I’m looking at it the opposite way. Harvard isn’t going to have the depth & breadth of upper level tech classes that MIT will for engineering/CS. And MIT won’t have the depth & breadth upper level social science classes that Harvard will if you’re a historian major or pre-law.
Math may be the exception.
Absolutely. The FB parent groups that I am in for my kids’ colleges are full of parents writing “with the amount of money I’m paying, we deserve…”this or that.
Probably a topic for another thread, but I would not be surprised if legacy admission hangs on indefinitely. Just a year ago, or even half a year, as a casual observer, I thought it was on its way out. Now not so sure.
These institutions really like it. I happened to chat with an admissions dean a while ago at a top liberal-arts college and asked about legacy policy—expecting a tentative answer, with reflections on equity and other matters. But she flatly said “maintaining our traditions is very important to us.”
I imagine its fundraising effects go from obvious to subtle. Obvious as in “my kid’s applying in a year, here’s an extra big gift.” But subtle like “maybe I’ll have kids one day, and it would be good to maintain connections (via donation) with the institution.”
Alums just like the idea of legacy, even if, as we’re told, it only helps (at many places) as a slight bump for legacy applicants already at the top of the pool. The idea of legacy may be more valuable than an actual prospect of a legacy admit.
And then there’s the “institutional-priorities” side of legacy which, however fuzzy, I believe is real—or believe admissions offices believe is real.
You’re probably right that legacy will stick around. FWIW, I don’t think it’s a “slight bump” - the various analyses I’ve seen (notably the one with Harvard’s data) suggest it’s a pretty big deal as an admissions tip.
The only highly selective private(s) that don’t use it, to my knowledge, are MIT and maybe CalTech. But MIT’s whole brand is kinda built on wanting the smartest and much less so on tradition, family ties, etc., so that’s why I think it works for MIT (and maybe CalTech). I could arguably see one or two others trying to copy this approach (Chicago? CMU?) but I think Harvard and the rest of the Ivy+ group will take the slight hit to admitted class quality (plus the P.R. hit) in exchange for the institutional value of legacy preferences.
(I’m not endorsing legacy preferences, personally, to be clear.)
(That said, maybe discussion of legacy preferences belongs in the other admissions thread, and this thread should stay mainly confined to SAT/ACT. Mods, feel free to move my post and others over, as you see fit.)
There are quite a few others that do not consider legacy – Amherst, CMU, Johns Hopkins, etc. However, it is true that the most highly selective private colleges favor legacies. Those that do not are in the minority.
Harvard may appear that way, but I have not seen any corners cut with legacy at other schools. They are remarkably strong candidates, even if the legacy status bubbled them up to the top of the pile. So it’s “slight” in that sense.
I really have not seen evidence of such “hit” (though people say the Harvard data suggests it). In fact, the Princeton class survey suggests the exact opposite.
The Harvard analysis suggests it is far more than a slight bump. Instead it was one of the strongest analyzed non-athlete hooks on average. With full controls, the regression analysis suggests 10x increased chance of admission for being a legacy and 22x increased chance of admission for being a double legacy. This regression had controls for admission reader ratings and other hooks, so it is similar to saying if the applicants with the same on paper, with the same with same admission reader ratings, then the legacy kid is 10x more likely to be admitted than the non-legacy kid on average. Legacy also wasn’t just a tie breaker, the admitted legacies tended to have weaker admin reader scores than the admitted non-legacies.
The analysis also found that the strength of the legacy boost is increasing over time. This strength increase relates to the % of legacies in class remaining relatively constant, in spite of the admit rate for non-legacies rapidly decreasing over time. The selectivity of the non-legacy pool appeared to be increasing more rapidly than the legacy pool.
Nevertheless, freshmen surveys at Harvard, Princeton, and other schools consistently show that legacies have higher SAT scores. For example, the Harvard freshmen survey shows the following average SAT scores. Legacies are often among the highest scoring kids at the college. This is not a contradiction. Like other high income groups, scores tend to be a strong point of the application for legacies. The boost primarily applies to other non-score admission criteria.
Harvard Freshman Survey: Mean SAT Score
Legacies = 1550
Very High Income = 1550
Had Private Admissions Counselor = 1550
Non-legacies = 1515
Low Income = 1480
Athletes = 1370
Princeton Freshman Survey: Median SAT Score
Legacies = 1545
No FA = 1545
Non-legacies = 1525
Full FA = 1490
Athletes = 1405
And, a least at Harvard, the legacies that attend are mostly very wealthy.
It seems to me that legacy alone does not give that much of a bump - the bump is for legacies from very wealthy families. The Class of 2022, the last time they didn’t take “I don’t want to say” as a response, had almost 50% of their legacies who were from the top 1% by income, and 87% were in the top 20%. According to the Chetty article, around 21% of Harvard graduates end up in the top 1%, and 65% end up in the top 20%. Of Harvard graduates, 9.2% end up in the bottom 20%, and fewer than 0.5% of the legacies in the class of 2022 were from that income percentiles (around 0.7% from the bottom two quartiles).
There is no reason to suppose that legacies from poor families are applying at any lower rate than very wealthy legacies. They know that it’s viable, and they know that Harvard is generous with FA.
The reason that the legacies have high SAT scores is because they are mostly very wealthy, and therefore have the benefits that wealth provide for SAT testing. Any difference in SAT scores between non-legacies and legacies for Harvard is easily explainable by the difference in income. The median income of non-legacy Harvard students is far lower than that of Legacies.
Sounds like a good result for Harvard. Smart and wealthy legacies
The Harvard lawsuit regression analysis that I quoted earlier controls for being a double legacy, being on the special dean/director’s list, being on the z-list, SES disadvantaged status, race, and other factors correlated with wealth. The listed legacy boost was in comparison to non-legacy applicants who have similar controls that are associated with wealth. Without these controls, I expect the legacy boost would be higher. This also contributes to why the income distribution legacy admits is so much higher than the income distribution of Harvard grads as whole – higher income legacies are more likely to have additional hooks like double legacy, special interest list, or recruited athlete than lower income legacies.
Students from lower income families are far less likely to apply to Harvard than students from higher income families. Combining the Harvard lawsuit admission rate by income to the Chetty study income distribution suggests lower income students are ~50x less likely to apply to Harvard than higher income kids. It’s not just a matter of lower income kids being less qualified. Among high achieving kids (when considering more than just having a top % test score), lower income kids are still far less likely to apply to highly selective colleges than higher income kids. For example, the study at https://www.nber.org/papers/w18586 begins,
We show that the vast majority of very high-achieving students who are low-income do not apply to any selective college or university. This is despite the fact that selective institutions would often cost them less, owing to generous financial aid, .
There are many reasons why this relationship occurs, and many of those reasons would also contribute to a lower income legacies also being less likely to apply. There is no reason to suppose that high income legacies and low income legacies apply to Harvard at the same rate.
For example, lower income legacies are more likely to attend a high school where the overwhelming portion of their friends and colleagues apply to in-state publics, rather than out of state privates, including high achieving kids; schools in which the counselors do not promote applying to x reaches, y matches, and z safeties; schools that are generally less like to focus on getting as many kids in to T## colleges as possible; … I’d also expect parents who are less financially successful after Harvard are less likely to strongly encourage their kids to follow in their footsteps than parents who are more financially successful after Harvard. I could continue.
However, most of the reasons that they give for low income kids to avoid applying to Harvard do not work when the parents are Harvard alumni. Poor legacies are attending Harvard at 5% the rate of wealthy legacies.
For the number to be as they are, they would have be be applying at 5% of the rate of the wealthiest legacies.
I’m not just talking about student from the bottom 20%. Among the legacies only 0.7% were in the $40K-80K bracket, and 10.4% were in the $80K-$125K bracket. $125K was around the 70th percentile. Overall, the poorest legacies were highly underrepresented, but so were middle and upper middle income Harvard legacies.
Those income brackets include faculty at regional universities and community colleges, high school teachers, librarians, and a whole host of people whose income is the result of their chosen profession, not because they weren’t successful.
They have no reason not to apply to Harvard, and yet they are highly underrepresented among legacies. Really, only the 1% are overrepresented, and pretty seriously so.
We are digressing, so this is my last post on legacy and wealth on this thread.
Might our friend @MITChris provide a glimpse into MIT’s grading rigor?
To be clear, by “easy” to earn a C, I meant there is no grade inflation and if you earn a C, you will get a C (not an A.).
It is not easy to earn anything above an F at MIT. Their grades are hard fought.
To be clear, by “easy” to earn a C, I meant there is no grade inflation and if you earn a C, you will get a C (not an A.).
It is not easy to earn anything above an F at MIT. Their grades are hard fought.
Is this based on a general feeling? Or is there an external reference?
MIT grading is indeed different from HYPS. Some differences include first semester being pass or no record, up to 48 units may be graded on pass or no record after first semester, no grades below C may be reported on transcript during first year, and using a 5.0 scale.
The more recent fraternity/sorority reports mention that students living in a fraternity/sorority average a ~4.5 GPA while at MIT. . I’d expect students living in campus housing average a bit higher, which would make the overall average in the A- region. This average has been creeping up over time, although 2022 is an exception with a drop from post-COVID 2020-21, likely due to removing COVID effects, such as remote classes.
I doubt that a large portion of students are receiving C and F grades. However, this does not mean that classes are easy. The MIT student body is composed of high achieving students who are expected to be successful in classes. MIT is known to be especially collaborative, rather than having a reputation for fighting for top grades or weeding kids out. I expect the grading policies above help facilitate that atmosphere.
Yes, but 4 is a B.
Not 100% sure what you are getting at. My student currently attends MIT.
This is due to MIT’s generous PNR system. Students have 48 credits (roughly 4 classes, but not exactly) to place on Pass or No Record (meaning a C or B will appear as a Pass on their transcript, a D or F will not appear at all.) I can assure you that students do make use of their PNR’s. That is why you will generally not see multiple C’s and F’s on MIT transcripts. But C’s and F’s are routinely earned.
This is 100% true. And part of what my kid loved(s) about MIT. They routinely work collaboratively with friends on PSETs, and there is no backstabbing (at least in my kid’s experience.) However, there’s also work that canNOT be collaborated on (ie your code).
The PNR system was introduced during COVID. The fraternity reports referenced earlier show a mean GPA of 4.49 and 4.46 in the 2 semesters prior to this policy compared to 4.49 and 4.56 in the most recent 2 semesters. There was only an average ~0.05 GPA increase, which is consistent with the historical grade inflation trend. Both before and after the policy the mean GPA appears to be A- (assuming non-fraternity students are slightly higher than fraternity students). It is not suggestive of “C’s and F’s are routinely earned”, as you state.
Isn’t MIT Pass/No Record then ABC/No Record for the first year? I am not saying there is inflation, but there is quite the on boarding. You don’t necessarily see this at other schools.
PNR existed before covid. However, it was edited to its current iteration for the class of 2024. In the past, students had to place a class on PNR prior to knowing their final grade. Starting with the class of '24, students wait until after final grades have been posted to make that declaration. The exception was spring 2021: every matriculated student was awarded one “PE” that semester but they had to apply it prior to finals.
My student was responsible for grading as a TA/LA/grader. Yes, students do earn C’s, D’s, and F’s. It’s fine if you don’t believe me though. I’ve said all I have to say on this topic so I’ll be moving on.