Does “Affirmative Action” Generate “Misunderstanding” Based On “Stereotypes”?

Critics of racial preferences often point out that admitting minority students with lower grades and test scores than their peers creates the perception that minorities students are less qualified. Moreover, we critics say, that “perception” is generally accurate at selective colleges; students admitted with lower grades and test scores, whatever their race, will tend to be less qualified than those with higher grades and test scores.

Today an article in the Michigan Daily, of all places, presents evidence supporting this criticism. Some of this evidence is of the familiar anecdotal variety, as when Michigan freshman Bradley Johnson “gets the impression from some of his classmates that they think he lacks the credentials to be at the University.” Johnson is convinced is that “the reason for this uneasiness is not that he lacks the qualifications, but rather the fact that he is black, making him a target of a stereotype that labels some minority college students as unworthy to attend the University.”

Now comes the funny part. How does the author, Michael Kan, a daily staff reporter for the Michigan Daily, treat this common discomfort? Like this:

Ironically, the underpinnings of the stereotyping Johnson faces appear to result from the University’s race-conscious admissions, the very policies that were meant in part to curb such preconceptions.

Since implementing its race-conscious admission policies, the University has been leading the charge to promote diversity in the country’s education system and affording educational opportunities to disadvantaged minority groups.

Yet for students like Johnson, the admissions policies have had an unspoken side effect — causing some nonminority students to believe minorities are getting a free pass into one of the nation’s top universities, regardless of their qualification.

That “ironically” is priceless. A much more apt term would be “predictably.”

But, as they say, wait! There’s more! Kan’s article referred to more than anecdotal evidence — specifically, to a study conducted by the National Research Center for College and University Admissions (Note: this study was not mentioned, or I couldn’t find it, on the NRCCUA site). That study, Kan writes, found that

78 percent of high school seniors nationwide said using race, ethnicity and religious background as admissions factors affects the way nonminority students feel about their minority classmates. The study also found 82 percent of students opposed race-conscious admissions, in spite of the 70 percent that also said attending a University with a diverse campus was important to them.

Put that “in spite of” aside, along with the earlier “ironically.” Preferentialists are, apparently, simply incapable of seeing that there is no inconsistency between liking “diversity” and opposing the racial discrimination that they think necessary to achieve it. (The One Florida plan is proving that diversity can be generated without racial preferences, but that’s for another post.)

The NRCCUA, however, seems as incapable of understanding its own evidence as is reporter Kan. Michael Fleischer, a spokesman for NRCCUA, is quoted as saying,

“This view of minorities unable to perform well at affirmative action universities is because a lot of (non-minority) students don’t understand it and have misconceptions about the term.”

….

Fleischer said the study did not directly quantify the pervasiveness of the stereotype, but he added that researchers found many students basing their responses on misconceptions of race-conscious admissions.

“Many students surveyed had that stereotype in mind, and now when they see minorities, they might think, ‘Did they take spots away from people who deserved to get in? Did they meet the same criteria as I did to get accepted?'”

At the same time, Fleischer said researchers noted that many of the students surveyed believe that race-conscious admissions are merely a quota system, whereas schools like the University only consider race as one of the many admission factors in a subjective process.

“Obviously, colleges aren’t doing a good job of communicating how they are creating a diverse student body,” he added.

“Don’t understand”? “Misconceptions”? “Stereotype”? It’s true that there’s a lot of misunderstanding here, but most of it is on the part of affirmative action’s defenders. Here is Kan again, trying to recognize some truth in the “stereotype”:

Other students say the stereotype does in fact reflect that many unqualified minority students are unjustly taking the spots of more-qualified nonminority students. Engineering freshman Stacey Young said she thought that the academic performance of her minority classmates has not been on par with many of the other students.

But note well: freshman Stacey Young did not say, at least as quoted here, that any of her minority classmates were “unqualified.” What she said is that they weren’t performing as well as the non-minority students. And indeed, the intelligent criticism of racial preferences is not that they allow for the admission of unqualified students but simply, but clearly, that those who are preferentially admitted, by definition (otherwise they wouldn’t need preferences), are less well qualified on the whole than those admitted without preferences.

One can imagine a theoretical college in a theoretical universe where that might not be true. Assume that Theoretical U. is looking for the most qualified students, judged by the traditional criteria of grades and test scores. Assume further that among its 20,000 applicants for a freshman class of 1000 are 2,000 applicants who have perfect (1600) SAT scores and 4.0 grade averages, and now assume that 200 of those 2,000 perfect scorers are “underrepresented” (that’s how they don’t count the Asians) minorities. If, to promote “diversity,” all 200 of those minority applicants were automatically admitted, it would not follow that they were less qualified than the 800 non-minorities selected from the remaining 1800 applicants.

But that is purely Theoretical. In the real world, at places like Michigan, it’s a far different story. Just how different has been revealed by studies conducted by the Center for Equal Opportunity. As quoted in my post here, Linda Chavez of CEO reported that when her organization studied admissions standards at the University of Michigan,

we discovered that the median SAT scores for black students who were admitted to the school were 230 points lower than for whites. What’s more, their high school grades lagged nearly a half point (on a four-point scale) behind those of whites. From the data we obtained under a Freedom of Information request, we calculated that the odds of being admitted if you were a black student with the same qualifications as a white applicant were 174-to-1.

It would appear that the students surveyed by NRCCUA suffer from much less “misunderstanding” of affirmative action than those doing the surveying, or reporting on the results.

Say What? (10)

  1. ThePrecinctChair October 4, 2004 at 3:47 pm | | Reply

    Why wouldn’t we think minorities are not as qualified, given the affirmative action plan? It’s sort of like presuming that women spend less on drinks at a bar on “Ladies Night.”

  2. LTEC October 4, 2004 at 4:28 pm | | Reply

    1) One more point. Perhaps the 70% of students who value a “diverse” campus have a broader notion of “diversity” than the author of the article.

    2) It is very hard to understand the exact meaning of CEO statements of the form:

    “The odds of being admitted if you were a black student with the same qualifications as a white applicant were 174-to-1.”

    I’ve looked at the CEO site, and their articles refer to a certain statistics text for the meaning of this. It would be good if CEO would at least attempt to give people an idea of what this statistic means. It is VERY far from obvious, even to a mathematician.

  3. Andrew P. Connors October 4, 2004 at 5:17 pm | | Reply

    I took it to mean that given a student population with the same GPA and test scores that the U of M would take 174 blacks for every 1 white.

    And I am a mathematician.

  4. Michelle Dulak Thomson October 4, 2004 at 6:00 pm | | Reply

    LTEC, Andrew, I agree with LTEC on this; it’s far from clear to me either, and though I’m not a mathematician I am a UC/Berkeley mechanical engineering graduate.

    I am wondering first of all if the words “black” and “white” aren’t reversed. That is the way the comparison is usually made, and it would make more sense that way. I would then read it:

    A white applicant with the grades and test scores of the [average?] black [applicant?] would have only a 1-in-174 chance of admission.

    Of course, what’s the “average”? Mean? Median? (Median makes more sense, surely.) And are we comparing all whites and all blacks who applied, or all who were accepted, or all who actually attended? (Actually the quoted graf seems to rule the last out, but I’m still not sure whether Chavez is comparing the grades and SATs of the admittees, or those of the whole applicant pool. It would make a huge difference.)

    But assume that “black” and “white” are in the right places; then, to me anyway, the statement looks like

    If 175 black students apply, all with scores and grades equaling those of the [average?] white applicant, 174 of them will likely be admitted.

    Which is quite different from Andrew’s version, and proof enough that the statement is confusing.

  5. Cobra October 4, 2004 at 6:44 pm | | Reply

    Michelle,

    I agree with you. The statement is confusing. 174-1 odds are extreme LONGSHOT odds. For example, two terrible teams in the NFL, the San Diego Chargers and the Arizona Cardinals were given pre-season odds to win the Super Bowl. They came out 150-1 and 200-1 respectively.

    –Cobra

  6. James October 4, 2004 at 6:57 pm | | Reply

    John-

    I know this post is completely off base but since we’re on the subject of Linda Chavez and the CEO has anyone else noticed her eery resemblance to actress Maggie Gyllenhall.(A younger version of Linda, at least)

    Maggie : http://allstars-online.net/bw-maggie05.html

    http://allstars-online.net/bw-maggie07.html

    Linda: http://www.eaglestalent.com/speaker-bureau,35,presenter,Cultural-Diversity-Media-Broadcast-Print-Politics-Linda-Chavez,speaker.asp

  7. Michelle Dulak Thomson October 4, 2004 at 7:00 pm | | Reply

    Cobra, I think they just misplaced the words “black” and “white” and no one noticed, because the formula is so familiar by now that the eye just glides over it. But if you fail to specify what “qualifications” are, and whether you are referring to applicants or acceptances, and how you are determining what the scores/grades of an average white applicant would be (they don’t even bother to say “average,” but surely they don’t mean any randomly-chosen white applicant at all), you are not making clear what the 174-1 means.

  8. LTEC October 4, 2004 at 8:51 pm | | Reply

    If you go to the next to last paragraph on page 17 of the CEO document at

    http://www.ceousa.org/pdfs/MDMED.pdf

    you can see a technical discussion of how CEO calculates these odds. A footnote on page 18 suggests:

    “For a discussion of logistic regression and a more complete discussion of odds ratios, see Alan Agresti, Introduction to Categorical Data Analysis.”

    I think it would be a good idea if CEO would present a few over-simplified examples to give the lay reader some sort of intuitive idea what these odds mean.

  9. John S Bolton October 5, 2004 at 8:51 pm | | Reply

    It shows the mendacity of the federal courts today; that they say they are in the business of breaking up racial stereotypes. They have the brighter college students in an environment that makes blacks appear as naturally unintelligent, lazy and lacking in intellectual interests, yet obsessed with the relative prestige of their people on these points of comparison. If one looks at these effects, and realizes that the courts must want to make people believe that these typologies are true, one might ask: what is the purpose of that? Would it be that officials believe that they must mount a propaganda campaign to persuade people that certain groups are utterly dependent on such public interventions? This has the effect of underhandedly making the case for aggression by public officials.

  10. Chetly Zarko October 10, 2004 at 11:05 pm | | Reply

    I believe the CEO statistic refers to the “grid system” U-M had in place before the point system, and is referring to a student only in one hypothetical GPA/SAT combination (a cell on the grid) situation (in this case, Jennifer Gratz’s position in 1996). CEO also did a statistic for Barbara Grutter, although the Law School’s system was “holistic,” so it simply analyzed the entire pool of candidates and compared their respective odds, and it came to a 230-something-to-one statistic (the weight of the “finger” on the scale was greater than the weight of measured preferences)…

    I agree CEO didn’t clarify this well.

Say What?