Inside Higher Ed has an
interview with Jerry Z. Muller, the author of
The Tyranny of Metrics.

As someone who works at a broad-access institution (which gets beaten up regularly for our "terrible" graduation rates) I want to print this on a t-shirt and wear it to our next Board of Regents meeting:
Q: Supporters of the use of metrics in higher education -- including Democratic and Republican politicians -- argue that metrics lead to improvement. Knowing that a college has a low graduation rate, in a frequently cited example, can spur improvements. How would you respond?
A: It may indeed spur improvement, if, for example, institutions can find the resources to assure that students are better advised, that the courses necessary for them to graduate are offered, etc. But one source of low graduation rates is the low level of preparedness of admitted students, so the most efficient way to raise graduation rates would be to stiffen criteria for admission. But then legislators (having been misled by organizations such as the Lumina Foundation into believing that more and more people should go to college and that state governments ought to engage in “outcomes-based funding”) complain about lack of “access.”
So the most frequent method to increase graduation rates is to lower the standards for graduation -- easier courses, more lax grading, etc. There’s tremendous pressure on instructors (more and more of whom are adjuncts) to do just that -- especially once one gets below the level of flagship institutions. By allowing more students to graduate, a college transparently demonstrates its accountability through its excellent metric of performance. The legislators are appeased. The fact that no one has learned much is beside the point.
(I don't have a beef with measuring things; I like measuring stuff, in fact. What's infuriating is policymakers' selective ignorance of cause and effect, and their refusal to engage in basic critical reasoning.)