<p>
</p>
<p>Ordinary English is good, really easy to understand.</p>
<p>“Lies, damned lies, and statistics”</p>
<p>Can anyone translate that to plain English?</p>
<p>
</p>
<p>Did those with <2400/36 got into as good or better colleges (if non Asians)?</p>
<p>
</p>
<p>Oh, so an “ordinary English” translation involves turning one straightforward compound sentence into two sentences, one simple and one complex, both laden with metaphor and meaningless to anyone who has not followed the discussion. Don’t we have a more appropriate term for this type of writing? We certainly do - obfuscation - and siserune is a master of it without peer.</p>
<p>
</p>
<p>Then why don’t you point out where Table 5 from the original 2002 version of the paper is in the later 2004 and 2005 papers? Wait, let me guess, I’m going to get a “do your own work, I’m not your unpaid RA” response. Alright, fine, here’s the answer: Table 5 from the 2002 version of Avery et al. does not exist in either the 2004 or 2005 versions.</p>
<p>This is just sad. If that discussion existed in the later versions, you wouldn’t have had to reference the “original 2002 edition” of the paper; tautologically, it would have BEEN in the later versions!</p>
<p>"Did those with <2400/36 got into as good or better colleges (if non Asians)? "</p>
<p>Not sure what the question is.</p>
<p>Only pointing out that there is no conspiracy out there to lock Asians out. However, perfect scores don’t always get people in or keep them out. It does matter what else they have on their resume.</p>
<p>The main impediment usually is how they want to balance the class racial profile.</p>
<p>Maybe it is all about “Asians have a small advantage in admissions”, but they deserve a greater advantage, based on their INDIVIDUAL applications? Or, to put it otherwise, even MORE Asians would get in if applications were reviewed without reference to race?</p>
<p>(I am not Asian, and am using this as an argument for INDIVIDUAL NON-RACE IDENTIFIED evaluation)</p>
<p>"
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
"</p>
<p>Can you put their gender next to each, texaspg? Are they all Asian? Which Ivy? Appreciate.</p>
<p>They are all Asian and refers to findmoreinfo’s post wondering whether having perfect scores is still not good enough for Asians to get in. I tried to put the gender down and noticed it breaks into 3 each.</p>
<p>I know more about these candidates but I can’t say more in an open forum since it is a privacy issue and can point directly to them.</p>
<p>
</p>
<p>In the statistical studies we are talking about the measured Asian effect <em>is</em> an estimate of the difference between the actual admission and its non race-identified, or non Asian-identified, evil/virtuous twin. A positive coefficient on the ASIAN variable suggests that Asian admission results would be worse under race-blinded evaluation. </p>
<p>There is a separate and crucial question of whether a nonzero Asian effect comes from actions that take place in a university admission office, and not peculiarities of the statistical methods and data that were used to calculate the supposed effect. I posted many times some statistical explanations of why, for example, Espenshade’s study using M+V as the weighting of SAT score when the admission office uses something like M + 2V or M + (1.8)V, would contribute about 10 points of Asian penalty to the analysis. And then there is the effect of omitting Early Decision status, athletics (recruited and not), and choice of major from various analyses. The omission of those would generate additional Asian penalties, or cause underestimation of the Asian advantage in studies that find that.</p>
<p><a href=“http://talk.collegeconfidential.com/12808256-post1613.html[/url]”>http://talk.collegeconfidential.com/12808256-post1613.html</a></p>
<p>Interesting analysis going on w.r.t. to race and sex and major determining who gets the 25 seats available for waitlisters. Collective wisdom seems to point to Asians not getting any since they are oversubscribed using the past baseline.</p>
<p><a href=“http://talk.collegeconfidential.com/harvard-university/1336101-official-2016-waitlist-discussion-thread-6.html?highlight=harvard+waitlist[/url]”>http://talk.collegeconfidential.com/harvard-university/1336101-official-2016-waitlist-discussion-thread-6.html?highlight=harvard+waitlist</a></p>
<p>Siserune,</p>
<p>After being away for some duties, glad to see discussions on this thread. My Quick Reply with automatic quote option doesn’t work for multiple browsers so I have to put your quote in " ".</p>
<p>" China Pride"? lol!!! China government is our rival.</p>
<p>“…one place to look is the original 2002 edition of the Revealed Preferences college ranking study by Avery, Hoxby, Metrick, and Glickman.</p>
<p>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. ”</p>
<p>What is the definition for ‘positive statistical effect’ in this research? What are the data and how do they calculated to come to the “conclusion”? Or this is your conclusion?</p>
<p>The 2002 research result (or any other year of this research) has nothing to do with the admission and qualification correlation by race. This is merely ranking of schools based on applicant preferences while they were in high school. You can see the new 2012 publication which has no indication of admission and qualification correlation by race at all, either. [A</a> Revealed Preference Ranking of U.S. Colleges and Universities by Christopher Avery, Mark Glickman, Caroline Hoxby, Andrew Metrick :: SSRN](<a href=“http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105#captchaSection]A”>http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105#captchaSection)</p>
<p>One has to see the data to know what was worked on and how. Many research collected good data but didn’t analyze them in a right way or manipulated them to get results they wanted. Like this one you referenced: (which also tried to argue that race has no effect)
<a href=“http://iacs5.ucsd.edu/~bbackes/pdf/bbackes_jmp.pdf[/url]”>http://iacs5.ucsd.edu/~bbackes/pdf/bbackes_jmp.pdf</a>
(This is a sloppy research with inconsistency and lack of statistics in those tables)</p>
<p>Page 38 - Obviously Asians stay in their old GPA average to be admitted to Berkeley but the bar for URMs is higher, white and Asians are now qualified with same GPA and even URM need higher GPA to get admitted
Asian: 3.7 and still 3.7 after AA ban
White: 3.6 became 3.7
URM: 3.4 became 3.5
But their conclusion was that race has no effect - that is completely blind to the data.</p>
<p>Besides,
Page 38 Table2 doesn’t show statistics by each UC campuses
Page 39 Table3 doesn’t include white
Page 40 Table 4 doesn’t show stats by each UC campuses
Page 41-44 Table5-8 doesn’t include white or Asian groups</p>
<p>What I would like to know is each top school’s change in percentage on admtted Asian/white/URM over the past 10 years. (Ivies, Stanford, MIT, Duke, and Johns Hopkins. Cal Tech is fine, we all know they go by numbers.)</p>
<p>I am interested in knowing what you are going to say about the Princeton Asian admission improvement after complains and law suits.
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%</p>
<p>Fabrizio,</p>
<p>"
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. (The first instance in this thread 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].”
"</p>
<p>I agree! When authors of an article removed/hid original factors, it’s because these factors could affect the result to the opposite direction or they were incorrect and the authors were willing to change. This disqualifies the original version for being used or cited. I am sure siserune knew it but still used it - for a purpose.</p>
<p>I like it that fabrizio sticks to facts, writes in a comprehensible way, and does not resort to personal attacks. If siserune is completely comfortable with his position, then why not a straightforward defense? Why the obfuscation?</p>
<p><a href=“fabrizio:”>quote</a></p>
<p>So if Antonovics and Backes show that the estimated change in admissions probability for Asians relative to whites after Proposition 209 all else equal
[/quote]
</p>
<p>That wasn’t the test being performed. The UC Asian discrimination theory predicts that the Before and After effects (on admission probability, of changing a white applicant to Asian) of prop 209 should satify:</p>
<ol>
<li><p>Before < 0 (effect of being Asian was negative; regression coefficient “Asian” < 0)</p></li>
<li><p>After > Before (prop 209 made it better to be Asian; coefficient “Ban*Asian” > 0)</p></li>
<li><p>After is small in absolute magnitude compared to Before (post-209, any effects caused directly by race are minor).</p></li>
</ol>
<p>fabrizio’s postings on this mis-stated what After means in terms of regression coefficients. It is the sum (Asian + Ban<em>Asian), not Ban</em>Asian. The results are then:</p>
<pre><code> UC-B UCLA UCSD UC-D UC-I UCSB UCSC UC-R
</code></pre>
<p>Before -1% +1% +2% -5% -2% +2% -2% -1%
After +1% +4% +4% -1% -4% -3% -2% -2%</p>
<p>change +2% +3% +2% +4% -2% -1% -0% -1%</p>
<p>Only UC Davis is consistent with the standard discrimination theory that would lead to “skyrocketing Asian enrollment after proposition 209”. The signs are somewhat random: the Before and Change coefficients are half positive and half negative (and no pattern in After either, though that is the expected result).</p>
<p>Another test of whether the measured Asian coefficients represent a discrimination effect is to see if the signs of Before and Change are opposite more often than not, so that a pre-existing race effect tends to be mitigated. This does not happen at all. At 4 of 8 schools, a positive or negative Before effect becomes noticeably stronger after prop. 209 and at a fifth, the same or slightly stronger. Only at one school, UC Davis, does the magnitude of the Asian effect (ignoring the sign) become smaller. </p>
<p>
</p>
<p>Here, fabrizio spins and stretches the material into a total dismissal of the overall results of the study as “statistical noise”. The actual comments were more limited and of course were accurate.</p>
<p>The term “small statistical noise” was used for the smaller regression outputs such as the Before/After effects of +1 and -1 at Berkeley which are in reality 1 + X and -1+Y where X and Y represent effects of the choice of model, which data were available, what predictor variables were formed from the data, etc – and where |X| and |Y| can easily be larger than 1: the noise can be larger than the signal. Note also that two of the “effect of 209 ban” Asian coefficients of 1 percent, the ones for UCSB and UCR, are not statistically significant, which is further indication that individual effects of this size, especially when small, are not all that meaningful by themselves (in this regression). </p>
<p>
</p>
<p>Here fabrizio stretches and spins my more specific, limited and accurate comments into a general false “rule” that is easier to attack than the actual comments posted. In addition to the reasons above, 1 percent effects on the estimated admission probability where most of the accepted students have a large admission probability (such as 30 to 70 percent a priori), are small and, given the inaccuracy of the model, likely to be less than model mis-specification error. Effects of 4 percent for UCSD and UCI are more meaningful, but both of them appear as “race” effects after a race ban, which is further evidence that one cannot read the regression coefficients literally without more analysis.</p>
<p>
</p>
<p>And there is a similar reason why people who understand the material do more than E-Z glance at the sign and number of asterisks next to one or two coefficients, before running victory laps and posting denunciations on the Internet.</p>
<p>
</p>
<p>Comedy gold! </p>
<p>
</p>
<p>This is the second time MisterK posts non-specific insinuations that smuggle fabrizio’s equally non-specific insinuations into the discussion as somehow being correct, reliable or accepted background. My reply is similar:</p>
<p>if anyone other than fabrizio and Findmoreinfo (whose recent postings I will get around to answering one by one, sooner or later) thinks that I have obfuscated any material point, just name the disputed item and quote the original statements, and I will be happy to clarify, elaborate or otherwise get to the heart of the matter. I don’t think I could possibly be more clear or straightforward than I have been about the mathematics, but maybe some people do not find it clear or consider some points unresolved. </p>
<p>On the subject of obfuscation, I think fabrizio owes the readers here a clear answer about the math he denounced as incompetent, but could never indicate an error once the discussion became extremely numerical in a way that sidestepped the various word games about “paradoxes” and “mechanical correlations”. He never did answer, though he did abruptly stop talking about it once the calculation was posted. Are the calculations in #419 correct, or not? Was the conditional distribution the one that corresponds to what an admission officer would use, or not?</p>
<p><a href=“http://talk.collegeconfidential.com/13591902-post419.html[/url]”>http://talk.collegeconfidential.com/13591902-post419.html</a>
<a href=“http://talk.collegeconfidential.com/13585598-post402.html[/url]”>http://talk.collegeconfidential.com/13585598-post402.html</a></p>
<p>I am referring of course to the “negative correlation between Talent and Asian given Score” discussion.</p>
<p>
</p>
<p>Antonovics and Backes conducted a differences-in-differences analysis. Therefore, the coefficient on the interacted term, Ban*Asian, is the one of interest. You’ll kindly note that I was careful with my language to emphasize this: “…the estimated change in admissions probability for Asians relative to whites after Proposition 209 all else equal…” (post 919, emphasis added).</p>
<p>
</p>
<p>Which contradicts your previous conclusion that “NOTHING HAPPENED as a result of the ban on race in California, where white vs Asian comparisons are concerned.” Thank you.</p>
<p>
</p>
<p>As compared to five that are, three of which have standard errors that are at least in the thousandths? That’s some ugly cherry picking there, siserune.</p>
<p>
</p>
<p>Oh, sure. That’s why [Bertrand</a>, Duflo, and Mullainathan](<a href=“http://economics.mit.edu/files/750]Bertrand”>http://economics.mit.edu/files/750) published their methodological piece on standard errors in differences-in-differences analyses in 2004; and why [url=<a href=“http://129.3.20.41/eps/em/papers/0508/0508018.pdf]Gabor”>http://129.3.20.41/eps/em/papers/0508/0508018.pdf]Gabor</a> K</p>
<p>
</p>
<p>I appreciate the compliment, but I must admit that I’m not above personal attacks.</p>
<p>
</p>
<p>The discussion never changed. To the contrary, [it</a> became almost identical](<a href=“http://talk.collegeconfidential.com/13591902-post419.html]it”>http://talk.collegeconfidential.com/13591902-post419.html) to the examples you humorously claimed were “from college admission.”</p>
<p>[Kevin</a> Murphy](<a href=“http://www.cs.ubc.ca/~murphyk/Bayes/bayes.html]Kevin”>Graphical Models)
[Judea</a> Pearl](<a href=“Causality: Models, Reasoning, and Inference - Judea Pearl - Google Books”>Causality: Models, Reasoning, and Inference - Judea Pearl - Google Books)</p>
<p>So there was no need for a reply on my part. From my perspective, you had already admitted that the negative relationship you so proudly displayed in support of your “meritocratic discounting” sham was an “accident of statistics.” But since you asked so nicely, I give you…the same reply I gave you last year. In the language of Murphy and Pearl, talent and effort became conditionally (negatively) dependent given that the score of 750 was observed since “either property alone is sufficient to explain [the score of 750]” (Murphy’s words paraphrased to suit your example).</p>
<p>But knowing siserune, such an answer is likely to be very unsatisfactory. Fine. I will demonstrate why siserune’s “extremely numerical” discussion is simply a modification of the examples in Murphy and Pearl:</p>
<ol>
<li><p>The key assumption in both Murphy and Pearl’s examples is that possessing at least one of two attributes leads to college admission. siserune assumed “SAT 750+ is attained exactly by those who possess Talent, or Effort, or both,” which is the same thing.</p></li>
<li><p>As a consequence of that assumption, the schools have seemingly weird characteristics, as “brainy” people are less likely to be “sporty” and vice versa (Murphy), and students who had high (low) GPAs are more likely to have poor (excellent) musical talent (Pearl), even though in the general population, the corresponding pairs of attributes are uncorrelated. In siserune’s case, he computes that if you select 100 Asians and 100 non-Asians with SAT scores of 750 or higher, there will be more non-Asians with talent than Asians. Why?</p></li>
<li><p>Well, siserune’s example is no different than Murphy’s or Pearl’s examples. So the explanation is no different: talent and effort “compete” to “explain” the score, in Murphy’s language. Just as Pearl noted in his example that “students with low grades are likely to be exceptionally gifted in music, which explains their admission to graduate school,” Asians with low talent are likely to have expended significant effort, which explains their 750+ score.</p></li>
</ol>
<p>
</p>
<p>Well, here’s an intentional obfuscation (a humorous one, I think):</p>
<p>
</p>
<p>Whatever that was supposed to mean, you’ll have to admit that it was obfuscation. </p>
<p>But mostly, your posts are off the charts in writing complexity. It’s not because they incorporate some deep mathematical ideas that require complex prose (actually, math lends itself to simple writing). It may simply be a personal style. But it raises one’s skepticism about your arguments.</p>
<p>The reader gets the impression that fabrizio is trying to win the debate with facts and interpretation, and siserune is trying to win with rhetoric and logical maneuvering. You’ve heard the phrase, “If you don’t have the law, you argue the facts; if you don’t have the facts, you argue the law; …” That’s the line of thought that’s setting off my skepto-meter. Can you support your position more straightforwardly? </p>
<p>
Perhaps, but I did actually did read this entire thread (!!!), and I noticed that despite a lot of personal bashing and baiting, you seemed to maintain good humor and continued to deal with facts. Maybe you slipped up here and there, I sure would have.</p>
<p>In all seriousness, how would a historically black institution such as Howard treat AA? Sorry if this has been addressed before, I just always wondered if these universities treat Affirmative Action in a traditional sense (URM = Minority in Total Population) or they consider minorities specific to the school demographics.</p>
<p>^I’m not sure I understand.</p>
<p>" I just always wondered if these universities treat Affirmative Action in a traditional sense (URM = Minority …"</p>
<p>How would that work? I know there were plenty of white students in my medical school class at Howard.</p>
<p>I can’t believe I just read through all 63 pages of this thread. I must admit the exchanges between Fab, Sis, lookingforward, perozziniman, Findmoreinfo, and the others were quite entertaining and enlightening at times.</p>
<p>In the meantime, I’ll offer my own take on AA, since I don’t know enough about stats to contribute to the immediate debate at hand.</p>
<p>I have many asian friends who were “shafted” this round of admissions (I’m an hs senior). For the record, I am also “asian.” Many felt that it was because of their race that they were denied spots at top (read: Ivy) schools. Was it really because of their race? No one knows for sure. Personally, I did not feel that they were particularly worthy candidates. They had respectable stats (22/2300 SAT, 3.8+GPAuw), respectable extracurriculars (All-Eastern Orchestras, national level AcaDec/SciOly/MUN teams), and I imagine decent essays and LoRs. In all respects they were ‘qualified.’ But were they ‘deserving’ of admission?</p>
<p>The thing is, they had nothing that made them ‘stand out.’ Nothing to set them apart from other kids who did the same things but better (ie, get 2400/4.0 and win those national level competitions, along with better essays etc.) Nothing to set them apart from…dare I say it? the other asians! Now when I say asians I refer to the common stereotypical study-work-Tiger student.</p>
<p>Do we need to identify them formally as “asian” to identify them as “upper class hard working” students? No. But why does this connection exist in the minds of so many people? I think that it is important that people realize that it is difficult to escape one’s own cultural legacy. It is a part of who we are. Are there exceptions? Of course. But as a whole, while it may not be PC, and while siserune may debate the quantity and quality of it, Asians work harder at academics. That’s the long and the short of it. People from the South tend to be more easily angered and tend to be more violent. Koreans (used) to have the worst airline in the world due to communication/superiority/hierarchal constraints in their society. Black/AA people are good at sports. Call it racist, call it stereotyping, but on average it’s all true. My source is a book by Malcolm Gladwell entitled “Outliers.” In it he cites most of the relevant studies. (I disagree with him on why asians work harder however; while he cites a rice-paddy culture, I choose to simply point out the fact that because of Confucius China was basically the first to offer the people a chance at a good job and life if they did well on a test. This is magnified even further in what Findmoreinfo said a while back, when he pointed out that the Asians who emigrated were basically the top 1% of their homelands).</p>
<p>Where am I going with this, you may ask. Well basically, I think that race in admissions does matter. It doesn’t matter on its own, and AA is at fault for giving such extreme advantages to middle/upper class minorities as cited by fabrizio. But it creates a context that adcoms can use to help fill in some of the blanks of the application. An Asian’s score doesn’t have to be ‘meritocratically discounted’ but race can be used to compare one asian to another. Why should a college want a plurality of students who all share Asian upbringing and ideals? That’s boring and inconducive to achieving ‘diversity,’ which I define as fabrizio’s mix of talents, backgrounds, and ideas. </p>
<p>However, we are moving closer and closer to a multicultural ‘browned’ society. Here is where it becomes the adcom’s job to carefully decide how race is used in admissions. The middle class african american growing up in my school district might as well be ‘asian’ for all the habits and characteristics he/she may show. Is it right for AA to reward him/her while edging out the Asian applicant with better/equal stats? Obviously not. Do I support racial preferences? Yes, but with a catch. The applicant should be viewed as an individual first, with their ethnicity/race acting as a backdrop. I find it a logical fallacy to assume that a person can contribute to diversity solely based on skin color. </p>
<p>Sorry if this isn’t very clear, it’s 12:30AM and I am tired.</p>