The University of Virginia’s “Finals Weekend” — what other schools call graduation — is upon us. Not far behind, no doubt, will be the annual accolades such as the Journal of Blacks in Higher Education’s report that “The University of Virginia consistently posts the highest Black student graduation rate of any state-operated university in the country.” And this has been true “over the past 20 years.”
There is a dark side to this accomplishment. Last June, for example, the African American graduation rate was 80.3% — as usual, the highest in the country among public institutions — but no one seemed to notice that the corresponding 19.7% rate at which blacks failed to graduate in four years was twice as high as the rate for Asians (9.5%) and whites (10.9%). Thus, according to UVa’s Institutional Assessment data, of the 218 blacks who entered in 2011, 43 of them did not graduate in 2015. The six-year graduation rates for the 250 blacks entering in 2009 were better (88.8%), but the rate at which they failed to graduate (11.2%) remained more than twice as high as the rate for whites (5.4%).
These most recent numbers are not unique. In fact, they are a bit better than numbers from the previous decade. UVa’s admissions data show that 3,048 blacks enrolled at UVa from the fall of 2000 through 2010, and its graduation rate data reveal that 789 of them, 25.8%, did not receive a degree after four years. 411 of the 2,844 who entered through the fall of 2009, 14.5%, did not receive degrees after six years.
The fact that blacks fail to graduate at twice the rate of whites and Asians at UVa even though it boasts the highest black graduation rate of any public university strongly suggests the presence and effect of “mismatch” that Richard Sander and others have documented — that any group admitted to selective institutions with much lower academic qualifications than their peers will cluster at the bottom of their classes and sustain higher failure rates.
The fact that UVa each year admits a much higher proportion of black applicants than Asians and whites indicates that it is indeed lowering the admissions bar for blacks. In the twelve classes entering from the fall of 2000 through 2011 (to mirror the four-year graduation rates discussed above), an average 33.9% of white applicants were offered admission, but for blacks the offer rate was 54.3%. In the most recent data available, which I discussed here, UVa offered early admission to 29.7% of the white applicants to the class of 2020 and 42.5% of the blacks.
It is theoretically conceivable — but highly unlikely — that there is a non-discriminatory explanation for the racial preference revealed by these admission proportions. But UVa, like most (perhaps all?) similar institutions, makes it impossible to analyze the nature and degree of racial preference it offers by refusing to publish SAT scores and other academic qualification data by race, even though it clearly has the data. Similarly, it publishes student GPA by gender but not by race. And, as I observed on Minding The Campus a few months ago, “it will be a freezing day in July in Charlottesville before UVa voluntarily releases test scores, etc., by race.”
Well, it is not yet freezing in July, but I have just received a small taste of SAT scores by race from UVa. I shared my Minding The Campus piece linked above with Steve Landes, my Delegate in the Virginia General Assembly, who is interested in this issue and who forwarded my piece with questions to the University’s government relations office. Since Del. Landes is Chairman of the House of Delegates Education Committee and Vice Chairman of the Appropriations Committee, his queries tend to be answered promptly.
Laura Fornash, Executive Assistant to the President for State Governmental relations (and former Virginia Secretary of Education), sent the math plus verbal results, by race, for the students admitted to the class entering next fall, along with the observation that not all students take the SAT and that UVa’s racial range is not surprising given the results nationally. And indeed the results are not surprising:
There is not, of course, a perfect correlation between SAT scores and graduation rates, but those scores are a significant component and indicator of academic qualifications, and there is compelling evidence that raising academic requirements raises the probability of graduation.
Peter Salins, former Provost of the State University of New York system, provided strong evidence for that correlation in an Op-ed in The New York Times. In the 1990s, he wrote, “several SUNY campuses chose to raise their admissions standards by requiring higher SAT scores, while others opted to keep them unchanged,” thus providing “a controlled experiment of sorts that can fairly conclusively tell us whether SAT scores were accurate predictors of whether a student would get a degree.”
The short answer, he stated, is “yes, they were.” Campuses that raised the emphasis on the SAT experienced “remarkable improvements” in graduation rates, especially at the more selective campuses, ranging “from 10 percent (at Stony Brook, where the six-year graduation rate went to 59.2 percent from 53.8 percent) to 95 percent (at Old Westbury, which went to 35.9 percent from 18.4 percent).” The results were clear, Salins concluded. “[O]nly those campuses whose incoming students’ SAT scores improved substantially saw gains in graduation rates.”
Worse After Prop. 209
The results in California after Prop. 209 outlawed racial preferences were even more dramatic, and more on point for UVa.
- For the whole University of California system: blacks who entered in 1996 (before Prop. 209) had a four-year graduation rate of 26.5%. For the class entering in 2001, after Prop. 209, the black graduation rate increased to 38.6%. Six-year rates increased from 64% to 70.2%. Looking at a wider swath of this data, Richard Sander noted that “For the six cohorts of black freshmen who started at UC campuses before Prop 209 went into effect (the matriculating years of 1992 through 1997), the average four-year graduation rate was only 22.2%. For the years since 1998 (matriculating years 1998 through 2005), the black four-year graduation rate across the UC system is 39.4% — a near doubling. For Hispanics the four-year graduation numbers are 27.2% for 1992-97, and 41.8% for 1998-2005.”
- UC San Diego. The four-year black graduation rate increased from 37.7% for the class entering in 1996 to 51.4% for the class entering in 2001. The six-year rate went from 69.8% to 78.8%.
- UCLA. The four-year black graduation rate increased from 31.1% for the 1996 class to 53.7% for the 2001 class.
- UC Berkeley. The four-year black graduation rate increased from 28.8% for the 1996 class to 35% for the 2001 class.
The improvement in black graduation rates at the University of California would likely have been even more dramatic but for the widespread “holistic cheating” and evasion. Indeed, in researching their “magisterial” book, Mismatch, Richard Sander and Stuart Taylor Jr. found so much cheating that they concluded outright bans of affirmative action are not likely to work.
“[B]ecause of universities’ determination to circumvent any ban,” they argue (p. 279), outlawing preferences would not end them but would lead — and has led — to universities evading bans, thus possibly making mismatch much worse, not better.” The post-209 experience in California, however, suggests that such bans are like speed limits: they may not restrict speed to the posted limit, but they do make drivers drive slower than they otherwise would.
High Cost to Students and Taxpayers
Admitting minorities with higher SAT scores after Prop. 209 was not the only cause of their higher graduation rates, but it would be unreasonable to assume it was not a significant component. Thus one of the most disturbing aspects of UVa’s — or any university’s — continuing year after year to admit cohorts of minority students with much lower SAT scores than their peers is that administrators know that a high but reasonably predictable portion of them will fail to get degrees, with a high cost not only to the students and their families but also the taxpayers who support public education.
The reason UVa and others continue to do so, even armed as they are with this knowledge, is, of course, their determination to provide a sufficient degree of pigmentary “diversity,” whatever the cost. A very large, very disappointing, part of that cost at UVa are the 832 black students who entered from 2000 through 2011 who failed to get a degree, a failure to graduate rate of 25.5% of the 3,266 blacks who were admitted over those twelve years. Assuming as I do that many if not most of those 832 students would not have been admitted but for their race, I believe they are just as much victims of “diversity” as the equal number of whites and Asians who would have been admitted but for their race.
Laura Fornash, UVa’s assistant to the president for state governmental relations, agreed with my February Minding The Campus piece that the reason for UVa’s higher admit rate for blacks was that its yield rate was lower than for whites (36.4% vs. 44.5% for next fall’s entering class, she reported), since blacks qualified to go to UVa were heavily courted by other institutions. But that reason is a justification only if one assumes that UVa must have a large enough number of blacks to meet its self-imposed “diversity” obligation.
‘Diversity Uses Blacks for the Benefit of Whites’
What these diversity-justified preferences amount to in practice is admitting a large cohort of blacks knowing that a high percentage of them will not graduate so that there will be a sufficient number of them to provide “diversity” to whites, Asians, and others. “Let me state bluntly,” I stated bluntly back in 2002, “diversity uses blacks for the benefit of whites.”
Minorities admitted who would not have been admitted but for their race (the purpose and effect of affirmative action) are not admitted, after all, to provide “diversity” to themselves. Whatever benefit they derive from being in a “diverse” student body they would also receive if they attended a less selective majority-white institution where they would have a better chance of graduating.
If UVa treated all applicants without regard to their race or ethnicity, it would have what it must believe are too few blacks and too many Asians. (Whether or not it has too few or too many Jews, Mormons, Muslims, Methodists, Missouri Synod Lutherans, or transgender males is not known, because apparently neither religion nor gender identification (at least for now) is regarded as a relevant enough source of “diversity” to count.
What Is to Be Done?
In an ideal world — or even one that merely attempted to live up to what Gunnar Myrdal called “The American Creed” — benefits and burdens would not be distributed (especially by public institutions) on the basis of race, a principle the citizens of California, Michigan, Nebraska , and Arizona enshrined in their state constitutions. But sadly, the world we live in is far from ideal.
Nevertheless, since most Americans believe that government should be transparent, a worthy and achievable goal is to require state institutions and even private universities that receive public funds to publish data, such as SAT scores and class rank, revealing how heavily they put their thumbs on the racial scale. “It’s outrageous that public universities are not transparent about whether, and how, and how heavily they treat students differently on the basis of skin color and what country their ancestors came from,”
Roger Clegg, president and general counsel of the Center for Equal Opportunity argues (in an email to me, quoted with permission) — especially because, he continued, “the victims of this lack of transparency are the supposed ‘beneficiaries‘ of the discrimination — who are not told that their chances of graduation and getting good grades will be substantially less than other students’.” Conveniently, Clegg and Hans von Spakovsky have written a draft model “sunshine bill” that provides a good starting point for consideration of such legislation.
Universities, in short, should at least be required to publish data that would provide prospective applicants with information about the fate of students with qualifications similar to their own. UVa has just released SAT scores by race for one year. It would be enormously helpful to prospective applicants, their parents, and school counselors to have that data for every year, broken down by ranges of scores, which would allow useful correlations with the graduation rate data already provided. I suspect students of whatever race or ethnicity with similar SAT scores would have similar graduation rates, but since UVa and most other universities refuse to release that data voluntarily, it is impossible to confirm that suspicion.
It is easy to understand the institutional reluctance. After all, some black students may be reluctant to attend UVa if they knew in advance that their chance of failing to obtain a degree were two times higher than whites or Asians and that they would have an equal chance at less selective institutions where their qualifications equaled those of their peers.
Randall Kennedy’s View
It is less easy to understand the willingness, even eagerness, of zealous affirmative action advocates to sacrifice those mismatched students who predictably fail to earn degrees. Consider the argument, for example, of Harvard law professor Randall Kennedy, whose book, For Discrimination, I reviewed here, “The Odd Career Of Randall Kennedy.”
Kennedy supports the preferential admission of black law students even if “mismatch” results in the production of fewer black lawyers because, he wrote, most of the preferentially admitted do graduate and “the cadre of black attorneys trained at the top-tier schools are more valuable to the black community than those trained at the lower-tier schools, and hence, if necessary, maintaining the numbers at the higher-tier schools would be worth sacrificing marginal members or potential members of the black bar.”
Perhaps many “marginal” potential UVa students would agree with that view, but since a few — and their tuition-paying parents — may not, UVa should publish the data allowing them to decide for themselves.