how much do "connections" help when applying to an ivy league?

Applying to College
41

<p>

</p>

<p>I’m not persuaded there is such a thing as a geographic advantage to being from Idaho. Princeton helpfully tells us where their entering class came from. The Princeton class of 2014 has exactly 1 Idahoan and 176 New Jerseyans. Now surely more New Jerseyans than Idahoans applied; we don’t have those figures. But we do have a loose proxy: the College Board tells us, for each state, how many college-bound seniors in the HS class of 2010 sent one or more SAT score reports to the most popular colleges. Since Princeton requires SAT Subject Test reports from every applicant, anyone who completes an application sent at least one SAT score report. No doubt some sent score reports but never completed their application, but we can say that the number of Idahoans sending SAT score reports to Princeton represents an upper bound on the number of Idahoans who applied; and the same for New Jersey. So . . . 52 Idahoans sent SAT score reports to Princeton, and 1 ultimately enrolled, so that 1.92% of Idahoans sending SAT score reports to Princeton actually ended up in the Class of 2014. Meanwhile 3,245 New Jerseyans sent SAT score reports to Princeton, and 176 ultimately enrolled, so that 5.42% of New Jerseyans sending SAT score reports to Princeton ultimately enrolled. Does that mean New Jerseyans were admitted at 2.5 times the rate as Idahoans? Well, we can’t conclude that; it could be that a lower percentage of Idahoans actually completed their applications, or that Princeton’s yield of Idahoans offered admission was significantly lower. But from the limited data available, it sure doesn’t look like there’s a significant admissions advantage to being from Idaho; if anything, the data suggest the opposite. And the same is true, by the way, for virtually every “underrepresented” state. I think the idea that applicants from underrepresented states have a big leg up in admissions is just urban myth. And the adcoms confirm that pretty consistently. They consider geographic diversity. They’d like to be able to boast at least one from each state, and to get one they might need to make offers to two. But once they’re reasonably confident they’ve bagged their one, they have no reason to systematically advantage applicants from any particular state. Not like legacies, for whom at Penn, at least, there’s roughly a 2-to-1 advantage. </p>

<p>

</p>

<p>So where do they say this? I haven’t seen this. And if it were the case, then why do they feel compelled to add a legacy preference–which they say they do? And even if the legacy pool is somewhat stronger than the general applicant pool, I find it hard to believe that at a school like Penn the legacy pool is TWICE as strong as the general applicant pool. Yet legacies are admitted at roughly twice the rate.</p>

42

<p>bc:</p>

<p>New Jersey residents sending SAT scores to P’ton also includes some faculty brats, which would tend to receive a boost (I would guess). Heck, New Jersey “residents” could also include faculty brats from Penn applying to P’ton.</p>

43

<p>Dean J:</p>

<p>That is nice to hear. </p>

<p>I’ve heard of other colleges where that is not the case - where legacy applicants are rated into categories depending upon whether the parents are regular donors and whether they are large donors. I read about one college where there is no legacy preference if the parent has not donated regularly to the college or been active in alumni clubs.</p>

<p>There was a study a couple years ago that said that if an alum’s child is denied admission, the parent is much more likely to stop giving to the college. That is why so many colleges have such absurdly huge waiting lists - they can put many children of alums on the waiting list and try to convince the parent that their child was ALMOST admitted - to let them down easier.</p>

44

<p>Sometimes the waiting lists are not just to let legacy parents down gently, they’re also to appease the high-end private feeder schools when an outright denial would provoke a severe protest.</p>

45

<p>Below i have taken the raw data that i’ve found about Penn’s class of 2010 (so a bit outdated as the school has become more competitive) and triangulated it with other data sources that i have found and here is how legacy admits shake out vs non-legacies for ED, RD, and Full-cycle. The big takeaway is that legacy status (even though the pdf link that was posted above says otherwise) does play an major role, not just in ED, but also in RD. From this data you can see that during ED legacies are admitted at a 1.9x multiple, while in RD they are admitted at a 1.4x multiple, and full cycle its 1.8x. </p>

<p>Class of 2010: </p>

<pre><code> Accepted Applied Rate Multiple
</code></pre>

<p>EARLY DECISION:<br>
Legacy 254 523 48.6% 1.9x
Non-Legacy 926 3,625 25.5%<br>
Total 1,180 4,148 28.4% </p>

<p>REGULAR DECISION:<br>
Legacy 137 736 18.6% 1.4x
Non-Legacy 2,305 16,964 13.6%<br>
Total 2,442 17,700 13.8% </p>

<p>TOTAL:<br>
Legacy 391 1,259 31.1% 1.8x
Non-Legacy 3,231 19,220 16.8%<br>
Total 3,622 20,479 17.7% </p>

<p>Having attended Penn I can tell you anecdotally that I was shocked my freshman year when I arrived at Penn thinking that a ton of people were going to be uber-smart and then I realized that there were a ton of legacies and wealthy non-legacies (i.e. father in an MD at some investment bank, brazilian billionaire, etc. that donated $) that were, in my opinion, not as that well qualified or smart. </p>

<p>However, eventually i realized that they provided the most valuable thing that I could have acquired at Penn (well in addition to the Wharton name on the CV) - an incredible network. From a business perspective, my network (particularly driven by these type of wealthy legacies/non-legacies) has enhanced my career significantly and has provided me with opportunities I could not have had without them.</p>

46

<p>oh yeah and to go back to the OP…

  • yes legacies help so she does have that going for her…especially if she’s applying ED.
  • based on anecdotal evidence from my experience at Penn I knew a family who had one member that donated a significant amount of money and during my 4 years there were 6 family members attending (2 of his kids and 4 of his nephews) - all of them complete idiots. :)</p>

<p>So yes it helps a lot if they donate a lot of money - even if its her uncle. :)</p>

47

<p>

If the things that support your thesis are true, even if schools deny them, and the things that don’t support your thesis are myths, even if the schools assert them, it’s kind of hard to discuss this with you.</p>

48

<p>A sibling may not officially be considered a legacy. However, I’ve read at least one admissions officer say that if their brother or sister is already attending a college and is doing well, they try to accept their sibling if they are qualified.</p>

49

<p>I also read about one admissions office where they keep careful track of how past students at various high schools have done at their college. If the students did well, they gave an extra bump to students from that school. That also can be self-perpetuating, because the students who were stuck in the worst high schools may have the most difficulty adjusting.</p>

50

<p>

</p>

<p>Hunt, the Penn data speak for themselves on the size of the legacy advantage, roughly 2-to-1. And I’ve talked to a lot of admissions officers about the so-called “underrepresented state” advantage. They all say the same thing: yes, they like geographic diversity but it’s just a very small factor, not nearly as big a deal as people from “overrepresented” states think it is, so if you’re from an underrepresented state don’t expect it to be much of a boost to your application. I can show you tons more examples from the Princeton data. Here are how many kids from the “underrepresented” Great Plains and Rocky Mountain states ended up in Princeton’s Class of 2014, followed by the number from that state who submitted SAT scores to Princeton, which we can again take as an upper bound on applications:</p>

<p>Idaho 1 / 52
Kansas 0 / 98
Montana 4 / 37
Nebraska 1 / 54
Nevada 3 / 114
New Mexico 4 / 84
North Dakota 0 / 15
Oklahoma 5 / 106
South Dakota 1 / 22
Utah 1 / 90
Wyoming 1 / <5
11-state total 21 / 676 = 3.1%</p>

<p>As for “overrepresented states”:</p>

<p>Connecticut 50 / (data not available but < 751) = > 6.7%
DC 10 / 161 = 6.2%
Maryland 59 / 1,064 = 5.5%
Massachusetts 53 / (data not available but < 1,454) = > 3.6%
New Jersey 176 / 3,245 = 5.4%
New York 105 / (data not available but < 3,358) = > 3.1%
Pennsylvania 62 / 1,830 = 3.4%</p>

<p>Data are not available for some states because for each state the College Board lists only the 49 schools receiving the most SAT score reports.</p>

<p>Now obviously the data are limited in what they tell us, but the pattern is clear and unmistakable: the percentage of SAT-submitters who end up actually enrolled as freshmen at Princeton is significantly HIGHER in the “overrepresented states” than in the “underrepresented states.”</p>

<p>You are of course free to refuse to discuss this further if you like; I know it’s sometimes upsetting to have your preconceived worldview challenged. But I repeat: there’s simply no statistical evidence of a significant admissions advantage to being from an “underrepresented state.” The colleges all want at least one enrolled freshman from each state, but once they’ve made that minimal quota, state of residence appears not to matter much, if at all. And sometimes they don’t even get that one. Just ask the 98 Kansans who got as far as submitting SAT scores to Princeton, some sizable fraction of whom no doubt completed applications, none of whom walked through the gates in this fall’s entering class.</p>

51

<p>

That’s *exactly *what they say about legacy.</p>

<p>Those data on states aren’t too helpful without knowing something about the stats of the applicants.</p>

52

<p>

</p>

<p>True enough, and we can say the exact same thing about legacies. What’s striking, though, is that when all is said an done Penn admits legacies at twice the rate of non-legacies; whereas the percentage of kids who ultimately attend elite schools from “underrepresented” states is vanishingly small. You posit the legacy advantage is due to stronger stats. I say, show me some data.</p>

<p>And the weak outcomes for kids from “underrepresented” states CAN’T be because they all have bad stats. Some of them need to be in the top 10% of their class. Some do get top SAT scores. We know that because to even apply to a school like Princeton they needed to take SAT Subject tests, and we have state-by-state data on SAT subject test score distributions. </p>

<p>Take Kansas, for example, which sent no one to Princeton this fall despite the fact that 98 Kansas kids sent Princeton their SAT scores. Only 430 Kansans took SAT subject tests–meaning nearly 1 in every 4 SAT Subject Test-takers in Kansas sent their scores to Princeton. The most popular was Math Level 2, which 293 Kansans took, scoring a mean of 691 with a standard deviation of 82. A whole bunch of those, 98 to be exact (as many as sent scores to Princeton, coincidentally), scored in the 750-800 range. The SAT Math Level 2-takers who also took the SAT Reasoning (SAT I) test had mean CR scores of 658, M of 705, and W of 651. So this group includes some good, some very good, and some outstanding students.</p>

<p>Were the New Jerseyans any better? No. Interestingly enough, the 7,009 New Jerseyans who took SAT Math Level 2 had an identical mean of 691, with 2,399 (substantially fewer than the 3,245 who sent SAT scores to Princeton) scoring in the 750-800 range. NJ SAT Math 2-takers who also took the SAT Reasoning Test had slightly lower mean CR (645) and M (703) and slightly higher W (661) scores than their Kansas counterparts. </p>

<p>The next most popular Subject Test in Kansas was Lit, where 202 test-takers scored a mean of 649; but on this test only 27 (13% of those taking the test) scored in the 750-800 range. In New Jersey that test was third most popular with 4,922 takers scoring a mean of 620; of those, only 320 (7% of those taking the test) scored in the 750-800 range.</p>

<p>Third most popular in Kansas was US History where 162 takers scored a mean of 672, with 41 (27% of test-takers) scoring in the 750-800 range; while in New Jersey that test was second most popular, with 5,730 takers scoring a mean of 650, and with 953 (17% of test-takers) scoring in the 750-800 range.</p>

<p>After those three popular tests, the numbers taking any of the other Subject Tests falls off dramatically in both states, and with somewhat more mixed outcomes—Kansans scoring better in Bio-M and Chemistry, New Jerseyans scoring slightly better in World History and Physics (though only 9 Kansans took the World History test so that comparison is pretty meaningless).</p>

<p>Conclusion: the number of SAT Subject Test-takers is much larger in NJ, but the score distribution is actually slightly stronger among the Kansans who took SAT Subject Tests. Nearly half of New Jerseyans who took SAT Subject Tests sent their scores to Princeton, whereas slightly less than 1 in 4 Kansans who took SAT Subject Tests did. Now from this data we can’t be absolutely certain that the Kansans who sent scores to Princeton had stats as strong or stronger than the New Jerseyans who sent their scores. It’s possible, I suppose, that the New Jerseyans who sent Princeton their SAT scores all came from the top half of the SAT Subject Test score distribution, while the Kansans who sent Princeton their scores came from lower down in that state’s distribution. But a more reasonable assumption is that both groups drew broadly from across the score distribution, but skewed toward higher scores as those would be the students most optimistic about their chances of gaining admission—and the ones who would be encouraged by teachers, GCs and families to go for it. And if that’s the case, then we can reasonably infer that the stats of the Kansas applicants were probably at least roughly comparable to the stats of the New Jersey applicants, or possibly even slightly higher. Yet a far higher percentage of New Jerseyans—actually, infinitely higher in this case, as Kansas’ rate was zero—ended up at Princeton.</p>

<p>Now I’m not suggesting that Princeton systematically FAVORS New Jersey residents–at least not consciously, though that could be the incidental effect of a bias in a favor of legacies, accommodations for Princeton faculty and staff, and affirmative action for URMs who probably make up a larger percentage of New Jersey’s college applicant pool. I am suggesting there is absolutely no statistical evidence of a geographic bias in favor of “underrepresented” states, at least not one that shows up in actual enrollment data.</p>

53

<p>The home state of a university will favor students from their state. Harvard states that they are Massachusetts’ university and will favor a student from their state. This helps when projects go to the state, city, other governmental bodies and business community.</p>

54

<p>Do we know anything about acceptance rates? I think that’s the most important piece that’s missing here. In my admittedly limited experience, it seems that students who come from more distant regions are less likely to attend an east coast school, especially when they are being wooed with scholarships and other incentives to schools closer to home.</p>

55

<p>The only data I’ve ever seen is the Harvard Fitzsimmons quote from some years back that the average SAT for legacy admits was with 2 points of the class in one of Golden’s articles. My simple assertion is that the legacy pool must necessarily be stronger if you can admit at a 3x rate and still get the same average. Try that from a merely equally qualified pool. The other explanation would be that legacy admissions are more heavily concentrated among the higher scorers. If so, thats not much of an “advantage”. </p>

<p>

Anyone care to list the Ivy’s official enunciations of this preference?</p>

56

<p>The number they used to throw around was that roughly 20% of the students at a school were legacy admits who go in there simply because they were legacy (this is back more then a few years). I had a cousin who attended Dartmouth who made a young fortune writing term papers for these kind of kids, got flown around in private jets, etc…and many of them were the stereotype, of the rich kid who got in because of their family…</p>

<p>And there is a pretty good example of this from that generation. If you look at the stats on the most recent past president, he got into Yale with C grades, and got into Harvard Business school with a C average…and I don’t think for his generation he was necessarily an outlier among legacy admits, either, I suspect there were a lot more of this type back then. </p>

<p>There has definitely been a trend away from this, but it still exists. Recently I believe the Times had an article about the changing nature of admits, and one of the reasons they were pulling back on legacy admits is that what they found that was with alumni giving, legacy admits didn’t seem to change the amount donated to the school, so one of the arguments about legacy admits, that it encouraged rich alumni to give money with the quid pro quo of getting their kid in to the school. The article also I recall gave indications that the old family legacy admits has been the scions of the many ‘instant’ billionairs of the tech generation, that they may be getting preferential admits because their family is well off and can be expected to donate…I will note that there was very little data about this, the evidence tended to be anecdotal. </p>

<p>I think it also depends on the school as well, it seems like some schools have been more vigorous in changing legacy admits then other schools, which doesn’t surprise me. I would hazard a guess that Harvard and Yale might be more likely then let’s say columbia to weight the legacy effect (note, it is a guess, not fact), simply based on the culture.</p>

<p>As far as the scions of Ivy league graduates being better prepared or ‘better students’, I think there is more then a bit of self serving elitism in there. Sure, if you are talking about the scions of the very well off, of the dynasties who go there and are worth more then Croesus, the preparation is better, these are the people whose kids get into the elite prep schools like Andover and Choate, and obviously have no trouble shelling out for tutors and such…but using that as an argument that the legacy admits are superior to others simply because that leaves out some very real facts, that the scions of the well off, despite all the advantages, often turn our mediocre, the same as kids of ‘regular’ families. There are plenty of Ivy league graduates of rich families and not so rich families who end up doing nothing important, there are plenty of kids who show promise, get into college and then coast, it is statistically ridiculous to claim that the children of alumni are necessarily better then other students…</p>

<p>The real question here is not if legacy admits are equally qualified (or better qualified) then a typical admit, and got a bump for being a legacy, but rather are they being admitted with scores and such less then the bell curve (obviously, there are going to be admits from any group that are at the top). And at least one study seems to imply that schools are admitting legacy kids with worse stats then the typical, that the argument Hunt and others made that legacy kids are going to be better prepared, simply isn’t true:</p>

<p>[News:</a> Legacy Admits: More Money, Lower Scores - Inside Higher Ed](<a href=“http://www.insidehighered.com/news/2008/08/04/legacy]News:”>http://www.insidehighered.com/news/2008/08/04/legacy)</p>

<p>A more recent piece, even more detailed:</p>

<p>[News:</a> Legacy of Bias - Inside Higher Ed](<a href=“http://www.insidehighered.com/news/2010/09/22/legacy]News:”>http://www.insidehighered.com/news/2010/09/22/legacy)</p>

<p>If what this says is true (and look at who wrote this, it is a right wing think tank, citing sources like the Wall St Journal), then I leave it to you to decide if legacy admits are fair or make sense, on the basis mentioned here.</p>

57

<p>From the Duke article</p>

<p>

</p>

<p>I can only laugh out loud at this. One wonders how the authors could have been so obtuse as to miss “students whose parents were chemical engineers with double majors” as the appropriate reference group against which Duke legacy admits should be measured.</p>

<p>Who pays for this “research” to be conducted?</p>

58

<p>Just to add to the above post, at the time he was admitted to Yale, George W. Bush’s grandather was on the Yale Board of Trustees (former Senator Prescott Bush) and his daddy was in the US Congress.</p>

59

<p>

</p>

<p>Did he say the average SAT for legacy admits was within 2 points of the average SAT for all admits—or within 2 points of the average SAT for the entering class? The way you phrase it makes it sound like the latter, and if so, that could be a very misleading statement. At most elite colleges and universities the average stats of the admit pool are significantly higher than the average stats of enrolled students, because many of those at the top of the admit pool decide to go elsewhere while those lower down in the pecking order among admits tend either to have fewer choices or are more likely to decide they’re “lucky” to be admitted and quickly accept. (I don’t know whether this is as much true at Harvard, though, which seems to be top dog in most of the cross-admit contests).</p>

<p>

</p>

<p>I don’t see how that follows. Doesn’t it depend on the size of the applicant pool, and how many well qualified applicants there are for each available slot? If the applicant pool is big enough, you can do a lot of things with it. Suppose College X has 1,000 places to fill, and gets 30,000 applications of which 2,000 are from legacies. Now suppose further the stats profiles of the legacy and non-legacy applicants are identical, and that the school’s yields on legacy and non-legacy admits are identical, in each case 50%. Then it’s easy enough for the school to fill 200 places in its entering class by admitting 400 legacies with stats exactly mirroring the overall admit pool (of whom 200 enroll), for a 20% legacy admit rate; while filling the remaining 800 places by admitting 1600 non-legacies whose stats also exactly mirror the overall admit pool, for a 5.7% non-legacy admit rate. Ex hypothesis same stats at both the applicant pool, admit pool, and enrolled student levels, but very different admit rates, HUGE advantage for legacies. Unequal treatment, one might say. </p>

<p>In my view this whole debate has nothing to do with whether unqualified or less-qualified legacies are being admitted. It has to do with whether there’s any real justification, as a matter of social policy, for awarding a legacy preference. Colleges are legally entitled to do it, of course. But SHOULD they do it, and if so, why? From the vantage point of the merely EQUALLY qualified non-legacy whose chances of admission may be half (or in my example, less than half) those of an otherwise identically qualified and credentialed legacy candidate, it seems manifestly unfair. And it’s hard to see any socially valuable end served by it, other than perpetuation of a hereditary elite which in my book is a social goal of dubious value.</p>

<p>

</p>

<p>dadx, maybe I’m missing something, but I don’t see the humor. Isn’t it the case that, by definition, legacies are almost invariably people whose parents have college degrees–making that a relevant comparison group? And isn’t it also the case that an extraordinarily large percentage of graduates of elite colleges like Duke—perhaps on the order of 85%—end up with graduate or professional degrees?</p>

<p>Again, I think this “lower average SAT scores” argument is a bit of a red herring. For reasons posted above, I think a legacy preference is unfair to non-legacies even if the SAT profiles of the two groups are identical, because it leads to unequal admit rates in a way that is not rationally related to the achievement of any legitimate educational or social objective. And it’s not just that legacies benefit; non-legacies are affirmative harmed because of the zero-sum nature of the admissions process.</p>

60

<p>Dadx-
I don’t know where you got the quote about ‘chemical engineers with dual degrees’, the study was trying to show how legacy admits compared to people from backgrounds where kids came from roughly the same kind of academic background in terms of the parental education level. </p>

<p>BTW, ‘professional degree’ could include engineering, IT, medicine or law, it isn’t ‘Chemical engineers with a dual degree’ i.e rocket scientists, and the comparison is about relative education levels.</p>

<p>Legacy admits, by their very nature, come from families who are college educated, and the differentiation is because there are different levels of education associated with going to college. Among other things, people with professional degrees tend to be at the higher end of the education scale, and their kids routinely tend to do much better then kids from parents at other levels…hence the comparisons. The point is those who claim that legacy students are better educated then others, grew up with more opportunities and the like, need to be compared against a comparable population, and the comparable population are kids who come from households where the parents are educated and presumably offer a comparable environment to the well off legacy kids, you need to compare apples and apples, not apples and oranges. </p>

<p>If they had done a study against all students, it wouldn’t be a valid comparison. The overall school population includes kids who came from tough backgrounds, who didn’t have the educational opportunities like tutoring and SAT classes in their lives to boost scores, or go to schools as good as those from educated backrounds go to. They may not have had access to cultural offerings, or may come from a home where English is not spoken, or may come from a background where they are the first to go to college, they don’t come from comparable backgrounds. I am sure that a kid who came from a legacy family would do a lot better then a kid who grew up in East Harlem or something like that, but it wouldn’t be a valid comparison, since the claim is that legacy kids do well because they have access to education and such…but to measure that means getting measured against non legacy kids with the same background…</p>

<p>And the people who did this if I read the second link correctly, were sociologists whose field is involved in relevant statistical research, they were the ones who culled the data, they are people a lot better qualified then yourself or myself in determining that (the sociologists report, btw, is peer reviewed and would not be published if the data were so faulty). </p>

<p>More importantly, whether it was professional degrees or general college degrees in the family background, the legacy students did worse as a whole, which blows the whole argument to kingdom come, literally. If in fact legacy students are admitted because they do better then other kids, they should be blowing out the bell curve, not lagging it, and what this shows is that many legacy kids get admitted even though they don’t measure up to their peers, which is the point. And if what these studies say is true, that legacy admits don’t bring in more giving, then basically it is nothing more then the old boy network, something most people find objectionable. </p>

<p>My favorite quote on the subject:</p>

<p>“You know, the rich are quite different then the rest of us”
“Yeah, they have a lot of money”</p>

<p>The first person was F. Scott Fitzgerald, the second was Ernest Hemingway</p>