In September, 2007, after seeing a significant drop in its African-American enrollment, UCLA adopted a new “holistic” admissions system.
As no one will dispute, the purpose of the system was to increase the admission rate of underrepresented minorities, especially African-Americans. Although it did little to increase the admission rates of Latinos, Chicanos, and American Indians, somehow it was spectacular at increasing the African-American admission rate, which increased to 16.5% during the first year of the holistic system, from 11.5% during the last year of the prior system.
At the time, I was a member of UCLA’s faculty oversight committee on admissions. During Spring 2008, I asked for a set of 1,000 random application files so I could perform some statistical analysis on UCLA’s admission process.
The admissions staff and their superiors—senior administrators at UCLA—denied my request. They said that the reason was to protect the privacy of the applicants. I offered to allow them to redact any personal or identifying information from the files. They still refused.
I suspected then, and I am near certain now, that the real reason was to cover up illegal activity. Specifically, Proposition 209, a provision of the California Constitution, disallows public universities to use race in admissions decisions. I believe that UCLA was violating that law and that it continues to do so today.
Shortly after I requested the data—I believe to deflect attention from my request—three members of my committee proposed instead that the committee should name an “independent researcher” to analyze UCLA admissions. I had no problem with that aspect of their proposal, although I did have a major problem with a second aspect of their proposal—that the data would be given only to the independent researcher and would not be available to members of our committee. That aspect of the proposal meant that none of us would be allowed to verify the analysis of the independent researcher, much less examine our own hypotheses about UCLA admissions.
I made further attempts to gain the data, all unsuccessful. Eventually, I resigned in protest from the committee. I wrote a report, which noted why I should have received the data and why I believe UCLA was covering up illegal activity. My resignation and the report were covered fairly extensively in the media. (For instance, see here, here, here, or here.)
My committee formed a subcommittee, which eventually named Robert Mare, a UCLA sociologist, to be the independent researcher.
Almost four years later, on May 17, 2012, Mare and UCLA released the report to the public.
That same day, UCLA released one of the most dishonest documents I’ve ever read, a press release that claimed “Mare’s report found no evidence of bias in UCLA’s admissions process.”
The report and press release went completely unreported by the media. It was not even mentioned by any blogs. This is strange given: (A) the many people and resources that UCLA devotes to public relations, (B) the controversy and criticism that UCLA received around the time that it commissioned Mare’s study, and (C) the report’s supposedly fantastic findings about UCLA admissions.
The problem is that the findings really are not so fantastic for UCLA. The report does find evidence of bias. I believe that UCLA administrators were well aware of this, and I believe that that is the reason they did not try harder to get the media to publicize the report
Approximately five months after Mare’s report was released, the report received its first mention by a journalist or blogger. This was by David Leonhardt, a New York Times reporter. On October 13, 2012, he posted an essay, “Race, ‘Holistic Admissions’ and UCLA,” on the Times’s Economix blog.
Leonhardt learned of Mare’s study, not from UCLA, but from Richard Sander, a UCLA law professor and frequent critic of University of California admissions. Leonhardt learned of the study while interviewing Sander about his recent book, “Mismatch: How Affirmative Action Hurts Students It’s Intended to Help, and Why Universities Won’t Admit It”.
The second mention of Mare’s study came on October 23, 2012, by student journalist Alexia Boarsky, who confirmed to me that she learned of the study not from UCLA but from Leonhardt’s blog post. Her article, “Findings by Law Professor Suggest that UCLA Admissions may be Violating Prop. 209,” noted Sander’s book and a recent report that he posted online.
“The book and report,” she wrote, “provide evidence supporting Sander’s claims that the UCLA admissions process violates Proposition 209, which passed in 1996 and made it illegal for California universities to take race into account during the admissions process.”
She also noted:
“What seems to be happening is that there is discrimination after the [initial] holistic scores are generated,” Sander said. “(Admissions officials) seem to be making discriminatory decisions with lots of black and Hispanic students with poor holistic scores being admitted.”
Data from a report published by UCLA sociology Professor Robert Mare in May also helps support this finding. The university commissioned Mare to do an independent review of the holistic admissions process. …
Mare’s statistical analysis shows that in each of the two years he examined—2007 and 2008—the university admitted more than 100 black students who would not have been admitted based on the holistic admissions process alone.
This is around one third of the total number of admitted black students, which was around 350 students in 2007 and 2008, but is relatively minor compared to the 10,000 overall admitted students each year, Mare said.
Today, the Daily Bruin published a response to Boarky’s article. It was written by Scott Waugh, the executive vice chancellor and provost of UCLA, and Janina Montero, the vice chancellor for student affairs at UCLA. Here are the first three paragraphs of their essay:
At UCLA, we believe in every one of our students because they are admitted on their merits alone. Consistent with state law, race and ethnicity play no role whatsoever in our holistic review admissions process.
Allegations to the contrary made by UCLA law professor Richard Sander and others are simply not true.
When in 2008 Professor Sander and others first made allegations that UCLA might be using race in admissions, UCLA’s faculty commissioned sociology professor Robert Mare to conduct an unbiased, independent study that found no evidence of bias in our holistic review admissions process.
Waugh and Montero, however, either (A) haven’t carefully read the Mare report, (B) are intentionally trying to deceive readers, or (C) both.
The following are some examples where the Mare report shows evidence of racial bias in UCLA admissions:
1) One instance is the .391 number in column F of his Table 10. Specifically, note that it is positive and that the z-score next to it is 4. This means that it is highly significant statistically, and it indicates the following:
Suppose you take a black and a white student who are identical on every other variable in Prof. Mare’s data set. That is, they have identical grades, identical SAT scores, attended high schools of identical quality, and have parents with identical incomes and educational backgrounds. They’re also identical on the “Previously Unrecorded Variables” that Prof. Mare describes in his Appendix Figure 3. This includes the “Limits to Achievement” variable that Prof. Mare created, which means that the two students are identical on such things as: (i) whether their life experience includes homelessness, (ii) whether their life experience includes incarceration, (iii) whether they lived in a dangerous neighborhood, and so on. (See question 16 in Appendix Figure 3 for more details on the questions that Mare used to construct “Limits to Achievement.”)
The .391 number means that the black student has a significantly higher probability of being selected for “supplemental review.” The latter is one of three “second chance” rounds in the UCLA process. That is, if a student did not receive a favorable holistic score in the first round, and if the admissions staff considers him or her worthy of supplemental review, then he or she is asked for additional information. A senior member of the admissions staff then conducts a second holistic review of the student and possibly gives him or her an amended holistic score. This is a case of treating students differently because of their race, a violation of Prop. 209.
2) Even more problematic is the -.706 number in column G of the same table. It indicates the following: Suppose you take two students who have been selected for supplemental review. Suppose one is black and one is white, but otherwise they are identical on all the variables that Prof. Mare included in his analysis. The fact that the number is negative (also, note that its z- score, -5, means that it is highly statistically significant) means that the black student is significantly more likely to receive a lower holistic score than the white student. (Lower scores are better, which means that the black student is more likely to be admitted.) Once again, that’s a violation of Prop. 209.
3) Another smoking gun in the Mare report is the -.865 number in column D of his Table 10. This number indicates that black students receive significant racial preferences in the “Final Review” stage of the admissions process. (The latter occurs when the first two readers give “discrepant” scores in the first round of the holistic process. “Discrepant” means that their scores differ by more than 1.0. When this happens, a senior staff member conducts a third holistic review of the applicant. The applicant’s final holistic score is determined by that senior staff member.)
4) On p. 74 of his report, Mare writes: “Absent the adjusted disparities estimated in this analysis [i.e. absent the apparent racial preferences given to Black applicants estimated in this analysis], 121 fewer Black applicants would have been admitted, which amounts to more than 33 percent of the actual number admitted.”
5) On p. 74, Mare also writes, “Absent the adjusted disparities estimated in this analysis [i.e. absent the apparent discrimination against North Asian applicants], 245 more North Asian applicants would have been admitted, which would be almost a 9 percent increase in the number admitted from that group.
6) Perhaps the most shocking aspect of the Mare report is his Table 8. The holistic system gives every applicant a score from 1 to 5, where 1 is the best score and 5 is the worst score. If UCLA admitted students strictly according to their holistic scores, then the cutoff for admission would be 2.5 or 2.75. Most important, all students with scores of 3.0 or higher would be rejected. Yet, that was not what happened. UCLA frequently violated that rule. And as Table 8 illustrates, the violations were correlated with race. For instance, of the African American students with a 3.0 holistic score, 26% were admitted. However, of the North Asian students with a 3.0 score, only 3% were admitted. Of the African American students with a 3.5 holistic score, 20% were admitted. Of the North Asian students with a 3.5 holistic score, only 1% were admitted.
Stay tuned. I expect this issue to play prominently in the news over the next several days, and perhaps several weeks.