Archive

Archive for November, 2007

THES – QS World University Rankings 2005 & 2006

November 5, 2007 Leave a comment

Okay, I ran into this site and I got a bit interested in their University Rankings. So I put the top 200 lists into Excel, and imported it into SPSS and then back into Excel with some formatting to create this.

“Huh?”

Don’t ask…

THES - QS World University Rankings Top 200 (2005 & 2006) Format 1

The first format (above) is sorted on Change, where it’s convenient to see which country has had the biggest growth. The second format (below) is sorted on percentage of 2006, to see how the top dogs are doing, so to speak.

THES - QS World University Rankings Top 200 (2005 & 2006) Format 2

Interesting statistics, heehee. Course, it would be even more interesting if I pulled the stats for population count for each of those countries. I guess I could get them from the CIA’s World Factbook.

“Why don’t you?”

Well, you know, I’m busy with my thesis and stuff. Serious person doing serious stuff and all that.

“But not busy enough to draw these sheets in the first place?”

Actually, these rankings are somehow related to peer reviews, as peer reviews are also used to determine university rankings, such as this one. How useful indeed! Anyhow, it seems we all have to work hard! But I’m glad the Netherlands is up there, not doing too bad if I say so myself!

For source data, go to the THES – QS World University Rankings site. And yes, there are actually 201 countries and not 200 countries listed per rankings. I believe this has to do with the fact that more countries can share a single spot/rank. I guess that makes sense, in an awkward way.

“No it doesn’t, since they skip every time universities share the same spot.”

Ah, well then it has to do with this I guess.

*Due to misinterpretation of supplied data, the published position at 165= of the Université Libre de Bruxelles was incorrect and the accurate position of 120 is reflected in this table. The originally published positions of any between 120 and 164 have been retained.

I still don’t quite get it, but I guess I don’t really care, as this error is seemingly consistent in both rankings. Relatively, it shouldn’t affect things.

On growing journal prices

November 2, 2007 Leave a comment

Oy,

Still alive and working on my thesis. In the meantime, some interesting stuff I have found.

http://www.libraryjournal.com/article/CA6321722.html

Distinguished economists Ted Bergstrom and Preston McAfee sent an open letter to university presidents and provosts last fall suggesting, among other things, that universities should bill publishers for faculty service if the cost of a journal exceeds a certain reasonable level (see “End Free Ride for Costly Journals,” LJ 12/05). To identify the worst offenders, Bergstrom and McAfee created a web site that charts the cost of around 5000 journals, using price per article and price per citation to rank each journal as good value, medium value, or bad value (www.journalprices.com).

The details have been debated, but one conclusion is unavoidable: an extremely high percentage of journals from the six largest STM publishers fall into the bad value category (74% on average), while an extremely low percentage of titles from the nonprofits are rated as bad (14%). Blackwell and Elsevier had the lowest percentages of bad titles (55% and 68%, respectively), while Sage, Springer, Taylor & Francis, and Wiley all had percentages in the mid-eighties. The data challenge librarians and scholars to reconcile price to value before renewing a journal or donating time and expertise to help a journal succeed.

http://www.journalprices.com/

This website represents our best attempt to compute the price per article and price per citation. Currently we use the ISI data for citations through 2005 and 2006 prices, which are the most recent data available to us. Not all journals report information the same way, and errors are possible. When reported to us, we correct errors. Moreover, prices per unit for journals that have recently expanded are underestimated. The coloration (red for very low value, yellow for low value, and green for good value) is computed by comparing the composite price index to the median for non-profit journals in the same subject. Be advised that price per citation, price per article and the composite index are not perfect measures of value. Neither of us are experts in most of the fields represented, and others may reasonably, or unreasonably, disagree with the value assessment. We have mapped a large set of journal categories into 17 areas.

“And?”

Well, just sharing some good sources. They can be very useful.