<p>"And the classification of Stanford and WUSTL, according to your chosen criteria, exemplified the silliness of the results. Perhaps not relying on questionable data might have saved some of it, but that was clearly the case here. Just think that students of Nashville who attend WUSTL are supporting the illusory notion of a national reach! How many miles separate both cities? "</p>
<p>About 300, just like the distance from Philadelphia to Boston, and just a little but less than the distance from SF to LA. I’d love to have the data take into centers of population or catchment areas too. I note you didn’t say anything when people talked about how people in NJ “go out of state” when that’s a heck of a lot less meaningful in small states. </p>
<p>“What if someone doesn’t want to apply to WashU because they <em>correctly</em> think that WashU has a far high proportion of students from the midwest than an NE school (or a west coast school, or a school in St. Louis, France), and that’s what they mean when they say WashU is too midwestern for them?”</p>
<p>Then whatever, since that would be a correct statement. It’s the incorrect statements I object to. </p>
<p>“But your thesis is pretty shallow if it doesn’t take into account these sorts of issues. Not only is some of the data just handled improperly, but there’s so much data that’s not here and thus not considered. At this point, it’s a cute statistic, but without the appropriate context, it’s not much more. It’s certainly not enough information to make judgments about which schools are truly “national” and which are not.”</p>
<p>In my very first post, I laid out assumptions that I explicitly said I knew were in here that might or might not be correct, such as that elite qualifications were equally distributed by region. I explicitly said that my guess is that these results reflect the regionality of the applicant pool. Throughout this entire discussion, I explicitly mentioned or acknowledged other assumptions or areas where I wish I had had the data cut that way but acknowledged that I didn’t. You would have a point if I pretended this was the data set to end all data sets. </p>
<p>"Internationals pay full tuition and schools need money. But I believe Pizzagirl and Poetgrl wouldn’t conider Columbia to be properly international till that number climbs to 95%. "</p>
<p>I never said anything about the % of internationals. I am not indistinguishable from Poetgrl. </p>
<p>There is such a thing as constructive criticism. One can express a contrary opinion without using dismissive, contemptuous language (“missing its mark by a mile and a half”, “utterly unconvincing”, “silliness of the results”.) </p>
<p>Virtually every college quality metric has problems with the choice of criteria or how we define the features we measure. Once we’ve eliminated as much of the slop as possible, do we wind up with plausible results? In this case, Harvard-Princteton-Yale are shown to have more “national” drawing power than Penn and Cornell. Pomona, Amherst, Williams and Swarthmore are shown to have more national drawing power than Colgate and Hamilton. These are plausible results, consistent with other measurements. </p>
<p>The more surprising finding is that WUSTL, NU, and Chicago seem to have more national drawing power than the Ivies. I suspect (assuming the data is accurate and ~appropriately categorized) this is more an accident of geography than an indicator of drawing power. It’s easier for more people to make their way to places near the center than to places near the edges. It may also be the case that the NE has greater concentrations of people who value what the most selective schools in their own region have to offer, while the most selective MW schools need to recruit farther afield.</p>
<p>Actually I would say WashU and UChicago do. NU’s pattern resembles that of the Ivies. See my post 2. </p>
<p>WashU and Chicago, like all schools, overindex to home region - but they don’t overindex AS MUCH as Ivies and NU do (in the 160 range versus the 200+ range). WashU and Chicago also have a secondary overindex - in the both cases to the Northeast at the 130 range. NU doesn’t; like the Ivies, it has average development in a second region and below average in the remaining two. </p>
<p>So, “WashU and Chicago do a better job of pulling out of home region than NU” is a true statement, which of course says nothing about quality or selectivity of any of those schools relative to one another, it’s just a commentary on the makeup of the student body. </p>
<p>“It may also be the case that the NE has greater concentrations of people who value what the most selective schools in their own region have to offer, while the most selective MW schools need to recruit farther afield.”</p>
<p>If I had been able to shape the data into “% of students coming from greater than x miles away,” I’d still have to normalize by % of the pop (or the college-age-pop, or the elite-worthy college-age-pop) who live greater than x miles away. In other words (and these numbers are ENTIRELY made up for illustrative purposes so no running with them to the bank) if you found that 80% of WashU students were from greater than 300 miles away but only 30% of Harvard students were, that wouldn’t mean anything if indeed 80% of the elite-worthy population lived greater than 300 miles from STL but only 30% of the elite-worthy population lived greater than 300 miles from Boston. You’d normalize, exactly like I did. </p>
<p>I agree that there are a lot of variables not considered in the simple data of the OP. Since we are talking about schools with high admissions standards don’t we need to look at population of students capable of handling the demanding workload, as opposed to straight population figures. Maybe the Ivies just can’t find enough qualified applicants from other regions. </p>
<p>I just find it so hard to draw any conclusions from this data other than “kids tend to go to schools closer to home.” Which isn’t exactly a shocking disclosure. </p>
<p>“For all the supposed johnny-come-lately status of WashU, it’s notable that they achieve a more national student body than ANY Ivy.”</p>
<p>This result is not surprising considering the fact the WashU is almost adjacent to the mean center of the United States population (currently located at Laclede County, Missouri).</p>
<p>As I said maybe 300 posts back, I wish I had the universe of those who were “elite-school” worthy. The universe I have here is “those who attended the top 20 or so schools” which is not exactly the same thing. </p>
<p>It may be that the Ivies can’t find enough qualified applicants from other regions because those regions are full of “dumber” kids, or they are equally smart but they simply go to the UCs, Michigan, Wisconsin and so forth, or that legacy, facbrat and/or “good neighbor” missions draw the pool back to the home region. All have some level of plausibility. </p>
<p>“This result is not surprising considering the fact the WashU is almost adjacent to the mean center of the United States population (currently located at Laclede County, Missouri).”</p>
<p>It’s a meaningless mean, though. Let’s say you restricted every college in the country to only admitting students who live with 300 miles (for the sake of argument). There are enough people in the northeast metropolis that stretches from BOS to DC to fill many of those schools. There aren’t enough people within 300 miles of STL to do so. </p>
<p>Yes, which is why states are imperfect and MSAs are better – which I’ve said repeatedly – and noting that “NJ and CT send a lot of kids out of state” is meaningless. </p>
<p>If I were to look only at applicant pools, I also wonder if the Boston B student who isn’t “Harvard worthy” is more likely to throw in an app to Harvard on the strength of “what the hell, why not give it a shot, nothing to lose” compared to, say, the Minneapolis B student who isn’t Harvard worthy. </p>
<p>Extrapolated more broadly, it would be interesting to know if the UNQUALIFIEDs (which is different from the merely not-admitted; I mean the “never really had a chance” crowd) has any skew. </p>
<p>In other words, I’d like to know a) at the MSA level, what % of elite-school-worthy students in that MSA applied to each college and b) at the college level, what % of the applicant pool they got from each MSA was elite-school-worthy in the first place. </p>
<p>Of course, even if you solved the question of using the SAT vs ACT, it begs the question whether you would just establish a uniform floor as being elite school worthy across all schools. </p>
<p>To come at this a different way, forget where students come from. </p>
<p>If you found that College A’s alums mostly settled in a handful of geographic locations, and College B’s alums were more evenly scattered all over the country, would that be a compelling piece to you, all else being equal between the two? </p>
<p>@Pizzagirl An earlier poster, @bp001, had thought that NE’erners were actually MORE likely to have left their home region when surprise! They are the least likely to leave their home region. But they also occupy proportionately more seats in elite schools, and those seats are disproportionately located in the NE.</p>
<p>Actually that is not what I said, but thanks for misquoting. What I said is more students were coming from the NE than were supported by the census numbers you were using. If this were true, your initial assumptions would be wrong. Your subsequent analysis proved that I was indeed correct that more students do come from the NE.(See post #264 If I compare them to the actual % of the population in each part of the country, I get these indexes:
Northeast 181, Midwest 74, South 66, West 91.
So … it is true that Northeasterners disproportionately “fill the seats” of the nation’s elite universities. The West is about average and the Midwest and South are below average.)</p>
<p>What percentage of out-of-region students does the AVERAGE school draw? Or even the average school in the bottom half of the US News top 100? I think you’d find that any of the schools we’re talking about (whether WUSTL or the Ivies) does have a lot more national drawing power than the typical university that the average student attends.</p>
<p>If we’re talking about effects on the social or academic atmosphere, it does not necessarily take too much yeast to leaven the dough. </p>