Do Elite Colleges Discriminate Against Asian Students?

Applying to College
741

<p>

</p>

<p>Well, no. Your calculation, despite its small sample size and “denominator of 6” problems, confirms the pattern I described, with the white to Asian (and especially the Jewish-to-Asian) ratio growing at each stage of selectivity.</p>

<p>Your numbers for East Asians: USAMO qualifiers, 53% (933/1759) ; selection camp qualifiers, 44.8% (74/165) ; IMO team 41.7% (10/24). The last figure is a slight overestimate and drops to 30-40 percent if you include more years. </p>

<p>The numbers from the study (link below) are as follows. For years 1988-2007, total number of students followed by percentage. In their table, “Asian” includes Indians, not only East Asians.</p>

<p>USAMO-preparatory summer camp (AMSP) : 83 percent Asian (not part of the IMO competition)
IMO Team selection camp. 216 Asian (42%) ~56 Jewish (11%) ~209 (40%)
US IMO team. 40 Asian (33%) ~25 Jewish (21%) ~50 White (42%)</p>

<p>Note the doubled Jewish representation, the rising non-Asian representation, and the slowly but surely declining Asian percentages, as the selectivity is increased. The numbers in my earlier post were quite realistic: in recent years non-Asians are about 35-40 percent at the USAMO qualifier stage and 60-65 percent of the IMO team, which is a rather large “white to Asian odds ratio” (between 2.25-to-1 and 3.45-to-1). Discrimination! </p>

<p>Your only remaining point may be that Indians do not do better than East Asians in the USAMO-to-IMO selection. Even if that were true, the picture after high school seems to show increased Indian representation compared to East Asians, and even if that were false there is plenty of evidence of East Asian underrepresentation (relative to numbers in college, grad school and professions) at the higher levels. As I mentioned, the Wikipedia has a large list of CS grad student and professor prizes, and if you wish to spend another hour researching those, or the list of winners or Harvard and MIT math prizes, or the national undergrad math awards, it would be fun to review your data, because it’s really quite interesting, and unlike you, I gathered the numbers before making any comments. Enjoy!</p>

<p><a href=“http://www.ams.org/notices/200810/fea-gallian.pdf[/url]”>http://www.ams.org/notices/200810/fea-gallian.pdf&lt;/a&gt;&lt;/p&gt;

742

<p><em>Hands clapping</em></p>

<p>Data is data. siserune has provided a link to the paper, and to the extent that “East Asian representation actually declines as one goes to higher levels of selection,” he is right.</p>

<p>Based on #716, this sub-discussion resulted from Hunt’s comment that the only implication of “Asians’ [possibly] perform[ing] better than whites on STEM majors…is that a higher precentage of the top group will be Asians than their representation in the population at large” (emphasis added). As shown on page six of the paper, Asians make up 10% of all tenured professors at the top five American math departments and 13% of all junior faculty. Given 4.5% representation in the population, Hunt is clearly right: a higher percentage of the top group will be Asians than their representation in the population at large.</p>

<p>To answer siserune’s tongue-in-cheek claims of “discrimination,” I find it highly unlikely that the IMO selection process practices racial discrimination of any sort. The U.S. IMO team cares about winning, not about how many faces of [insert racial classification here] are on the team.</p>

<p>Now, if siserune could just produce the links to the papers he mentioned several pages back, we can have more productive discussion!</p>

743

<p>siserune:

[QUOTE=Keilexandra]

Regarding USAMO, why is it strange that Asian representation would decrease as difficulty increased? Asians are overrepresented at the beginning levels because the pool of qualified talent is deeper–but as you narrow up, even a less qualified group will have outstanding talent at the top, and overrepresented groups will become less overrepresented. The highest level of USAMO is basically measuring mathematical genius–since mathematicians are born, not made–and there’s no reason to think that true math talent is distributed other than equally across the races.

[/quote]
</p>

<p>10char</p>

744

<p>As long as admissions criteria in elite colleges are unclear, it is impossible to conclude that admissions discriminates. Given the statistics, it is only fair to conclude that some of the criteria used by admissions are unfavorable for Asian students. </p>

<p>I don’t think that including these ‘unfavorable aspects’ is part of a master plan set up to discriminate Asian students (unlike past plans to discriminate Jewish students). The reason is much simpler. Getting into an elite college in the US is not a reward for an A+ past; it is recognition of (academic) potential. However, HS stats and even EA’s tell more about working habits and memorization skills than about potential. Concerning ‘potential’ a small remark on a recommendation, a few words in an essay, something said during an interview can make or break a student’s chance. As there are many Asian students with outstanding HS stats and EA’s, there will also be many Asian students who do not benefit from this way of working. </p>

<p>I wholeheartedly agree with StillGreen (#740) that what needs to be done is “to improve transparency in the selection process.” However, colleges will not be overly eager unless they can get hold of the ‘potential’ information in a more structured way. Having entrance exams could be an option, but this would upset the whole admissions system too much. Maybe a more sophisticated test could do the trick, a test that does not measure things one can prep for (like the SAT), but a test that measures the ability to think independently, critically, analytically, creatively, i.e. a test that would add information to the HS stats (and to AP’s and SAT II’s). Such a test could give talented underachieving HS students a chance to prove themselves in a college environment that fits their needs. Of course the less talented HS students having worked day and night to become valedictorian will be less enthusiastic.</p>

745

<p>

</p>

<p>Before I answer this post, let me make a comment about your other post, #732. IMHO, it is the finest piece of analysis on this topic I have seen on CC. Nothing else comes close. It is a ringing endorsement for a math-focused liberal education that I have always advocated. I wish more students here on CC can see it, and more folks here can understand it. LOL</p>

<p>Now, let’s get back on topic. ;)</p>

<p>My position is that the elites have failed America because they have always been the handmaiden of the ruling class. It is not so much their definition of merit has swayed from academics as much as the fact that they don’t seem to have a definition of merit at all. </p>

<p>If we don’t even have a definition, then how can we be in a position to modify the process when the “products” (think Iraq, Wall Street etc.) are clearly unsatisfactory? My working hypothesis is that the elites are only interested in protecting their brand by catering to the ruling class on one hand and project the image of a meritocracy to the public on the other. The government can force them to make a choice, cater to the rich and powerful and do without public funding, or take public research money and turn the schools into true meritocracies. (I am not holding my breath here, BTW).</p>

<p>My other point is that there are those who seem to think the Adcoms have some kind of secret method of detecting future potential. I am unaware of any work in the social sciences that make it possible to do so at this time. If such a method exist, ones who have it would not be working on college admission, elite or otherwise; they would probably be working with the political parties picking their next presidential candidate, the President picking his administration, Wall Street with their next batch of bankers, Fortune 500 companies with their senior management, or NFL teams with their draft choices.</p>

746

<p>can someone please just close this thread…</p>

747

<p>Re: post #741</p>

<p>Let me address the numbers and statistics first.
Below is a slightly modified table in post #738, with the percentages calculated and included.</p>

<p>YEAR _____ USAMO _____ MOSP _____ WINNERS(TOP 12) ____ IMO
2006 ___ 204/404 (50%)<strong><em>21/54 (39%)</em></strong> 8/12 (75%)________ 3/6 (50%)
2007 ___ 226/459 (49%)<strong><em>24/55 (44%)</em></strong> 5/12 (42%)________ 3/6 (50%)
2008 ___ 258/465 (55%)<strong><em>29/56 (52%)</em></strong> 4/12 (33%)________ 1/6 (17%)
2009 ___ 245/431 (57%)______ n.a. ________ 6/12 (50%)________ 3/6 (50%)</p>

<p>Consider the 2006 data, if we randomly take 54 students from the 404 USAMO qualifiers to go to MOSP, what is the probability that the percentage of Asians among the chosen will be less than 39% or above 61% (away from the population mean of 50% using two tailed analysis)? Answer: 9.1%. And if we randomly chose 12 members from the 404 qualifiers to be named winners, what is the probability that the percentage of Asians will be less than 25% or above 75%? Answer: 25%. (If you know your combinatorics or statistics, you can try it out as an exercise.) Neither of the two sub-samples can be considered statistically different the original population beyond random variations by strict statistical standards.</p>

<p>Same exercise can be done on all other data points, and you will find that virtually all the variations are not statistically significant.</p>

<p>Though neither is statistically significant, siserune, you seem to be very fond of the lower asian numbers but somehow don’t like higher figure, because you think it “is a slight overestimate” and you would like to “include more years” to show a lower figure. Sorry, there is a name for that kind of practice – scientific fraud. As a matter of fact, that is what you have been practicing here.</p>

<p>Your whole exercise on this and some other threads seems to be an effort to justify the notion that Asians are somehow intellectually inferior to some super race/ethnicity. And you have a tendency to make erroneous assumption about what other people have and have not said. You stated:

I never made such a point at all. If you want to know my opinion, I can tell you that I believe the representation of south Asians in academic competitions as well admission to elite colleges will grow the fastest in the next decade because of the demographics.</p>

<p>You used Andreescu et al. paper as the source of your following statements:

</p>

<p>This is not what the paper said.
Your number for AMSP is for the year 2007 only, not for 1988-2007.
The IMO team member number is for 1988-2007 (and you only counted males). But the MOSP member number is only for 1998-2007 not for 1988-2007 (again, you only counted males).</p>

<p>The directly comparing a 10 year data set with a 20 year data set but claiming otherwise, to put it generously, is scientifically dishonest. If you truly want an apple to apple comparison, you can easily find the names of 1998-2007 IMO members. In those 10 years, Asian IMO members competed a total of 25 person-times, which was 40% of the total.</p>

<p>If you have something to show everyone, please go ahead. What I have seen from you so far are just a vivid imagination and some repugnant ideological believes.</p>

748

<p>siserune, I think that you really missed the bigger picture. Compare the IMO Gold Medalists from China and India starting from the 1980s when China started to participate, the results are very clear. If the Chinese can not do well here, that may be simply because they do have the “best” math kids here. That is all. Also, only the Gold Medalists count.</p>

749

<p>I am not a subscriber to theories that only a very small percentage of the population, a particular gender, or only some specific superior ethnic groups are endowed with the abilities to solve some of the harder math problems. I believe that the USAMO and even IMO level intellectual capacity can be achieved by most youngsters of any race and gender, given the right environment, nurture, opportunity, and with persistence and hard work. And I find the suggestion ([post #716](<a href=“http://talk.collegeconfidential.com/1063412783-post716.html”>http://talk.collegeconfidential.com/1063412783-post716.html&lt;/a&gt;)) that only some super race/ethnicity is capable of performing in certain high-level “IQ-taxing fields” repugnant. I thought that the concept of a super race had died a long, long time ago with Adolf Hitler.</p>

<p>The Andreescu et al. paper, referred to in post #761, is neither very interesting nor informative. It gave a snapshot of the profiles of preteens to college students involved in math competitions for the period of 1988-2007. But what the authors and most readers don’t realize is that the picture was very much skewed by the changing demographics over time. The authors did not acknowledge and were probably completely unaware of the history, politics, and the dynamic demographics of immigration. The picture of 1988 is very different from the one of 2007. </p>

<p>siserune did not like some of the data in my table (post #738) from the last 4 years, because s/he thinks they overestimate Asian USAMO winners and IMO team members. S/he would like to go back a few more years to show that fewer Asians are capable of such “IQ-taxing” tasks. S/he is probably right that if we go further back to 1960s or 1970s, you will find few if any Asians even competing at USAMO. And if we were to go back 25 years, there were probably not that many competitive asian pupils at National mathcounts either. But it does not say anything about the superiority or inferiority of any ethnic group but rather reflects the social, cultural, and the demographic changes over the last 20-30 years. And hopefully over time, this super race nonsense will also be buried once for all.</p>

750

<p>telling yourself that no they dont discriminate against asian kids would be lying to yourself. Ethnical factor is definitely taken into account no matter what the colleges states. If completely fair, MIT would have 100% Asian population, and can they do that? Though I will say that discrimination is not the right term to use. Rather, it’s just a way trying to balance things out. what can you say about that? nothing in the world is 100% fair.</p>

751

<p>Re: post #745 “My other point is that there are those who seem to think the Adcoms have some kind of secret method of detecting future potential. I am unaware of any work in the social sciences that make it possible to do so at this time.”</p>

<p>Canuckguy: Correct! It does not work! It more or less used to work before things became so competitive. In the meantime the number of applicants has increased and the number of applications per student has increased, while the admissions process remained the same. No way that admissions can do a fair job anymore. Students prep for SAT’s (this type of test is not really fit to test potential anyhow), HS teachers or advisors do not have the time to write detailed recommendations anymore because there are too many to write, grade inflation is out of proportion, parents or professionals write or over-edit application essays, professional companies are hired to guide students through the application process,and so on, and so on. This ridiculous situation plus the fact that admissions does not have secret methods of detecting potential is making it virtually impossible for admissions to select the applicants they are really looking for. I am convinced that elite colleges are still much more interested in students with (academic) potential than in students who have gone overboard in HS (HS stats tell more about working habits and memorization skills than about potential). Elite colleges are interested in getting future gold medallists on their list of alumni. (Math) Gold won during a HS past is far less relevant. This applies to both Asians and non-Asians and I do not think that race is an issue. However, there will be an increase in unfair decisions as the present situation is making it virtually impossible for admissions to select the ‘potential’ they are really looking for. Lacking reliable information may result in a somewhat sophisticated form of a lottery system. Not necessarily large groups will benefit from this kind of system. And I have no doubt that the group of students with outstanding HS stats is not only a large group, but that is also a very Asian group.</p>

752

<p>

Well, then doesn’t the heavy overrepresentation of Asians at the earlier stages of these competitions suggest concentration of Asian interest in math out of proportion to innate ability?

Why would you think this? You think the students with the best stats are all Asians? That simply isn’t the case.</p>

753

<p>

</p>

<p>Do you mind to provide some support for this claim?</p>

754

<p>re: post 750 “telling yourself that no they dont discriminate against asian kids would be lying to yourself. Ethnical factor is definitely taken into account no matter what the colleges states. If completely fair, MIT would have 100% Asian population, and can they do that?”</p>

<p>Kalynguazi: Your statement is arrogant, discriminative, incorrect and assuming something that may not be shared by everyone. There are plenty of non-Asian students having superb HS stats and EA’s. Excluding them entirely, even as a possibility, is the kind of prejudice that is usually called discrimination. You did not mention the criteria that would make MIT’s admissions system ‘completely fair’. Not knowing your definition I can only guess what it stands for. Assuming that its quality is in line with your other remarks, MIT has some very good reasons to ignore your definition of ‘completely fair’. Never forget that MIT’s college program does not just focus on the sciences and that it offers highly respected majors in non-sciences as well.</p>

755

<p>Oops! I think my post way back in #673 where I attributed an observation to Rachel Toor may actually have appeared in Michelle Hernandez’s “A Is For Admission”. I remember that Hernandez, in a fictionalized committee discussion about a student not to be admitted, had an admissions officer comment to the effect that “I don’t get the love of learning here. Just another dutiful Asian daughter”.
I think I recalled the original observation as Toor’s because I generally dislike Toor and like Hernandez. But it really doesn’t matter.
What strikes me about this thread is how heated it is, particularly in the smarmy way in which academics go about things.
Never likely to be a smoking gun on this issue. Watergate took massive resources and ultimately great luck to prove (tape recordings). Even if you wired admissions folks, they are unlikely to be as unguarded as Hernandez was 10 years ago. On the numbers, there are countless arguments to explain them away- building a diverse class, legacies, athletes, need those classics majors, etc., etc.
Time may be the ultimate healer for those who have patience. As a meaningful percentage of alums are Asian-Americans, the scale will balance. My pet theory is that Jim Kim was drawn particularly to Dartmouth because it is in some ways the most American of the Ivies. I have to believe that in the end, talent will win out.</p>

756

<p>

Well, it’s not a claim, but rather a speculation, but here goes: if Asians are highly overrepresented at the early stages of a competition based on an objective test, but are not overrepresented at the later stages–also based on an object test, what does that suggest? To me, it suggests that a lot of Asians are interested in trying to participate in the competition. What does it mean to you?</p>

<p>I’m a little perplexed by the reaction to the hypothesis that Asians are disadvantaged by concentration in specific major areas and ECs (as well as geographic concentration). It’s not an insult; if it’s true, it’s simply an artifact of cultural norms. If you don’t think it’s true, why don’t you?</p>

757

<p>Asians are also disadvantaged by excessive chasing-of-the-same-schools because of a cultural mindset that makes them think that upper-middle-class-economic-success-in-America is only achievable at those handful of schools and that the would-be engineer who doesn’t make it to MIT or Caltech is going to be flipping burgers the rest of his life. That’s a cultural artifact too.</p>

758

<p>“if Asians are highly overrepresented at the early stages of a competition based on an objective test, but are not overrepresented at the later stages–also based on an object test, what does that suggest?” </p>

<p>This NCL’s table: </p>

<p>YEAR _____ USAMO _____ MOSP _____ WINNERS(TOP 12) ____ IMO
2006 ___ 204/404 (50%)<strong><em>21/54 (39%)</em></strong> 8/12 (75%)________ 3/6 (50%)
2007 ___ 226/459 (49%)<strong><em>24/55 (44%)</em></strong> 5/12 (42%)________ 3/6 (50%)
2008 ___ 258/465 (55%)<strong><em>29/56 (52%)</em></strong> 4/12 (33%)________ 1/6 (17%)
2009 ___ 245/431 (57%)______ n.a. ________ 6/12 (50%)________ 3/6 (50%)</p>

<p>Hunt: You are not refuting NCL’s table, but claim a different conclusion. In 3 of the last four years, East Asians appear quite consistently making up more or less 50% of the rosters from USAMO qualifiers to IMO members. This analysis is discounted somewhat by excluding South Asians and Southeast Asian members. If all asians are included, IMO teams in the last 4 years will have at least 50% or more asians each year, and that is quite consistent with the asian shared number of USAMO qualifiers.</p>

759

<p>

</p>

<p>That’s a disadvantage only if the admissions system is structured such that Asian applicants compete only against other Asian applicants. Unless you have evidence that that is the case, “excessive chasing-of-the-same-schools” is not a disadvantage.</p>

<p>Your claim is a major case of “behold[ing] thou the mote that is in thy brother’s eye, but consider[ing] not the beam that is in thine own eye.” Tim Wise, an outspoken advocate of positive discrimination and racial preferences, has claimed that one of the reasons why race-neutral admissions is unacceptable is its creation of what he considers a two-tiered system, where whites and Asians are at top schools (e.g. Berkeley) and “underrepresented” minorities are at bottom schools (e.g. CSU East Bay, whose freshman class is [over</a> 40% “underrepresented”](<a href=“College Search - BigFuture | College Board”>College Search - BigFuture | College Board)). Clearly, he believes that unless “underrepresented” minorities graduate from top schools, they’ll “be flipping burgers.”</p>

760

<p>I’ve appreciated the close reading of the thread-opening study presentation provided by several participants here, and the back-and-forth in general. But is there anything else data-based to say in this thread, or is it time to close it?</p>