<p>
</p>
<p>No, you didn’t. You said “under a plain reading the sentence expresses an idea that is incorrect, misleading, contestable, or (at best) oversimplified, i.e., the summary of Asians’ academic behavior as “working harder”.” But no, that’s not how the sentence is “supposed” to be read because you never used the word “supposed.” Right.</p>
<p>My magazine may be empty, but yours was never loaded to begin with, sis.</p>
<p>With historical hindsight, we would see the early 20th century medical schools’ admission quotas for Jews as a grave injustice (it is the reason why schools like Brandeis were created), yet for some reason it is acceptable to practice an analogous practice today against Asians.</p>
<p>If a bunch of Asian community leaders got together and founded various universities (pushing for top notch faculty) that gave a large admission bonus to Asian applicants, it would be readily litigated under the 14th amendment, and there would be an immense outcry against the practice – we could readily predict that the argument that “it is a private institution, it can accept whoever it wants” would be rejected in the courts – so long as they were pro-Asian institutions, favouring a community with little political power.</p>
<p>If Affirmative Action is to compensate for privilege, then it would be only just target socioeconomic privilege. pre-Holocaust admissions policies at medical schools unjustly affected both rich Jewish children of bankers who grew up in privilege and children of Russian Jewish farmers who got chased out of Russia in Stalinist pogroms.</p>
<p>*but of course all you white privileged people can do is perceive us as some faceless “yellow horde” because your racist psychology prevents you from seeing our internal diversity. *</p>
<p>Say What? Who’s the faceless yellow horde and who used the expression? The phrase is embarassing. The insinuation- no, the declaration- that we, your faceless, mostly private about our own details fellow CC posters have repulsive personal attiitudes-- well, what point are you making and what sentiments are you revealing?
Name colleges today that discriminate against a high-match Asian for being Asian. No reverse proofs from publics. Find one with a policy. Show us this debate is based on reality.</p>
<p>The problem with citing UC is that Asians make up a large percentage of the population in CA.</p>
<p>
</p>
<p>Does this mean you don’t want to be my friend? ;)</p>
<p>
</p>
<p>it is implied in your language and your unconscious stereotyping. of course you may not think you are being prejudiced, but prejudice can be subtle (at least to the prejudger, not the receiver).</p>
<p>
</p>
<p>A tall order, when colleges are not wont to spit out the exact formula (or formulaic guidelines) for admissions. This is like saying, “find Walmart branches today that discriminate against women for being women” – when it’s obvious from statistics that discrimination is occurring, and the company may not necessarily have an explicit policy they are willing to reveal. Cultural prejudice especially, can be implicit yet so in-your-face.</p>
<p>
</p>
<p>I’m talking about privilege; a system; a culture; I don’t mean to attack individuals.</p>
<p>lookingforward - Are you a current adcom for an Ivy?</p>
<p>I am hard-pressed to figure what I have “implied in [my] language and [my] unconscious stereotyping.” What stereotypes? That the 100+ apps from TJ are all high-performers, with similar ECs? Yes, I wonder why people insist there is Asian discrimination but can’t source it. Note that I have not name-called or hyper-generalized under the blanket of defending my position.</p>
<p>Yes, I am a reader, not an adcom. Thank heavens. It’s a brutal job and people underestimate adcoms’ goals and sensitivity.</p>
<p>
</p>
<p>Impressive! If your goal is comedy then never mind, just carry on, but if not, then you might want to consider the possibility that you are flailing against basic (undergraduate) textbook level material via misunderstanding of far more basic (AP Statistics) vocabulary. What you just posted as a loud correction is the kind of gaffe for which doctoral students get exposed and defrocked (un-funded, in current parlance). Lucky it was done anonymously!</p>
<p>
</p>
<p>I talked about selection given the known information (the test score, in the admission example), and said so in the first posting and the one that followed. This can only mean “conditional”, as I spelled out in the followup.</p>
<p>
</p>
<p>Reviewing high school AP Stats might be in order. Your comments are badly confused.</p>
<p>That two variables here are the “independent variables”, with some other variable “dependent” (meaning, a function of) those two, does not mean that the “independent variables” are independent in the statistical sense. When talking about whether pairs of variables are correlated, whether in general or conditional upon the test score, correlation can only mean a lack of statistical independence.</p>
<p>It is actually unusual, though not impossible, for the “independent” variables to be independent statistically. Talent and effort certainly are correlated in real life (we spend more time on things we are good at, and become better at things we practice). This has no effect on the argument for why being a “hard worker” implies underperforming one’s credentials, since that argument assumed only that more talent and more effort tend to improve the outcomes. It does, however, contradict the statement that these variables cannot be correlated because they are the “INDEPENDENT variables”.</p>
<p>"Yes, I am a reader, not an adcom. Thank heavens. It’s a brutal job and people underestimate adcoms’ goals and sensitivity. "</p>
<p>lookingforward - can you clarify what a reader means? </p>
<p>My pet peeve is that the adcoms are usually too young, a lot of times fresh graduates. I suspect that probably gives them an edge in identifying what is a fit between their school and the applicant but as a parent I feel a bit apprehensive about entrusting people that young to be deciding my kid’s future.</p>
<p>I think a lot of the rationalization would have melted away if folks only remember this Duke study; the graph leaves very little to the imagination:</p>
<p>[Testing</a> for ‘Mismatch’ | Inside Higher Ed](<a href=“http://www.insidehighered.com/news/2009/04/20/mismatch]Testing”>http://www.insidehighered.com/news/2009/04/20/mismatch)</p>
<p>
</p>
<p>I certainly agree with the political power part. Almost all human interactions can be easily explained away by looking at the power relationship. The Economist has this to say about Karabels The Chosen:</p>
<p>*the concept of meritocracy itself is strategic and flexible, and often in outright conflict with egalitarian aims. “Those who are able to define ‘merit’,” he writes, “will almost invariably possess more of it, and those with greater resources–cultural, economic, and social–will generally be able to ensure that the educational system will deem their children more meritorious.” *</p>
<p>Lani Guinier agreed, calling The Chosen a refreshingly candid account of the admissions madness at elite colleges, where merit often functioned simply as a handmaiden to power.</p>
<p>
</p>
<p>That is not what I said. You (continued to) misread and misquote, and (continue to) maintain a conspicuous silence about the factual matter that prompted this exchange. </p>
<p>Misreading: the words “plain reading” are a reference to your term (“plain English”, post #286) for your reading of what you quoted from Steinberg et al. Where I wrote the words “plain reading”, they are interchangeable with any of: “fabrizio’s reading”, the “most literal reading”, a “literal (out of context) reading”, or the like. Nowhere did I say or imply that any of those readings is the only plain-English reading, or the reading that would first come to mind when reading the book, or the likeliest possibility for what the authors meant.</p>
<p>Misquotation: you quoted only a part of my sentence. It actually began “I stated [that]…”, so that the whole thing is a paraphrase of earlier statements, not an autonomous comment on how to read Steinberg et al’s text. Those earlier comments again are in response to your reading (i.e., the most literal reading) of the sentence. What was stated in those comments, and reiterated and paraphrased later on, was not that your reading is right or wrong as a literary assertion about what the authors meant to say, but that the statement you attributed to them is factually incorrect and not supported by their book, even if you are right that it is what they intended to say.</p>
<p>Misdirection: still no evidence or argument about the factual question in dispute, which is whether Asian academic behavior includes “hard work” (i.e., significant difficulty) distinct from “spending more time”.</p>
<p><a href=“Canuckguy:”>quote</a>
[/quote]
</p>
<p>greetings from warmer climes!</p>
<p>
</p>
<p>I realize you remember this, but for the benefit of “the viewers at home”…</p>
<ol>
<li><p>There was a longer discussion of the Duke study in a predecessor of this thread, in which I posted calculations that abruptly silenced the Chinese sparring partners who had been citing it as evidence of anti-Asian discrimination at Duke.</p></li>
<li><p>The paper on which the Inside Higher Ed is based was not a study of discrimination, but a fairly subtle application of theory to test whether applicants or universities have “private information” relevant to predicting applicants’ grades at college. Their conclusion was that applicants do not have any usable information not contained in the material they submit in the application (such as knowledge of their own ability) but that the admission office does have additional power to predict grades, not explained by the shared information. I don’t know how generalizable the study is, but it is most directly relevant to the second-guessing of admission decisions by applicants and their families, not to discrimination as such.</p></li>
</ol>
<p>By the way, I saw recently that Yale published its SAT averages broken down by race and recruited athlete status for a three year period not long after the Espenshade study, and the Asian/white SAT differences were a lot smaller than the Espenshade 50 point discount or the Duke 47 point difference, with most of it disappearing when athlete status was taken into account. The differences between men and women were considerably larger, which as you may recall was my basic interpretation of Espenshade: a strong finding of gender discrimination in favor of women (with no credible interpretation except as discrimination) and a weak finding of an Asian “effect”, wherever it may come from.</p>
<p>
</p>
<p>So talent is one of many variables that “cannot be influenced much by effort,” yet we “become better at things we practice.” Great.</p>
<p>My apologies for not being clear with which meaning of “independent variable” I used. Before your revelation that “talent and effort certainly are correlated in real life,” I was working off your talent “cannot be influenced much by effort” line, and I too strongly interpreted that as statistical independence, wrongly ignoring the “much” and conflating two distinct concepts.</p>
<p>Your supposed “negative correlation” of Asian and talent given a fixed score is not a correlation. It’s nothing but a mechanical relationship. If you fix 750 and assume that “effort” is high, “talent” has to be low to maintain equality, but that in no way means that “talent” and “effort” have a negative correlation conditioning on a fixed score.</p>
<p>With comments like that and a hat tip to Freud, I can guess why you’re so concerned about my funding status.</p>
<p>
</p>
<p>Go check the last paragraph of your post #324 and then come back and tell me that you did not say, “under a plain reading the sentence expresses an idea that is incorrect, misleading, contestable, or (at best) oversimplified, i.e., the summary of Asians’ academic behavior as “working harder”.”</p>
<p>^^Hi, siserune and fabrizio. Yes, winter is here. I just backed into somebodys parked car outside my driveway the other day and am too much of a man to just drive off…</p>
<p>My posting of the Duke study was not aimed at your dispute, but to inform other posters that there are information are out there that dispute their claims. In other words, I am not interested in the study per se so much as I am interested in the data provided by Duke. They are most certainly controversial if not downright subversive. For example, I heard a million times that URMs only do worse on the SAT, or that they only do worse on the SAT plus GPA, or that they write great essays that floored the adcoms…etc.etc. Sorry, folks. They aint true, at least not according to the Duke data.</p>
<p>I do have a question for you, siserune. Your theory concerning how SAT scores over-estimate the ability of Asian applicants is interesting. I would expand your theory and argue that the scores certainly over-estimate the ability of the URMs. If the elites compensate for this with the Asians, why are they not do so for the URMs as well?</p>
<p>My answer, as you already know, is power.;)</p>
<p>Now, back to mourn my financial losses… BTW, do you have the site for the Yale SAT data? I would love to see it.</p>
<p>Oh Fab. In a way, I SO get you! You are so impassioned about this, I can imagine you must feel like I do. Like the very survival of your people is at stake. I think I know how you must feel. Our reasons are different, sure, as is where we “sit”. I am California, Middle aged, with young adult kids, with advantages for which i am eternally grateful. I am looking at infant mortality, high school graduation, health disparity, poverty, ect. You know the story, and no, it does not belong here.I do not believe affirmative action is a facet in most people’s lives. I don’t believe it saves lives. I don’t know that racism against Asians is a facet in most people’s lives but I am willing to learn. I come here and see the kids who beat the odds. I want to know how and why. I don’t believe there is an easy answer. I dont believe there is a single cause. I am not sure what drives you, but I am sure it is just as compelling. I stick around trying to get a glimpse. I know there are Asian students who post here, wishing they were black. I swear, that blows my mind! And I imagine you really believe the answers lie here. </p>
<p>Good luck!</p>
<p>As an update, the Arcidiacono et al. paper mentioned in Inside Higher Ed has been published in Quantitative Economics, an open access journal. Everyone may view the published article for free [url=<a href=“Does affirmative action lead to mismatch? A new test and evidence | The Econometric Society”>Does affirmative action lead to mismatch? A new test and evidence | The Econometric Society]here[/url</a>].</p>
<p>
</p>
<p>And the implication of their conclusion is that while their paper does not prove that affirmative action leads to mismatch, mismatch is nevertheless plausible. They show that if mismatch exists, then selective universities possess private information; equivalently, if selective universities do not possess private information, then mismatch does not exist.</p>
<p>From the article,</p>
<p>
</p>
<p><a href=“fab:”>quote</a></p>
<p>My apologies for not being clear with which meaning of “independent variable” I used. […] I was working off your talent “cannot be influenced much by effort” line, [interpreted] as statistical independence,
[/quote]
</p>
<p>The comments about variables not being able to acquire a correlation (upon condititioning) are equally nonsensical if effort cannot affect talent and if it can, so your confusion about that is independent of anything I have said.</p>
<p>
</p>
<p>Exactly what is the difference? You appear to be trying to replace standard math/statistical vocabulary with an ill-defined replacement terminology that banishes the phrases (that is, correct mathematical statements) you don’t like, such as in this case, “negative correlation between ASIAN and Talent given Score”. If there is a definition that other people can use to determine what is a legitimate “correlation” and what is an illegitimate “mechanical correlation” then of course the situation would be different, but it is not possible. There is no there, there.</p>
<p>
</p>
<p>It doesn’t? Because that is exactly how the rest of the world that uses conventional mathematics would describe it: a negative correlation conditional on the score. Do you know of some book or paper that uses fabrizio-math terms like “mechanical relationship” or “mechanical correlation” and explains what those are, because they are presumably different from ordinary non-mechanical correlations, but I’m not sure anyone can quite explain how.</p>
<p>In fact, some of the more common examples to illustrate the effect of conditioning are from college admission:</p>
<p>
</p>
<p>Maybe a textbook would help? </p>
<p>"For example, if the admission criteria to a certain graduate school call for either high grades as an undergraduate or special musical talents, then those two attributes will be found to be correlated (negatively) in the student population of that school, even if those attributes are uncorrelated in the population at large. " – J Pearl, Causality, (2000) p.17</p>
<p>
</p>
<p>Nice way to sweep your self-contradiction under the rug there, sis. Remind me again, is talent one of many variables that “cannot be influenced much by effort,” or do we “become better at things we practice”?</p>
<p>
</p>
<p>I’m so thankful, sis. I learned something new today from you, and while you were trying to embarrass me, I’m not sure it worked seeing as how you ever so conveniently ignored the name of the concept you’re using with your conditional correlation assertion: Berkson’s paradox.</p>
<p>So, uh, yeah, you’re right. There is a negative correlation between being Asian and talent when you fix a score at 750. But that negative correlation arises from a paradox / a fallacy / selection bias. I think Berkson would be rolling over in his grave if he knew someone was advocating for “meritocratic discounting” based on a PARADOX.</p>
<p>
</p>
<p>Sounds good, until…</p>
<p>
</p>
<p>By golly, that’s it! Nothing wholesome can come from a correlation “arising from a paradox”, how could I have missed that. The Fruit of the Poisonous Paradox. </p>
<p>Does it mean anything to say that something “arises from a paradox”? Is that another way of saying you don’t like it, or does it have a translation into ordinary people’s math language (the kind used in textbooks, papers or academic web sites) that can explain how the illegitimate paradoxical correlation differs from a patriotic, upstanding ordinary correlation? </p>
<p>A concrete example: in statistical regression models, bread and butter of econometrics, there are many similar “paradoxes” such as positive coefficients becoming negative when more variables are included. It occurs frequently in practical, numerical, empirical work. When that happens, do you have some criterion for deciding which regressions are bogus paradox-based garbage and which ones are acceptable in polite company? “The ones that I like” is not an answer.</p>
<p>
</p>
<p>Let me understand your logic better, since it seems a bit beyond the ordinary realm. If this type of result were called Berkson’s “phenomenon”, it would be legitimate to draw conclusions from it, but because it is sometimes called a “paradox”, or Berkson’s “bias”, it is not. It’s the name of the theorem that determines whether the application is valid. Is that the principle you are stating? In other words, depending on whether people call it a “selection effect” or “range restriction” or “sampling bias” (all three are sometimes used) or a “paradox”, the same piece of mathematics can become OK or not OK. </p>
<p>This seems like an important new breakthrough and I would like to make sure I quote and credit you properly on it.</p>