@Pizzagirl - You’re right. That’s part of why the “add 45.5% model” is so crude and why I pointed out the actual rate was 40.6%. A few comments to address the issue:
His data is 8 years old at this point, and admit rates were higher.
Hurwitz’s sample is of 30 colleges. We don’t know who they are, but by the time you’re talking about Princeton you’re in even more rarefied air acceptance-rate wise than for the 30 colleges, so you’d expect Princeton’s rate to be lower.
If we use 7.4% non-legacy acceptance rate for Princeton, then doing all the math I did above with the 7.63X model gives a 37.8% primary legacy acceptance rate. Princeton says the rate for all legacies is 30.8% I think Harvard is at 30% too. Yale says “20-25%”. Could easily be higher for all the 30 colleges.
You really have to believe that the legacy applications have a big enough stink bomb in them (that isn’t related to SAT, URM, gender, recruited athlete status) to reconcile the model’s 64.4% with the sample’s 40.6%. Maybe they write awful essays because they think they’re getting into mom or dad’s alma mater anyway. As I said above, I’m a bit skeptical, but what do I know?
Had 31 members at the time of Hurwitz study; has a few more now.
Biggest disconnect in the numbers is that the acceptance rates at these schools have declined significantly over the past 8 years. And the 3X is an average. So Princeton’s data won’t be the same necessarily as Mt. Holyoke’s.
But any way you slice it, legacies get many more ping pong balls in the hopper than similarly qualified candidates. If you are smart enough to get into the lottery, it is a big boost. Going from a 10 to 30% admissions chance is a huge delta. And that would generally square with where the HYP type schools are today.
This could be true, and I confess my impressions are based on anecdotal evidence–but I have to say I’ve been reading results threads here on CC for six or seven years, and it is not my impression that there are very many students who get admitted to a high-reach legacy school, and are then rejected from all peer schools. Maybe kids for whom that happens don’t post here, of course.
What if being a legacy at a highly selective school gives you an advantage over similar students who aren’t legacies, even at highly selective schools where the first student isn’t a legacy? What would that mean? It might support the idea that these schools are using legacy as a surrogate for likely full pay.
What if the “big negative something” is the legacy status at the legacy college? It’s a benefit at the legacy, but a disadvantage at the others.
The one way around this would be to look at ED results. Are they better for non-legacy applicants in ED, where other legacy ties don’t matter? For example, let’s consider Princeton legacies. Princeton knows they’re more likely to yield if admitted than other applicants. Duke knows that too. How do Princeton legacies fare in Duke’s ED round?
Not aiming at you on this, Hunt. To get behind this, I think one has to stop assuming legacy itself is the primary determinant or even a last minute call point. I know I’m a broken record on that. But I still believe many assume all apps from all top performers are created equal. And then that legacy is the icing that makes Susie more desirable than Sally. It’s not that flat. You’ve got to suspend disbelief for a moment, ask yourselves why a legacy might present as a better fit, what a good fit really is, at various colleges. It goes way beyond full pay. Or payback for donations or alum work.
Many on CC claim not to help their kids with apps. I think that can be true (I only got random moments when mine let me offer an editorial opinion.) BUT, we had plenty of talks beforehand, discussing the values of different colleges (the colleges’ values, not the value of a degree from X to get into IB,) what my kids’ strengths were that might attract a college’s attention, how different schools would navigate them closer to their own goals, etc.
Now imagine how that could help a legacy kid form the application to the alma mater, versus the kid who knows media ranking and wants to go to a top college, the most top or any top, but doesn’t now much more.
Looking- you raise a wonderful point. I remember a kid I interviewed for Brown several years ago who was interested in Life Sciences although not sure exactly what path she wanted to take- bio, neuro, cog sci with a double major in chem, etc. We had a great conversation about Brown’s flexible curriculum, ability to do inter-disciplinary work without a lot of departmental politics, etc. And then I said, “and as an undergrad you might find some of the public/community lectures or a symposium sponsored by the Med school to be of interest” and she said, “Brown has a Med School?”
This is a conversation which I doubt I’d have had with a legacy applicant. A parent who went to Brown who had a HS senior interested in something science/life science/cellular in nature, at some point would have said “go check out the website of the Med School”. Not because the kid would be applying to the combined program (this kid wasn’t set on an MD so it wouldn’t have been appropriate) but because that’s what you say to your kid if he/she is interested in your alma mater.
And if I had a buck for every kid I interviewed who wanted to go to Brown’s Law school…
@lookingforward, I have a lot of sympathy for your argument. But if it’s true, why do colleges explicitly say that legacy status is considered as a positive element for admission? Why not say that they don’t need to consider it, because legacies do very well for many other reasons?
There’s been three academic professional studies on elite legacy admissions that I’m aware of.
Espenshade 2004 concluded legacy status was equal to adding 160 SAT points to an app.
Bowen 2005 concluded that legacy boosted admission chances by 20 points (not 20 percent but 20 points). Bowen, fyi, was a Princeton faculty member and later Princeton president. So I’d suspect he knows what he is talking about.
Hurwitz 2011 concluded legacy on average triples the chances of admission. Based on an analysis of 60,000 applications. In a study which supposedly controlled for most other variables in the applications (like grades, test scores, ECs, etc.). Hurwitz also said that the legacy boost was increased by the use of early admission programs. So he was also controlling for ED.
Plus, the schools themselves affirmatively say that there’s a legacy boost, although they usually downplay its significance.
Come on guys, there’s a significant advantage. I think the evidence for evolution and climate change, in contrast, is much less certain. : )
But didn’t Hurwitz find the exact opposite of what Espenshade did with respect to SAT?
I believe there is a legacy advantage, and I’m prepared to believe that it can be substantial in terms of admission to the specific legacy school. But what I would really like to know is how much of an advantage it really is in terms of gaining advantage to at least one highly selective school.
Brown is a great example because of the curriculum. Legacies will know about it. Still, it’s up to them to make sense of their apps and interviews.
And, yes, legacy can be a positive element or a boost. But not for the simplistic, superficial reasons that the college “prefers” them or somehow thinks all its grads are wealthy, full pay families. Especially not those who are need blind and don’t get a financial peek. Many of the top colleges build movers and shakers, sure, and are proud of that. But many also broadly define what an influencer is- and it’s not as simple as salary or the number of people you hold power over.
We agree plenty of legacies get eliminated. It’s not as simple as their stats didn’t measure up. It’s an application and review process.
Northwesty, you can go on with the pursuit of the studies, that’s more your thing than mine. But many of them are narrow or limited in ways that suggest this is calculable, forgetting the holistic “more” that plays. I keep mentioning what Espenshade said about his own study.
I’ve found the study, which turned out to be available (free) if you’re a member of certain organizations, including certain alumni associations. It took a little doing, but it was available to me.
The bad news is that I haven’t had time to parse through it in detail. My up-to-dateness on the math that he uses is not good, although I’m beginning to understand the use of the “odds” factor he employs. I’m not sure I agree with it as a way of expressing his findings, though.
“Because conditional logistic regression analysis requires students incorporated into the sample to have submitted applications to multiple colleges, I have removed from the sample students who were admitted through early decision processes at the sampled colleges.” I’m not sure what the landscape of admissions options looked like at these 30 schools at that time, so it isn’t clear how that affects the data.
One thing that I found unusual: he states that, among the applicants he used (about 133,000), that only 47% of them submitted applications to more than one of the 30 schools. This strikes me as being at odds with what I think I know about todays college admissions. I’m not sure what would explain that. If I’d had to guess at that number, I’d have guessed it would be more like 70%.
There are other questions that I have about the results, but it will take more time for me to pose them.
LF – so why do you think the schools bother with legacy admissions at all?
I think it is pretty clearly about finding full payors. Others think it is about finding donations. If it isn’t about those things, why do it at all? All the elite schools have tons of extremely qualified applicants to pick from after all.
If the legacy kids are so smart and such awesome fits, they’d get in in high numbers without any legacy boost, right?
Colleges have a history of being coy about why they have a legacy program at all. According to some, legacy and holistic admissions started as a Trojan horse to limit their enrollment of smart Jewish students.
“Going from a 10 to 30% admissions chance is a huge delta. And that would generally square with where the HYP type schools are today.”
I guess it depends on how you see these things. I see them as - you’ve got a 70% chance to be rejected, versus a 90% chance to be rejected. Either way, don’t start buying dear alma mater sweatshirts and booking the Thanksgiving plane tickets quite yet. A 70% chance of rain and a 90% chance of rain are both enough for me not to plan the outdoor picnic. I just can’t get ‘excited’ over increases in chances until / unless they take me over the 50% mark.
From my perspective, NW, it’s fair to acknowledge that the parent, grandparent, sibling, is already part of the fold, has experienced that school, made it through. Imo, has little to do with the fact that the Ivies used to build dynasties of grads. That’s then, not now.
And so is the Jewish thing at Harvard, 70 years ago. I don’t care why it began or who has what to say about it. We cant keep coming back to that or Bush jr. Much has changed, if not nearly everything, about admissions- and including the ferocious competition. I really advocate people stop tying to game this and fit it into neat little slots. Your legacy kid has 0 advantage, zero, if he can’t manage a good application. No matter how many other legacy kids did get in.
No way is it about finding full payors-what are you assuming? That they go look like at your financials? They don’t have time. They do have an app in front of them, a kid’s fair shot to present him- or herself. And the expectations are not as simple as CC believes, not just stats, rigor, a few titles, some awards, and the English teacher thinks you wrote a great essay. Not curing cancer.
“Because conditional logistic regression analysis requires students incorporated into the sample to have submitted applications to multiple colleges, I have removed from the sample students who were admitted through early decision processes at the sampled colleges.”
Anecdata only, but most of the NU-legacy-kids I know applied ED (as did my son). I do know 3 RD legacies (all 3 double legacies) and all 3 did get in - one chose West Point, another chose USC for a specific popular music program (he’s wildly talented and already on the road to a career, and being closer to LA was a major advantage over NU’s music program), and the third was actually wait-listed, planned to go elsewhere and got pulled in at the very end.
I absolutely advise any legacy to apply ED if possible, I surmise the ED pool is “thicker” with legacies than the RD pool at most schools, and I think it’s difficult to sort this out if this person’s analysis doesn’t include the ED kids.
I personally wonder about the impact of single legacy vs double legacy (for obvious reasons, as my kid was a double legacy). I suspect there may be some difference there as well.
I’ve always thought the reasons were more fuzzy. People like the idea of tradition and passing on values. I think there’s a notion that legacies do that. I also think there’s a hope that admitting legacies will keep old boy networks intact whether that is Boston consulting firms, New York investment banks. I think it dates back to the tradition of New England prep schools essentially sending the colleges lists of the boys that they thought were suitable for each college.
My younger son was not a legacy at U of Chicago, but a lot of relatives had gone there. It made it easier for him to write a tongue-in-cheek why Chicago essay, and also to dare to do it.
I think the argument is that experience teaches that legacies are much more likely to be full pay than non-legacies, and that one way to bump up the number of full pays without directly looking at finances is to admit more legacies. I think colleges might have the motive to do this; whether they actually do it is not so easy to answer.
I don’t know where that comes from, Hunt. In any given year, the legacy applicants to Yale are not all the kids of neurosurgeons, CEOs, and investment bankers. There is no loss when the kid puts down his parents are, say, teachers or one is unemployed. From my perspective, they do not assume. Neither that the doctor is wealthy, nor that the teacher has a trust fund.