Harvard’s freshman class is more than one-third legacy—here’s why that’s a problem

@yucca10 , you say the numbers suggest that relatives other than parents play a role in admissions. Please explain the mechanism by which you believe this happens, as the application doesn’t ask for the information anywhere.

Here’s an alternate explanation for the numbers. Families that produce one Harvard student are likely to produce more progeny that are qualified for Harvard and have a good idea of what the school is looking for.

There is a lot of overlap between different hook categories. For example, the lawsuit analysis considered the following criteria as hooks that were excluded from the “baseline” sample. I realize applying early is not a “hook”, but it was associated with an increased chance of admission on par with some of the weaker hooks.

Applies Early – 34% of Admits
Legacy – 14% of Admits
Recruited Athlete – 11% of Admits
Dean/Director’s Special Interest List – 9% of Admits
Staff Child – 0.8% of Admits

Faculty Child -- 0.3% of Admits

Total of Above Categories = 47% of Admits

The sum of 34% + 14% +11% + 9% + 0.8% +0.3% = 69.1% of admits., yet the total was instead 47% due to admits falling in multiple categories. A more complete list of categories of preferred admit categories from the lawsuit sample is below. Note that the lawsuit sample does not include internationals, which as noted above is not a hook. What I’m calling “hook strength coef” is the regression coefficient for admission decision in the original Plantiff analysis with full sample and full controls. A strength of 0 means no effect. A higher number means a greater chance of admission among applicants who had similar reader and interviewer ratings, similar stats, similar hooks, similar concentration, and similar dozens of other criteria. Increasing academic rating from 3 (“excellent grades and mid-600 tolow-700 scores”) to 2 (“superb grades and mid-to high-700 scores”) was associated with a 0.9.

Applies Early – 34% of Admits (24% admit rate, Hook strength coeff = 1.3)
SES Disadvantaged Flag – 16.5% of Admits (10% admit rate, Hook strength coeff = 1.1)
Legacy – 14% of Admits (34% admit rate, Hook strength coeff = 1.8)
Black – 13% of Admits (8% admit rate, Hook strength coeff = 2.7)
Hispanic – 12% of Admits (7% admit rate, Hook strength coeff = 1.4)
Recruited Athlete – 11% of Admits (86% admit rate, Hook strength coeff = 7.8)
Dean/Director’s Special Interest List – 9% of Admits (42% admit rate, Hook strength coeff = 2.3)
Faculty/Staff Child – 1% of Admits (47% admit rate, Hook strength coeff = 1.7)

The only vehicle you have is your application. You have to ask yourself why some legacy apps might be more attractive. You don’t want to be the average kid who assumes it’s just stats, a few titles, and/or other shiny (but often meaningless) bullets, like “founded the pie club.”

And that’s where legacies can be better informed, show their match, and compete. It’s not just what you want or how your hs did things. Nor is it the simplistic view that all H alums are wealthy and the school just wants their money. (Not.) None of that is the level of thinking H looks for. Or the energies and other traits.

But yes: " a very good illustration of why it is often a good idea to check the original sources." And apply a little critical thinking. Just because you see the Crimson was involved doesn;t make it an ultimate authority.

@millie210 Obviously, nobody is preventing students from mentioning a Harvard alum relative in their essays.
The explanation you provided is plausible but I have hard time believing that it accounts for more than 20% undergraduates with Harvard-affiliated relatives. If not direct relevance to admissions, it suggests that Harvard has a preference for applicants from certain social circles, which doesn’t really contradict what you said.

Regarding the numbers from the Harvard survey, the sum of the percentages of the different family relationship categories is below by year. It’s possible to be related to more than 1 person who attended Harvard, so totals are expected to exceed 100%. In the years 2018-21, the average total of categories is 117%, with a +/- 3% range. This seems like a reasonable amount of overlap between categories to me. However, in the referenced class of 2022 year, something different happens. The category totals sum is exactly 100.0%. All categories decreased in a similar proportion as if someone scaled the them to make the sum 100%, rather than using a poorly thought out survey that did not allow a student to be related to more than 1 person who attended Harvard. Random chance would also likely have the totals have +/-0.1 or 0.2% off from 100.0%, again suggesting they were intentionally scaled to equal 100.0%. For this reason, I’d consider the class of 2022 legacy numbers referenced in the news story to be invalid.

2018 – 73.1% not related + 43.9% sum of relationship categories = 117%
2019 – 72.2% not related + 45.1% sum of relationship categories = 117%
2020 – 73.0% not related + 41.7% sum of relationship categories = 114%
2021 – 70.7% not related + 49.3% sum of relationship categories = 120%
2022 – 63.2% not related + 36.8% sum of relationship categories = 100.0%

Looking at the 4 valid years. It looks like an average of 72.3% of students say they are not related to anyone who attended Harvard, meaning 27.7% have 1 or more persons in their family who attended Harvard. If I assume the 1 parent who attended category Harvard means exactly 1, then an average of 16.4% of students are what Harvard officially considers legacies (either one or both parents attended Harvard), and another 11.3% of students are not traditional legacies and instead have a non-parent relative who attended Harvard such as a sibling, grandparent, aunt/uncle, or more distant relative. 11% seems reasonable to me when considering the far greater chance of applying, far greater chance of being well qualified, far greater of having non-legacy hooks, etc; so I see no reason to assume there a strong boost for relationships besides what Harvard officially considers legacy and flags as such in the student files.

My kid was a legacy but didn’t tell them.

Kids from all social circles can present a good app, show their match. Or not. Some legacies are more aware of what match is, in lots of respects. Many kids do not. They stop at their own wants.

You indicate your parents’ educations on the CA. It’s not what the essay is for.

Not sure why people focus on this, as a private institution they can admit whomever they want.

And as someone whose kid was waitlisted by my alma mater (not Harvard), I can say there is an impact on the school. Can’t imagine my view of my alma mater will ever be the same, and I gave them a piece of my mind. If my kid didn’t belong there academically I would not have encouraged him to apply at all. Funny but my kid ended up in a better school for his major that is MORE competitive, which only pissed me off even more at my alma mater. They will never get a nickel out of me.

@blevine I agree.

I’m not sure what all the indignation is about in general when it comes to college admissions. Is there more value to giving a leg up to an underepresented minority or a first gen college student or a recruited athlete? Why shouldnt legacy applicants be welcomed? They bring a richness through their family’s history and commitment and they are more likely to become donors…which benefits the whole population. And likewise with wealthy applicants, regardless of legacy status-wealthy kids have the right to learn, too, and donations benefit the whole school.

We have become really tunnel visioned in the past 10-15 years over social issues and while increased awareness is a good thing, the whole victim/need to revise history mentality is unhealthy. Let’s just learn from the past and move forward with increased knowledge and awareness to be the best version of who we can be.

I dont think anyone has the right to a seat in a college classroom and schools have the right to balance their admitted student populations however they see is in the best interest of the school overall. IMHO.

I think that the debate is perfectly valid. However, I think that we should use actual data and proper analysis.

We can’t add up several categories that have overlapping members and assume that they aren’t overlapping. In other words, many individuals might have two or more hooks, or people might have multiple relatives who are alumni. You can’t simply add the numbers and come to a valid conclusion. In addition, Harvard has stated publicly that only children of Harvard College alumni get the legacy tip, so in this case, we need to use their definition.

At the Visitas program today, a student asked the university president a question, which he prefaced with the comment that legacies account for 1/3 of the college. The president and the dean of admissions corrected him and said that 12-13% are legacies. I wrote to the authors of the CNBC article, and they ignored my email. Mr. Berry started this thread with a title that is factually incorrect. We need to work with actual facts, not alternative facts, if we are to have a legitimate debate.

The Harvard Crimson presented the results of a survey to which about 60% of the freshman class had responded, and it indicated that only 63.2% of the respondents said they had no family connection with alumni or current students at Harvard College. Unless the Crimson was lying, or grossly misrepresenting its survey, that should count as a fact.

The survey’s number for relationships that Harvard counts when it reports on legacy admissions was margin-of-error consistent with what Harvard usually reports about legacies: 14.5% (in the survey) vs. 12-14% or “less than 15%.”

I am a non-legacy and non-hooked student at Harvard and I do not see what is wrong with giving legacies special preference if they apply Early. It makes sense to increase yield and maintain a long-term Harvard community. I am glad my kids will one day (hopefully) benefit from legacy status and they should not have to feel bad about that in my opinion.

studentathlete18, Legacy preference is an English 1800s thing, not a USA 2019 thing. It won’t persist for long. Maybe your kids will have to achieve at high levels if they want to attend a highly selective school.

Every year Harvard admits the 2,000 kids that it thinks will be most valuable to Harvard, based on the many needs Harvard has and the constituencies it needs to satisfy. Some of those admitted are legacies; who they and their families are, and how important they are to Harvard, are factors that help push them into the admit pile, but everything in the app matters, and the vast majority of legacies are denied.

Harvard’s probably in the best position to judge which applicants are most valuable to Harvard, it’s a private institution that can admit who it likes so long as it does so in accordance with the law, and unless the value of applicants’ family connections to Harvard declines to zero, or the government imposes a (probably unconstitutional) requirement that legacy status not be considered, being a legacy is going to continue to provide an admissions boost for the foreseeable future, IMO.