Do Legacy Preferences Count More Than Race?

<p>Look at comment 18 attached to the article in the second link for a clear statement as to the author’s findings.</p>

<p>tsdad, thanks for pointing us to the comments…#17 appears to be from Theda & William Skocpol, sociologist and former dean of the Graduate School of Arts & Sciences at Harvard/BU physics professor, respectively.</p>

<p>I’m not sure I buy the rejecting-your-rival’s-legacies theory, but if my son got into Brown and rejected by Michigan, I’d be looking for non-linear effects, too.</p>

<p>Military Mom – Your daughter sounds wonderful, and I’m sorry she was deferred. And I know it must hurt – and it probably hurts more when you see studies like this. But I think it’s highly unlikely that the reason she was deferred was because of her religion or her family’s military background. In fact, I find it inconceivable that either of those would be the cause for deferral at any school.</p>

<p>While many legacies are accepted at these schools, an even higher percentage are denied. I’ve known several extremely qualified double legacies who did not get into their parents’ school. Why? Who knows. Fifteen years ago they would have been accepted. It’s gotten so competitive – especially when you are coming from suburban DC.</p>

<p>JHS, I think you’re doing the math wrong. If 10% of Harvard applicants get accepted, and legacies have a 50% better chance of being admitted, that would mean 15% of legacies get accepted. Not 60%.</p>

<p>I read the original paper and it is convincing to me. If anyone would like a copy, send me your email and I can forward it to you. It doesn’t make a lot of sense to debate the results based on a magazine article’s version.</p>

<p>Racin: The Chronicle article is very clear that it means a 50 percentage-point increase in the rate of admission, i.e., from 10% to 60%, not a 50% increase in the rate from 10% to 15%. As I said above, I wouldn’t reject the latter as absurd at all. Comment #18 to the article, cited by tsdad, suggests that the author of the article (and by extension the separate author of the blog post about the article) misread the study, and that it claimed the more believable 50% (or so) increase in admission rate. If I made a mistake, it wasn’t in math, it was in trusting that the Chronicle author knew how to read a paper published in an education journal.</p>

<p>Of course, starbright is correct: debating the newspaper article or the blog is sort of pointless.</p>

<p>JHS, I think that you are misinterpreting the numbers. (Or relying too much on someone else’s misinterpretation)</p>

<p>That is, the article states, " if a non-legacy applicant with a certain set of credentials has a 15 percent chance of admissions, a primary legacy applicant with identical credentials would have a 60 percent chance of getting in." – but I do NOT interpret think the study meant that if school A. has a 15% overall chance of admissions, that legacies get in at a rate of 60%.</p>

<p>The article made very clear that they controlled for multiple variables – SAT scores, GPA, EC’s, etc. </p>

<p>One of the very common fallacies that seems to permeate CC is the idea that if a given school admits 15% of applicants, that means that each applicant has a 15% chance of admission. Of course the reality is that some applicants to that school have a -0- chance of admission (clearly unqualified) – some have below 15% (school is a big reach) – and some highly accomplished students probably have very high odds of admission. </p>

<p>So the study is saying that they developed a formula to project odds of admission based on numerous factors (test scores, GPA, class rank, etc.), that if a student with a given set of credentials has a 20% chance of being accepted, then the legacy student has a a 62% chance of admission - but that if a weaker student’s credentials leave them with only a 5% chance of admission, the weaker legacy applicant would only have about a 15% chance. That hypothetical student with the 15% chance of admission might have a 45% chance.</p>

<p>I get this from going to the abstract of the actual STUDY - at [url=<a href=“http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VB9-51R0790-1&_user=10&_coverDate=12%2F16%2F2010&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1598653271&_rerunOrigin=google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=4fc64ab8bc30351d900feaf7c499648e&searchtype=a]ScienceDirect”>http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6VB9-51R0790-1&_user=10&_coverDate=12%2F16%2F2010&_rdoc=1&_fmt=high&_orig=search&_origin=search&_sort=d&_docanchor=&view=c&_searchStrId=1598653271&_rerunOrigin=google&_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=4fc64ab8bc30351d900feaf7c499648e&searchtype=a]ScienceDirect</a> - Economics of Education Review : The impact of legacy status on undergraduate admissions at elite colleges and universities<a href=“*I%20have%20this%20annoying%20habit%20of%20preferring%20to%20read%20the%20actual%20studies%20rather%20than%20what%20news%20reports%20say%20about%20them”>/url</a>. The abstract says,

</p>

<p>So I am just taking the 3.13 multiplier the author suggests and applying that. But given that he also says that the advantage depends on selectivity and academic strength, that 3.13 multiplier itself is probably just a ballpark figure, variable depending on the school and the applicant qualities. </p>

<p>I’d prefer to read the actual study – I generally don’t trust abstracts all that much – but it’s not worth $20 to me to find out.</p>

<p>One thing that is interesting to me is this notion of “secondary legacy.” This sort of contradicts what I’ve often read on CC - that the majority of these so-called secondary legaciy relationships are not preferential legacies but merely something called “tip factors.” Not sure if this is just a semantic difference.</p>

<p>I love writers that use studies that “conclude that after better controlling for variables than previous researchers did” … nice way to predefine the quality of work from which one wants to write an article - no discussion of why the study had better control - no discussion with critical researchers that the study was actually better - just take the study’s results and write the article one wanted to write in the first place, so that the predefined conclusion of “the case for eliminating ancestry discrimination in college admissions continues to grow” can be stated, in the same tone of the writer’s other articles.</p>

<p>calmom: As I said already, it seems pretty clear the Chronicle writer misread the paper.</p>

<p>bovertine: A number of universities do say they extend legacy preferences to children of graduate school alumni, or to grandchildren of college alumni. Stanford and Penn both do this explicitly. In the case of Penn, that has probably been a good strategy, since in my generation the prestige of the professional schools – Business, Law, Medicine – outstripped that of the College, and I think Penn got some mileage out of marketing to the children of Wharton grads (like, for example, Ivanka Trump). Remember, folks, that for 95% of colleges (or more) legacy preferences are pure marketing. </p>

<p>As for sibling legacy, I thought everyone more or less recognized that. Clearly the study says otherwise, but when I look at the world I see qualified siblings getting accepted by selective colleges at a much higher rate than the children of alumni.</p>

<p>thrill: My understanding based on the press stories is that the study controlled for all those factors by looking at multiple applications by the legacy kids, which if true is pretty ingenious. It’s not that the study rated every applicants essays and controlled for that. It just assumed that the same kid had the same quality essays (and grades, and ECs, and recommendations) when he or she applied to 8 different schools.</p>

<p>(Back in the day Harvard made clear it considered me a legacy on the basis of two parents in the Law School, a grandmother at Radcliffe, and about 20 uncles, great-uncles, and cousins at Harvard College.)</p>

<p>I’ll try to provide some of the key parts of the primary question:</p>

<p>

</p>

<p>

</p>

<p>

</p>

<p>

</p>

<p>I’m enjoying the NFL playoffs and editorial assistance provided by Ken Jackson, but I would like to note the quote from the Stanford official. More comments later. </p>

<p>

</p>

<p>^ They are, but only slightly. In this dataset, the average was 10 points on CR and 6 points on Math. Because conditional logistical regression is used, the specific variance explained by legacy is unique and independent of that explained by SAT scores.</p>

<p>In today’s admissions cycle, you absolutely, clearly need to be Hispanic. This is the true advantage in admissions. Colleges see the demographics and note the compound increase in Hispanic populaiton and clearly want to put their markers down and find and bring in as many as they can in order to grasp a leadership position in the most important growth demographic. You really need to be Hispanic to have Ivy hopes these days.</p>

<p>

</p>

<p>That may be. From what I hear they admit only 25% of SAT 2400 and GPA 4.0. How much stronger legacy applicants to get over 50% admit? What defines strength?</p>

<p>Question: Does having a parent who went to a university’s grad school make me a legacy? I’m a junior and wondering if my dad’s degree from Duke’s Fuqua School might help me next year if/when I apply to Duke. If it makes any difference, he makes annual gift contributions (not nearly enough to make me appealing to the development office, but it can’t hurt, right?).</p>

<p>It would seem to be a tip: [Duke</a> University Admissions: Ask a Question](<a href=“http://admissions.duke.edu/faq/indexb571.html?iQuestionID=518%20&iCategoryID=1]Duke”>http://admissions.duke.edu/faq/indexb571.html?iQuestionID=518%20&iCategoryID=1)</p>

<p>Re: question about grad school: it did make a difference at UNC, at least in 2003 when my son applied as an out of stater. Both my wife and I have graduate degrees from Carolina, and he was put in the admissions pool made up of children of out of state alum. Students in this pool receive special consideration. Those admitted have credentials closer to, but still better, than in-state applicants. </p>

<p>These things differ from school to school so you need to ask the college directly.</p>

<br>

<br>

<p>Depends on the polices of the school. For Stanford the answer is yes. For Harvard the answer is no.</p>

<p>How much are donations scrutinized? On one hand, it may be that being a legacy helped S. OTOh, we aren’t “important” alumni from the standpoint of donations, so why should they care?</p>