Sorry, my bad. I meant to say Harvard first-years representing the bottom 60th percentile of U.S. household incomes, which I believe peaks at ~72k a year. Again, according to your own sources, they represent almost 30% of the incoming class.
If we’re talking the “cultural vibe” of a college, just looking at the income distribution is only part of the picture.
I am more familiar with the Yale resources for kids from the lower income brackets- but I imagine Harvard’s resources are even more robust.
Kid needs an emergency bus ticket home because grandma’s in the hospital? The Head of College (formerly the Master) has discretionary funds. Kid needs a tuxedo for an orchestral performance? Ditto. Kid wants to take a low paying or no-paying internship for the summer? There are dozens and dozens of paid fellowships which allow the kid to earn what he/she would have earned in a paying job, many of which include a travel stipend. Kid wants to go with the a cappella group on their European tour during spring break? There is a student activities fund which will quietly pay airfare and all expenses.
So the days when you could readily distinguish between the kids who had discretionary funds and those that didn’t… not so apparent now. No, Yale is not buying bottle service at fancy clubs in NYC for the kids on financial aid. But for the elements which represent “social capital”-- performing arts (which incurs expenses), internships, travel, etc.- those get baked in for kids who need and want them.
I think the income disparities will be more pronounced at the colleges which don’t have the endowments and generous alums that the HYP’s have, but still have a fair number of high income kids. I’ve known plenty of kids from bottom quartile earning families at the HYP/similar colleges and they lead pretty enviable lifestyles while living on campus!
5 Likes
I looked up the Chetty data on Harvard myself, and found that 67% of their students come from the top 20% of the income distribution, whereas 4.5% come from the bottom 20% of the income distribution, and the rest in between. This is much more beach-cliff shaped than barbell shaped. And as the previous poster said, Harvard has more kids from low-income families than most selective 4-year colleges. I did some extra poking around on Chetty’s research organization websight — opportunityinsights.org — and found that you do get a barbell shape IF you hold SAT scores fixed. So for the subset of kids with a 1400 SAT, there are more low-income and high income kids at the most selective colleges than you might expect, and fewer middle income kids. However, numerically speaking, there aren’t very many low-income kids with 1400 SATs…
( see Education | Opportunity Insights for data)
As for the OP’s question, my sister and I both grew up in a middle class income family (Albeit with well-educated parents - a JD and an Ed.D - so you might call us upper middle class, culturally?) We both went to well-known LACS on the East Coast. My sister followed up with a public health Ph.D at a top-20 university and she definitely became more normalized to things like jetting off to Prague for friends’ birthday parties, or the fancy wedding things… but once she had kids that stopped. I think the main difference is —as others have mentioned — she definitely has ideas about how things should be done, and is not shy about getting them done. She is very active in her local community with schools, sports leagues, etc. Admittedly, it was her drive that got her into her top-tier Ph.D program in the first place… so which came first? Hard to tell…
3 Likes
As did I. It’s already been established that the data is more than ten years old and has been superseded by what @Data10 and I both refer to as the Harvard Freshman Study:
I think the phrase refers to the point in anyone’s thought experiment where the folks from families earning below the median U.S. household income ends and where the folk earning high six figures begins to explode. Yes, you can design any number of bar graphs and line curves that render FGLI students all but invisible, but a barbell is the image that persists.
As typical, per highly selective college admissions, I think the problem is parents/students lack detailed information about the admission process and why students are accepted/rejected, so they make guesses and assumptions, which are often incorrect. If there was a public, histogram that showed the shape of the income distribution, I expect the barbell myth would quickly become far less common.
However, without adequate information about income distribution, the myth persists. Parents/students see colleges making a big deal about portion of Pell grants, first gen, and other groupings associated with lower incomes, and parents are aware of large portion of high income students and favoritism given to certain high-income groups. So if a well-qualified student who is not in these favored lower income and favored higher income groups is rejected, the parent/student might assume the rejection occurred because of the students’ income – the college accepts low and high income kids, but they accept very few kids in their income group (which is often better described as high income than middle income). The lack of kids being accepted in their income group creates the image of a barbell shaped distribution.
Lack of knowledge about FA can also create this type of rumor. Some believe that low income kids get enough fA for a free ride or near free ride, and the rest are full pay or near full pay. Only the wealthy can afford the $80k/yeer price tag, and only the low income kids get a pass on the cost; so middle income kids can’t afford and do not attend, creating a barbell shaped distribution (at college without merit scholarships).
3 Likes
My kid is at a respectable private college in PA, receiving FA. He was honestly just floored by the wealth. Kids who spent summers in Europe, expensive cars, contacts with influential people etc
But it didn’t change him in any way. They became his friends and that was that.
I am confused about this whole barbell discussion though I think the original question is interesting. How does the barbell image relate to the “donut hole” image that I see mentioned on these pages occasionally? I am following this discussion but I feel like I am misunderstanding the relationship between a barbell (or other shapes) and the effect of college on socioeconomic expectations.
It seems like the distribution in family income on a college campus could be caused by any and probably all of the following.
- cost of attendance
- need based financial aid policies
- merit aid policies
- number of students in each income bracket who apply
- number of applicants in each income bracket who are admitted
- which admitted students choose to attend (either for affordability reasons or school culture/fit or academic programs) --are admitted students from some incomes brackets more or less likely to enroll and why?
My vague impression (since I don’t really understand the metaphor) is that concerns about a barbell or other shapes are usually used to critique the high COA at so many colleges and financial aid policies that don’t meet perceived need --that is whether admitted students can afford to attend. But it also seems likely that that the other items that I listed also play a role even if it is not as large as the first two.
1 Like
All this talk of barbells has me thinking of Culebrita, a tiny island-paradise off the coast of Culebra. Culebra is a larger (but still small) island off the coast of Puerto Rico.
Culebrita is shaped like a barbell – a 40-20-40 barbell. hehe
To make this post apropos of the thread, I will say that my friends at UW were generally upper-middle or middle class – two sons of a well-off farmer (that is a thing in Wisconsin), the son of a mid-level IBM employee, the daughter of a personal financial planner. I was the son of a health care administrator (native American tribal clinic) and high school choral director. This was pre-smart phone, so the toys we sought then were the latest video game consoles and a decent laptop with internet capability. That was a big thing in the mid-90s. Nobody drove cars on campus until we were seniors – then it was my '81 Skylark, an early-90s Integra, an old Ford Escort with a slipping transmission. None of us were quite from jet-setting families. The boys of the farmer probably could have been, but nobody from Hillsboro is flying to Cape Cod weddings.
2 Likes
I always thought the “barbell” was supposed to refer to the probability of admission (presumably for qualified applicants)? So the probability of an applicant getting in to a top college was highest for those with the lowest and highest incomes.
2 Likes
Oh, gotcha! Thanks. That makes sense. So I totally misunderstood. I was thinking it was about who could afford to attend, but now that you put it that way, this back and forth is clearer. And that seems to be what the poster above was saying with the 1400 SAT (yes, I picked up on the fact that this data is out of date, but it is interesting).
However, the probability of a student from a low income family getting a 1400+ SAT is lower than for an equal ability/motivation student from a high income family, due to the latter having better support at school (teachers/counselors/etc.), school courses and curricula in typical neighborhood schools in high versus low income areas, and more parental support (money for test prep and multiple test sittings, greater likelihood of parental knowledge and encouragement about college in general).
I think this is spot on, only in this context it applies to admissions, and not just paying for college.
Parents who aren’t getting their kids into the schools they think they should seem to be increasingly bitter and resentful at those supposedly stealing their spots. Perhaps because of the futility at punching up, they choose to punch down. Blame the low FGLI kids. Blame the URMs. Push them off the bar all together.
A better metaphorical shape for the curve would be a club.
3 Likes
For my son, Princeton '22, it wasn’t so much a “shift” as a broadened perspective that came from being exposed to an environment of a hitherto inexperienced level of wealth. Coming from an upper middle-class suburbanite kid, he saw and experienced a display of wealth at an entirely different level. For my other son, flagship public university '20, his experience wasn’t anything even remotely resembling. Both sons are still the same with their values, fiscal management and discipline, but there’s a big chasm in the way they understand wealth and what possibilities wealth can bring.
6 Likes
Son is a freshman at UD and there is definitely a lot of wealth there. We are from MA and he has made a lot of friends from New England. He doens’t ask about heir wealth but they are from the wealthiest towns and several went to boarding schools. His roommate’s dad owns a Lamborghini. He spent a lot of $$ first semester trying to “keep up with the Jones’s”. After having to work 7 days a week on his break, we think he has learned a lesson. We are not poor and are funding his education but spending $$ is on him. We’ve met a lot of his friends over break and they are really nice, down to earth kids. But they’ll go back to school wearing their new Patagonia flannel shirts and he’ll be in his Walmart knockoff. Luckily that stuff doesn’t bother him at all. If anything, I think its motivating him to work harder as he seems to like how the “other half” lives. He’s been invited to some really great restaurants when his friends’ parents have come to town and he’s really enjoyed talking to the parents about their careers. I’ve also reminded him to always write a thank you note, I’ve stressed to him that any of these parents could be a great connection once he gets into the business world. I
I’m sure the next 3.5 years will be a challenge as we will have to say no to a lot of things he’ll want to do but that’s part of life.
4 Likes
The poster who first mentioned the “barbell” implied it was reference to the distribution of students at highly selective private colleges – “very low and very high income, with little in between”. The large number of students an either spectrum of income with little in between creates a barbell shaped distribution, and this bi-modal barbell distribution impacts the SES expectations that are the topic of this thread. As discussed, the reality is quite different, without a disproportionately large number of low income students.
Prior to this thread, I’ve never seen barbell or donut hole refer to probability of admission. In general, probably of admission increases as income goes up. However, a small minority of colleges do give a strong admission boost to lower income kids, one of which is Harvard. The Harvard lawsuit docs presents an internal Harvard OIR study that shows predicted admit rate steadily increases as income increase. Higher income applicants have better average reader ratings than lower income kids, as well as a higher rate of ALDC hooks. Test scores were an especially weak point for lower income applicants. At the lowest income groups, ~15% of applicants scored 1500+ while at the highest income groups who applied for FA, ~40% of applicants scored 1500+. However, the actual admit rate did not following the predicted admit rate based on scores, reader ratings, and race. Instead, the actual admit rate was fairly flat across all income groups that applied for FA, presumably due to Harvard applicants who received the “SES disadvantaged” flag getting a significant boost in chance of admission.
Rather than low income groups having a low admit rate, the disproportionately small number of low income kids at Harvard more related to a disproportionately small number of low income kids applying to Harvard. This effect has been noted in many studies. For example, the Harvard-authored study at https://www.nber.org/system/files/working_papers/w18586/w18586.pdf begins by stating, “We show that the vast majority of very high-achieving students who are low-income do not apply to any selective college or university.” If academically qualified low income applicants generally do not apply, it is not surprising that they compose a disproportionately small portion of matriculating students.
1 Like
The study reported kids with 1400+ SAT had the following income distribution:
Top Quintile Income – 59% of kids with 1400+
Upper Quintile Income – 20% of kids with 1400+
Middle Quintile Income – 12% of kids with 1400+
Lower Quintile Income – 6% of kids with 1400+
Bottom Quintile Income – 3% of kids with 1400+
Few low income kids score 1400+. The lower the income, the more rare a 1400+ SAT becomes. That increased rarity increases chance of unique circumstances besides just score that were not controlled for and may be more of a driving factor. For example, perhaps a larger portion of low income kids with a 1400+ attended a selective HS, such as selective magnet or private, than middle income groups with comparable score. The study linked in my previous post found that attending a selective magnet/private HS was a strong predictor of applying to Ivy+ type colleges, including among low income kids. Or perhaps the rare low income kids with 1400+ are more likely to be recent immigrants from a country that generally places a strong value on attending highly selective colleges, which could also increase chances of applying to Ivy+ colleges over other income groups.
The reasons for the results is not obvious, but what is obvious is only a small portion of 1400+ SAT kids are low income, and only a small portion of kids attending Ivy+ type colleges are low income.
1 Like
Thank you for the insight! Unfortunately I do think our D23 would be bothered by the constant difference even if in theory she shouldn’t be. It’s more the day to day little things like someone mentioned up thread: going out to eat, going out for coffee, getting a hotel room somewhere for an event, concert tickets, exchanging bday gifts, etc. It’s not about the friends themselves at all. We have no problem with wealth at all. We are are plenty privileged ourselves already. Plus honestly it’s actually pretty fun to have some wealthy friends here and there.
It would just be nice for her to have a few friends who are similarly financially limited in college.
4 Likes
But, there will never be a disproportionately large number of low income students at Harvard. That is what makes this entire sub-thread an example of reductio ad absurdum. For Harvard to perfectly represent the socioeconomic diversity of the United States, it would have to shrink the size of both its student body and faculty by about half and devote all of the putative “savings” to need-based FA and some sort of politically feasible lottery system to gain admission. In effect, it would have to become a “high-class” version of Berea. Which has some attractive possibilities, I admit. 
1 Like
Public non-selective colleges and 2-year colleges tend to have a SES distribution that is a reasonably good reflection of the US population. Selective colleges generally do not. Nobody claimed that Harvard has, should have, or will have a SES distribution that is a good reflection of the US population – only that Harvard does not have a barbell shaped distribution that was initially described as “very low and very high income, with little in between.”
I don’t want to go in to a lot of detail because this is off topic from both the thread and barbell distribution tangent, ,but many options were proposed in the lawsuit that do substantially increase portion lower income students. The Plantiff proposed this was an adequate substitute for racial preferences. For example, under simulation 6 proposed changes, an estimated 65% of the student body would be flagged as “SES disadvantaged”, which corresponds to below ~median US income. I don’t think this type of proposal is realistic, but it is theoretically possible, with Harvard’s large endowment.
Yet at least two posters have consistently made it the crux of their critique of the barbell metaphor: