Leaving aside the suggestions of fraud, this is also a good demonstration of Goodhart's Law: "A measure that becomes a target becomes a bad measure".
There are good reasons to believe that, for example, selectivity of admissions or graduation rates are some kind of quality signal. But as soon as it becomes strategically important to 'score well' on those metrics it becomes well worth universities doing things that will improve the metric without improving the quality - such as rejecting qualified applicants or graduating those who ought to have failed. And now the quality signal is much less useful if you're simply trying to use it to understand what's going on.
I can't speak for the math department, but I've noticed that there are general contentions between Columbia's faculty and administration that seem somewhat unique to the school. Professor Martin Chalfie has noted how cramped and run-down his lab is in an interview, which seems like soft-spoken criticism at the way Columbia has failed to use its enormous endowment towards acquiring more research and teaching spaces.
Perhaps someone at Harvard or MIT can chime in with whether such contentions exist outside of Columbia, but I'm not too surprised to see a professor criticize the school for somewhat blatant wrongdoing.
In 2022, MIT is a cesspool of fraud and corruption. This wasn't the case a quarter-century ago.
The extreme competition seems to be overtaking all of the elites. MIT is the #1 university brand in the world, but it spent most of its integrity and soul to get there.
There's a fine line you need to walk to become a faculty member at elite schools.
You need to lie a little on grant applications to align when you want to do to what will be funded. You need to lie a little bit in publications so they have impact. You need to fight for credit, sometimes on work you didn't do. As these become normalized, winners do these more and more; otherwise, you won't get that faculty job.
The culture slowly trickles down. I think most grad students at MIT are still honest, but not the most successful ones (most of the ones who find faculty jobs are at least a little bit crooked). Second-tier school faculty slots are filled with graduates of first-tier schools.
MIT has a traditional hacking culture which emphasizes breaking rules. This worked well when this involved climbing on rooftops, but it works less well when the endowment is O($100M) per faculty member, and there's money to embezzle through complex corporate schemes and financial games.
Contentions don't exist too much between faculty and admin right now, but they definitely do between students and Institute. MIT grad students are working to unionize, and the Institute, to union-bust.
Can you provide more context on how MIT spent most of its integrity and soul to build its brand? I went to MIT over a decade ago and I always appreciated how honest, hardworking, and genuinely curious the student body was. Professors always seemed brilliant as well.
Not to diminish what you wrote, but almost every single bit of it could be said of almost every research-heavy university in the US today. It's academics in general today, not just MIT.
> I think most grad students at MIT are still honest, but not the most successful ones (most of the ones who find faculty jobs are at least a little bit crooked).
While this might be true, I suspect the statistics are not that different from the general population.
I don't think so. In my experience, most people are pretty decent and honest.
There's something which extreme competition does which breaks that, both by whom it selects, and by how it changes culture. MIT a quarter-century ago was much more honest than MIT today. Peer institutions -- one stop down -- are also still much more honest, although I suspect that will change in a decade or two.
The same change happened in several of MIT's peer schools.
Having worked at several F500 multinationals, I can say confidently that, yup, no different from gen pop. They were corrupt as hell.
Big SaaS providers offered us 40k each to get on board with some all-in Cloud offerings. I'd bet my hat some of the offshore IT companies we used were throwing kickbacks to Ops Managers.
Not speculation, either -- there were executives fired for such things. One set up a shell company and was billing it for consulting services to the tune of 400k per year. He got busted and probably faced charges, probably.
FYI: Executives of F500 multinationals are not the "general population."
The competition needed to be an executive at a F500 multinational extreme. There are many good studies on the topic, from "Dictator's Handbook" to "Power." The latter should be read with a grain of salt, but has helpful insights.
Elected politicians -- above some level -- operate under analogous constraints too. If you don't take an election donation in return for political favors, your opponent will.
I think the decline of the integrity of MIT (as for executives and politicians) became inevitable once competition and $$$ reached a certain threshold.
The rankings lag reality by quite a lot. I think MIT probably was the #1 university in the world a couple of decades ago, but wasn't ranked ultra-high then. It is ranked ultra-high today.
I'm talking more about brand/'hype' in particular. 15 years ago when I was starting my PhD, everyone adored MIT and wanted to be them.
Now I'm faculty at a reasonably good university in Europe, and obviously MIT is a fantastic place and I'd probably give a finger to get a tenured job there. But I think Harvard/Oxford/Princeton/Stanford have slightly eclipsed MIT in pure brand power. Stanford especially gets more of a 'wow!' reaction these days.
MIT's hay day in terms of research was decades ago -- before I got there -- sixties and eighties. The book Hackers by Levy is a good read on MIT Classic. It took a while for the work from there to be recognized outside in other institutions; you can give a 10-20 year window there. In the general public, virtually no one had heard of MIT.
By the late nineties, MIT research was just past its peak. Peer reputation peaked then or a little bit later. The general public was just starting to notice it existed.
In the 2010's, MIT research quality and integrity was in freefall. General public noticed MIT was #1, perhaps a half-century after-the-fact. Peer reputation was just past the peak.
I expect it will take another 10-30 years before the general public notices MIT has declined.
Stanford gets a "Wow" reaction for entrepreneurship more so than research.
I definitely wouldn't give a finger to be at MIT or Stanford. Of the ones you'd mentioned, Princeton, I think I'd be pretty happy at. If I had my druthers, though, I'd pick a school which is on the rise, basically where MIT was in the sixties or eighties. Mostly, I want a place which has a lot of freedom, integrity, and an open, accepting culture. Georgia Tech seems like a decent place right now. Yale would be nice. ASU would be a uniquely good fit. There are a few state schools I like too.
Schools in Europe are a mixed bag. I've considered going to a less-well-known school in one of the poorer countries in Europe, where there are smart people, and where I could buy a home for cash, and have the perfect freedom of being independently wealthy. If you're at MIT or Stanford, you need to do consulting, startups, or similar to have a decent standard of living, and that brings a pile of conflicts-of-interest. Entry level mortgage requires $200k+ in income in either of those housing markets, which is more than junior faculty are paid. Tenured faculty -- including consulting and outside interests -- make a mint.
It's not that I disagree with you exactly, but I think you're overgeneralising from tech. In my discipline (in the humanities) MIT is still top-notch, though has never quite been on a par with Princeton. Research integrity isn't really an issue, and there's no outside or consulting income to be had, so it's all salary. Housing is complex because rich universities often own property that they let you live in for free.
The humanities? Like education? Where MIT acquired a bunch of IP through fraud, lies, and sometimes intimidation, bundled it up into edX, and sold it for $800M, with money lining the pockets of well-positioned faculty members?
Or Stanford? With their school of ed? Baking data to support politically-popular causes, and gaining "impact" when fraudulent research is adopted by virtue of reinforcing what teachers want to hear, with the only victims being the students?
Yeah. Please. Do some research and come back another time.
If you want high-integrity ed research, you can look towards ASU, WPI, or many other schools one tier down in brand, and two tiers up in integrity.
No, not education. You guessed wrong about which discipline, and then went on a crazy rant... not a good look for you.
There is one of the humanities disciplines that MIT has been excellent in for several decades... if you don't know which one it is, then you don't know as much as you think you do.
(As I said, 'integrity' isn't really an issue in my discipline. Outside possible plagiarism, I guess. Another way in which your rant was a bit silly.)
Also, ASU and WPI one tier down in brand from MIT? Dream on, try three or four tiers.
For your reference, tiers are traditionally defined by Carnegie. There are around 5000 schools in the US. A couple hundred are in the top two tiers, R1 and R2. Here is a list:
You started out talking about '#1 university brand', and now you want to pretend you mean the Carnegie categories? Come off it. Not only are they much too coarse-grained, but they're also (by design) US-centric.
I'm at somewhere way less prominent than either, but even I can see that there are way more than 1 prestige tiers between MIT and ASU.
But of far greater concern is your bizarre condescending tone. You even told me to do my research. Are you an antivaxxer?
I can tell you that the student body at MIT generally had a pretty dim view of the admins when I was there, always trying to kill the things that make MIT an amazing place, while building stupidly overpriced vanity buildings. Not to mention the incredibly shameful way they handled the Aaron Swartz case.
I think there were a mix of cases when I was there, with admins sometimes supporting the wonderful weird and other times not. Often when they did, it was because the weird was only a background level of annoying/risky and when they didn’t, it was often that the level of cacophony had exceeded that admin’s tolerance for it.
It got worse: while the admissions office and museum were using MIT hacking culture to attract applicants and advance its brand, the school changed both the type of officers hired for the campus police which is necessary due to its inner city location and started arresting students for attempting to pull off harmless hacks. I also heard some reports they got generally more aggressive towards the student population.
When I was there they hired older, very experienced police officers who were looking for a good, not normally intense final job for their career. So they were generally understanding and fit into the culture.
Aaron Swartz was an entirely different thing. Instead of hammering JSTOR from his own institution of Harvard to the point it blocked the whole campus, he disguised himself demonstrating mens rea, a guilty mind, let himself into a machine room and left behind the laptop to do his act of civil disobedience or whatever. And then proved chronically depressed people have no business committing such crimes. From beginning to end he demonstrated ill intent towards MIT.
I was formerly a PhD student at Columbia in the social sciences. The faculty I interacted with didn't seem to care a whit one way or the other about the school at large, they were just really invested in the department's competitiveness in attracting faculty, grad students, etc.
I can't speak to any run-down labs, but regarding cramped working conditions -- space is just always going to be at a premium in Manhattan. Arguably Columbia is going above and beyond to acquire more space by opening up the Manhattanville campus [0,1,2].
In general, when academics complain about things, I take it with a grain of salt. Having said that, I don't doubt for a minute that the administration juked the stats to get a better USN&W ranking.
The Chemistry Department specifically has horrible infrastructure and a huge part of the problem is that construction is very difficult. NYC isn't a college town like Cambridge or New Haven and Columbia has to pay more to get stuff built and on top of that it's a long process involving multiple constituencies. Just look at how long it took for Columbia's Manhattanville campus to come together.
There are plans in the works to fix these infrastructure issues but the timeline is frustratingly long.
Yeah, on one hand I don't think admin are to blame for the delays in acquiring space and development. They have to compete against the enormous cost, enormous time delays, and even their student body who despise gentrification. That said, these delays have definitely contributed to the tensions, as we can see in the difference in relationships with admin held by BME, neuro, and CBS, who were granted space in Manhattanville, vs everyone else.
I don't know too much about NYC development, but Manhattanville seems somewhat unambitious in height, I don't know if that was the limit to Columbia's air rights or what, but it seems like a waste of limited footprint.
Generally for a paper if you don't actually care to see the author back up their arguments (which is not a great way of fully ingesting a paper, but lots of things don't need to be fully ingested) you can just read the intro and the conclusion.
> §8: CONCLUSION
> No one should try to reform or rehabilitate the ranking. It is irredeemable. In Colin Diver’s memorable formulation, “Trying to rank institutions of higher education is a little like trying to rank religions or philosophies. The entire enterprise is flawed, not only in detail but also in conception.”
Got to appreciate an essayist who'll enter the conclusion section guns a'blazing.
"As someone who knew a little math, what really drove me bonkers about the college guide was:
(a) the logical absurdity of adding together completely unrelated statistics to produce a single measure of merit — the key point being that you can produce an astonishing range of different results depending on the relative weight each component factor is assigned. And there is simply no logical, a priori basis for establishing such a weighting objectively. Do SAT scores count 30% of the total score? 32.2%? 18.78234%? (How about zero?) It's the classic apples + oranges – bananas/kumquats = fruit salad approach to statistics, and is completely meaningless."
General rankings have limited use. The value of data depends on the interests of the reader, which for college 'rankings' vary widely. For example, if you are interested only in the status of your college degree among the public, then the ranking itself over time and the credibility of US News are probably most valuable, regardless of how they are calculated. But if you are interested in the status of your degree within academia then the 'peer assessment survey' seems most important. Your interests and thus the value of different data will vary by what you want to do (study? work? donate? evaluate research? hire researchers or teachers?), what field(s) you are interested in and at what level, how you learn/work, etc. For many people, the 'ranking' might rely almost completely on one department or just a few specific faculty members.
The US News & World Report's ranking system apparently depends (at least in part) on class size (8% of ranking), proportion of faculty with terminal degrees (3% of ranking), proportion of faculty who are full-time (1% of ranking), student-faculty ratio (1%), financial resources per student (10%), retention and graduation rates (35%), student debt (5%), and "'peer assessment survey' [20%] in which college presidents, provosts, and admissions deans are asked to rate other institutions."
All that data can be valuable but says little directly about the quality of education received by students or the quality of research, though the latter might be outside USN's scope. Sometimes, objectivity is a distraction because the objective data is too limited to inform us. For example, objective data about a play - number of words, etc. - doesn't tell you much about it.
In those cases, expert judgment is an excellent tool and I think USN's 'peer assessment survey' is the right approach for many purposes, but USN surveys people with narrow knowlede - administrators, including admissions officers.
I might look at Times Higher Education, which performs a 'reputation survey' that asks (IIRC) thousands of published, tenured faculty about the quality of other schools' departments in their field of expertise. These are people with expertise and significant access, though of course their understanding of the student's perspective will be limited. (Columbia rates around 10th, IIRC.)
One interesting thing is the US News College Rankings sues a completely system of measurements than the field-specific rankings, e.g. the US News Computer Science Rankings. The latter is based only on "the judgments of deans and senior faculty members of computer science departments at institutions around the country."
I've been obsessed with computer science rankings for a while, and worked with others to compare US News Computer Science rankings with other computer science rankings: https://drafty.cs.brown.edu/csopenrankings/
Some rankings are naturally harder to game, like that your bachelors or doctoral graduates get hired as professors at other research universities. Or that professors at your universities win best paper awards at leading conferences. So I've been looking at whether various rankings are "biased".
There's some clear biases with US News, like it ranks CalTech as 11, but CalTech is listed as rank 39 in the aggregate because it does poorly on faculty publications and best paper awards. Yale CS is another example of a highly ranked department by US News (rank 20), that has an aggregated ranking of 35. Harvard CS does amazing with placement (ranked 6 for their undergraduate and doctoral students becoming professors), but has an aggregate ranking of 23.
That's a weird metric for judging education quality (which is at least what college rankings purport to do). I can see cases where either high or low values are markers of either good or bad education quality.
I go to washu - which does decently well on the rankings. There's a tradition every fall for the incoming freshman class to read a selected book, and my year they chose Cathy oNiel's weapons of math destruction, which goes over instances of data being used for bad outcomes, including a chapter devoted to how gaming the university ranking system had become common practice. I found it a very interesting choice on the part of the uni.
It is important to remember that universities, like most organisations, are not monolithic entities. It is almost certain that the person who selects the book for the common reading program is not the same person who is responsible to optimise the university’s ranking.
I took a class with Northwestern's last president Morton Schapiro. He admitted that the school lies about standardized test numbers in one of the lectures. Can't say it was too surprising. It seems like every school in the top 50 is claiming that at least 25% of their students get 34+ on the act.
I think U.S. News Ranking only really has merit because it's been around for a long time and has also done a good job of transferring its brand into good SEO.
“ In each previous section, what was at issue was a discrepancy between two figures, both obtained from data provided by Columbia. Regarding class sizes, the information provided to U.S. News conflicts with the information in the Directory of Classes. Regarding terminal degrees, the information provided to U.S. News conflicts with the information in the Columbia College Bulletin. Regarding full-time faculty, the information provided to U.S. News conflicts with the information provided to the Department of Education. And so on.”
The big conclusions are that Columbia seems to be providing inaccurate data, and that one of the outcomes of chasing rankings are that transfer students end up as second class students, at least at Columbia.
I think it’s a data driven case of how elite universities can perpetuate a system of reduced social mobility. For Columbia, the objectively more poor transfer students support the more wealthy non-transfer students, and the graduation rate shows that disparity.
> one of the outcomes of chasing rankings are that transfer students end up as second class students, at least at Columbia.
Interesting — when I was in law school, transfer students benefitted from the fact that they avoided our harsh first-year grading curves (20% A's, 60% B's, and 20% C's). The curve for second- and third-year classes was much more lenient, and it was relatively easy to get a 3.5 average during those years.
As a result, the transfer students disproportionately ended up with the highest GPAs, which seemed unfair to those of us who were there all 3 years.
I couldn’t find his data set available, so his assertion about terminal degrees is curious to me.
He seems to consider Ph.D. the only terminal degree, and specifically calls out the Arts school for having faculty with Master’s. But in most fields in the Arts, an MFA (though usually not an MA) is terminal.
Not only are they apparently lying to US News, they appear to be running a scheme of enrolling an unusually high amount of transfer students (who are not counted in the US News graduation rates) for the purpose of soaking them for tuition revenue.
Per the article, transfers get much less financial aid than non-transfers at Columbia and also have much lower graduation rates, which is the opposite of what is seen at other schools with large transfer populations like in the UC system.
Prof. Thaddeus is very courageous for bringing these facts to light, and/or has a deep love for his university that he feels is being corrupted by their pursuit of a higher ranking.
There are good reasons to believe that, for example, selectivity of admissions or graduation rates are some kind of quality signal. But as soon as it becomes strategically important to 'score well' on those metrics it becomes well worth universities doing things that will improve the metric without improving the quality - such as rejecting qualified applicants or graduating those who ought to have failed. And now the quality signal is much less useful if you're simply trying to use it to understand what's going on.