Sixty-second
Here's an interesting approach to ranking schools:A Revealed Preference Ranking of U.S. Colleges and Universities. It's vulnerable to a number of critiques, such the fact that location often plays a significant role in college choice, not quality. This is vulnerable to the counter-attack that those who look at this ranking will be thinking about location as well. Clearly, this typical economist retort would be half-right. A more interesting comparison might attempt to remove the variable of distance by looking only at students who travelled out of their region to attend universities. It all comes down to the East Coast: the location of the majority of the best students (maybe that's not true, but it would seem so) and the majority of the best colleges. I'm starting to re-think the above paragraph but do not have time to correct it. Commenters, please do so.
Another critique would be that we shouldn't trust kids in their choices of colleges. The relatively high ranking of Arizona State (80) seems to speak more to a desire among high-school seniors to get "git drunk and partay" more than anything else.
In the end, anything I say will be a response to the relatively shitty ranking of Wash U (62). We probably deserve this because we are ratings whores; on the other hand, I can't imagine receiving a better education than I am now at University of Miami (55) or even UCLA (36).
Last but not least the paper is on the SSRN website. This is a repository of working papers in the social sciences. Check it out.
Another critique would be that we shouldn't trust kids in their choices of colleges. The relatively high ranking of Arizona State (80) seems to speak more to a desire among high-school seniors to get "git drunk and partay" more than anything else.
In the end, anything I say will be a response to the relatively shitty ranking of Wash U (62). We probably deserve this because we are ratings whores; on the other hand, I can't imagine receiving a better education than I am now at University of Miami (55) or even UCLA (36).
Last but not least the paper is on the SSRN website. This is a repository of working papers in the social sciences. Check it out.
posted by Austin 5-000 at 9:22 AM
9 Comments:
one word:
landlocked.
we're like the uzbekistan of academia
Why not Kazakhstan? It's bigger (by a whole lot), it's badder, and it's still landlocked.
Because Kazakhstan is for wangsters
you are for za vangstahs
i poot za vang in za asshole des sheriffs
yes, I have woken soo fery early. ees necessary, you see. I ham at verk.
Oh verk ees zoo hard.
Forgive me for intruding upon ze slavic accent (or whatever that is) wars, but it just so happens that I wrote my econometrics research paper on how successfully schools can manipulate rankings from places like USA today. A couple interesting issues:
Austin 5-000 is defintely right to point out that the revealed preference model might put too much trust in high-schoolers (and their guidance counselors, who, if they are like mine, didn't know anything at all about the schools they were sending us to). The model gives a better account of their expectations, and probably of the quality of their sports programs, than it does of the actual quality of the schools, and unless you have a convincing argument to show that these all line up, it isn't too useful.
The USAT rankings have their own host of problems though, some of them similar to the ones you'd get with a revealed preference ranking. Part of what USAT uses to determine rankings is the "peer assessment," essentially a popularity poll distributed to the deans of various top-50 univerisites, and it's worth 25% of the total ranking. How honest or well-informed their responses are is anybody's guess.
Even with that criterion aside, there are other oddities. "Alumni giving rate" is a significant factor, and I'm not convinced that it's a very solid link to increased academic quality. One could argue that better schools produce more talented and wealthy students, but a lot of the school's fundraising income can be attributed to Telethon-like programs instead. A school could actually spend something like $100,000 paying students to make annoying phone calls to alumns, get $50 donations from 1500 people, lose money, and increase its ranking.
A couple other fun things schools do to mess with the USAT scores include not counting verbal SAT scores for foreign students while still counting math scores (increases score percentile ranges) and soliciting applications from underqualified students (increases selectivity).
You are now free to return to slav-talk.
Ooh zees ees great noows! Zou my language does not zay zo, Your post vaz eerudite
Post a Comment
<< Home