<p>“I have some interest in discussing regional preferences and in noticing that schools many on this board claim to be national turn out to be just as regional as the rest.”</p>
<ol>
<li><p>I don’t really consider this “analysis” to be methodologically worthwhile. GIGO. I keep mentioning that even though I live below the Mason-Dixon line, we’re getting counted as northeasterners. That’s just one methodological problem, and it’s a big one. This “analysis” has taken three of the highest scoring jurisdictions (which is kind of the playing field for the northeastern elite schools) and stuck 'em into the region that already starts out with the highest test scores. Just the admits from the Washington, DC area would have shifted 3% of total admits from the “northeast” (and they’ve put Texas, TEXAS in the southeast!! LOL!!). I’m sure there were other admits from the state of Virginia.</p></li>
<li><p>As well, related to the GIGO theme, I’m not sure the numbers match up with those of the most authoritative source, which would be Harvard’s admissions folks. Here’s a link:</p></li>
</ol>
<p><a href=“Admissions Statistics | Harvard”>https://college.harvard.edu/admissions/admissions-statistics</a></p>
<p>They cut the country differently, into more regions. As you can see from the link, the New England area accounts for 17% of enrollment, as does the SOUTH. The MID-ATLANTIC region, which is where I’d place DC, Maryland, and Virginia, has 23% of enrollment. Why, that’s even more than New England!</p>
<p>It’d be nice if Harvard’s table told us which states went into which regions, because then we could map their actual results with what is in this thread.</p>
<ol>
<li> I wouldn’t define a “national” university by such a trivial metric, even if I thought the data analysis was worth anything. As I said previously, to maximize geographical representation, you’ll have to give something else up. Here’s one: African Americans comprise 14% of the US population. Harvard this year had an admission rate for African Americans of about 12%. WUSTL was something less than 7%. Hispanics comprise 17% of the US population. Harvard’s admitted class is 13% Hispanic. WUSTL, less than 5%. Which do you think really represents national diversity?</li>
</ol>
<p>How does ethnic diversity intersect with geographical proportionality? Is it possible that maximizing one reduces the other?</p>
<p>What other factors does one sacrifice to have more proportional geographical representation? Which of those factors are actually more important to “national” status than more precisely proportional geographical representation?</p>