Want diversity w/o Affirmative Action? Don't rely on the SAT

<p>epiphany,</p>

<p>Given how you have misunderstood my post #500, I’ll take some responsibility for not wording it clearly enough. However, I believe that had you read it more calmly and carefully, you would have responded differently or not at all.</p>

<p>I wrote that “…Espenshade and Chung found that being Asian was the equivalent of a 220 point SAT deduction.” That is, the deduction doesn’t actually happen. It’s as if it happens, hence equivalent. My statement would have been defamatory if I had written, “…Espenshade and Chung found that being Asian caused a 220 point SAT deduction.” However, that is neither what they found nor what happens in practice. As you can plainly see, I did not use such wording.</p>

<p>Later on, I wrote “…the practice of certain private schools actually artificially reducing the scores of certain applicants as acceptable.” Notice how I said “certain private schools.” I did not mention Princeton, and for that matter, I did not mention any university or school by name in my post #500. I concede that I should have been more specific. cheers has told me that certain K-12 private Catholic schools engage in this practice, and after she described it, she did not disapprove of its “deliberate fraud.” I’m glad that you disapprove.</p>

<p>I understand that you and most other supporters of race-based admissions weren’t too excited about Dr. Espenshade and Professor Chung’s research. I remind you that after publishing their papers on this issue, both publicly reaffirmed their commitment to affirmative action. (Political correctness?) Please do not impugn their characters by suggesting that they are not true researchers. I never stated that any student “‘has’ to have a certain score to be accepted to any U in this country,” and I have never disputed what you wrote. I thus fail to see why you felt the need to say that.</p>

<p>I have not “confuse[d] the existence of high scores as a trend among high percentages of Asian students, with a REQUIREMENT that any institution has toward <em>Asian</em> applicants.” I see it as more of a case in self-selection.</p>

<p>I remind you that I’m not for an admissions system that is based purely on statistics. That is a gross caricature that supporters of race-based admissions bring out when their opponents propose race-neutral admissions.</p>

<p>I contest your statement that “The trend toward higher scores is the result of expectations among Asian families…” The causality doesn’t make sense. You say that the expectations among Asian families cause the trend toward higher scores in this group. Yet, many non-Asian families have similar expectations regarding high scores. So, under your logic, why don't all groups exhibit "Asian-like" scores?</p>

<p>I hope you’re not suggesting that these Asian applicants are absolutely weaker in these “other factors” compared to their peers. In spite of Berkeley and LA’s efforts to make their admissions more “holistic” while still race-neutral, Asians still make up the plurality on both campuses. Maybe they’re not quite so bad in the “other factors” department.</p>

<p>Luckily, in academics, statements like your penultimate one are of little worth. An empirical paper must be refuted with modeling and statistics, not hubris. If you’re not able to find a paper besides Professor Kidder’s, I suggest you minimize your dislike of Espenshade and Chung’s findings. A paper that uses law school data to refute a paper based on undergraduate data is inadequate as a response.</p>

<p>There’s no letter ‘c’ in the Greek alphabet. Since there is a k, Perikles is more accurate.</p>

<p>It seems a lot of people have been citing Espenshade and Chung's paper a lot, and I don't know if they have all read it. </p>

<p>They published their paper in the Social Science Quaterly. Take a look at this rankings of journals in economics and social science and determine for yourself where this journal ranks;</p>

<p><a href="http://www.econ.vu.nl/econometriclinks/rankings/eerjournalsranking2002.html%5B/url%5D"&gt;http://www.econ.vu.nl/econometriclinks/rankings/eerjournalsranking2002.html&lt;/a&gt;&lt;/p>

<p>You get the idea. I don't know of anyone working hard to publish in this journal in the profession.</p>

<p>There are so many things wrong with this paper no wonder it was only accepted by the Social Science Quarterly.</p>

<p>"Here we use micro simulations to illustrate what the effects might
be if one were to withdraw preferences for different student groups."</p>

<p>The study is a SIMULATION of what will happen if race is removed from the application process. I think they forgot that in AMERICA the economic and social variables are functions of race, where you leave, how much education your parents have etc, this is especially true for blacks. There are so many things you have to control for I don't think this study can be done.</p>

<p>Logistic regression? For this kind of study. No wonder they got everything to come out right. The results is driven by the methodology and the fact that Asians on average have higher scores than the general popluation. It does not mean thant Asians are held to a higher standards. Everything will be overestimated I quit reading after the description of the methodology. This will have been a total reject if I was peer reviewing this paper.</p>

<p>No wonder the paper was published by the Social Science Quarterly. I guess Princeton carries a lot of weight for this journal. This paper wont pass the first round of any peer review from a reputable journal.</p>

<p>tega,</p>

<p>I disagree that the ranking table you referred to is appropriate in evaluating the quality of Social Science Quarterly. I believe it is not.</p>

<p>The table comes from the paper, “Rankings of Academic Journals and Institutions in Economics.” As the web page mentions, it is a ranking of economic journals. The phrase “social science” is not found anywhere on the page.</p>

<p>Using similar logic, I could show that few in the “profession” are “working hard to publish in” the American Economic Review because almost nobody in art history cites from it in their papers.</p>

<p>“Wait a minute,” you might say. “We’re talking about economics here, so why did are you bringing up art history? It is irrelevant.”</p>

<p>To which I respond, exactly. A table ranking the quality of economic journals is irrelevant in determining how good Social Science Quarterly is. The table concerned itself exclusively with “academic journals and institutions in economics.” I believe it is inappropriate to use it to gauge the level of Social Science Quarterly.</p>

<p>Even if Social Science Quarterly were a “lesser” journal in sociology, I would still disagree with your suggestion that to be of any worth, a paper must be published in a top-tier journal. I believe that it is disrespectful to the hundreds of professors who publish in these “lesser” journals to insinuate that their work is not important.</p>

<p>Furthermore, in the paper that is the subject of this thread, Professors Alon and Tienda wrote that their data confirmed the results from Espenshade, Chung, and Walling in 2004, a closely related but different paper from the controversial 2005 Espenshade and Chung paper. About half a year ago, Professors Massey and Mooney published a paper in Social Problems that cited the 2005 paper from Espenshade and Chung without any mention that it was inferior or had been refuted.</p>

<p>I understand that many persons, particularly those who support racial preferences, strongly dislike the research done in this area by Professors Espenshade and Chung. However, I fail to see what good comes from disparaging either their academic integrity or the quality of the journal they published in.</p>

<p>Social Science Quarterly publishes research that deals with social science. It accepts any work done in economics. Economics is a social science.</p>

<p>"Nationally recognized as one of the top journals in the field, Social Science Quarterly (SSQ) publishes current research on a broad range of topics including political science, sociology, economics, history, social work, geography, international studies, and women's studies. SSQ is the journal of the Southwestern Social Science Association."</p>

<p>AER does not accept art history because it is not for art history. </p>

<p>“Professors Massey and Mooney published a paper in Social Problems that cited the 2005 paper from Espenshade and Chung without any mention that it was inferior or had been refuted."</p>

<p>I am not aware of this paper. Could you please provide a summary of their methodology and data? OR you can post a link to the paper so that I can read it myself. In data analysis, most of the time, your answers are driven by the methodology used. That is why there are ways to check for the robustness of results. For example, you can check your results using subsets of the data to see if the results are still consistent with using the full sample. </p>

<p>The logit regression that they used influenced their results. Please find a good econometrics book and read about logit estimation, maybe, it will be clear to you why this estimation is not what they should be using for the kind of question they want to answer. IMHO the question they want to answer can’t be solved using the data set they have because they don't have a lot of the variables they need to control for.</p>

<p>"I understand that many persons, particularly those who support racial preferences, strongly dislike the research done in this area by Professors Espenshade and Chung. However"</p>

<p>Although I am not particularly in support of racial preferences (I believe socio-economic circumstances should play a significant role), I will be naive to expect a lot of Blacks in America to all of a sudden break free from centuries of oppressed racist agenda to achieve at the same level as Whites and Asians. Blacks in America have only been "free" since 1965. They have suffered from years of institutionalized racism that have made it difficult for them to be fully invested citizens of America. I support affirmation action because I think we as society have to help Blacks overcome some of the burdens they have had to endure over the years.</p>

<p>"I fail to see what good comes from disparaging either their academic integrity or the quality of the journal they published in."</p>

<p>I am not disparaging their academic work. I don’t know if you are in college yet, but in ACADEMIA you question the work of people all the time; are they using the right data set, is the results driven by their estimation procedure, did they do any robust test, etc. This is because we keep building on what has already being done and we have to make sure that we are building on a solid foundation. Their results were driven by their estimation procedure. It is clear that they have an agenda from reading the first couple of pages of their paper. I do not dislike their research; they just did not use the right estimation.</p>

<p>I'm curious, which of the conclusions of the paper do you think are invalidated by their methodology. I think this is an interesting aside in a long thread and I was hoping to avoid rereading the whole thing. You state "IMHO the question they want to answer can’t be solved using the data set they have because they don't have a lot of the variables they need to control for." I am also curious what kind of data you think would be required to answer the question. I'm not arguing anything right now just interested.</p>

<p>tega,</p>

<p>I agree that economics is a social science and is therefore appropriate for publication in Social Science Quarterly. However, from the description you quoted, the journal is not focused exclusively on economics. It also includes papers from political science, sociology, history, social work, geography, international studies, and gender studies, basically any social science. Given that its scope of topics is very broad, it is not surprising that compared to journals that deal exclusively with economics or subfields within economics, Social Science Quarterly does not have a high cross citation impact.</p>

<p>I have taken both introductory micro and introductory macroeconomics. Furthermore, my uncle is an economist, so I am somewhat familiar with the subfields. In the top twenty journals listed in the rankings, only one – the Journal of Human Resources, ranked 17 – stands out as not being an “obvious” economic journal. With the possible exception of this journal, all the ones ranked concern themselves with economics-related papers. As a point of reference, I don’t think there are any gender studies papers in the Quarterly Journal of Economics, but such papers can be found in Social Science Quarterly.</p>

<p>I still do not believe that the rankings are appropriate to judge the quality of Social Science Quarterly.</p>

<p>The recent paper from Professors Massey and Mooney is titled “The Effects of America’s Three Affirmative Action Programs on Academic Performance.” The abstract is [url= <a href="http://caliber.ucpress.net/doi/abs/10.1525/sp.2007.54.1.99%5Dhere%5B/url"&gt;http://caliber.ucpress.net/doi/abs/10.1525/sp.2007.54.1.99]here[/url&lt;/a&gt;].&lt;/p>

<p>To clarify, I do not “expect a lot of blacks in America to all of a sudden break free from centuries of oppressed racist agenda to achieve at the same level as whites and Asians.” I’m not interested in equality of result. I’m far more interested in equality of opportunity. I care greatly if students are denied a chance to compete in the first place. I also think “we as society have to help blacks overcome some of the burdens they have had to endure over the years.” I simply disagree that considering race is a necessary help.</p>

<p>I expect academic work to be questioned. That is proper and wholly appropriate. However, I do not believe that dismissing a journal as not reputable is either proper or appropriate.</p>

<p>Curious14, They are assuming that race, SAT, GPA, legacy and being an athlete are the most important things that addcoms look at. Elite schools consider more than these in making their decisions. </p>

<p>They need info on the income levels of the students, their essays(they have to find some metric to compare all the essays written by the students), where the students live etc.</p>

<p>The logistic regression predicts the probability of being accepted to the three schools schools given race neutral, athletic neutral and legacy neutral admission process. </p>

<p>Asians as a group applying to the three schools they considered might have on average higher SAT score than Blacks, but very few Asians are athlete, so if you run a regression where athletic preferences are eliminated you are going to confirm what is already in the data, that Asians have higher SAT score and they more will be admitted if much emphasis is placed on SAT. This does not mean that Asians are being descriminated against. They should be holding the income level of the applicants constant, their essay scores, where they live, etc all these variables play a significant role in the admission process, and they did not account for these variables in their paper.</p>

<p>Fabrizio,</p>

<p>"I’m far more interested in equality of opportunity. I care greatly if students are denied a chance to compete in the first place."</p>

<p>Blacks in America have never had an equal opportunity to compete from the word go!</p>

<p>tega,</p>

<p>I disagree that American blacks have never had an equal opportunity to compete “from the word go.” I have not read any recent newspaper account of a proctor refusing to allow a black student to take a standardized test. No agency has recently reported about any governor blocking the doorway so as to prevent black students from entering the school.</p>

<p>I do agree that our country used to absolutely deny them equal opportunity. However, I respectfully disagree that they’ve never had equal opportunity. The amount of opportunity now is greater than at any time before, and there is still more room to go.</p>