New USNWR rankings live now

Because Princeton calls itself a university and has a grad school. Of course according to that standard, why is Wesleyan considered a LAC? Size might make the difference, I guess? I admit there is a gray area.

It does make sense, though to separate LACs from universities (however we’re defining them), because the experiences and resources are very different. People often seek different kinds of college experiences from one category or the other, so it makes sense to assess them on different scales.

1 Like

sure, but in many fields (biomedical science) the graduate presence is TINY. for example, NIH funding for Princeton is equivalent to one floor of a med school building at Hopkins or UNC etc. . If a kid wants to do some research and apply to med school or graduate school in biomedical science there’s not a great reason to rank Princeton in a completely separate category from lets say swarthmore, williams or amherst… if the goal is to be a guide for choosing a college

undergrads at Universities are assigned to either older undergrads, grad students or technicians. Could argue this is worse than being directly supervised by faculty at an LAC.

1 Like

On citation impact, most would agree with you. But on whether reputation is relevant when choosing an undergraduate college, the camp that believes so and the camp that believes otherwise will never be able to convince the other party. Articles after articles have been written on this, and the subject has been beaten to death on CC, Reddit, Quora, and elsewhere.

2 Likes

LOL. I would bet being in the lab with the highest citation impact at Harvard would be a horrible experience for an undergrad… would probably convince them to skip research/PhD and go to Med school.

Did you just contradict yourself?

I personally don’t mind a small reputation ranking, I just think it could be much, much smaller and not the single biggest individual criteria. But I understand why they do it.

The ranking companies would like you to believe they develop a methodology based on merit and that it is applied objectively to all colleges and the chips simply fall where they may, with it just happening to be the case that the most well known schools tend to dominate. But we all know this isn’t the case and in fact the opposite is. USN’s ranking is clearly massaged to end up with at least enough of the results matching preexisting expectations that we lend it “credibility.” If HYPSM were not in the top 10, even people less critical than this group would inherently suspect it was “off.” So whenever they tweak a methodology, I guarantee you they run simulations of the impacts and internally negotiate and try different variations until they arrive at a carefully crafted output that they believe strikes the balance of being believed by the masses, suiting their institutional objectives (for example, in the latest case demonstrating they are listening to the criticism about being too elite-biased) and throws in just enough unexpected changes to generate headlines and buzz. There is absolutely nothing organic about the process or the methodology. It’s a business. So in their case, that 20% is what guarantees the level of elites outcome believability they feel they need.

8 Likes

no. labs that churn out lots of high profile/citation stuff can also be snake pits and not friendly places for undergrads. Therefore although USNWR would rank that school highly, it would not necessarily be fun for an undergrad to be in that lab. (Undergrads are RARELY on publications that result in high citation papers because they take many years from start to finish and would not likely justify authorship for an undergrad.)

3 Likes

Citation impact is a new criteria. Reputation was part of the old calculation at a similar weight.

1 Like

One hundred percent agree with this. If you look at the T25, it’s hard to take issue with the ranking, even with some movement, all of these schools have legitimate claims to being academically excellent. All of the somewhat crazy swings of 10, 20, or 30 plus places occur further up with list.

USNWR supposedly got rid of rejection rate a few years ago; if they get rid of reputation that would be a big improvement. no one knows who fills out these surveys; the response rate is dropping, and hosting conferences with free swag and inviting other universities’ admin is a way to boost it, as Northeastern did.

1 Like

They might as well address why the career outcomes of the 1/3 receiving federal aid are worse. If the outcomes were good, it would not affect their ranking, like the other ivys above them. Anyway they are still #18. Not sure why there are naming some publics ranked higher than them in the statement.

someone’s going to tweak their algorithm (A stat professor from Reed figured out the old one that included rejection rate) - and get rid of citation impact and reputation and put out a ranking I hope.

I understand why we don’t want rejection rate to be a big factor in rankings, but what in these ranking methodologies measures the academic stength of the students? boards scores are being left behind, I don’t see high school class rank or GPA…wouldn’t many students and parents care about the academic strength of their student body peers? that seems important to me- although hiow to measure it is still an issue (and I don’t think graduation rate alone is enough).

5 Likes

On the flip side, undergrads at top places may be participating in truly cutting-edge research projects funded by NSF, NIH, DARPA, DoE, ARPA-E, etc., thus gaining valuable knowledge and perhaps publication experiences that set them up for success in graduate school applications. Co-authoring papers as upperclassmen isn’t that rare at top places.

In contrast, undergraduates at places not known for research may also be working on research, perhaps even in friendly, less cut-throat environments. But the research topics may be minor, inconsequential extensions of something already well-understood and thus are only good for lower-tier conferences/journals, or they may be practical problems some local companies pay professors 50k a year to solve that have little intellectual/publication values.

1 Like

these schools can fill their classes many times over with perfect test scores and grades. They chooese not to do that on purpose; they want future bankers and donors too.

the most “meritocratic” school is probably Caltech. of the other ones probably MIT. beyond that there’s donor list students, legacies, and athletes to make sure future donors are made.

For anyone interested in continuing the conversation about colleges and enrollments, demographic cliffs, type of colleges affected, or regional differences/impacts, I’ve started a separate thread so that this (meandering) thread doesn’t get derailed: College Sustainability: Enrollment, Types, & Regional Impacts

3 Likes

Can you say that the school deserved the ranking from US News if its a member of AAU, consistently high global ranking, top ten in government research funding? Some of the global rankings are legit although some of us wanna diminish it’s importance because it doesn’t apply to undergrads but it can enhance your reputation as a school.

High school rank was part of the criteria until this year. Not sure why that was dropped but test scores were kept in a test optional and test blind world. Selectivity had been a factor up to a year or two ago but is no longer. Definitely seems like current version overweighs outcomes and underweighs actual undergrad academic experience and peer group.

2 Likes

For the USNWR, are you looking only at national or both national/global ranking?
Sounds like both of them have great reference values and the rankings are alot of difference.

1 Like