This is what I don’t get, and why I’d like to see the whole study (but not enough to pay for it). This seems to mean that of two similarly situated candidates, the legacy is 3 times as likely to be admitted to the legacy school. But if both of those candidates are admitted to multiple peer schools, does it really matter that one of them is much more likely to be admitted to the legacy school?
In other words, if Joe is a Yale legacy, and Jim is a Harvard legacy, are we concerned that Jim is rejected from Yale, and Joe is rejected from Harvard–especially if both of them are also admitted to Brown?
I can’t get too excited if I learn that, due to special privilege, a kid gets into Harvard who would “only” have been able to get into Cornell without that advantage. Is that what we are talking about here?
Off topic but kind of on point, my old employer - a global class law firm, would entertain candidates only from the tippy top of their class at Top 10 schools including Harvard but never Yalees, ever. Not being from the east coast, I don’t get the distinction, sorry to run off course, but I thought it was curious and downline of our topic…
It would be fascinating to see a study of the transmission of educational status in Ivy League families. There is reputed to be a high degree of intermarriage. How does Hurwitz account for the progeny of Princeton & Harvard marriages?
I think most legacy applicants apply early if they like their parents’ college. One place I think legacy really helps is that the child is more likely to have visited campus. The potential legacy applicants who don’t like mom’s college are more likely not to apply than someone who wants to apply to “top 10 colleges.”
“In other words, if Joe is a Yale legacy, and Jim is a Harvard legacy, are we concerned that Jim is rejected from Yale, and Joe is rejected from Harvard–especially if both of them are also admitted to Brown?”
Assume everyone has the exact same qualifications.
Joe will have a 45% chance of getting into Yale but only a 15% chance of getting into Harvard, Princeton and Stanford. Jim will have a 45% chance of getting into Harvard but only a 15% chance of getting into Yale, Princeton and Stanford. Non-legacy Julie is 15% at every school.
Since we are talking about probabilities of less than 100%, you can generate anecdotal evidence for every possibilitiy. And those one-off outcomes are what each of us see. But we don’t see the entire pool. So sometimes Joe gets in nowhere and Julie gets in everywhere. Because 15% probabilities happen 15% of the time. Sometimes Jim gets rejected by Harvard but gets into Yale. But that isn’t what usually happens year after year with thousands and thousands of applicants.
Based on the entire pool, more often Joe gets into Yale but not Harvard, Jim gets into Harvard but not Yale, and Julie gets in nowhere because of how legacy preference works. Because if you are in a close or tied up game, having a tie breaker is big.
All of them may get into Georgetown. None of them may get into Cornell.
This is a fair point, Hunt. I once heard an admissions officer at a highly selective LAC say that when he first began that job, he would lose sleep agonizing over all the highly qualified applicants who ended up in his “reject” pile, as well as all those he advocated for but failed to convince the admissions committee to admit. Then he had an epiphany: he realized that, because this is America where there are many very good schools and students may apply to as many of them as they like, they were all going to get into very good schools. It’s just that they couldn’t all get into his school, but he didn’t need to worry about them; they’d be fine.
It’s a nice story, but I’m not sure it’s quite as simple as that when it comes to legacy preferences in admissions. First, even if you have a strong non-legacy applicant who applies to 4, 5, or 6 extremely selective schools and has strong enough credentials to be in the lottery pool at each of them, it’s far from guaranteed that she will be accepted to any of them. It’s not as if every strong candidate, legacy or not, is admitted to “multiple peer schools.” It could be that the legacy who is admitted to H would not have been admitted to Y, P, or S where she is not a legacy but instead is just another well-qualified member of the lottery pool, who are admitted at extremely low rates in part because Y, P, and S are all admitting their own legacies at 3.13 times the rate of non-legacies. Or it could be that the non-legacy who is not admitted to H in part because legacies are being admitted to H at 3.13 times the rate of similarly credentialed non-legacies, is also not admitted to Y, P, or S, which again are admitting their own legacies at 3.13 times the rate of non-legacies.
Also, you have some applicants who don’t even apply to multiple peer schools at the most selective level. Pizzagirl just gave us an example: her son was a legacy at Northwestern, which was both his highest reach and his first choice. He was admitted. No doubt many other similarly credentialed non-legacies (and probably even some similarly credentialed legacies) weren’t admitted. Some of the non-legacies might have had Northwestern as their highest reach and #1 choice, just like PG’s son, but if they’re not admitted, they needed to dial down to a somewhat less selective school. Is that the end of the world? No, of course not; if they constructed their lists well, chances are they all ended up at good to very good schools, as PG’s son would have, had he not been admitted to Northwestern, and they can still have very fine lives. Still, I could never bring myself to say to those kids, “Look it just doesn’t matter.” These are young people with hopes and dreams, who have worked hard, sacrificed, and invested heavily in pursuit of those dreams. Just as it’s perfectly appropriate to celebrate with them when that acceptance letter comes from their dream school, so, too, is it appropriate to feel their pain when they’re rejected by their dream school. In the long run, it probably matters less than they imagine whether they’re accepted at their dream school or not; I think most of the putative advantages of being at a super-elite school v. a merely elite or merely very good one are just pure hokum. Still, it does them a disservice to pretend as if the game isn’t rigged. It’s rigged; some people with credentialed equal to but no better than theirs have an inside track that dramatically boosts their chances of admission, and that is true at each and every school they apply to that uses a legacy preference in admissions. Let’s be honest about it and say so. That way they can at least have more realistic expectations going in. And it may take some of the sting out of that rejection.
It sheds some light on a question I had, and exposes that the results of the study aren’t being reported in an accurate way. This stuff always annoys me. Given that the reporter had access to the author, I don’t think this is likely to be wrong, although her verbal clarification is inartful. (My underlining)
If correct, most of the above discussion is based on erroneous understanding.
If you want a college to show you the love because you’re a legacy, you had better give them the ultimate indication of demonstrated interest – an ED application.
Besides, the remarkably frank Penn admissions office says openly on its Web site that it gives a bigger legacy boost to ED applicants, and I think most kids who are legacies at other schools expect that this is probably a universal practice.
Hurwitz says that “The legacy admissions advantage is further enhanced through early admissions programs.”
Back to the original question. Why do schools give legacies preference?
I say for the exact same reason they do early admissions – to enroll full payors. Legacy is mostly about tuition dollars. Only a little bit about donations and tradition.
“Still, it does them a disservice to pretend as if the game isn’t rigged. It’s rigged; some people with credentialed equal to but no better than theirs have an inside track that dramatically boosts their chances of admission, and that is true at each and every school they apply to that uses a legacy preference in admissions. Let’s be honest about it and say so. That way they can at least have more realistic expectations going in. And it may take some of the sting out of that rejection.”
Two reactions:
Life is rigged. Your kids and mine were born into intact families that both valued education and had the means to pay for good educations, as well as the “auxiliary” things (sports lessons, academic camps, music lessons, etc.) that made our collective kids look appealing to elite colleges.
I think the problem with this is … Just as you wouldn’t want a white or an Asian student specifically thinking that a black or Hispanic candidate “stole” his rightful spot (for all you know, he was turned down for another white or Asian kid! it’s not like he had a lock on spot #242 and someone else swiped it) … I don’t think you can conclude that because your non-legacy kid was rejected from Elite U and some legacy you have heard of got in, that legacy kid “stole” your kid’s rightful spot.
It’s not because Daddy is on Wall St and junior will be, too, they’ll be full pay and donate lots of money. I’ve never seen an adcom comment in that respect.
First, I’m still uncomfortable with the word “preference.” It implies they prefer to take a legacy over anyone else. How do we get into their heads and say that? As if they cull though, come up with a few hundred legacies, fill those seats and then grudgingly look at the rest.
I think what’s frustrating for some, including myself, is that the top universities go MOST of the way, but not all of the way, towards academic meritocracy.
i.e. We know that GPA, courses & rigor, test scores and the like are very important. Most of the students admitted to the highly selectives are near the top on those criteria.
Furthermore, there is signalling going on in multiple directions. The highly selectives are signalling that they have the cream of the crop. The admitted students are signalling that they ARE cream of the crop. The rejected students, unfortunately, are signalled to that they are somewhat lacking. When they’re 25, the graduates of the highly selectives list it on their resumes to signal how smart and dedicate they are.
The admissions process goes MOST of the way, based on various measures of academic merit.
But then you get this stuff like legacy preferences (among many other tips), that mean that while Betsy got accepted to Harvard and Susan got rejected, there’s at least the chance that Susan was actually a stronger student in H.S. than Betsy.
i.e. The signal is no longer so clear.
If Harvard admitted on a true lottery system (like, I think, some charter schools do for younger students), but built into its lottery was a tip factor for legacies, I don’t think folks would mind so much. But it’s that stamp of approval - THIS kid is Harvard material, THAT kid is not, that makes the legacy preferences and other non-academic tip factors frustrating…
“It implies they prefer to take a legacy over anyone else. How do we get into their heads and say that?”
If they didn’t have a preference for legacies, then they wouldn’t have a legacy policy obviously.
If two kids are equal, they have a tendency to prefer the legacy. About 3X more according to Hurwitz. If the legacy is just a little behind the other kid, they also do usually prefer the legacy. If the legacy is more than a little behind, they don’t prefer the legacy. They also prefer athletes and URMs.
We know why they prefer athletes (competitive teams) and URMs (diversity). But they are evasive about why (and how much) they prefer legacies. I don’t think it has much to do with donations or community or tradition.
You can give a little tip without insisting it’s affixed preference.
I don’t think people can advocate some revised “meritocracy” without trying, again, to either leave holistic behind or dictate, based on personal, outsider opinions, what distinguishes one 4.0/top score kid from another. In the latter case, even if you could come up with some new hierarchy, class president versus research intern, there would likely still be too many good kids.
Lots of factors (as PG often mentions) go into choices. So, if two kids are otherwise equal (assuming you can identify this without simply relying on stats alone,) would you take the umpteenth kid from Mass or NY…or the kid from Wyoming? The thousandth kid who was that class president or the one who did something equal, but different? Lots of examples. Few absolutes. Too many applicants.
To my understanding, if you’re looking for highly selective, elite schools with admissions meritocracy based strictly on numerical stats and not on a holistic reading that considers other factors like hooks, then you’re considering oxford and Cambridge. Someone correct me if I’m wrong about that.
@dadx: Actually, @northwesty is right, or at least more right. I have a copy of the original journal article. northwesty’s description of the model specification (i.e. 3X for all legacies, both primary and secondary) is right. However, in the abstract when Hurwitz says “odds” he’s referring to something called an “odds ratio” which is not the same as “admission chances.” It’s like in horse racing … if a horse has a fair payoff of 3:1, then the odds ratio is 1/3 while the chance the horse wins is 25%.
The 45.1% add-on factor for primary legacies’ “admission chances” was an attempt to make it reader friendly, but would make a horrible statistical model. It’s not a good way to summarize the results other than at the coarsest level.
For those interested, the multiplication factors that Hurwitz found for all legacies is 3.13X, for primary legacies is 7.63X, and for secondary legacies is 2.07X. To be honest, I found these numbers quite a bit higher than I expected. I have no reason to think that Hurwitz did sloppy work, but a different approach where Hurwitz explicitly controls for SAT score, URM, gender, and recruited athlete status lowers the legacy advantage from 3.13X, 7.63X, 2.07X to 2.31X, 3.89X, 1.82X. These last numbers seem more reasonable to me, but what do I know?
P.S. The difference between these two sets of numbers of one of Hurwitz’s primary contributions to the literature. His claim is that legacies look great on an SAT score, etc basis but there’s a big negative something that lurks in their applications that other colleges see but the legacy college overlooks. Personally, I’d like more evidence for this.
I’ll have to spend the $ to find the article, I guess, if I want to argue over questions of fact. I’m having great trouble reconciling the numbers in your article description with the exchange between the smart commenter and the author of the report I cited.
I have a great suspicion that his way of describing these facts is different than any of us would guess.
Here’s how to reconcile. The average acceptance rate in the sample of 30 colleges for non-legacies is 19.2%. This gives an odds ratio of (.192)/(1-.192) = .237. Multiply this by 7.63 (primary legacy) gives an odds ratio of 1.81. Converting this back to acceptance rates gives 1.81/(1.81+1) = 64.4%. So 64.4% is (crudely) the average acceptance rate for primary legacies. Finally, 64.4% - 19.2% = 45.2%, which crudely gives the model’s prediction of the average advantage of a primary legacy over a non-legacy.
(In reality, the acceptance rate for primary legacies in his sample was only 40.6%, not 64.4%. Hopefully you see the issue.).
If the math is obtuse - you can see that northwesty simply copied and pasted from Hurwitz’s abstract, so that’s how Hurwitz describes his own results when writing for a non-laymen audience.
Except plenty of elite schools have published their legacy acceptance rates and they are nowhere near 60%. Nowhere. The highest I’ve see was Princeton which I think was ~40% though I may be misremembering. They’ve certainly been “published” on CC.
that is written by someone who might be brighter than the average reporter.