The payoff for a prestigious college degree is smaller than you think

College Search & Selection
582

There are a lot of problems with this type of analysis, most of which I have covered in this thread already. You can’t just compare raw averages because there is so much self-selection at both schools as to make the averages meaningless. For example, many young people attend state schools with the goal of becoming a teacher or working in law enforcement. I can’t imagine there are that many that do the same with privates.

The only relevant analysis is to compare those that got into Top 20 privates that attended those privates vs. went to state schools instead, and track that cohort through their careers.

1 Like
583

I am embarrassed to admit I did not know until now that Sergey Brin went to UMD College Park. They do have an excellent CS department and I knew several of the professors and grad students there when he was an undergrad. That’s one that’s going to stick in my mind. (Which is why I don’t think I knew and forgot somehow.)

You make a valid point, but you could equally argue that the other two founders were able to springboard effectively off an Ivy League undergrad education, while the founders of Google only had entrepreneurial opportunities after starting grad school at Stanford. I find the latter point compelling. There are significant advantages to going to an elite school that go beyond the education you’ll receive.

1 Like
584

At full pay, I don’t think you’ll get a better ROI on Rice, Wake, Wash U, CMU, Emory, ND versus Florida, Georgia, UT, UCs, Illinois, Purdue. However, on average, the starting salaries will be higher at the higher ranked privates than the #30-70 publics. I have found that the quality deviation is lower at the higher ranked schools.

The concentration of top jobs is higher at top privates. For example, a much greater percentage of Rice students go to top-tier IB, consulting, tech and med programs than at UT. However, Turing Scholar, Plan II or CBHP at UT? Pretty clear that you’ll have as many opportunities as a Rice grad, if not more, percentage wise.

Georgetown international relations vs Maryland? Go with GTown if can swing it without hardship.

1 Like
585

A relative of mine started working in investment banking after his UC undergrad education and is now, many years later, getting his MBA at Berkeley in order to pivot into something in the health care and/or environmental management field. I can’t remember exactly TBH. But he grew tired of of investment banking. Anecdote of 1.

1 Like
586

Here is some more analysis…

A study by Caroline Hoxby … suggests that graduates of elite schools do earn more than those of comparable ability who attended other colleges. Hoxby studied male students who entered college in 1982, and … projected that among students of similar aptitude, those who attended the most selective colleges would earn an average of $2.9 million during their careers; those who attended the next most selective colleges would earn $2.8 million; and those who attended all other colleges would average $2.5 million.

I am bumping into a fire wall but will try to provide the full study.

587

Did you read the study? It is saying that any earnings difference is eliminated by an elite graduate degree for those with a BA from Tier 1 (elite private), Tier 2 (elite LAC) and Tier 3 (public R1). The only gap is with Tier 4 (others, generally non-selective public/private).

So this does nothing whatsoever to invalidate the strategy of saving money for an elite graduate degree by attending a state flagship.

Specifically:

“Finally, consider only those with graduate degrees from the most selective institutions that are grouped into tier 1. Figure 11 demonstrates that undergraduate tier continues to exert a strong influence on earnings even among those with the most elite post-baccalaureate degrees. The pay gap for men who have bachelor’s degrees from institutions in tiers 1–3 is largely eliminated, but the earnings gap between men in tiers 1–3 relative to tier 4 remains substantial”

image
image742×493 55.6 KB

2 Likes
588

If grads from Tier 1 (elite private), Tier 2 (elite LAC) and Tier 3 (public R1) go to elite grad schools. Knowing that elite privates have graduation rates in the low to mid ~90s, they seem to retain students at a much higher rate than other categories of colleges.

589

This is the conclusion from the author…

The objective of this paper is to identify whether your less prestigious undergraduate
credential can be “scrubbed” by earning an elite graduate degree. If so, then talented graduate school- bound students could choose their undergraduate institution based solely on cost or for any other reason. This paper shows the perils of this strategy. The substantial premium to elite undergraduate education is not overcome for those with a nonelite undergraduate degree even by earning a post-graduate degree from an elite institution.

Are you suggesting the author doesn’t understand the data? The post I responded to said don’t pay for an elite undergrad…take the money and go anywhere for undergrad, then get your “branding” in grad school. This clearly suggests that’s a dangerous strategy. Doesn’t it?

1 Like
590

I think this is a great strategy (and Larry Page and Sergey Brin provide an extreme example of payoff). But to evaluate it completely, don’t you also need to take into account the probability of getting admitted to an elite graduate program with a public flagship undergrad degree? While I also think this is an attainable step that many have made, I suspect there’s still an advantage in starting out in an elite university.

All other things equal, the student attending an elite school will have better advisors, possibly better access to research (granted, public universities do a lot of research), and a greater likelihood of getting letters of recommendations from superstars in their field. Nobody with stellar performance is going to be ruled out of the best grad program by attending a state university, but among students with strong performance, the better university will still give them an edge. I think there are exceptional cases like Berkeley, which at least in computer science would be treated as equivalent or better than any other top program.

2 Likes
591

The issue is that the author uses the term “less prestigious” to refer to Tier 4 institutions. You are misinterpreting the paper as saying prestigious = Tier 1, whereas to the author, “prestigious” = Tiers 1-3 and “elite” = Tier 1.

I don’t think anyone is saying “go anywhere” for undergrad. A for profit diploma mill is not helpful for anyone. The usual dilemma is state flagship at $0K to ~$30K per year (depending on merit) vs elite private at ~$75K per year. The study does nothing to suggest that choosing the state flagship is a “dangerous strategy” if you go on to do an elite graduate degree. The chart above shows that (at least for men), that strategy (Tier 3 undergrad followed by Tier 1 grad degree) delivers higher average earnings than staying at Tier 1 for both. It makes sense intuitively if some elite students hang around for a grad degree (eg because they enjoy the college) rather than using it as part of an earnings maximization strategy.

4 Likes
592

As twoin18 also mentioned, for the author, non-elite means tier-4 and those colleges are conveniently omitted and tiers 1-3 are elite, which are listed.

“A study by Caroline Hoxby … suggests that graduates of elite schools”

Again, for her, elites include Tiers 1-3, so like 250 colleges, Tier1 includes Syracuse, Saint Louis University, and Rockefeller.

593

I agree that saving $$ at undergrad to fund grad school at a more elite school makes sense, and more so if that’s what’s financially needed to get there. I can’t help but think that many schools seeing their top undergrads leaving for greener grad schools is a bitter pill to swallow. But students need to do what’s in their own best interests - and that’s most likely found at the more elite programs.

594

A few observations:

  • Averages aren’t really meanful because your student isn’t an average.

  • A few wildly successful graduates from any one college aren’t representative of that college.

  • If where one gets her/his K-12 education matters whether s/he goes to college or not, then where one gets an undergraduate education also matters whether s/he goes to graduate school or not.

  • Some students will, and some won’t, succeed no matter where they go to colleges.

  • Some other students will be more successful because of the colleges they attend, but how much of that success is due to their college education isn’t quantifiable because it’s highly dependent on each student.

5 Likes
595

I don’t think it does, she’s only saying to attend a Tier 1-3 college, as even Tier 2 and Tier 3 undergrads make more than Tier 1 undergrad after grad school.

“She (Joni Hersch) is:”

She isn’t looking at outcomes related to undergrad degree, only the effect of a Tier 1 grad school. So my point of going to undergrad at a Tier 3 place like Arizona State and then a Tier 1 grad school is supported because their earnings are similar if not higher than a Tier 1 undergrad going to a Tier 1 grad school ($239K to $229K). I just picked ASU because it was first alphabetically.

1 Like
596

@Twoin18, great recommendation letter. So, let’s make the assumption that luminary advisors can help change the probabilities of admission to the best grad schools (and open up possibilities beyond). Moreover, there is a higher concentration of prestigious professors at prestigious schools. Undergraduates at those schools are more likely to come into contact with them and, if they can take advantage of the contact, derive benefit that increases the probability of success at the next step and even the next several steps. [In my case, it helped with grad schools, post-doc and assistant professorship at prestigious schools].

The primary point that I was making it is that is a bit facile to say that it doesn’t matter what undergraduate school you go to if you end up in a prestigious PhD program because it is plausible that where you go for undergrad affects the probability of where you will get in next. And, it may affect the probabilities of subsequent steps as well.

However, as you point out, there are luminary profs whose recommendations can have an outsized influence at a number of schools. If there is a higher level of concentration at certain schools, then which undergraduate school you attend may increase the probabilities of subsequent positive events.

597

Circular reasoning again, just because a professor is at a prestigious college does not mean they’re prestigious.

“if you end up in a prestigious PhD program because it is plausible that where you go for undergrad affects the probability of where you will get in next”

Sure, but the study at Vanderbilt showed that not to be the case, yes where you go matters but as long as you’re in the top 250 (Tiers 1-3), you’ll do fine at least wrt earnings. In fact the Tier 2 LACs and Tier 3 public universities do slightly better than Tier 1.

598

That references this paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2473238

It uses four tiers of undergraduate college prestige, with the colleges listed in Appendix Table 1 (page 46):

  • Tier 1: Private Research I and II schools, including the usual super-selective universities (not LACs), but also including Boston University, CWRU, Howard, NYU, University of Miami, University of Rochester, Rockefeller, Yeshiva, Brandeis, BYU, GWU, Lehigh, Northeastern, RPI, Saint Louis, Syracuse, and Tulane.
  • Tier 2: Private Liberal Arts I schools, basically a large selection of private LACs, probably many of which are not that selective and not well known (even among the LAC advocates on these forums).
  • Tier 3: Public Research I schools, mostly state flagship level universities of varying admission selectivity and associated prestige.
  • Tier 4: all others.

Given that each tier contains schools with a large range of admission selectivity and associated prestige, the claims made are not all that strong.

599

It’s good to read the paper and draw conclusions, rather than take isolated quotes. The paper divides colleges in to selective and non-selective tiers. The definitions of selective/“elite”/“prestigious” are different than the ones used elsewhere in this thread. Some examples are below. Some of the selective colleges in tiers 1-3 have >90% acceptance rate. Others have <10% acceptance rate.

Selective Colleges
Tier 1 = Selective Privates. Examples include St. Louis U, GWU, BYU, Miami, and Howard
Tier 2 = Selective LACs. Examples include Agnes Scott, Alibion, Albright, and Allegheny
Tier 3 = Selective Publics. Examples include Arizona State, Florida State, and Iowa State

Not Selective Colleges
Tier 4 = Anything no listed in tier 1-3

The paper compares earnings among students who received a post graduate degree from tier 1-3 college, and tries to isolate whether it depends on the student attending a selective tier 1-3 undergrad college vs non-selective tier 4 undergrad college.

The author found that with no control for things like college major, professional degree, and employment characteristics that there was a difference in earnings. Kids who attended tier 1-3 for undergrad had higher earnings that tier 4.

After controlling for college major, professional degree, and parents education, the difference generally remained significant. I say “generally” because there was not a significant difference between tier 2 (LACs) and non-selective in some cases.

After controlling for employment characteristics including things like field of work (for example government vs finance), the difference between tier 3 (publics) and non-selective also was not statistically significant in some cases. However, tier 1 (non-LAC privates) still remained significant. Some specific regression coefficients for males attending a tier 1-3 for graduate with all the discussed controls are below:

BA at Tier 1 (privates): +0.13 (0.01) – Significant at 1% level
BA at Tier 2 (LACs): +0.02 (0.01) – Not significantly different from 0
BA at Tier 3 (publics): +0.01 (0.01) – Not significantly different from 0
BA at Tier 4 (non-selective): 0.00 – Reference for comparison

Other variables that were significant at 1% level in earnings include

  • Not being Black or non-US citizen
  • Father having professional degree
  • Mother having any college degree (higher degree level the better)
  • Studying tech or econ/business
  • Working for a larger size company (the bigger the better)

It’s unclear to me exactly why attending a selective non-LAC private was associated with a higher earnings after controls than attending a selective LAC or public. As noted, the “selective” definitions are questionable to say the least, which may contribute. However, I suspect the bigger issue is missing key controls. A key one that is missing is a measure of quality of student, such as test scores or GPA. It might be a relationship like the “tier 1” kids have a higher concentration of academically stellar students than the other tiers, with better stats and such. And the academically stellar students tend to have higher earnings on average.

600

I always thought that in the traditional academic approach, it was considered beneficial to move students between schools as a kind of cross-fertilization. It may vary by department and by school. I think I remember MIT having a reputation for students who stay there for grad school.

As long as they get equally promising students, I think most departments would prefer applicants from elsewhere. (My direct experience is limited to computer science.)

601

The study finds “that the substantial premium to an elite undergraduate degree remains even among those who earn graduate or professional degrees at elite institutions.”

Also, “Among those who earn a graduate degree from an elite institution, the present value of the earnings advantage to having both an undergraduate and a graduate degree from an elite institution generally greatly exceeds any likely cost advantage from attending a less prestigious undergraduate institution.”

And don’t attend a Tier 4 institution if you have high stats. At least reach for a public flagship or get some OOS tuition breaks at another state’s higher ranked flagship if yours is poor.

Funny that it’s Vandy doing this research. A c-suite guy at my old company bragged that his son turned down Vandy for UT-Austin business and can spend the “extra money” on a top MBA. That assumes his son will do well in undergrad and desire a top MBA.