"Race" in College Admission FAQ & Discussion 9

Findmoreinfo · April 24, 2012, 11:59pm

These are my findings:

SAT 2400er’s
Asian Americans/Asians

radiant_tragedy - rejected by all ivy league schools, accepted by Rice U.
Yodastreet - rejected by all ivy league schools, accepted by U. of Washington
norcalguy - rejected by all OTHER ivy league schools, accepted by Cornell
christiansoldier - rejected by all OTHER ivy league schools, accepted by Princeton
skiiz21 - rejected by all ivy league schools, accepted by UC Berkeley

It would be much more efficient if CC stats profile has race in there. It’s hard to find clues from individual posts. Can’t find much info. for other races.

But still, there are people like kevinjimba whose stats is clearly to be fake (don’t know how many self reporting scores are fake). I can only conclude that the data on CC can’t be used for discussion on this matter.

siserune · April 25, 2012, 5:43am

Based on your list I had a look at the CC stat profiles. They are unusable as an indicator of admission rates because the list of schools in the profile does not reflect applications really sent (only a list of “intentions to apply”), and very few people update the profile after admission. At best the profile is a pointer to users whose postings, if any exist, might provide some data. Still, let’s look at what was offered:

Some problems with your method.

<ol>
<li>Exaggerating the rejection rate by writing “all” Ivy League schools without evidence that applications were sent to most of those universities. For example, the self-reported results for ChristianSoldier, Princeton class of 2014, (posted on 4-01-2011 when decisions came out) were “ACCEPTED: Princeton, Stanford, UC Berkeley - WAIT-LISTED: MIT - REJECTED: Harvard, Yale” which is a 50% acceptance rate at HYPS, 40% acceptance (and only 40% rejection) rate at HYPSM. Your list makes it appear as though he was accepted at 1/8 and rejected at 7/8 of the Ivy League schools, which is an admission rate several times lower than what was actually observed.</li>
<li>A waitlisting may not make much difference for most applicants, but when gathering data to measure the universities’ admission process, it is quite different from a rejection. Waitlist is a frequent outcome for students with high grades and test scores, so is a fate that will disproportionately afflict the Asian applicants whose results particularly interest you. Jian Li, as a famous example, was falsely reported in many news articles as having been “rejected” by numerous top universities, but was rejected only by MIT. </li>
<li>Your research is far from complete even on its own terms. Other recent stat profiles not on your list are Asian acceptees to the same schools, and additional users have posted results that do not appear in stat profiles. This includes some non-Asian 2400 scorers rejected at all the top tier schools to which they applied.</li>
</ol>

Methodology aside, your question is interesting, but the answer can be predicted as a “thought experiment”, because…

<ol>
<li>Using the Ivy League type of admission criteria, under race-neutral or race-blind admission it is still expected that Asians (especially those with 2400 SAT) will have a lower admission rate than whites with the same scores. No discrimination is required, and no particular discrimination can be inferred unless the disparity in admission rates is gigantic. One reason is that if you imagine that there is an Asian test-prep or academic striving effect that (on average, not in every individual) raises the Asian SAT distribution by something modest like 20-40 points per applicant over whites, due to the normal distribution of the test scores, this small shift will have a huge influence on the upper tail at the perfect and near-perfect scores. The percentage of Asian perfect scores that are attributable to preparation more so than ability will be substantially higher than for whites, simply as a statistical consequence of the undisputed sociological observation that some immigrant groups tend to strive academically. Applicants with “false positive” SAT scores are less likely to have other material that confirms ability at the level suggested by the SAT, and this will depress their rate of admission.</li>
</ol>

Findmoreinfo · April 25, 2012, 8:46am

siserune ,

I said CC data can’t be used for this discussion. Didn’t you read that?

The ones I listed are those I could find info. for their race, schools applied, accepted/rejected from a sea of their posts.

I went through profiles of a portion of 2400er’s but didn’t find much info. for those who didn’t list information in their profile (especially not of the race field). Going through their posts are too much to do and also not exactly reliable. And I already indicated the data cannot be used.

siserune · April 25, 2012, 2:20pm

The CC data is informative if:

<ul>
<li>the algorithm you used to determine the data set is disclosed to the point of being replicable. (Ex: “class of 2014 or later stat profiles having 2400 SAT and one or more posts in a Results thread indicating application/admission outcomes and ethnicity, with the latest Results posting for each such user taken as the data source”.)</li>
<li>results are given as a detailed list of School/Decision for each username, treating waitlist as a different outcome than rejection. The list should be for the individual’s full set of applications or at least the ones in a given tier, because even if the goal were to gather data on HYP applicants only, outcomes at similar universities would affect the interpretation of each case, and the total set of data is too small to disregard the extra information. This would also prevent the problem of subjective or slanted summaries like “rejected from ALL other Ivy league schools” to describe someone with a 40+ percent success rate at HYPSM and no applications to the other schools.</li>
<li>if the sample is tiny, such as 2400 scorers, it is important to list every username you found so that further investigation is possible. Some data points may turn out not to be valid, such as international students, or are different in some other way that should be accounted for. (Recent example: user “20more” provided data on several high-scoring Asians rejected at the top ranked schools, asserting this as evidence of impossible admission standards for Asian males, but later it was disclosed that most of them were internationals.) </li>
<li>since your project is to get the results by ethnicity, gender also plays a role (it “interacts” with race when counting raw admission rates), and the complete self-reported race/gender information should be given.</li>
</ul>

One of those non-Asian rejectees with 2400 (white/Hispanic, now at Brown after being shut out of the uppermost admission targets) went to the trouble of compiling some data from the annual Results threads for Harvard and several of its rivals, including admission rates and average SAT scores of accepted, rejected and waitlisted, for each ethnicity. His numbers were for all applicants, not only those with perfect SAT, and there was nothing obvious in the tables that suggested different admission standards for Asians. The China Pride warriors who populate these CC discussions could, and would, have posted links to that Results census for ever after, if only it had confirmed any of their theories about the admissions process, but that has not happened. You can find the material and form your own opinion by searching for postings of user “silverturtle”.

Shrinkrap · April 25, 2012, 7:23pm

[Cricket</a> Sound - YouTube](<a href=“Cricket Sound - YouTube”>Cricket Sound - YouTube)

Findmoreinfo · April 27, 2012, 9:05am

“since your project is …”

I think you make yourself confused. When a person is to start a research, he/she need to start to look into senarios and what is available at hand for furthur investigation.

It is not a “project”, “research”, “thesis”,“disertation” topic YET. If the data is not good and the person thinks it is not an achievable direction then he/she can stop there and change/adjust to another senario. If the person still thinks the senario is possibly right, he/she can start his/her own data collection and continue on it.

CC data is certainly not good no matter how you twist it. The self reporting way can give much untrue data while no name/responsibility associated. Some people even conflict their own admission claims in some of their posts.

I’ll try to see if I can get official data from target colleges. And if I do, it will be something to go with. (and if it is data starting from say 10 years ago, it will be better than data from the past two years)

everett1 · May 2, 2012, 8:37am

These universities should never even inquire.

Ethnicity has no bearing on diversity in any way. Just the opposite. I can guess the political party of a black or Jewish student with 95-100% certainty, without asking one single question!!!

Sex of applicant obviously brings diversity; substantial diversity as opposed to manifestly superficial color mixing.

Diversity?

What do college administrators do when a Conservative speaker is scheduled and the entire student body vetoes, while insisting the President of Iran be welcomed. Well, censorship of the “non diverse” conservative is OK if the students say so, while the diverse dictator who appeals to a lockstep (opposite of diverse)campus is good?

There are other reasons universities shouldnt be in the superficial diversity business, but not enough space here to discuss.

Gotcha; all a y’all.

tokenadult · May 3, 2012, 11:40am

As a reminder, as I have time to rewrite the FAQ posts off-list, with updates, I’ll be replacing this FAQ and discussion thread on the ever contentious issue we are discussing here with a new-and-improved FAQ and discussion thread, updated for the current admission season.

StitchInTime · May 3, 2012, 12:38pm

Via the Boston Globe: [Directories</a> identified Warren as minority](<a href=“http://articles.boston.com/2012-04-30/news/31500872_1_law-professors-minority-status-elizabeth-warren]Directories”>http://articles.boston.com/2012-04-30/news/31500872_1_law-professors-minority-status-elizabeth-warren):

US Senate candidate Elizabeth Warren - who said Friday she didn’t realize Harvard Law School had been promoting her as a Native American faculty member in the 1990s - was listed as a minority professor in American law school directories for nine years before she landed at Harvard, documents show.

The Association of American Law Schools desk book, a directory of law professors from participating schools, includes Warren among the minority law professors listed, beginning in 1986 and continuing through 1995. The years include time she spent teaching at the University of Texas and the University of Pennsylvania, before she joined the faculty at Harvard Law…

…Warren’s unexpected minority status sparked controversy last week, when the Boston Herald reported that the school had named her a minority professor in the 1990s at a time when the campus was facing criticism about preponderance of white men on the faculty.

In a 1996 article, the Harvard Crimson quoted a Harvard Law School spokesman saying that the faculty of 71 included one Native American - Warren - in addition to a few black and Hispanic professors and 11 women.

The claim was repeated in a later Crimson story that called Warren the first woman with a minority background to receive tenure…

performersmom · May 3, 2012, 12:39pm

Thanks for the hard work, tokenadult- much appreciated.

Findmoreinfo · May 5, 2012, 3:42pm

I’ve tried to get official data and was told to use official data online for target universities.

Since Princeton and Harvard are two that are sued for discrimination in admissions I tried these two first and then some other top colleges. Didn’t find much for Harvard.
Princeton:
Percent of Asians among admitted students: (I compiled them from 11 documents. All are free to find info. for other races.)
2001-2002 12.5%
2002-2003 13.4%
2003-2004 13.2%
2004-2005 12.7%
2005-2006 13.3%
2006-2007 14.4%
2007-2008 14.9%
2008-2009 16.5%
2009-2010 17.5%
2010-2011 18%
2011-2012 18.7%

One may probably guess which year Princeton started to get complains and when the lawsuits began.

The most COMPLETE information based on race is Cornell’s own university freshman survey. (this one is for 2003)
<a href=“http://dpb.cornell.edu/documents/1000075.pdf[/url]”>http://dpb.cornell.edu/documents/1000075.pdf</a>
One can see details in qualifications of ethnicity groups and how the freshman students respond to abolishing Affirmative Action while they were in school. Also it is obvious about Asian’s interests in Medicine/Engineering and less interests in Law, like I indicated in an older post. This will take time for anyone to digest.

Asian Americans need to keep up the fight to stop affirmative action in college admissions. Princeton showed improvement after lawsuits since they need to provide data to the court. They may slip back to the old way once lawsuits are done.

siserune · May 9, 2012, 11:15pm

We’ve seen this movie before. If you want to know how it ends, one place to look is the original 2002 edition of the Revealed Preferences college ranking study by Avery, Hoxby, Metrick, and Glickman. They had an unusually detailed and well vetted data set on several thousand applicants with high class rank at top 500 high schools in the USA in the 1999-2000 admission year. In addition to a richer data set than the Espenshade 2009 study, their method was better: they used the applicant decisions (of which college to attend) and college decisions (complete admit/deny result for all applicants) to calibrate each other, estimating the desirability of the colleges and of the individual applicants. This controls for the effect of additional material outside the study, such as interviews, letters of recommendation, essays, intangibles, etc, by using the collection of all college admission results to estimate each applicant’s desirability.

The result? A substantial positive statistical effect of being Asian on an applicant’s desirability to colleges, controlling for academic credentials, parents’ income and education, gender, geography, in-state residence, legacy status, and many other variables known from the data set, which as I mentioned was extremely rich by the standards of admission studies. That’s among elite applicants to a mostly high-ranked set of colleges, with schools like the Ivy League, MIT, Stanford, Duke, Caltech and upper LACs drawing the lion’s share of applications.

The Revealed Preferences authors called the Asian result “surprising” given the idea that Proposition 209 at the UCs benefited Asians — just as the studies that find Asian academic underperformance after admission (e.g., at Berkeley, at Duke, and in Espenshade’s sample) invariably call the negative Asian academic effect a “surprise”. If anyone had paid closer attention to the UC data, or considered the possibility that striving implies both success and underperformance, there would have been fewer surprises.

UC Statfinder has searchable data on all UCs and you can duly check the non-effects of Proposition 209 on Berkeley admission at a gross level, such as the ratio of white and Asian admission rates staying the same from 1997-8. This is the first sign that one should look further.

Then there are the Center For Equal Opportunity studies (authors were Lerner and Nagai, see ceousa.org) on UC Berkeley and UCSD admission data. The first study was a simulation of what race-blind admission at UCB would have looked like, based on a sample of applications a few years before Prop 209. Result: fewer Asians, more whites, if admission were done purely on an academic index derived from SAT + GPA. The UCSD study was based on application + admission data for UCSD and also found Asians had been admitted at higher rates compared to whites (I don’t recall what controls were used, if any).

A more recent, large-scale and direct test of the UC Asian discrimination theory was performed in a 2012 paper by Antonovic and Backes, “Were Minority Students Discouraged From Applying to University of California Campuses After the Affirmative Action Ban?”. Table 3 estimates the effect of race, math SAT, verbal SAT, GPA, income level, and parental education on admission probability to the UCs before and after Prop 209. Of the 8 campuses of the UC, only the UC Davis results are consistent with the idea of Asian discrimination being present before Prop 209 and absent or greatly reduced afterward. At other campuses either 209 seemed to have a negative effect on Asian admission chances, or there was a pro-Asian effect before 209, or the numbers before and after 209 appear to be small statistical noise. Despite a huge sample size of around 100000 applications in the analysis of each school, of the 8 (before) + 8 (after) Asian effects measured, 3 are not statistically significant and 2 were not significant beyond the 95% confidence level. Nada, yet again.

Another study (Contreras et al) compared post-209 enrollment levels to numbers of California high school graduates and found no change in the white/Asian numbers at the UC campuses considered. I think it was UCLA, Irvine and one other.

To summarize: as far as multiple studies can determine, NOTHING HAPPENED as a result of the ban on race in California, where white vs Asian comparisons are concerned.

That’s in California. Other logistic regression analyses (i.e., Espenshade’s methodology) performed by the Center for Equal Opportunity on UVA, U Wisconsin Madison, and other state flagship campuses showed pro-Asian affirmative action, and analyses at other campuses such as U Michigan showed a lack of discrimination. Studies of state law schools showed stronger pro-Asian affirmative action, which is not surprising in light of the lower average LSATs (and bar exam pass rates) for Asians compared to whites.
Speaking of lower LSATs, had the current China Pride guerillas looked into that data (ref: lsac.org) there would have been no dispute about the Baylor spreadsheet, which looks completely plausible.

I’ll get to the Princeton data later, but for now, suffice it to say that the movie does not end particularly differently in that case.

HomeSchoolNewbe · May 10, 2012, 12:11am

@siserune^^
Thank You.

fabrizio · May 10, 2012, 11:14am

[Here</a> is the paper](<a href=“http://iacs5.ucsd.edu/~bbackes/pdf/bbackes_jmp.pdf]Here”>http://iacs5.ucsd.edu/~bbackes/pdf/bbackes_jmp.pdf) siserune referred to. Table 3 is on page 39 of the PDF.

To follow siserune’s first statement, the relevant coefficients are the -.05 (intersection of Asian and UC D), indicating that relative to whites, Asians were less likely to be admitted to UCD before Proposition 209, and the .04 (intersection of Ban*Asian and UC D), indicating that relative to whites, Asians were more likely to be admitted to UCD after Proposition 209, all else equal. I do not dispute siserune’s first description.

siserune’s second statement is a bit more problematic. He states the results can be grouped into three categories: (1) Proposition 209 hurt Asians, (2) Asians benefited before Proposition 209, or (3) the coefficients are statistically insignificant from zero.

The only school that fits into category one is UC Irvine; note the -.02 coefficient (intersection of BanAsian and UC Irvine). Three schools – UCLA, UC SD, and UC SB – had positive coefficients on Asian, indicating that relative to whites, Asians were more likely to be admitted before Proposition 209. And the remaining three schools can be put into the third category, since the coefficients on BanAsian are statistically insignificant.

Wait, the remaining three? I only listed seven schools in the previous paragraph, but the authors examined eight. Which one was missing? Berkeley. It does not fit into any of siserune’s three categories. The coefficient on Ban*Asian is positive, so it’s not in category one. The coefficient on Asian is negative, so it’s not in category two. And both coefficients are significantly different from zero at the five percent level or better, so it’s not in category three.

To quote Rick Perry on behalf of siserune, “Oops.”

fabrizio · May 10, 2012, 11:35am

siserune was pretty careful with his words here; he referred to "[the</a> original 2002 edition](<a href="http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.197.8685]the">http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.197.8685)" of the Revealed Preferences college ranking study by Avery, Hoxby, Metrick, and Glickman. Indeed, on page 30, the authors write, “Being Asian is a mild advantage in admissions. (This is interesting, because there are anecdotes that suggests that being Asian is a disadvantage in admissions at California’s public universities. Evidently these anecdotes are just anecdotes or whatever systematic phenomenon exists is specific to the University of California.” The table itself is on page 45, and a 95% confidence interval for the coefficient does not contain zero.

But note how siserune said “original 2002 edition.” That ought to suggest that there were later versions of the paper, and indeed that is the case. One can verify that a [2004</a> version of the paper](<a href=“http://theunbrokenwindow.com/Higher%20Ed/Higher%20Ed%20Course/Prestige%20and%20Rankings/Hoxby%20Quality%20Measure.pdf]2004”>http://theunbrokenwindow.com/Higher%20Ed/Higher%20Ed%20Course/Prestige%20and%20Rankings/Hoxby%20Quality%20Measure.pdf) no longer contains any reference to “price variables.” The most recent version of the paper is from [2005[/url</a>], and it too does not contain any discussion of “price variables.”

Why was it removed? Only the authors know. The point I want to make here is that this is yet another case of siserune spinning a paper’s results and in the process ignoring what the paper’s actual contribution was. ([url=<a href=“http://talk.collegeconfidential.com/college-admissions/1228264-race-college-admission-faq-discussion-9-a-24.html#post13561900]The”>http://talk.collegeconfidential.com/college-admissions/1228264-race-college-admission-faq-discussion-9-a-24.html#post13561900]The</a> first instance in this thread](<a href=“http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105]2005[/url”>http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105) was when siserune described the results of “the Duke study.”) The Avery et al. paper is about using fundamental microeconomic theory to produce a ranking of colleges. It has nothing to do with his war against “China Pride guerillas [sic].”

texaspg · May 10, 2012, 11:55am

findmoreinfo - I will give you some data points on Asians that I know locally with a 2400 or 36

2400 - not sure of ivies and whether applied - got into Rice, Berkeley etc.
2400 - got into Stanford (not sure where else the person got in)
36 - got into Harvard and MIT and few other well known top 20 schools. 
36 - got into Stanford
36 - Got into one Ivy
36 - got into a few top 20s but no ivies

In most of these cases, it mattered how they did in school and what else they did (ECs, volunteer work etc).

In many cases, Asians who were not perfect in their scores but presented much better resumes overall did a lot better in their admission cycle.

OTOH, I know of several Asians who went to top private schools, did extremely well with test scores, have outstanding resumes (well prepared for college app process since 6th grade or earlier - ECs, top summer programs, research etc for three or more years), who pretty much got into several Ivies, Stanford, MIT etc.

performersmom · May 10, 2012, 11:55am

I know I will get in trouble for my lack of statistical support and references, but I will try to express this simply, as this issue relates to hard numbers:

How can and should we be measuring unfair admissions practices by race/ethnicity, when different groups have different average stats (not mention the unmeasurable subjective differences) in their applications?
Is it unfair when the percentage accepted for each group varies from the the percentage in the general population (of the state or the nation)? from the representation of that group in the applicant pool?
Or is it unfair when the percentage accepted does not reflect the superiority of inferiority of the stats and other qualifications of the INDIViDUAL applicants? And what happens when the average qualifications for one group are higher than those of another group?

Nationally, Asians are being accepted at a much higher rate than their representation in the general population. Are they being accepted in line with their percentage of applications? And please remind us of the same for California.

Asian representation at top colleges appears to be higher than in the general population. But is it high enough to be “fair”, i.e. did a lot of well-qualified Asians get turned down, too, to keep that percentage from getting even higher? Yes, the colleges can say they are over-accepting Asians in response to the quality of their apps, but is this true in terms of number of Asians applying? In terms of number of Asians who are really qualified? That is where the Asian anger is directed. This is why I think Siserune’s analysis is a bit fallacious.

Do Asians want to be considered as individuals or as a members of an ethnic or racial group? what about URM’s?
And, ironically, Asians are URM’s in the national population, and many do have less opportunity than the average citizen…

How important and effective for society is RACIAL BALANCE in education? SOCIAL ENGINEERING in education? That is, of course, one underlying question.
Obviously, a diverse student population is another tenet held by these colleges as providing the optimal learning environment, and is a big factor in these results.
But the colleges need to admit that these are the reasons for the feelling that many qualified Asians are not admitted.

siserune · May 10, 2012, 4:01pm

Explicit agreement that my first quoted statement and at least 7/8 of the second are correct. OK, and then…

Uh-oh. A sure sign that another E-Z Read skimming and flailing demonstration is coming next.

Here my correct statement about “small statistical noise” is reworked into an incorrect one about statistical significance. Statistical insignificance refers only to one particular type of noise. My remark was mainly about other sources of noise that are larger in this case. Though not as large as one current, local source…

Think harder. At Berkeley the estimated Asian effect went from 1% lower admission probability before Prop209 to +1% higher afterward, where we also know that second number is the result of a race-blinded admission procedure. This should tell you that a 1 percent effect is noise from the limitations of the data and the model, and there is no reason that should not be true for both the Before and the After estimates.

After 209, personally identifying data fields are removed from the application before the UC admission readers see it, and use of race in the decision is illegal. Despite this, the estimated post-209 effects on admissions probability from being Asian at the UC schools were nonzero: +1%, +4%, +4%, -1%, -4%, +1%, -2%, -2%. At six of the schools there is an Asian effect larger than 1 percent under race-blinded admission, and the signs vary. At the low resolution of the model and the data set, 1-2 percent effects in either direction are small, nearly meaningless, and can come from the choice of model, changes in the overall UC selectivity level and applicant demographics, different availability of majors over the years, and many other factors not necessarily related to prop 209.

Which is to say that most of the 16 before and after Asian effects seen in the regression are (when considered individually) meaningless, as far as the study can show, and are not a matter for fine parsing. The parsing was necessary only to reach that conclusion.

As a comparison, the opposite is true for the underrepresented minorities. At 7 of 8 UC schools there was a giant decline in the magnitude of the URM coefficient (or it went from positive to negative), and a smaller decline at the 8th UC. This very clearly fits with the standard Prop 209 narrative. For Asians only 1 of the 8 schools, UC Davis, has regression results supporting the narrative, and the magnitude of the pre-209 Asian effect is almost the same (-5% versus -4% to +4%) as what is seen at other UCs post-209 sans race, so even there it is not clear that the 5 percent disadvantage was a racial one.

fabrizio · May 10, 2012, 5:20pm

Do I think your repeated references to “China Pride guerillas [sic]” and “China Pride warriors” are childish? Yes. Does that mean I always disagree with you? No.

Oh this is funny. You maintain that “NOTHING HAPPENED as a result of the ban on race in California, where white vs Asian comparisons are concerned” (direct quote). So if Antonovics and Backes show that the estimated change in admissions probability for Asians relative to whites after Proposition 209 all else equal is statistically nonzero at five out of eight UC schools and positive in four of the five schools, their results must be “statistical noise”!

There is no rule anywhere that “1-2 percent effects…are small [and] nearly meaningless.” There’s a reason we compute standard errors instead of solely eyeballing the coefficients and asking ourselves whether the results are “statistical noise.”

siserune · May 12, 2012, 4:37am

(fabrizio : )

A translation to ordinary English, as the crickets continue chirping:

“The bombs were aimed accurately, and exploded at the intended location. Instead of acknowledging the damage, or else credibly demonstrating that the targets were only superficially harmed, we will inspire confidence with a week of nonstop military parades.”

Carry on.

The “mild” advantage can be converted to SAT points on the Espenshade-Chung (2004) scale by comparing the Black, or Hispanic, or Legacy effects, or the Asian odds ratios, between the two studies. Result: 20 to 45 points of “chill” Asian bonus.

As always, I’ll let fab choose the terminology and decide whether these mild, groovy Asian coefficients are “substantial”, but I will point out that he is currently arguing for effects equivalent to 9 and 18 SAT points at UC Berkeley as evidence for the UC discrimination theory, and that 50 points is apparently the gold standard given the publicity surrounding the E&C study.

That’s very interesting, because I was under the impression that price variables are absolutely central to the 2004 and 2005 papers and are the very same ones from the 2002 article. Therefore, at least one of us must be making very confident assertions that are dead wrong. Could it be that someone in CC is asserting knowledge of a paper based on a 30 second EZ-skim word search? Who, who, would ever do such a thing?

"Race" in College Admission FAQ & Discussion 9

CONNECT WITH US