Want diversity w/o Affirmative Action? Don't rely on the SAT

Parents Forum
21

<p><a href="epiphany%20wrote:">quote</a>
Here's the key phrase for me, in the article:</p>

<p>"doing so would not result in a diminution of student quality."

[/quote]
</p>

<p>Note that this study uses a very similar methodology to the Princeton study (of race in admissions) that you claim is garbage. In fact, some of their earlier work with the 1982/1992 dataset was done at Princeton in the same department. </p>

<p>There is one big methodological problem that this study faces that did not trouble the Princeton admissions study. Both papers develop a model of admissions and use it to predict who the admitted students would be under different policies. The difference is that the earlier study, of race in admissions, only had to guess who would get in, and look at their SAT scores, which were known. The newer study has to also speculate on the chances of the hypothetical new group of admits to graduate. This is quite difficult and I'm not sure it can be done reliably, in part because the population of no-SAT admits has to include a lot of people with statistics that would not have made it into the with-SAT admit pool at each given college, so there is little or no data (at that college) from which to extrapolate their graduation rates. </p>

<p>So what is likely is that the race-in-admissions study is fine, while this study is deeply flawed. We will see when it's published.</p>

22

<p><a href="curious14:">quote</a>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.

[/quote]
</p>

<p>Using graduation rates guarantees the results before the study is performed.
Just as everyone knows more minorities would be admitted without the SAT,
everyone knows that SAT has more effect on admission than on graduation. So if the goal is only to maintain certain rates of graduation, there is always room to reduce the weight of SAT and increase the weight of factors more advantageous to minorities: grades, class rank, recommendations, etc. This would lead to admission of more minorities.</p>

23

<p>I think diversity (or lack thereof) has nothing to do with SAT scores and everything to do with economics, opportunity and resources. Even discussing diversity in the strict context of $45K+/yr, super-elite schools demonstrates a lack of understanding of the challenges facing low-income African American and Latino students (you could add rural white kids in there too, but they are not considered URMs in spite of facing many similar circumstances). </p>

<p>I've always thought that SAT/ACT scores are helpful in offering many students another tool to prove their worth. They can be part of a picture - not the dominant part, but perhaps important anyway.</p>

<p>Of course the true bottom line to diversity in higher education is to offer equal education opportunities to everyone in this country from pre-school on up, but I don't think that's going to happen anytime soon. :(</p>

24

<p>
[quote]
I'm that in "hard disciplines" the number of really low grades is highter but so is the number of really high grades.

[/quote]
</p>

<p>Agree. If I were bright enough to be in the top 1/2 or 1/4 of one percent of population, I would certainly major in the physical sciences/mathematics. On the other hand, if I am just an average student for the school I am in, humanities would be a safer choice. This type of self-selection goes on everyday. </p>

<p>
[quote]
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?

[/quote]
</p>

<p>It is razor-sharp analysis like this that put a lot of silliness to rest. Well done.</p>

25

<p>Thanks, Canuckguy.</p>

<p>
[quote]

So what is likely is that the race-in-admissions study is fine, while this study is deeply flawed. We will see when it's published.

[/quote]
</p>

<p>To my knowledge, the Espenshade / Chung study on racial preferences in admissions hasn't been refuted.</p>

<p>It took me way too long to see that the Kidder rebuttal is deeply flawed.</p>

<p>Espenshade and Chung used undergraduate data from several schools and showed that the removal of racial preferences barely affects whites, negatively affects "under-represented" minorities, and positively benefits Asians.</p>

<p>Kidder used data from Boalt Hall and showed that Proposition 209 resulted in many more white students being admited, many fewer "under-represented" students, and a trifle more Asian students.</p>

<p>For many, it seemed like E&C were soundly rebuked. I even thought so, until I finally saw that Kidder used POSTGRADUATE data to refute a study that used UNDERGRADUATE data.</p>

26

<p>Regarding post #20,</p>

<p>Grade inflation may be a problem, but i.m.o. it has never been the main problem. The main problems are the vast differences in the following, across the country, and even across the globe:</p>

<p>(a) different expectations & standards of scholarship, school to school, merely in the curriculum itself: in other words, the very content of the curriculum -- not just the texts & other readings, but the performance requirements. In some schools these are minimal; in others, much stiffer; in still others, there is no 'maximum' possible level of performance, & that creates a significant curve effect, by which it is amazingly difficult to achieve an A unless you are a fanatic about grades & eliminate most of your e.c.'s.</p>

<p>So grade inflation has to be seen not just as a reflection of performance standards but as it relates to the quality of the curriculum. What difference should it make if a student at a poor performing school does really, really, well in a course from the American Guidance Service curriculum AND from a "hard"-grading teacher --> as opposed to a student doing equally well in a high school course that is using a college text and requiring annotated research papers?</p>

<p>This is why the whole "corrective" aspect of the SAT is so phony. It is NOT corrective for high school curriculum. It has little to do with high school curriculum.</p>

<p>The student getting an A in AGS curriculum cannot be compared with a student getting an A in the "same" course at a school so different it might as well be in a different country.</p>

<p>(b) Different peer environments. A student who shines in a basic U.S. History course may look especially good versus other students in that class, of a "poorly performing school." But that star student may have little in common with even the 2nd- or 3rd-top student's coursework in a much more rigorous school, where most of the peers are HIGHLY performing. There is an up and a down to peers. They both stimulate & advance a student, and they correct or diminish the grade of that student.</p>

<p>Please do not get me wrong. I teach more of the underperforming students & environments than of the maximally performing, and I support the former's needs, achievements, and college admissions efforts. I'm glad that there are at the least, public options for these students. (Most state schools have an 'eligibility index' of some kind, & highly performing students of any environment should be rewarded.) </p>

<p>(c) Five broadly scaled grade indices vs. numerical grade indices. Further, esp. problematic is that in some letter-grading schools, 90 percentage points in the classwork = an A; in others, only 95+ = an A. In my D's school, this scale even differs teacher to teacher! There is one teacher who reflects the east coast standards in which he taught & was raised; a teacher of the same subject at the same school uses a very different scale.</p>

<p>Regarding the attempt to once again bring up the specious "Princeton study," I never brought it up in the context of this thread and this article. I don't think it's an appropriate comparison in this context, even though there are posters who love to bring it up as often as possible, preferably once a week.</p>

<p>I do agree with posters who object to using graduation rates as the sole index of "student quality." My cited comment, above, was one I offered if in fact the full methodology & results of the study <em>could</em> be determined (and IF that including an assessment of student qulaity, which I should have made clear). I don't think we know all the factors that went into the study, but graduation rates alone would not, for me, determine "student quality."</p>

27

<p>
[quote]
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.

[/quote]

I partially disagree with this. Colleges aren't going to take unqualified white kids who don't add to their diversity just because they can pay. Schools trying to increase their minority enrollment would most prefer upper class urm's who don't require financial aid. All the better if they have high SAT scores. Then they will give financial aid to be sure they have enough racial diversity, and/or SAT requirements may be relaxed to assure this goal is achieved.</p>

28

<p>
[quote]
Colleges aren't going to take unqualified white kids who don't add to their diversity just because they can pay.

[/quote]
</p>

<p>Unfortunately, even schools with the best intentions have to do this all the time. Very few schools have the funds to be entirely need-blind, so at some point, there are tough choices: do we take a minority kid who needs significant FA or a full-freight student, often academically somewhat stronger (because the strong URMs have been admitted) who does not add diversity? If the FA budget is spent, as it sometimes is as adcoms work their way toward the bottom tier of admits--even at schools with nice endowments- there's not even much to discuss.</p>

29

<p>I ask that supporters of "diversity" keep in mind Justice Kennedy's concurring opinion, which emphasized that certain kinds of diversity are indeed both desirable and legal.</p>

<p>Unless supporters of "diversity" are interested in another Supreme Court rebuke, I suggest that they preface the word diversity in sentences like "a full-freight student...who does not add diversity?" with racial.</p>

<p>The addition of a student, who is inherently unique, by definition adds diversity.</p>

30

<p>
[quote]
The addition of a student, who is inherently unique, by definition adds diversity.

[/quote]

Absolutely! This makes perfect sense as we all have special gifts to offer. :)</p>

<p>Is racial diversity the certain kind that Justice Kennedy said is both desirable and legal? Frankly, since I am the hard-working middle class parent of hard-working "white" BWRK's, I am in favor of letting a kid's accomplishments speak for themselves. There is no check box for any diversity my kids might add. Maybe not pc, but there it is.</p>

<p>I guess when they near the end of filling a class, ability to pay does come into play. But I bet there are enough qualified students of adequate means that the decision is not based only on money.</p>

31

<p>Maybe this is a moot discussion considering African American and Latino high school graduation rates lag behind whites by over 20%. </p>

<p>Personally I think if we truly fixed education in this country, we wouldn't have to even think about "college diversity" - which has really turned into just so much window dressing. And until rich white families are willing to welcome poor kids of color into all their schools, the colleges will be left squabbling over the same small group of high performing "diversity" students. </p>

<p>Isn't it funny how all of the sudden rich folks want to see minority faces in their kid's college, but they sure don't want them in their elementary/middle/high schools? (And my reference to this is specifically in the extreme opposition - usually from exclusive suburbs - to school choice and changing the way school districts are funded and district lines are drawn.)</p>

32

<p>The more I think about it the sillier using graduation rates as a stand alone proxy for quality seems. For example, schools with rampant grade inflation will tend to have higher graduation rates, all other things being the same. But does this mean they are "better" academic environments. Is UVA better than U of Chicago because it has a 93% graduation rate compared to Chicago's 91%.</p>

33

<p>
[quote]
Is racial diversity the certain kind that Justice Kennedy said is both desirable and legal?

[/quote]
</p>

<p>Unfortunately, he didn't give examples of what is and what isn't both desirable and legal regarding diversity.</p>

<p>I'm for racial diversity, but I'm against policies that result in dismissing capable and interesting students who magically fail to "add diversity."</p>

34

<p>first point to fabrizio...</p>

<p>you continue to use the word "preference" when, in fact, this is not the same thing as affirmative action. you do this ad nausea without seeming to understand exactly what kind of weight the word "preference" has. NO COLLEGE IS PREFERRING ONE RACE OF STUDENT OVER ANOTHER. it does not work that way - affirmative action and those of us who consider it while making admissions decisions do not work in the way that many folks who disagree with race-based affirmative action assume it does/we do. </p>

<p>secondly to fab and everyone else...
race is not the only thing that will get a kid admitted to a most selective school in much the same way that a high sat score or a special artistic or athletic talent will not. we look at the total package, period. when evaluating a student's "credentials," there is a lot of hard work, research, and experience backing it up. every admissions officer i know who works at a "most selective college or university" in this country (and let us remember that there are really very few of these schools - only about 50 schools out of the 4000+ in this country admit less than 30% of their applicants; most schools in this country are non-selective and don't use affirmative action, period) - tries their best to evaluate a student's quantifiable acheivements WITHIN THE CONTEXT OF THEIR EDUCATIONAL OPPORTUNITIES AND WITHIN THE SOCIAL CONTEXT THESE STUDENTS ARE LIVING. regardless of one's economic resources, racism, discrimination, and prejudice exist. even if a student is wealthy and attends an elite high school and has a lot of advantages, i am still going to give that student <em>some</em> benefit of the doubt if i know they are facing some form of discrimination - be it based on their race, gender, sexuality, or whatever. ignoring the fact that a student may have faced the psychological and social impacts of such discrimination (which we know impacts educational achievements and learning) is foolish and irresponsible on moral and pedagological levels if we are claiming to be "fair" or "meritocratic". this applies to students of all races, not just black/African American and Latino students. </p>

<p>simply put, setting the bar at one level and then expecting everyone to jump over it when there are impediments restricting how high each student can jump is not fair and not meritocratic. there are many under-represented students (both racially and economically) who are put at a disadvantage in the admissions process at many elite colleges because their high schools and elementary schools have not been able to adequately prepare them for the rigors of an elite college. and a lot of these students - actually more of these students, in real numbers - are white. this problem is symptomatic of the inequalities that exist in our k-12 system and now, more than ever, the gap between what is required to graduate high school and what is required to be prepared to do well in and graduate from college, academically speaking, is growing. however, there are a lot of economically disadvantaged students who do have adequate preparation (though not excellent or outstanding) as well as the motivation and drive to do well at elite colleges who are given a "bump" in the admissions process - and a lot of these students, like many of those mentioned in the winston and hill study, are white. however, just because we don't call this "bump" "affirmative action," do not assume that we aren't also giving white kids some benefit of the doubt in the admissions process in a similar way that some black/African American and Latino are given a "bump". </p>

<p>i would also like to remind everyone that white students have had access to BASIC public education for years that took them to grade levels higher than most racial minorities in this country. it was not that long ago that high schools were segregated and many of the non-white schools did not go up to the 12th grade (this was back when a high school diploma actually meant something). and even today, when looking at elite boarding and private schools in this country - schools that send huge numbers of their graduates to highly and most selective colleges and universities - one finds that racially, their students are OVERWHELMINGLY WHITE. many of these schools have been around for generations and have had special relationships with elite colleges. the students who have attended and who are attending these schools (again, overwhelmingly white) have so much advantage in the admissions process that it is astounding - between the college counseling, the phone calls made to college admissions offices on behalf of these students, the academic preparation, and the assumptions that are made about the quality of work these students are doing at their schools, these kids are hooked up and have higher admit rates than others. and again, most of them are white. </p>

<p>in addition, most students who have athletic, development, or legacy "hooks" are overwhemingly white. in fact, in some sports like squash and crew (which are often only found at the varsity level at elite schools) i have never seen a black/African American or Latino student recruited (although I'm sure there may be a few somewhere!). again, it is not like white kids do not get their "bumps"! </p>

<p>the fact of the matter is this: the actual number of black/African American and Latino students applying to our nation's most selective colleges and universities is staggeringly small, regardless of income level. in addition, affirmative action is not always used simply because a student is black/African American or Latino. when you actually chrunch the numbers, we are talking about a minute number of students in a given school's applicant pool - which itself is very small relative to the number of high school graduates every year - for whom affirmative action is "helping". when you consider that an academically competitive white student's "chances" of being admitted to one of these schools is only increased by around 2% if race-based affirmative action is not used (see bowen and bok's work), this, coupled with how overwhelmingly white athletic, development, and legacy hooks are; how white students have for generations had educational opportunities not present to others; and how white students also get consideration for the economic or social impediments that may have affected their educational achievements illustrates that race-based affirmative action 1) really "helps" black/African American and Latino students little and 2) really "hurts" white students little. it does not have the awesome, tidal wave effect many assume it does...especially when only a minute number of colleges and universities are using it in the first place!</p>

35

<p>Wow. So many words! Let’s imagine a mental experiment in which we take identical applications and randomly assign the race of “White or Asian” to one, “Black or Hispanic” to the other. Is there anyone who doubts that the “Black or Hispanic” candidate would be accepted in all cases in which the “White or Asian” candidate is accepted and in many cases in which the identical “White or Asian” candidate is rejected? Is there anyone who doubts that there would be no case in which the “White/Asian” candidate is accepted and the “Black/Hispanic” candidate is rejected? Sophistry aside, that is “preferring “ “Black and Hispanic” candidates to “White and Asian” candidates.</p>

36

<p>Admitting a few candidates that happened to be family friends of an admissions committee member would have an immeasurable small effect on the probability of admissions for all other candidates but that wouldn't make it right.</p>

37

<p>You want to get rid of "hooks" (preferences in standard english) for recruited athletes, developmment candidates, and legacies because they are anit-meritocratic, please do so. But don't use them as an excuse for other forms of preference.</p>

38

<p>curious,
you can "wow" all you want to, but the number of words in AdOfficer's post is minuscule compared to the number of words of opposers of any form of diversity admission (AA or anything related to opportunity-admission) on CC alone. "The opposition" complains often of the absence of authoritative posting of policy, of rationale, etc. But when an authority does actually post, with the supporting rationale, the post is nevertheless often dismissed or ridiculed.</p>

39

<p>Before this becomes just another AA thread, let me summarize what I took from the full article, which is now available on line:</p>

<p><a href="http://www.asanet.org/page.ww?section=American+Sociological+Review&name=Homepage%5B/url%5D"&gt;http://www.asanet.org/page.ww?section=American+Sociological+Review&name=Homepage&lt;/a&gt;&lt;/p>

<p>It turns out to be more sophisticated than the summaries. (The graduation rate as a measure of academic quality is not much more than a brief aside about the U-Texas top-10% policy, for instance.) </p>

<p>In a nutshell: From the 1980s to 90s (much of the data is pre 1992), emphasis in admissions shifted dramatically toward standardized test scores and away from high school achievement. This "shifting meritocracy" effect (the tendency to define merit in college admissions mainly through high SAT scores) becomes more pronounced the more selective a school is. Not surprisingly, the tendency to stress test scores made it impossible to enroll racially diverse classes without the use of race-sensitive admissions practices--particularly because racial differences in SAT scores are greater than differences in class rank (and these differences are widening). </p>

<p>The authors then pose the key question: suppose that you gave less weight to SATs and more to HS achievement to everybody? What if you rolled back the "shifting meritocracy" in other words? If you rolled back admissions practices to 1982 levels (when tests were weighted less), you don't need to give URM's as much of an admissions "bump." If you simply throw SAT scores out of your admissions model, the preference (their term ;) ) that you need to give URMs is dramatically smaller. </p>

<p>Finally, their main recommendation is modest: more institutions should use "full-file reviews" that look at standardized test scores in the context of the applicant's background and opportunities. One way to read the article, I think, is to say that URMs are getting a huge admissions advantage only if you define merit solely as high test scores. If you define merit as high class rank--which is a better predictor of college grades anyway--or in other broader and more generous ways, the amount of preference turns out to be quite small.</p>

40

<p>AdOfficer has addressed all the <em>niche</em> situations...URM, recruited athlete, legacy, etc.</p>

<p>What about the above average white male who does not fall into any groups? </p>

<p>Basically, it seems like because so many "overwhelmingly white athletic, development, and legacy hooks " are white , the unhooked, unathlete, non-development, non-legacy white kid gets left out of all the <em>bumps</em>. </p>

<p>Is that true? If so, it IS more difficult, in terms of grades, scores, ECs, everything, for people in that group (can you guess which group my kids are in?). </p>

<p>If what AdOfficer says is true, it's not the URMs that present my kids with a harder road to hoe, figuratively speaking. Apparently my sons are competing with the athletic, development, legacy whites.</p>

<p>Hmmm, that's makes me feel a lot better.</p>