|
|
|
|
|
|
|
Jens 'n' Frens
Idle thoughts of a relatively libertarian Republican in Cambridge, MA, and whomever he invites. Mostly political.
"A strong conviction that something must be done is the parent of many bad measures."
-- Daniel Webster
|
|
|
|
|
|
Monday, August 25, 2014 :::
Vox has an article about an attempt by graduate schools to have the Taft administration rank undergraduate colleges. It elicited blowback:
Public universities argued that looking only at grad school preparation was myopic, since their mission was to serve the needs of the state.
This seems like one of the least valid complaints about the whole enterprise.
One of the most important things to do when using metrics is to attend to their limitations; accordingly, if you are creating a list ranking the best X, especially when X is very diverse along many dimensions, you have to say what you mean by saying that one X is better than another; usually, at least at a first pass, this involves stating the reason for constructing this ranking in the first place. If you are constructing a ranking of colleges so that graduate schools have a guide as to which colleges best prepare their students for graduate school, you should construct the ranking that serves that purpose and be perfectly clear that that's what you are doing.
Perhaps, at the time, Virginia Tech was doing an excellent job of preparing students to be engineers and a terrible job of preparing them for graduate school — perhaps in part because it had no intention of preparing students for graduate school. On a ranking of "What school should you attend if you want to be an engineer?", perhaps it should have rated quite highly. On a ranking of "What school leaves you best prepared for graduate school?", perhaps it should not. On a ranking of "Colleges and Universities" that purports to measure graduate school preparation, "the needs of the state", the caliber of the football team and the preparation of students to be good citizens and/or employees, perhaps it would rank somewhere else — and perhaps the group of people who should care where it ranks would be reduced from few to none.
Indeed, these days there are already numerous private rankings, some less laughable than others, with the least laughable ones ranking schools in different subjects — a school that leaves its engineering students well-prepared for graduate school may or may not leave its history students as well-prepared for graduate school. Indeed, the very wrong-headed counterargument
The only reason to oppose transparency, some proponents argue, is that reveals facts colleges would rather conceal.
falls to the rebuttal that what the Education department should do is make the data available, without pretending to find a Platonic ideal ordering on that data that applies a proper weight and consideration to each element to recognize a single abstract notion of "College".
The justification for a new set of rankings from the federal government seems to be
... Obama wants federal financial aid to depend on how colleges do in the new federal quality rating system.
To the extent that there will be federal money going to these colleges and it is appropriate for the federal government to monitor colleges that should be subsidized in this way, it makes sense for the government to produce a metric of how creditable one school is over another, and this aim should be an important input into the design of such a metric. Once it is produced, it would be irresponsible for the department of Education to ever be unclear about what it has measured or why; indeed, if
College presidents argue that the ratings effort itself is misguided, or that measuring graduates' incomes is the wrong way to measure the value of higher education, or that there is no way a rating system will capture the different missions of community colleges and elite private universities.
those college presidents may be correct, but especially as regards the last two points, graduates' incomes may well be a good measure of which students' loans should be guaranteed, but nobody should pretend that that is the only valid measure of the value of higher education, or that doing well on any single rating system should be the primary goal both of community colleges and of AAU universities.
::: posted by dWj at 11:59 AM
|
|
|
|
