3global rank`tables #volatility

 


See also

QS: https://www.topuniversities.com/university-rankings/world-university-rankings/2021
THE: https://www.timeshighereducation.com/world-university-rankings/2021/world-ranking#!/page/0/length/25/sort_by/rank/sort_order/asc/cols/stats

— the fast ascent of NTU, HKUST, CMU, and several other tech universities

Science/Tech college ranking is mostly based on objective criteria, allowing these young colleges to become pre-eminent and world-class within a few decades.

CMU is now ranked #1 in CSci.

moat — their reputation is hard to protect because newcomers would challenge them constantly. In contrast, the slower domains have more of a moat.

— Comparing big-three international publishers 1) QS 2) THE, both UK-based, and 3) ARWU i.e. the Shanghai ranking
ARWU is mostly about the A in ARWU — “academic” i.e. research output, which is hardly visible to the lay public including those keen parents. ARWU doesn’t consider an institution’s reputation among academics or employers, nor international faculty or student representations, for example. In ARWU algorithm, I feel humanities are given insufficient credits. Some disciplines are national in nature, without a credible international journal (or conference) that’s comparable to the sci-tech disciplines. Wikipedia lists other criticism. For example, “a merger between two institutions will significantly increase the merged institutions’ score and give it a higher ranking”

QS and THE both consider reputation among academics and employers, what I call professional reputation, different from mass-media reputation. Mass media influences the lay public’s opinion.

https://www.topuniversities.com/student-info/university-news/comparing-world-university-rankings-qs-shanghai offers a simple yet useful summary:

  • A) If you’re looking for a university that has an enviable record of consistently producing high-quality research, try the Shanghai ranking.
  • Q) If you’re looking for a university which is well-regarded by employers and other academics, and will set you up perfectly for a well-paid graduate job, the QS World University Ranking should be your first port of call.
  • T) And, if you want to see a ranking which attempts to reflect both a university’s research impact and its reputation, the Times Higher Education ranking considers more variables than the other two.
  • Lastly, it should be noted that both the THE and QS rankings make an effort to consider how international a university is (in QS’s case, this is done by measuring the proportion of international students and faculty members). If you’re looking to study abroad, both these rankings will potentially have more to offer you. Personally, I(Victor) think this item reflects an inclusive, diverse culture.

==== overall validity of ranking across publishers
https://www.elsevier.com/research-intelligence/university-rankings-guide is the most detailed overview I have seen. It lists 5 leading publishers including CWTS and US-news, beside the big-3.

Every reader (except the UChicago graduates) of these rankings complain about, criticize or ridicule the discrepancies, even within a single system over a 10Y span — ridiculous volatility!

The more criticism, the more eyeballs. It reminds me of the Option pricing theories — grossly imperfect, but largely validated by years of observation in the field. If there’s some fatal flaw, then people would stop following.

#1 Target audience — aspiring students and their parents. As such, top universities are really luxury brands , partly for vanity. Kyle said college ranking is mostly about prestige, which is mostly based on research output.

— volatility in a ranking table .. The #1 visible, decisive quality indicator of any ranking table. [Analog: when you compare two hospitals, you look for reported malpractices; when you look at two religious groups, you look at violence tendencies; when you look at two software products, you look for instabilities, crashes;]

I feel ARWU is slightly more stable [accumulation], but all big3 league tables show very high volatility. A reputable university often has 100+ years of history, but its ranking seems to depend on the last few years. The volatility partly reflects 1) evolution of ranking criteria and 2) fast change in real research output — the reality we must accept.

I tend to focus on the research criteria. A university can churn out lots of publications for a few years but its reputation, its academic/research tradition takes decades and won’t rise so fast.

http://rankingwatch.blogspot.com/2021/03/energetic-and-proactive-leadership-or.html had a case study on UWI
http://rankingwatch.blogspot.com/2015/03/malaysia-and-rankings-saga-continues.html had a case study on UM

— CWUR .. not a well-known ranking consultancy.
— CWTS ranking.. By default, ordered by P[i.e. total publication], I found strange things like

  • 上海交大 ranked global #2, above all China universities, all UK universities, and all US universities except #1.
  • 四川大学 #8, U Sao Paulo #9, 吉林大学 #11,  above all UK universities, behind only 2 US universities (Harvard#1/Michigan#6)

Enough reason to say CWTS ranking is contradicting conventional wisdom and consensus.  THE and QS also shows ridiculous rankings (volatilities) in specific indicators, but not in the most visible, most scrutinized league tables.