<p>An interesting study suggests that selective schools' increasing reliance on SAT scores, due to concern for college rankings, is tied closely to increased need to use affirmative action to achieve diversity: colleges could actually increase diversity, the authors claim, by not relying on the SAT for admissions decisions; if they did so, there would be no decline in students' academic achievement. </p>
<p>
[quote]
The key finding, Alon said, is that if one accepts the need to place considerable emphasis on the SAT, there is a tension between merit and diversity. But the paper goes on to argue that this is a false choice. The two sociologists ran counterfactual simulations in which they use their findings to imagine an admissions world without the SAT or affirmative action. Running the numbers from the actual admissions years and removing the additional emphasis that they found colleges gave for high SAT scores and for being a member of certain groups they found that the two were balancing each other out, and that colleges need affirmative action to preserve current levels of diversity only if they stress the SAT.
<p>The summary of the report doesn't answer a couple of key questions: would the diversity w/o the SAT come at the expense of some groups? And do the results imply that SAT underpredict the performance of underrepresented minority students? (Other studies suggest that if when they do, the effect is small.)</p>
<p>Is it possible that competitive colleges are using high SAT scores of one group of applicants to counteract the low scores of another so the average SAT remains high for ranking purposes?</p>
<p>Assuming that GPA and graduation rate remain constant, is there a shift in the %tage of students entering humanities and away from the physical sciences and mathematics?</p>
<p>The authors seem to confuse cause and effect. Demographic changes in the US were responsible for the increased competition in the process. If the colleges simply tracked the SAT's of their students and did not use them in the process of admissions they would have observed the SAT's rising and they would have (absent affirmative action) observed a decline in URM's. </p>
<p>Making the claim that eliminating standardized testing (the SAT II's and AP's would have to go too) and relying on grades and class rank alone would not result in a decline in academic quality is preposterous on its face. Using graduation rates as a surrogate for academic quality is silly. Chosen majors have a lot more to do with graduation rates. Those who choose harder majors frequently take longer to graduate. No doubt abandoning standardized testing would mean less need for affirmative action, but we hardly needed their study to prove that. Another blinding glimpse of the obvious. WOW!</p>
<p>"An interesting study suggests that selective schools' increasing reliance on SAT scores, due to concern for college rankings, is tied closely to increased need to use affirmative action to achieve diversity: colleges could actually increase diversity, the authors claim, by not relying on the SAT for admissions decisions; if they did so, there would be no decline in students' academic achievement."</p>
<p>Gordon Winston and Catherine Hill's paper (Hill is now President of Vassar), "Access to the Most Selective Private Colleges by High-Ability, Low-Income Students: Are They Out There?" proved conclusively that these colleges could be much, much more diverse WITH using the SAT, and without AA - that there were between 2 1/2 and 3 times as many high-qualified lower income students available as were actually attending these colleges. When they don't have lower income students, it is because they choose not to.</p>
<p>Not that this is a great problem. Williams' loss is UCLA's gain.</p>
<p>C anuckguy: I am a bit puzzled by your post. Is your point that it is easier to earn high GPA's in the Humanities than in the sciences? This has not been my experience. Whenever I go to parties when the evry accomplished other party-goers discover that I am an English prof. the comment usually is, "English is the one college subject I could never getA's in." </p>
<p>I'm that in "hard disciplines" the number of really low grades is highter but so is the number of really high grades.</p>
<p>"At the same time, he said, the colleges that have students admitted without the SAT “perform just as well as those who submitted test scores.” This may possibly be true in a less-competive college not a top one."</p>
<p>The most competetive colleges for which there is extensive data on this question, Bates and Mt. Holyoke, show very little difference in academic achievement between submitting and non-submitting students in SAT-optional contexts. </p>
<p>The study above seems to be driving at a somewhat different question, though: URM's, on average, score lower on the SAT (probably because of lack of access, in many cases, to educational opportunites); so, if you emphasize SAT scores strongly, you need "affirmative action" of some type to get decent diversity. What the study seems to suggest--though this is never quite stated explicitly--is that if you look for students who are high achievers in high school, you wind up identifing many qualified URM students who might not be admitted if you focused mainly on their SAT scores--students who do just fine once they're enrolled. What's more you identify them in about the same numbers as the numbers that you'd admit by giving them special preference under a system that weights the SAT strongly.</p>
<p>Your post #10 is exactly my take on the article, too, MM. I'm not sure why all these side issues are being discussed, because side issues were not the point of the article. I think this is an important article.</p>
URM's, on average, score lower on the SAT (probably because of lack of access, in many cases, to educational opportunites)
[/quote]
</p>
<p>How do you reconcile your explanation with 1995 SAT data, which showed that black students from families earning more than $70,000 in 1995 dollars scored lower than white students from families earning less than $10,000 and barely scored higher than Asian students from families earning less than $10,000?</p>
<p>Do these poverty-level whites and Asians have access to educational opportunities that the upper middle class blacks do not?</p>
<p>
[quote]
I'm black and I got a 2370 SAT. I want AA and emphasis on high SAT scores to stay.
[/quote]
</p>
<p>If it turned out that Asian enrollment decreased under race-blind admissions, I would have no problems with that. Asians would just have to work harder.</p>
<p>By contrast, it is impossible for a supporter of racial preferences to make a similar claim.</p>
<p>If race-blind admissions results in fewer "under-represented" minorities, then they won't support the policy.</p>
<p>I, an opponent of racial preferences, have frequently written that I accept the possibility of decreased Asian enrollment through a race-blind policy.</p>
<p>I challenge a supporter of racial preferences to state that fewer "under-represented" minorities is an acceptable outcome of race-blind admissions.</p>
<p>"Gordon Winston and Catherine Hill's paper (Hill is now President of Vassar), "Access to the Most Selective Private Colleges by High-Ability, Low-Income Students: Are They Out There?" proved conclusively that these colleges could be much, much more diverse WITH using the SAT, and without AA - that there were between 2 1/2 and 3 times as many high-qualified lower income students available as were actually attending these colleges. When they don't have lower income students, it is because they choose not to.</p>
<p>Not that this is a great problem. Williams' loss is UCLA's gain."</p>
<p>The anti-SAT crowd rather ignores the fact that the test was a liberal innovation designed to ferret out high ability students amongst the great unwashed who did not attend prep schools. It was developed to increase diversity at Americas best schools not to decrease it. Whether it does that job or not depends on how it is used but I tend to agree with mini on this one.</p>
<p>Additionally college graduation rates are a poor proxy for academic excellence. America's best colleges and universities are looking for people who in the course of their lives can greatly advance the sum of human knowledge not merely master a subject area. There is a huge difference between competence and excellence. What does the Phi Beta Kappa chapter at a given school look like compared to the entering class? That would be a better proxy for excellence than anybody who managed to get through the Business School and yet even here kids in the most difficult subject areas would be under-represented.</p>
<p>The paper in question makes two points. The first point is that eliminating the SAT's (and presumably all other standardized testing) from the college admissions process will result in more URM's being admitted. I doubt there is anyone in the country that would have disagreed with this statement before they did their research. The second point is that the academic environment would be unaffected by the elimination of standardized testing. To prove this they use graduation rates as a proxy for quality. I wonder how many other criteria they had to dismiss before they found something that would support their point statistically. They make no attempt to defend the selection of this criterion. I am amazed this kind of stuff gets published.</p>
<p>Here is a link to a press release that provides a bit more detail, but frankly the research doesn't sound any more nuanced than curious 14 suggests. </p>
<p>Still, I think that there are interesting questions being addressed, maybe a bit obliquely: are there significant differences in ethnicity between groups of high achieving, but low SAT scoring and low-achieving but high-SAT scoring students? And how do students with these profiles perform in college? What are the implications of all this for schools at various levels of selectivity?</p>
<p>I believe that money is a bigger impediment for elite schools trying to increase their minority enrollment than the SAT. They don't take more because the prefere to take full pay non-minorities. The same is not necessarily true for either public schools or schools outside of the top 50 or so.</p>
<p>The truth is you don't need a 1550 SAT to do well at Williams but you are probably not in a position to benefit from a Williams education if you show up with a 1000 either. You might still graduate however.</p>
<p>One of the problems with American HS education is that grade inflation has compressed the range of GPAs. That coupled with the huge disparity between schools makes it very difficult to tell anything about a kid just based on grades. All 4.0's are not created equal. Class rank is probably a better indicator but even that leaves a lot to be desired as schools increasingly reward compliance over actual intellectual aptitude.</p>
<p>A rigid reliance on SAT scores is not necessarily good either but taken in context they can be very valuable indicator that you have found someone special. If you have some kid posting a 1250 in a school where the average is 900 and he/she is at the top of the class that is more impressive to me than a 1450 for some kid in a prep school. How good it is you won't know until you get it out and polish it up but it is sure worth spending some of those billions in the endowment on. Especially if the alternative is to give Al Gore $100K to give a speech promoting his latest book.</p>