campus.

So, individual minority students with low SAT scores perform as well as their counterparts with average or higher SAT scores, but schools that have a large “minority-majority test score gap” have a “social context” that “lower[s] the grades earned by black and Latino students as a group on that campus.”

As I’ve said, I must be missing something here, because this makes no sense to me. Nor do I understand how the authors can be sure that the lower grades earned by minority students at schools with a large test score gap are the result of a bad “social context” rather than their lower qualifications.

Say What? (2)

  1. Ivy March 24, 2009 at 7:46 am | | Reply

    That ladies and gentlemen is an intellectual rape.

    I think before writing this, Massey constantly repeated to himself: “No it MUST be ‘social context'”.

    To hell with Occam’s Razor, right?

  2. mj March 24, 2009 at 11:17 am | | Reply

    Today, we emphasize the “stereotype threat” and denigrate competing explanations for performance discrepancies.

    Tomorrow we claim that since stereotype threat explains all performance differences and public discussion of race preferences increases stereotype threat, such public discussion is a form of racism.

Say What?