<p>I believe I was trying to measure student body diversity (from the perspective of geographic diversity, fully understanding that there are other measures of diversity beyond geographic). I wasn’t trying to measure drawing power, because I’d need the applicant pool data for that. Besides, all I need to do is look at # of applicants per bed in the aggregate to know what the drawing power is. </p>
<p>"There are two different hypotheses working at cross-purposes.</p>
<p>One is that the kids in the midwest and the south are poorer / dumber (lower SES as a whole, poorer NMF PSAT qualifiers). They might potentially be <em>interested</em> in elite NE schools, but they simply wouldn’t qualify for elite NE school admittance at the same rate as their northeast cousins.</p>
<p>The other is that the kids in the midwest and the south are equally as smart / qualified on a per capita basis, but they don’t BOTHER to apply to elite NE schools because a good portion of them are “snapped up” by Michigan, Wisconsin, etc."</p>
<p>Was the goal of your analysis to prove or disprove either of the above? Then you have to design your experiement in a more robust way. </p>
<p>For hypothesis 1, define an objective metric for lower SES (average household income adjusted for cost of living will do) and another objective metric for elite academic performance (could be AP score distribution, national award distribution, extracurricular performance etc.). AP score distribution and national award distributions are easy to get, extracurricular performance, not so much. But it is possible to set up a robust framework and conduct the analysis. Are you interested in doing that?</p>
<p>For hypothesis 2, again, you have to first objectively define “snapped up”, and then create a metric for it. Since “snapped up” is an imprecise scientific term, I will let you define that.</p>
<p>However, why does it matter which hypothesis is true if kids from all regions have access to quality education across the country?</p>
<p>“Was the goal of your analysis to prove or disprove either of the above? Then you have to design your experiement in a more robust way.”</p>
<p>It wasn’t to prove or disprove either of those; I see these as explanatory hypotheses now that I’ve played with the data.</p>
<p>BTW, I haven’t forgotten the spreadsheet I owe you; in the spirit of transparency, I noticed a mistake that I’ve since changed. Remember how I said that I was classifying W&L as a southern LAC even though the census data I’m using puts VA in the NE? Well, I had marked Rice as a western uni but the census data is classifying Texas (Texans) nto the South. I wanted to see what happened if I then made Rice a southern university. I just ran it all, and nothing really changes materially, except that the West now “loses” seats at the elite u table (since now there are only 2 schools comprising it, USC and Stanford). I only have two choices: Count Rice as Western but Texans as Southerners, or count Rice as Southern and Texans as Southerners. I can’t break out Texans from the rest of my southerners, unfortunately. And I need to set up some anonymous email, which I have to figure out how to do.</p>
<p>So then I would have expected the regional indexes to census to be evenly skewed – everyone at equal levels to home region and below average to other regions. But I didn’t find that. I found Brown, for example, to be more regional than Harvard. I found NU and NDame to be more regional than UChicago and WashU. I found differences in the LACs. Etc. </p>
<p>Oh, by the way? Everything that I said about WashU’s more-national student body compared to Harvard?
The regional geographic distribution of WashU and UChicago are virtually identical. Everything I said about WashU could be said about UChicago from a geography standpoint.</p>
<p>So UChicago has a more-national student body compared to Harvard. Is that equally as “controversial”? Somehow I’m betting that’s not as controversial, because for some odd reason UChicago is not the “third rail” that WashU is. </p>
<p>If colleges are happy to oblige qualified local kids, that doesn’t mean they won’t reach out to qualified non-local kids. That would turn private college into state universities. Why do you take such extreme positions?</p>
<p>"For hypothesis 1, define an objective metric for lower SES (average household income adjusted for cost of living will do) and another objective metric for elite academic performance (could be AP score distribution, national award distribution, extracurricular performance etc.). "</p>
<p>I wonder if the right objective metric would be average HH income adjusted for COL, or rather % of a region (or MSA) that makes over $X per year (the $X adjusted for COL).</p>
<p>Given that only about half of high schools in this country even offer APs, that seems a very imperfect metric.</p>
<p>Extracurricular performance would be a really difficult one to quantify. Who’s “smarter,” the class president or the newspaper editor? As for sports, certain sports skew to / from different parts of the country.</p>
<p>If we could agree on a “floor” of ACT/SAT, I’d be willing to accept students who scored over that as “elite-school-worthy,” but then I’d need access to their choices in totality - because so many will be captured by state flagships once you’re outside the NE. </p>
<p>Yes, it is indeed true that more than half of high schools in the country don’t offer APs. This merely means that the students of these schools (unless they take AP courses on their own) are not elite-college material. Remember, we are not talking run of the mill college here. We are talking about the best of the best, the student population that are likely candidates for admission to the most elite schools. Hence, you would want to count which state sends most kids to the Math Olympiads, if you want to talk getting into an elite math program. </p>
<p>SES goes up with disposable income. It is not cut off based. </p>
<p>“My opinion is that people can’t handle that WashU might reject them when an IVY won’t. Who do they think they are?”</p>
<p>I must say that I never even heard about Washington University till I read this thread. It’s interesting. I have worked with graduates from every single school in the top-20 US News list except WashU.</p>
<h1>618 There are two different hypotheses working at cross-purposes.</h1>
<p>I think it is possible both that:</p>
<p>1) Southern and midwestern students with the potential to excel academically may come from backgrounds that don’t put an emphasis on creating the sort of profile attractive to the most competitive colleges, and thus they aren’t competitive for those schools whether or not they are interested in them.</p>
<p>2)There are southern and midwestern students from extremely enriched backgrounds, who probably do have the academic profiles attractive to the most competitive colleges, but they are only interested in their state flagship.</p>
<p>adding: I can imagine other possibilities as well. I am restraining myself. ; ) but do wonder if you really meant to write “dumber” : (</p>
<p>“There are southern and midwestern students from extremely enriched backgrounds, who probably do have the academic profiles attractive to the most competitive colleges, but they are only interested in their state flagship.”</p>
<p>Or they come to the Ivies. Remember, the Ivies still take a large number of Southern and Midwestern students.</p>
<p>As for WashU - again, so what? That’s likely a function of where you live. Of the NE schools, my circle is relatively heavily loaded to Penn. Of the midwest schools, my circle is heavily loaded to U of Chicago and NU – indeed, I’ve hired 2 U of Chicago people in the past year and just interviewed a U of Chicago grad for a position yesterday. So you have different circles. Hunt’s circles likely load heavily to Yale. And so on and so forth. So it goes. </p>
<p>“As for WashU - again, so what? That’s likely a function of where you live. Of the NE schools, my circle is relatively heavily loaded to Penn. Of the midwest schools, my circle is heavily loaded to U of Chicago and NU – indeed, I’ve hired 2 U of Chicago people in the past year and just interviewed a U of Chicago grad for a position yesterday. So you have different circles. Hunt’s circles likely load heavily to Yale. And so on and so forth. So it goes.”</p>
<p>I agree. People go where they want to go, and maintain the circle that they want to maintain. Why so much sensitivity about what NE folks think about WashU? It doesn’t bother me that someone in Missouri haven’t heard about the academic institutions that I attended. </p>
<p>On another note, you earlier wrote:</p>
<p>"I think most normal people would say - if 23% of the population of the country is found in the NE, and 41% of College X student body from the NE, then College X skews to the NE. "</p>
<p>I learned mathematical methods/modeling from an Ivy. Given your academic background above, perhaps there is a difference between an Ivy education and a top-20 education, after all.</p>
<p>*Or they come to the Ivies. Remember, the Ivies still take a large number of Southern and Midwestern students. *</p>
<p>Right. I was trying to be brief and not spell out all the possibilities. I’m pretty sure PG has no difficulty with the idea students aspire to the Ivies. </p>
<p>“There are southern and midwestern students from extremely enriched backgrounds, who probably do have the academic profiles attractive to the most competitive colleges, but they are only interested in their state flagship.”</p>
<p>Or they come to the Ivies. Remember, the Ivies still take a large number of Southern and Midwestern students."</p>
<p>Oh yes! Looking at my data, of the 58,000 Ivy undergrad seats, a full 5,780 are taken up by midwesterners and 7,890 by southerners. Don’t have my calculator handy, but that’s roughly 10% and 13% - so 23% of the Ivy bodies come from the midwest and south. That’s not all that impressive when those two regions together account for over 50% of the population. And yeah, yeah, they’re dumb and poor and deluded by football and all, but still. </p>
<p>You would have a better argument if you focused on MIT, since MIT does a somewhat better job than the Ivies in attracting a nationally representative student body. </p>
<p>"Oh yes! Looking at my data, of the 58,000 Ivy undergrad seats, a full 5,780 are taken up by midwesterners and 7,890 by southerners. Don’t have my calculator handy, but that’s roughly 10% and 13% - so 23% of the Ivy bodies come from the midwest and south. That’s not all that impressive when those two regions together account for over 50% of the population. "</p>
<p>Why is this surprising? The remaining 27% could go to the state schools like you suggested. That’s an even split.</p>
<p>You are really hung up on proportional representation, aren’t you? </p>
<p>* And yeah, yeah, they’re dumb and poor and deluded by football and all, but still.*</p>
<p>PG: that’s more than a little ugly and doesn’t really even make any sense… dumb and poor or smart and rich - if they want to stay close to home, even if it is because they value college sports, what is the problem? Just because someone doesn’t value what you value doesn’t make them stupid or wrong… just different than you.</p>