Based on not particularly systematic research, Vassar and Grinnell may be vulnerable under the revised methods of analysis.
The problem is that earnings information is generally not available for the listed colleges. For example, you mentioned Claremont McKenna. Earnings info is available for 3 of the 40+ majors. How do you compare for the other 40 majors for which earnings are not available? My guess is USNWR will handle this by not including the new earnings by major metric in their LAC ranking formula. It will instead only be used in certain other rankings.
The only majors for which comparison is possible are the ones with high enrollment. For the listed LACs, the most enrolled major is usually economics. The order of College Scorecard graduate earnings for economics at 4 years out was:
1 . Williams (highest)
2. Claremont McKenna
3. Amherst
4. Hamilton
5. Colgate
6. Swarthmore
7. Middlebury
8. Bowdoin
– Gap –
9. Richmond (lowest)
Computer Science also has a high enough enrollment to compare for a good portion of the colleges.
1 . Williams (highest)
2. Colgate
3. Swarthmore (different major name in College Scorecard)
4. Amherst
5. Bowdoin
6. Middlebury
NA. Claremont McKenna – No earnings info
NA. Hamilton – No earnings info
NA. Richmond – No earnings info
I expect the primary reason why Richmond ranks lower is it is less selective. More selective colleges tend to have a higher concentration of high academically achieving students, which is correlated with higher earnings after controlling for major. For economics specifically, there is also an influence in portion that want to work in finance/consulting and how well of a target school they are for this path. I expect selectivity of the college will be far more correlated with the new ranking than portion that majors in CS, business, or similar.
Therefore, schools from which the earnings of graduates depart from a selectivity extrapolation will be impacted by the introduction of earnings information.
More precisely, the schools from which the earnings by major ranking departs from existing USNWR ranking will be impacted by the introduction of earnings information.
This might include colleges that are selective enough to have an especially high concentration of high achieving students, but have something else pulling down their USNWR ranking more than is typical for that selectivity. GeorgiaTech is an example, with their co-op program and impact on graduation time pulling down USNWR ranking. CMU also meets this description. CMU is ranked #20 on USNWR, but their earnings by major are usually much higher.
It might also include colleges that have a larger than typical portion of students within a particular major who go in to fields that are associated with relatively lower earnings. Smith College and most other all women colleges are an example (if LACs are included). Smith is currently ranked #13 on USNWR LAC list, but their earnings by major are consistently far lower. Many public colleges and HBCUs also fit in to this category.
I did a little AI work. 
To predict which universities will dominate the Top 20 of the overall U.S. News & World Report (USNWR) National Universities leaderboard with the addition of the new Earnings by Major factor, we must blend the College Scorecard’s major-level salary premiums with USNWR’s remaining legacy ranking pillars. [1, 2, 3]
The Interplay of Old and New Factors
The Earnings by Major metric replaces the volatile Graduate Indebtedness factor. When compiling the overall National Universities ranking, USNWR does not only look at money. The new salary data will be combined with heavy legacy variables: [1, 2, 3]
- Peer Assessment (20%): A reputation survey voted on by college presidents, provosts, and deans. This heavily favors historic Ivy League, elite private, and top public flagship brands.
- Graduation and Retention Rates (~24-30%): Measures how many students stay and graduate on time.
- Faculty & Financial Resources (~20%): Instructional spending, faculty salaries, and class sizes. [1, 2, 3]
Because raw tech academies lose their degree-mix advantage through major normalization (e.g., comparing a math major only against other math majors), the ultimate Top 20 winners will be universities that possess elite multi-major salary outcomes combined with unmatched reputation and graduation data. [1, 2]
Projected Top 20 National Universities
Taking into account the multi-variable formula, these 20 institutions are uniquely positioned to command the top spots on the flagship USNWR list:
1. The Big Three Juggernauts (Projected Ranks 1–3)
- Princeton University: The perennial #1 remains virtually unshakeable. It pairs elite graduation rates and peer reputation with immense major-normalized earning power. It suffers no penalty from the elimination of student debt data due to its generous no-loan financial aid. [1, 2, 3, 4, 5]
- Massachusetts Institute of Technology (MIT): Even when normalized by major, MIT graduates routinely out-earn their direct discipline peers nationally. MIT combines this with a near-perfect peer assessment score. [1, 2, 3]
- Stanford University: Benefiting heavily from regional wage scaling in California, Stanford’s humanities, sciences, and STEM tracks all out-earn national benchmarks, cementing its lock on the top tier. [1]
2. The Legacy Elites with Surging Outcomes (Projected Ranks 4–10)
- Harvard University: While some niche majors have lower outcomes, Harvard’s unmatched peer reputation (20% weight) and flawless graduation metrics insulate it, keeping it firmly in the top 5. [1, 2, 3]
- University of Pennsylvania (UPenn): UPenn wins heavily under the new framework. Its pre-professional ecosystem (like Wharton and engineering) yields massive 4-year salary premiums for aid recipients, pushing it past peers that lack structured pipeline programs. [1, 2]
- California Institute of Technology (Caltech): Already dominant in raw earnings, Caltech will easily maintain a top 10 footprint because its student outcomes are stellar across every singular department it offers. [1]
- Yale University: Yale’s immense academic spending per student and elite peer reputation counteract any slight major-normalization penalties from its large humanities cohorts. [1, 2]
- Duke University: Duke blends high graduate outcomes in economics, biology, and computer science with robust legacy institutional wealth metrics. [1, 2]
- Brown University & Dartmouth College: Both schools see a boost from the elimination of the Graduate Indebtedness metric. Their strict undergraduate focus drives exceptional graduation and retention scoring. [1, 2]
3. The Major Multiplier Winners (Projected Ranks 11–15)
- Carnegie Mellon University (CMU): CMU is a prime candidate to leapfrog into the stable top 12. Its computer science, design, and drama programs command extreme marketplace salary premiums, pairing nicely with strong peer assessment.
- Johns Hopkins University (JHU): Elite medical, cellular biology, and biomedical engineering pipelines ensure its federal aid recipients scale past early-career wage expectations.
- University of Chicago: UChicago benefits from high-earning economics and data programs, which will offset any minor corrections from its more theoretical academic tracks.
- Northwestern University: Balanced beautifully between an elite journalism/humanities reputation and immense corporate placement power in finance and engineering.
- Columbia University: Despite past reporting turbulence, Columbia’s core metrics—fueled by Wall Street and NYC tech placement—keep its major-by-major returns exceptionally high. [1, 2, 3, 4, 5]
4. The Flagship & Co-op Disruptors (Projected Ranks 16–20)
- University of California, Berkeley: UC Berkeley is the public university most likely to challenge the top 15. Because the new metric isolates federal aid recipients, Berkeley’s massive economic mobility engine will score maximum points, neutralizing the legacy financial resources advantage held by smaller private schools.
- Georgia Institute of Technology (Georgia Tech): Georgia Tech will be a massive beneficiary. Its mandatory, structured co-op tracks mean graduates hit Year 4 with professional seniority over their peers. This structural salary boost, blended with an ascendant national reputation, pushes it toward the Top 20.
- Cornell University: Cornell’s highly specialized practical colleges (Engineering, Industrial and Labor Relations, Hotel Administration) give it an immediate advantage in major-by-major early career salaries.
- Georgetown University: Driven by top-tier placements in international relations, finance, and political science, Georgetown’s specific major outcomes heavily outpace national medians, solidifying its place in the top flight.
- University of Michigan—Ann Arbor: Alongside Berkeley, Michigan represents the peak of comprehensive public flagships. Its global alumni network pulls humanities and STEM salaries alike well above national baselines. [1, 2, 3, 4, 5, 6, 7]
We are doing all this analysis here and … the average person will just, as usual, open the rankings page and look at the numbers, and not dig any further…
3 Likes
Like typical AI responses, it sounds right to someone who is unfamiliar, but there are a lot of issues with the details. For example, if the weighting continues to be only 5% like graduate indebtedness, then that greatly limits potential effects. Even if GeorgiaTech had a perfect score in this metric (it won’t), the new metric would not lead to the large jump from #33 to #16 that the AI response lists.
I think the far more likely outcome is the rankings do not change much by design. USNWR tweaks the formula in such a way that the expected names appear towards the top such that the rankings look right to readers, but there are enough generally minor changes in specific colleges to give persons a reason to buy the new rankings instead of rely on last year’s numbers.
I suspect the larger year-over-year changes than in the previous 2 years that USNWR mentions more relates to reclassifying colleges on different ranking lists, such as Babson moving to regional ranking list, as discussed earlier.
As an example, Bentley could lose its #1 spot because of this change. This wouldn’t be a large change when considered statistically, but it would be a significant change perceptually.
The press release implies to me US News might already know there will be some “news” made by some rankings changes. I doubt anything as radical as some of what that AI predicted, but enough to be considered significant in US News superfan circles.
1 Like
I disagree to a certain extent. The concern that a ranking is becoming stagnant and obsolete is real. I do believe they need to and will shake it up as well. Time Magazine have just launched their inaugural college ranking a few weeks ago which was actually interesting and quite excellent. It partnered with Statista to launch its own data-backed project which places a heavy 75% weight on student outcomes and return on investment. Which I personally believe is truly what people are looking for in rankings and in selecting colleges in this current era. Time’s new rankings have been described as a “more modern, outcome-focused lens compared to traditional methodologies used by older ranking publications”
So, I do definitely think there will be movement and corrections and USNWR needs a new splash, so to speak . With regard to your specific mention of Georgia Tech. I believe they will definitely jump in the overall ranking as a correction of sorts with the addition of the new factor. We will see though by how much as well as others.
1 Like
Finally, someone who agrees with me Others on here continue to post that no one cares about stats and people don’t send their kids to college to find jobs.
Clearly, there is a diversity of opinions in society but it always seems like on this board, people largely disagree with this thinking.
But I can’t imagine for society as a whole - on a mass level, what else could it be?
Whether Time or any other ranking truly captures what the best college is - and if a ranking can be done for the masses when everyone’s situation is different - I don’t know.
But I’m glad there’s finally some rationale thought on the board
1 Like
Sorry, but there is no universe where the same people applying to LACs are going to suddenly start sending their kids to Brigham Young and Colorado School for Mines simply because they have higher (arbitrarily determined) ROIs than Swarthmore. None.
7 Likes
Agreed - but I said society as a whole - meaning - just as a state may have one lead party (like the Republicans in TN), it still has voters from the other. But on the CC, it seems nearly a one party system to use the analogy. Here, you’d think no one cares about job outcomes but more about learning to read and write and having a good experience.
Those things happen, or should, at every college btw - but at differing levels depending on major.
I do think, as costs go up and incomes don’t keep pace, more and more (that don’t qualify for need based aid) will likely lean toward those Mines type schools (well affordable Mines type schools) than in the past.
And let’s be honest - it’s not just the Swats of the world - but lots of LACs - from a Bucknell to Richmond to Dickinson and more - have some of these “outcome based” majors thus differentiating them from the top schools - that most can’t sniff anyway (like a Swat).
I hire for a living. Have done so for over 35 years. I have mostly worked for large multi-nationals, with a focus on US professional hires, with a few stints at smaller organizations.
I have seen trends come and go. I lived through recessions–software engineers driving taxi cabs- who’d have thunk it?, Math PhD’s trying to get “alternative certifications” to teach HS algebra, petroleum engineers who were frantically enrolling in their state’s “workforce development” program to get low cost training in something- anything- so they wouldn’t lose their home.
I cannot predict the labor market. I say that with 100% certainty. I was part of a global task force to figure out how many people from our “e-commerce” team (all hired over a period of 14 months) we could save and relocate to other functions, parts of the world, etc. so we didn’t have to let all of them go in one fell swoop. I had run a recruiting initiative to hire many of them which is what added insult to injury.
I am also humble enough to admit that today’s trend becomes tomorrow albatross. Back in the early days of the internet, I remember conducting an exit interview with a high performing young employee who was leaving a good job with one of those “Top employer” designations (not a flash in the pan- earned over decades) to join a start-up which was going to “disrupt” the pantyhose market. He had never worn pantyhose-- I had, since my very first job out of undergrad, and even earlier as a waitress where hose were required along with the polyester uniform and hairnet. I tried to explain to him that the pantyhose market didn’t need disrupting. Women were disrupting it all on their own-- person by person- by wearing pants to work, by wearing bare legs, by just NOT BUYING the itchy and uncomfortable garments which had replaced the even more uncomfortable nylon stockings. He wasn’t buying it. He was employee number 8 or so, was getting “equity”, very confident that every single market out there was ripe for disruption and that online shopping for pantyhose was going to be a hit. The enterprise lasted a year. I now buy pantyhose a few times a year (mostly to go to funerals) instead of every week. And most of my friends buy them-- never.
I think of him every time I read or meet someone who is confident in their ability to predict ROI, or the success or failure of a startup, or who “knows” what AI is going to do, need, require. I think of him every time I talk to a parent who is convinced that medicine is the only “safe” career (my radiologist is based in Mumbai- has been for over a decade, so that’s not due to AI, just due to better digital imaging). And I think of him every time someone tries to convince me that spending four years in college learning to do a discounted cash flow analysis and understand a balance sheet, or code in C++ (remember when everyone “knew” that was the key to lifetime employment?).
Many of these “safe careers” are going to be blown away. I don’t know which ones. I’m not that arrogant.
But I do know that with every recession and every “Black Swan” event that hits the professional labor market, the drive towards the basics- analytical reasoning, the ability to write convincingly, skill at interpreting events and translating those events to smart people who may or may not be subject matter experts- these skills remain evergreen.
I’ve written before about the best finance hire I ever made. A geology major who knew nothing about business, finance, had never examined a balance sheet-- ever. But his senior thesis had involved building a model to predict seismic events (tsunamis, earthquakes, etc.) and the MD’s went nuts over this kid. Who knew that predicting the unpredictable was a core skill at a hedge fund? More important than a degree in finance or econ; more important that getting an “A” in Valuation or passing all three levels of the CFA on the first try???
Stay humble my friends. You can’t beat the markets long term- they will beat you first.
18 Likes
This seems a misrepresentation of the position of most posters who do not see things as you do. Most do not argue no one cares about stats, but rather that the “stats” do not actually show what they purport to show because of various issues in the data and methodologies.
11 Likes
ehhhh - they say that but also that’s not why kids go to college - i.e. to get a job but rather to grow, etc.
But it’s fine - all can believe as they want.
Basically, with that statement though, people are saying schools lie - and yes, some point out differentials (like cost of living, which is totally fair).
And the truth is, not everyone is after money - and that’s pointed out too.
But I don’t think the CC is representative of society.
But back to rankings
It’s funny - there’s so many - and they get us all riled up.
Yesterday, @barrons tried to insinuate certain schools don’t deliver - when in fact, I showed they delivered data.
Others don’t.
btw - here is Wisonsin (which is @barrons school) - they don’t provide a knowledge rate but for Mech Engineering, are right in the range of other schools - from UMN to Bama to Ga Tech to Michigan State, etc. I do like that they list the job title and company captured, which is helpful, like for English when you see someone is managing a coffee shop - and btw - there’s nothing wrong with that - and there’s nothing wrong with making less $$. But I don’t think the masses want to see less $$ - and that’s what surveys are looking to attract.
I can see one dismissing this data to an extent because we don’t know what % is captured - but to act like it’s fake seems silly - to me.
But we’ll always agree to disagree. And all can comment on US News until Sept 22- when all the comments will be null and void and whatever results they come out with - will be analyzed, celebrated, ridiculed and more.
First Destination Survey – Data, Academic Planning & Institutional Research – UW–Madison
I would love a stat which shows “More kids who graduate from U Mass end up happily employed in their chosen field, vs. kids who graduate from U Oklahoma”. That would be nice.
But showing that salaries along the Eastern seaboard are higher than salaries in Oklahoma, Arkansas, Kansas and rural Texas-- nah. I already know that. And not because I studied statistics or econometrics in college. But because the things I studied taught me how to interpret data, understand the dangers of projecting from a small, self-selected population, etc.
The really wonderful thing about Classics is that history is written by the survivors. And understanding how much bias is baked into the documentary evidence we have–that’s a skill that is taught. And that using other forms of evidence to corroborate or repudiate documentary evidence (artistic, archaeological, anthropological, architectural, all the “A’s”) is critical for an understanding of the past.
I have mentioned this at my interviews for every single job offer I’ve ever gotten. One interviewer cried (he was a Renaissance Studies major, now a CEO). This is not to say how wonderful I am. It’s just to point out that fact- based analysis is not just taught in a chem lab. And speaking of- remember when “cigarettes cause cancer is just a theory”? Lots of chemists have a lot of “splaining” to do when they reach the pearly gates!
5 Likes