<p>
</p>
<p>You could think of yourself as asserting your common humanity with all other human beings, if that helps your conscience. I’m not actually aware of any college in the land that matches whatever ethnic categories it considers in the admission process exactly to the varied categories that applicants think about in their own personal lives.</p>
<p>
</p>
<p>How would they know that? Where is there evidence about this? (On my part, I can well believe that a majority of most subsets of college applicants would be white, because a majority of most college applicants are white, but it would be an issue of fact, to be determined empirically, which ethnicities are most associated with declining to answer questions about ethnicity.)</p>
<p>
</p>
<p>Historically and mathematically correct.</p>
<p>^Read my post again. I didn’t say there wasn’t discrimination. Every race discriminates against every other race to a degree. I just don’t think it’s a big deal. </p>
<p>I attend a public school that has a large Asian population. I estimate the top 10 percent of my class is 90 percent Asian. I do not know everyone, but I am friends with pretty much everyone who is in the top 10 percent. None – and I repeat this for emphasis, none! – of my Asian friends failed to get accepted to several great schools. My friends will be attending six Ivy League schools and schools like Chicago, Duke, Georgetown, Cal Tech, Pomona, CMC, UCLA, Berkeley, UCSD, Cornell, Williams, etc. To my knowledge, there wasn’t a single Asian friend who got skunked or who was forced to attend a safety. </p>
<p>Now let me talk about another friend I have who is African American. She was the only student in my school accepted into Stanford this year. She also applied to and was accepted to a local Cal State school. Going to Stanford is going to give her an opportunity to escape her economic situation, assuming she works hard. Going to Cal State Los Angeles wouldn’t have opened many doors for her. My best friend (an Asian) was rejected by Stanford but was accepted to UCLA, UCB, and Dartmouth to name a few. She is going to go to Dartmouth. Token and Fab could argue that Stanford was wrong in accepting my African American friend over my Asian friend (and my African American did have much lower numbers, fewer ECs, etc.), but how was my Asian friend harmed? She is going to Dartmouth and will accomplish everything she wants to accomplish. Had my African American friend not been accepted to Stanford, who knows? </p>
<p>I’m no social scientist and I can only argue my personal experiences but even if Asians, on average, lose a few points on their SAT and GPA, so what? They are savvy enough to apply to many school and are getting into great schools nonetheless. The same is true for my white friends. Again, my eyes were opened on April 2nd when I saw how everyone eventually received good news from great schools. I can’t believe I worried about this issue for two years. What a waste of energy.</p>
<p>These statistics that show that Asians lose points remind me of the statistic that the average American family has 2.5 kids. It’s a nice number but wholly inconsequential to real life. </p>
<p>Be proud that you’re Asian. Don’t worry about these fearmongers. If you’re a top student, do a good job on your apps, write good essays, you will be accepted into great schools even if you self-identify.</p>
<p>^ I don’t argue about individual harms from college admission practices. Applicants protect themselves from harm in the college admission process by </p>
<p>1) preparing well for college, and </p>
<p>2) applying to more than one college. </p>
<p>Note, however, that whatever I think, courts have found examples of applicants being unfairly discriminated against in some past cases, and some college admission practices are plainly illegal, whether your ox or my ox is gored or not. </p>
<p>I do think there is warrant </p>
<p>[Group</a> Classification on National ID Cards - Jim Fussell - 15 Nov 2001](<a href=“http://www.preventgenocide.org/prevent/removing-facilitating-factors/IDcards/]Group”>Group Classification on National ID Cards - Jim Fussell - 15 Nov 2001) </p>
<p>to consider carefully, in any society, whether more ethnicity-consciousness as a matter of official data-gathering or less is desirable. It might be better for all of your friends, for you, and for me and my children for the United States to operate on as “color-blind” basis as possible. I say more about my personal history of thinking about this issue back in [post</a> #10](<a href=“http://talk.collegeconfidential.com/1064033898-post10.html]post”>http://talk.collegeconfidential.com/1064033898-post10.html) of this long thread.</p>
<p>
</p>
<p>My actual message was that it is egregious and transparent for one to persistently attempt to draw others (but never oneself) into costly (to them, but never to oneself) battles in the hope that this would advance one’s own agenda. </p>
<p>
</p>
<p>What I wrote included the words “even when groundless”. Filing complaints as a form of speculative fishing for a problem that might exist or not, and that someone else would like answers about, is very different from seeking to correct a known (or partially known) problem.</p>
<p>
</p>
<p>Espenshade’s data and models, as discussed at length in his 2009 book, indicated: </p>
<p>-Asians being favored over whites in admission in the lower- and working-class income/SES range.</p>
<p>-lower Asian performance at university, compared to whites (both in the raw data and with all other factors being equalized using regressions). This is a sign of higher admission rates for Asians than under a pure academic merit selection.</p>
<p>-higher Asian utilization of SAT coaching, summer schools and academic camps, and proxy-academic extracurricular activities (high school math team or similar). This upward-biases the SAT scores insofar as those are utilized as indicators of other factors, such as IQ.</p>
<p>
</p>
<p>Before I elect whether to act on your personal likes and dislikes it would be good to understand what, concretely, you claim is offensive. Getting A’s, dominating elementary school math competitions, and relatively intensive SAT preparation are examples of chasing the low-hanging fruit. On the other hand, if the (US, East) Asian representation is, for example, 50+ percent at some lower stages and 20-30 percent at higher stages of selectivity, that would also suggest some feasting on the more easily attained outcomes. I don’t think I commented on how much of this is, shall we say, the fruit of parental encouragement.</p>
<p>Hope Full, first of all, let me offer you my congratulations. You know as well as anyone that it’s not easy to garner admission to any of the schools you listed. But I must say that your post comes off as, “It didn’t happen to me or my friends, so while it might happen, it isn’t important.”</p>
<p>I reiterate the history of Jews at Harvard in the early twentieth century. Lowell set a maximum quota of 15% for Jewish students, which meant he was willing to live with “overrepresentation.” Moreover, that quota was always “in effect,” meaning that but for the quota, more Jewish students could have been admitted. I can only hope that the lucky students did not feel that anti-Semitism suddenly became unimportant simply because they got in.</p>
<p>siserune, my apologies, but I do not understand your “actual message.” Could you clarify using different language? I fail to see how Li did not draw himself into the “battle,” since he was the one who filed the civil rights complaint and opened himself to a torrent of abuse from angry people here at CC and elsewhere. Many of these angry users predicted that Li’s complaint would kill his future career. Sounds like a big personal cost to me, but I acknowledge that I do not understand your “actual message” as you wrote it.</p>
<p>Who is this “someone else” you refer to? Are you suggesting that Li himself did not want an answer to his own complaint? And what does “partially known” mean? This thread is in its seventh incarnation partially because so many people, rightly or wrongly, believe that there is “negative action” against Asian applicants in the admissions process. Prior to the creation of these threads, you would see the same questions asked over and over again come every admissions season. Please, please tell me what “partially known” means, if the behaviors I just described do not count.</p>
<p>Thanks for telling me about the “elements.” I guess I’ll have to read the whole book since you didn’t provide links or page numbers, which I also requested. No complaints, though I hope you aren’t like that SEIU guy I spoke to who redirected me to sections of the NLRA which supposedly “proved” that workers in Georgia have a “right” to their jobs.</p>
<p>What I claim is offensive? Which claim is that, siserune? Perhaps we should take a look at what you wrote in a previous version of this thread:</p>
<p>
</p>
<p>So what, siserune, I was supposed to assume that white, Jewish, and South Asian kids use “more sophisticated educational strategies” than East Asian kids? That these “more sophisticated educational strategies” were not the “fruit of parental encouragement”? I kindly request your clarification.</p>
<p>^ ^ The historical example is a good one, because it certainly seems odd to make blanket prejudicial statements (as were made about Jewish people in an earlier era in the United States) about some group of people simply because it is observed that many individuals in that group do something that results in high rates of college admission. One mature response to such an observation might be to ask, “What do they know that I don’t know?” Another mature response might be to point out that there is a lot of opportunity to attend college for most Americans, even if most Americans have essentially no opportunity to attend the most highly selective colleges, so that everyone can make good use of early adulthood to pursue higher education. Yet another mature response would be to observe that after all, in America, there are many interesting careers and many ways to contribute to civil society that don’t strictly require a college degree at all. </p>
<p>But it would be immature and illogical to say (as evidently was said about Jewish students two generations ago) that measures to keep ethnic numbers “representative” are fair simply because the successful college applicants have figured out how to “easily” attain college admission. Applicants differ in their goals, and they differ in the individual backgrounds they are born into when pursuing their goals. Overly Procrustean group categorization has the ill effect of distracting individuals in American society from appreciating the accomplishments of their neighbors and from seeking their own paths of achievement.</p>
<p>I don’t have the ERC book that siserune mentioned in hand, but Google Books provides a “limited preview,” and what I found is quite interesting.</p>
<p>Yes, siserune is indeed correct that there is “higher Asian utilization of summer schools and academic camps” than whites. (I couldn’t find anything about SAT coaching, but it might be there.) What siserune doesn’t say, however, is that there is also “higher black and Hispanic utilization of summer schools and academic camps” than whites! On page 47 of ERC, the authors write</p>
<p>
</p>
<p>Where’s the upward bias for blacks and Hispanics, siserune? My goodness, are you suggesting that SAT scores overpredict performance for these “underrepresented” minorities, too?</p>
<p>As for the “lower Asian performance at university, compared to whites,” on page 252, ERC write that “Even Asian students, long believed by some to be academic superstars, end up graduating with a class rank that is on average 10 percentage points below the class rank of similar whites.” Is siserune vindicated in his assertion? Yes, and no. No because there is an immediate footnote following that sentence which states, “It is not clear what contributes to underperformance by Asian students. Some of the standard explanations advanced for blacks do not seem to apply to Asians.” siserune’s “low hanging fruit” hypothesis is his and his alone.</p>
<p>I posted a link upthread that went dead after the government agency that posted it revised the document. I’m posting again because the federal government figures here may inform several aspects of the recent discussion. </p>
<p><a href=“http://nces.ed.gov/pubs2010/2010152rev.pdf[/url]”>http://nces.ed.gov/pubs2010/2010152rev.pdf</a></p>
<p>^If Token and Fab were speaking before a court, I would welcome and applaud their arguments, since a court is supposed to address macro-level problems in society. But you guys aren’t talking to judges. You’re talking to a bunch of high school sophomores and juniors. Many of those Asian students truly believe that they won’t get into a good schol because of their race. Asian students post on this board all of the time trying to come up with clever ways to not reveal their ethnicity. That’s terrible. They absorb these fears and start to think it’s bad to be Asian. I know this is true because I was one of those students who thought I was going to have a very difficult time getting into a good school because of my race. But that’s simply not true. My race may have hurt me with one or two schools, but as long as we apply to multiple schools, our race is a non-factor, at least that what it looks like from my perspective. So all I am saying is keep in mind who your audience is.</p>
<p>
</p>
<p>“It is egregious for person A to wage a campaign attempting to incite person(s) B to draw agency C, into battle against institution(s) D, so that this might advance the interests of person A.” </p>
<p>Relevant interests of A might include data disclosure about D, generating publicity about contested political subjects of interest to A, having some (real or perceived) influence on other people and institutions, being a spectator at a good fight, or any number of other hypothetical possibilities. It matters less what A’s specific motivation is, than the fact that A eggs on others into battles where all the cost and risk is borne by those others and not at all by A. It’s not all that different conceptually from the ethical position of an arms dealer who publishes war-mongering editorials. </p>
<p>In this case, the mysterious “A” does this using not A’s own resources, but the resources provided by another unwitting and uninvolved party, one that we might call “EE”.</p>
<p>Clear enough now? </p>
<p>
</p>
<p>I was referring not to the Li case, but to the persistent ■■■■■■■■ for its would-be successors.</p>
<p>
</p>
<p>Bad sampling technique. These last seven threads, as far as they concern the Asian/white admission comparisons, are driven by (and under the control of) one side of the discussion, which, feeling aggrieved, naturally tends to be louder. </p>
<p>There is also some information asymmetry: the media played up Espenshade’s “3-to-1” Asian discrimination odds ratio from one of his regressions but did not disclose the 100+ SAT point Asian underperformance found in his other regressions, or the lower class rank observed in the raw data for these supposedly overqualified, 50 SAT points higher Asian matriculants. Saying that the status quo might be fair makes for a less sensational story than reporting discrimination, so reporting these findings is boring. It is interesting that no newspaper or media outlet has reported that Espenshade found what appear to be signatures of affirmative action in favor of those highly qualified Asian applicants, because there have been stories about athlete and legacy preferences, but Espenshade found that Asian underperformance effect was stronger than that for legacies and athletes.</p>
<p>siserune, yes, I understand who you are referring to now. Thank you for the clarification.</p>
<p>
</p>
<p>ERC do not comment on why there is Asian underperformance relative to that of whites, a finding which I do not contest. To argue, however, that the 9.9 percentage point difference in class rank at graduation is suggestive of “affirmative action in favor of…Asian applicants” is far-fetched, to put it mildly. I challenge any admissions official at any of the schools that were in ERC’s sample to publicly state that Asian applicants benefit from racial preferences, where Asian refers to East and South Asian students, not Southeast Asians or Pacific Islanders.</p>
<p>
</p>
<p>The claim about Espenshade-Chung-Radford was not that their models and data suggest overt affirmative action for Asians in the sense of a minimum quota, a lower objective baseline for admission, or anything like that. They do suggest that processes statistically equivalent to pro-Asian affirmative action were going on (as reflected in the academic performance data and regressions), just as the admissions regressions (and to a lesser extent, data) indicate something statistically equivalent to anti-Asian discrimination. The causal mechanisms might be totally different, such as not deflating the Asian credentials as much as historical performance data might indicate is warranted, or in the case of admission, taking the effect of other variables into account that correlate with being Asian. </p>
<p>For example, if engineering and pre-med admission is harder and requires higher SAT and high school grades at the time of application, and Asians disproportionately concentrate in those majors, that would show up as a “negative action” SAT reduction for Asians in Espenshade & Chung’s 2004 regressions, unless they control for intended major (which they did not).</p>
<p>As I have mentioned, the “US East Asians decimated under a meritocracy” is the only hypothesis floated so far in these discussions that is consistent with all of Espenshade’s findings. Discrimination alone can explain the admissions regressions but it is contradicted by the academic performance regressions, which suggest either over-admission (affirmative action), or more plausibly, under-discounting (relative to pure performance-predicting admission) of the Asian credentials. </p>
<p>There are other explanations consistent with both phenomena but they have the flavor of conspiracy theories — for example, admissions sabotaging the Asian performance by admitting low-performing Asian groups to keep down the number of super-qualified Chinese that whites are allegedly afraid of competing against — but these are inconsistent with other data, such as the math competitions and other high level measures (valedictorian numbers, summa degrees, elite grad school fellowships, and others). Such mechanisms would imply incredible overperformance by US East Asians at lower-ranked schools, similar to the Jews at CUNY getting preposterous numbers of PhD’s and Nobel prizes in the older generation. Nothing like that seems to have actually happened in the last 10-20 years. </p>
<p>
</p>
<p>“You challenge”? The admission offices are reading CC and this thread specifically, on call to answer your every data request? I guess you can expect a press conference from the Ivy League by the end of next week, if not this one due to the busy season.</p>
<p>
</p>
<p>Translation: No admissions official is going to publicly say that Asians benefit from affirmative action, and as I, siserune, already explained in detail, I was exaggerating when I used the phrase “affirmative action” to describe Asian underperformance as measured by the 9.9 percentage point difference in class rank when compared to white students in ERC’s sample.</p>
<p>By the way, siserune, am I going to get any further explanation regarding your “low hanging fruit” assertion?</p>
<p>siserune, since you continue to bring up the subject of high-level math competitions, would you care to refute what NCL wrote in post #802 of [an</a> older but similar thread](<a href=“http://talk.collegeconfidential.com/college-admissions/790609-do-elite-colleges-discriminate-against-asian-students-54.html]an”>http://talk.collegeconfidential.com/college-admissions/790609-do-elite-colleges-discriminate-against-asian-students-54.html)? To my knowledge, you never responded to that post. Perhaps I should ask what exactly you were trying to assert?</p>
<p>
</p>
<p>I can’t imagine what you consider exaggerated when the performance measures themselves show Asians being decimated (in pure merit selection such as math competitions, etc) with **odds ratios worse than in the 2004 Espenshade & Chung study that Jian Li and others claim is signature of affirmative action for non-Asians<a href=“within%20the%20US,%20non-URM%20pool”>/b</a>.</p>
<p>Standard assertions that you don’t contest when they are posted, such as “athletic and legacy preferences are affirmative action for rich whites”, do use the words Affirmative and Action, but do not imply a direct or even a conscious preference in the admission process for rich whites (such as a minimum white quota, 40 extra SAT points, or a lower admission baseline as with blacks or Hispanics). They simply express the idea that whites, or rich whites, are overadmitted relative to a pure academic selection under a particular type of admissions process.</p>
<p>To say that Asian academic underperformance is a “signature of affirmative action” is exactly the same type of statement, except that the hypothetical affirmation mechanism is different. If the end result is overadmission relative to pure merit selection (or pure merit within the US, non-URM, population) it is standard to speak of that as “affirmative action” even if this is only de facto AA and not an official preference category acknowledged by, or even known to, the admissions officers. For instance, if, to use Espenshade’s SAT terms, Asians were getting a 50 SAT point deduction but a 90 point penalty were in order based on predictable academic performance, that is 40 points of affirmative action in exactly the same way that the 50 is a form of “disaffirmative action” (or equivalently, affirmative action for non-Asians).</p>
<p>I’m also pretty sure I used the more precise phrase “statistical signature of affirmative action” to describe this in the earlier messages thatyou (Fab) read and replied to without incident, so this language usage issue seems to be a new and pointless way to avoid discussing the data.</p>
<p>
</p>
<p>You got the 9.9 from Espenshade’s regression, but (again), try to read more carefully. Espenshade’s models predicts “only” about a 5-point difference for most Asians because of a partly compensating positive effect of being an immigrant. </p>
<p>This was also true of his 2009 admission models, though not reported in “3-to-1 discrimination” media circus accounts: Asians are advantaged over whites in the lower income categories, disadvantaged in the higher ones, but at all income levels, admissions chances increase for immigrants. </p>
<p>Lesson: study the regressions carefully before drawing conclusions.</p>
<p>
</p>
<p>I refuted it prior to his posting, in #782, where I listed and anticipated the data-fudging techniques that NCL would employ in his “rebuttal”. Did you not catch this earlier comment?</p>
<p>
</p>
<p>As far as the statistical significance discussion is concerned, in #802 NCL ignored the above, and only posted more of the single-year significance calculations. He did not address, and certainly did not refute, the significance calculations I mentioned in #782, giving lower probabilities (p-values) of 1 to 5 percent for the entire data table, not the single years’ figures which of course are likelier to occur by chance.</p>
<p>I do thank you (Fab) for introducing the expressive phrase “denominator of 6”, which capture in three words the sort of tricks that are used for downplaying unpleasant findings.</p>
<p>NCL’s other point, which I did not have the opportunity to rebut because the earlier thread was locked, was that (according to him) the Asian decline patterns are nullified by year to year growth in the Asian population. In other words, Asians have a higher share of USAMO qualifiers compared to IMO training camp selectees or team members, not because of any tendency to be decimated by the selections, but because there are more Asian students coming up through the ranks in grades 8 to 10 than in grades 11-12 which are where most of the strongest students are at any given time. </p>
<p>It’s a nice theory, but it doesn’t match the data. The “US East Asian decimation” pattern survives just fine when taking population growth into account. NCL didn’t present population figures to support his claims, but there are enough other data to dispel any glib reference to demographics as an explanation of the Asian declines. One example is IMO medals. At the time of the last thread, I made a count for 1998-2009 and there was a rather large white-vs-Asian effect for the chance of an IMO selectee to get a gold medal, with whites being 50 to 80 percent likelier to achieve the gold. This is a small sample but it is one of a number of similar indicators that as you follow the same cohort down the academic pipeline, the decline proceeds. Other examples are the Asian qualification levels for the 10th grade olympiad vs the 12th grade olympiad (the second rate is lower, with odds ratios unexplainable by two years of population growth), or the enormous decline from the 50+ percent level of USAMO qualification to the 20-30 percent of Putnam competition representation within the same cohort, i.e., East Asians who attended US high schools. That’s an odds ratio between 3-1 and 4-1, and if the E.Asian share is dropping to such an extent for the same cohort even after several years of college, that indicates that something else is happening besides demographics, and its effects are strong.</p>
<p>If the believe of Asian intellectual inferiority can help some people sleep better at night, Ill just say, hey, whatever works. But it gets annoying when someone just keeps on selectively picking, distorting, and sometimes just making up evidence to support his view.</p>
<p>Post [url=<a href=“http://talk.collegeconfidential.com/1064804317-post909.html]#909[/url”>http://talk.collegeconfidential.com/1064804317-post909.html]#909[/url</a>] and a few later posts made repeated assertion on the following points.
</p>
<ol>
<li><p>It is true that for the lower- and working-class students, Espenshade et al. find that asian students had higher probabilities of admission. But siserune chose not to present the large quantities of other related facts in the book that form the entire picture. An average Asian applicant had higher class rank (or GPA) and SAT scores, took more AP classes, and had more leadership positions and academic awards. Isnt it possible that in the lower- and working-class group, the asian students are simply academically stronger than their white counterparts? And most importantly, siserune chose to ignore the fact that lower- and working-class students constitute only a very small sliver of the total applicants, about 8 percent of the white and 13 percent of asian applicants.</p></li>
<li><p>Group academic performance in college is difficult to capture with some simple metric. I think that is why different measurements have been used in the past, including graduation rates, freshmen GPA, and even starting median salaries. Clearly these measures all have their limitations. Espenshades model using class rank at graduation is much more problematic, and his implementation made it worse. There are two obvious flaws with this model. (1) The model basically compares the group class ranks at graduation with their group ranks (a virtual rank based on SATs etc) at admission. This is not an apple to apple comparison, because these are not the same groups of students (many in the admitted groups did not graduate). Espenshade clearly showed that asians had the highest graduation rate. If all of those did not graduate were included (in the bottom quintile), the class rank of asian would certainly go up relative to all other groups. (2) The model did not consider the effect of group aggregation into different majors, asians are heavily concentrated into engineering and natural sciences, and others are more into humanities and social sciences. To see this problem, just perform the following simple mental exercise: upon admission the incoming students are divided into two classes, the top half (in rank) of the students are grouped into one class and the bottom half into another, by graduation the two class will have the same academic class rank and the top half group will underperform by a whopping 50 points (out of 100) regardless how they were admitted.</p></li>
</ol>