To the extent that this paper’s evidence of misrepresentation remains plausible and not substantively refuted, this should, in principle, lower Columbia’s peer assessment score. If, however, future respondents to USN’s peer assessment survey appear neutral to this new information, this would seem to reveal a lack of attention that would further diminish the credibility of USN rankings in this scenario.
1 Like
Believe it or not, this is not new. People on CC have been calling attention to the way medical school expenses always get folded into a R1 university’s total financial resources per (undergraduate) student calculations for years.
However, reporting practices differ substantially:
The serious concern seems to be that Columbia reports all figures in a way that’s advantageous to its ranking to the point of transgressing common understanding. Most notably, this appears in its defining patient care as an instructional expense for its undergraduates.
5 Likes
Great thread. Interesting discussion & observations.
While it seems clear that Columbia has gone to excessive & inappropriate lengths to game the US News ratings & rankings, this university remains an outstanding institution. Probably better to assess Columbia University by one’s intended major or preferred area of study than by its rankings’ prestige.
Not disagreeing with you. But, the “normalization” of medical school expenses being included in the total USN financial resources calculation began with including the salaries of some of the most highly paid medical specialists in the world. From there, it was only a matter of time before some R1 university took it to the extreme by quietly folding patient care expenditures into the mix. Cue the Claude Rains line from Casablanca: “I’m shocked. Shocked that gambling is going on in this establishment!”
2 Likes
No, it is far from clear what happened. Without having the exact breakdown of data no one should assume anything .There is no proof yet about how and if data was unfairly manipulated. The President of the university is on record as backing the numbers.
Columbia has different schools. The College is very different from the School of General Studies or tghe School of Professional Studies.
No one can assume anything yet about the professor’s article until one can see exactly what and how the data is used.
Nothing so illustrates the old adage, “Those who live by the sword, die from the sword” than a college single-mindedly pursuing the rankings game only to find itself defending the way it collects its data.
You seem to be cherry picking one stat out of many dissected in the article. The article didn’t suggest the numbers were totally made up. What it demonstrated is that by (almost uniquely among its peers) declining to publish public CDS data they appear to be self-interpreting what they get to count instead of relying on the criteria that would be dictated and scrutinized by a public filing that their peers use, to their significant benefit. Their total instruction spend, for example, is not possible without including patient care based on their own other filing.
To your comment about proof, totally agreed. Columbia should publicly publish how it came up with its data or publicly report CDS like almost everyone else. If their numbers and criteria for how they count are accurate they should have no problem demonstrating it. So why aren’t they proving it?
12 Likes
So why doesn’t Columbia clear it up?
1 Like
Except that probably all colleges and universities do it to present itself in the best light. Just like companies presenting to investors, etc.
But everyone is obligated to present the truth.
Is the way Columbia presents any different from how the University of Chicago or NYU present data ?
Also, the adjective ‘dizzying’ is misleading. First, this development occurs over34 years. That is a long time ago. Second, in New York City, progress can occur like an oil boom - very sudden with opportunities, land value, growth. Third, Columbia has always been a world-class educational institution, and before 1988 paid no attention to the criteria in USNWR because this report didn’t exist. before that time, Columbia was still ranked among the elite and was in the 'most competitive ’ category in Barron’s and Princeton Review list of colleges and universities.
As a post-script. Agree 100 % transparency is essential for any institution. It has an active Office of Research Management : [Research Misconduct | Columbia | Research ]
So, it must be consistent with the guidelines of its own office on research output- as should every college and university.
Does Harvard and other universities do the same with its data ?
With respect to a comparison to NYU, the substantial difference was explained in the paper:
3 Likes
The CDS dataset specifies how to interpret what to include for its categories – this was all covered in the professors analysis. So for most of Columbia’s peers, if they selectively misinterpreted what to include in their data, people could clear see that and call it out. By withholding public CDS data, they are basically asking US News and everyone to trust them that what they reported was apples-to-apples with what the standard is, as opposed to self-interpreting their own standards. And their own professor found places where it was clearly apples-to-oranges, such as with the example of including all of their medical center patient care cost as “student instruction.” The college has the power to clear this up by being 100% transparent with its data. It so far has declined. Their simply saying their data is accurate is akin to President Clinton’s “I didn’t have relations with that woman” defense where he selectively chose to interrupt what they did as not fitting his definition.
4 Likes
The someone should write the President of Columbia and insist on clarification on this issue. This would be consistent with its own Office of Research Management.
How many other universities do the same ?
Crucially, at the primary level the main issue surely dwells with USNWR. Why does it allow this omission of insisting on the same criteria for all entrants in its yearly list ? If omission of CDS is allowed by USNWR then this omission should be corrected so that all must comply with the same criteria if they want to be considered for the rankings.
This article IS asking the President for clarification. And I am sure many others have done so since, as well, including journalists and concerned alumni, etc. Crickets.
This article has gotten a lot of publicity. The President is a smart person - he clearly knows we would like transparency and answers to these questions. Avoidance itself is concerning for an institution that claims integrity.
8 Likes
Per your question, it’s very unusual for a top university to not participate in the CDS.
Columbia, however, does not issue a Common Data Set. This is highly unusual for a university of its stature. Every other Ivy League school posts a Common Data Set on its website, as do all but eight of the universities among the top 100 in the U.S. News ranking.
3 Likes
Absolutely US News shares responsibility. I suspect they allow Columbia and a few other colleges to get away with non-transparent data because their list would lose stature if there were glaring omissions of top schools on it. And US News doesn’t have the resources to audit or validate what the colleges submit. The college rankings started as a little side feature of the US News magazine, and now that’s the only reason the company is still in business. They ceased publishing a magazine years ago and absolutely no one goes to their website for news. They are simply milking the reputation of their rankings and related subscription website as long as possible.
1 Like
USNWR actually does have resources to check on validity of submitted data, and several staff have said this on television interviews.
Blame USNWR for being lax about this insistence on CDS. Until then, neither Columbia nor any other institution is obligated to submit CDS if USNWR says it is not necessary.
No one denies Columbia has been around for a long time. You’re mostly arguing with yourself. Pandaboy1, meet Pandaboy1:
6 Likes