Are graduation and retention rates the right measures?

From today's Chronicle of Higher Education. A couple of quotes:

In the ongoing conversation about institutional quality, I continuously see references to retention and graduation rates as “the” -- or at least "a” -- key indicator of institutional quality. But what if we have it all wrong? What if measuring graduation and retention rates as evidence of institutional quality is like taking a person’s temperature to evaluate the efficacy or effectiveness (these are experimentally different measures) of his blood pressure medication?

Schools that serve a large number of wealthy students can win the numbers game when graduation and retention rates are reported as averages among the entire student body. Conversely, schools that serve large numbers of disadvantaged students have nowhere to hide.

Perhaps the real reason that retention and graduation rates are so low among disadvantaged students is that the Federal financial-aid system is based on the principle of giving a little bit of money to a lot of students rather than giving enough money to any students.

And now, for the rest of the story...