I've recently posted a few times on the SAT score controversies currently swirling around some of the University of California campuses; in particular, Berkeley, Los Angeles, and San Diego have come under fire. The UC president, Robert Dynes, has publicly defended their admissions program, which let low-SAT-scorers in through "comprehensive review."
All week, I've been itching to post on the updates to the news that vastly-underqualified students were admitted to these challenging universities. The main update was the recent LATimes article that insists that race was not a factor in the admission of low-scoring students. Well, John of Discriminations doesn't agree with the LATimes conclusion, and he did a great job of dissecting the article this week. Such a great job, in fact, that Mickey Kaus of Kausfiles took notice. Kudos to John for the mention!
Anyway, here's John's post:
Despite the length of this article — over 2200 words — neither the numbers nor the differential rate of admission to Berkeley and UCLA for low-scoring minorities is given. Thus we are told that Berkeley admitted “low-scoring blacks and Latinos at twice the rate of Asians and whites with similar scores” and also that Berkeley accepted “only 8% of all low-scoring applicants” (emphasis added). We are pointedly not told what the admission rates were for low-scoring whites and Asians or for blacks and Hispanics, nor the absolute numbers.
When critics of race preferences argue that high standards and thus relatively fewer minority admissions to Berkeley and UCLA are not discriminatory because minorities are able to attend other, less selective campuses of the University of California system, they are often called racist. Now the Los Angeles Times argues that, despite highly disproportionate admissions of low-scoring blacks and Hispanics over similarly low-scoring whites and Asians at Berkeley and UCLA, there is no discrimination because in the UC system as a whole low-scorers from all groups are accepted at about the same rate.
Nice try.
Here's what Kaus had to say (scroll down to the second posting for November 5th):
I actually don't understand the entire basis of the LAT story. Does it tell us anything important if one ethnic group with low scores is admitted at a higher rate than another group with low scores? Doesn't the rate depend on the number of low-SAT applicants, which could vary for all sorts of reasons?
Suppose, for example, that members of ethnic group A know that if they have low SAT scores they are unlikely to get in. Since the combination of SAT scores and G.P.A. normally required to ensure admission is published on the Web, those whose scores are low just won't bother to apply...Now suppose many members of ethnic group B know that they have a credible claim of having overcome race discrimination--and that this might get them in under the university's "comprehensive review" policy, in which overcoming hardship can outweigh low SAT scores. These group B students are likely to apply in very large numbers even if they have low SATs. As a result, their rate of acceptance may be no higher than those of the few low-SAT applicants from group A. But that rate doesn't tell us much about whether or not university officials are bending over too far to admit applicants from Group B.
In fact, although the Times doesn't discuss it, the paper's own data shows that low-SAT "underrepresented minorities" (primarily blacks and Latinos) do apply to UC in relatively great numbers--so many that, whatever the success rates, 65% of the students actually admitted to Berkeley and UCLA with low SATs are "underrepresented minorities." Even at Riverside, the least selective UC campus, 49% of low-SAT admissions are "underrepresented minorities." How does this show that, as the Times says, "UC admissions did not appear to be racially biased" or that the the "comprehensive review" program isn't a backdoor scheme of racial preferences? It doesn't.
Kaus then says that if ignoring the SAT in favor of "comprehensive review" results in the admission of worthier applicants, he's all for it, but he points out something I've mentioned before - if the UC system is downplaying the SAT because it allegedly doesn't predict success in college, then they'd better be prepared to show us that the comprehensive review process does.
Kaus also mentions this Oakland Tribune article, which does a good job of crunching the numbers:
According to an analysis by this newspaper, 90 percent of the 332 students admitted to Berkeley in fall 2002 with SAT scores 1000 or below were minorities. In 2001, 89 percent of the 388 students with low scores were minorities...
In fact, both UC San Diego and UCLA -- the UC system's two other most selective campuses -- last week reported they had also admitted a small percentage of students with low test scores. UC President Robert Dynes has said he will convene a study group to review the admissions process.
Berkeley officials said SAT test scores alone are not a good indicator of who will succeed and who will fail at the university...
Altogether now - is there any proof that comprehensive review is a better indicator?
None of the students has left Berkeley due to academic deficiency, officials said.
Okay, so at least they're trying to support it, although this isn't enough. I want to see how long it takes those students to graduate, what GPA they have when they graduate, what majors they're enrolled in, and how they do in the GRE. In other words, I want to see the type of data that the College Board routinely rolls out in support of the SAT.
...an additional look at data provided by UC headquarters shows that most of the low-scoring students are minority. In 2002, 63 -- or 19 percent -- of the students were black and 149 -- or 45 percent -- were Latino. Those are minority groups that are underrepresented at UC Berkeley and other UC campuses. Another 83 students (25 percent) were Asian, 5 (1.5 percent) were Native American and 23 (7 percent) were white. Another 9 students were categorized as "other."
In 2001, 66 -- or 17 percent -- of students admitted with scores below 1000 were black and 170 -- 44 percent -- were Latino. Asian students numbered 110 (28 percent); 25 students (6 percent) were white and 17 students (4 percent) were "other" or didn't provide the data.
Supporters of comprehensive review insist that this doesn't mean they're doing an end run around Proposition 209. The data suggest otherwise.
Others say no one should be surprised that minorities score poorly on the SAT -- and those scores alone don't mean a student is underqualified for the university.
A report released Friday by the San Francisco-based Equal Justice Society refutes Moores' analysis. A bevy of research indicates that socioeconomic status is closely correlated to SAT scores.
"With every $10,000 increase in family income, there is a lock-step increase in SAT scores," said the Equal Justice report, which was prepared by a coalition of Berkeley faculty, civil rights and education groups.
Two things fascinate me about this type of research:
(1) These anti-testing groups don't seem to notice that SES is also correlated with many indicators of academic achievement. Thus, the argument for not using the SAT could be extended to not using grades, not using AP test scores, not using AP class enrollment, not using number of years of math courses, etc. Why stop at opposing tests? Why not insist (as some do) that kids from poor backgrounds can't be expected to do well on any measures of academic achievement?
(2) The correlation of SES with SAT scores tells us absolutely nothing about whether the SAT is measuring something real. In fact, in our capitalistic society, it would be very odd if SAT didn't correlate with SES. The fact is that kids who come from wealthy homes are more likely to have had access to tutoring and better schools; are more likely to come from homes with educated parents; are more likely to have spent time around books and libraries and enriched environments. Thus, the SAT, which is related to SES, is telling us that kids who grow up around more money learn more, and do better academically as a whole. The faltering K-12 public system is complicit in this, because kids from poor backgrounds often get stuck in the worst schools, with the teachers most willing to lower standards and make excuses.
I agree that this sucks, and in some sense it is not fair that some kids get the shaft economically. However, this reality does not lead logically to the conclusion that colleges should feel obligated to pass over kids from wealthy homes, who are well-prepared for college, in favor of kids who just never had the opportunity. There is no indication that "leveling the playing field" in this respect is actually giving more opportunity to worthier candidates. There's no data to show that it make sense to admit students with poor grades and low SAT scores, under the assumption that these students will automatically flourish in a challenging college environment.
How colleges have reached the point of justifying the admission of students who didn't really accomplish anything in high school, simply on the basis of the fact that those students didn't have the same economic opportunities as other, more accomplished students, is beyond me. It's as though these policies are determined to punish some kids for having had more opportunities than others, and downplay the very real accomplishments of kids who took full advantage of their parents' generosity.
Posted by kswygert at November 7, 2003 10:36 AM