The Wall Street Journal's economics blog reports on a Webber and Ehrenberg paper that looks at the way higher education institutions allocate resources and finds that more money spent on support services improves graduation and persistence rates, especially at institutions that have low rates to begin with. They suggest that at such institutions, spending more on services and less on instruction might be a beneficial reallocation of scarce resources.
Now I haven't read the paper other than the abstract and what the WSJ reports (once victim of reduced budgets at state universities - no access to NBER working papers), and I know Ron Ehrenberg and he is an excellent economist (and one of the nicest guys you'll ever meet), but one problem jumps out at me immediately and it is not one that has an easy control in the statistical examination of the data. It may well be that increased spending on student services, especially for schools with a large population of struggling students can lead to a dramatic improvement in drop out rates and graduation rates. I have no problem believing this finding. At such schools, this effect may be larger than the impact of an extra dollar spend on instruction. But making the logical leap that perhaps resources should be taken away from instruction and put toward student services is tricky.
Here is the problem in my mind: an instructor with too many class and no job security has a very clear incentive to make the class easy for students. It significantly reduces headaches from having to give extra help to struggling students and has a direct and dramatic impact on student evaluations of the teacher - which help when it is time to renew contracts. [As an aside, I once asked a group of students who were all raving about a particular class and instructor what they liked about the class so much, their response was that it was the easiest class at OSU and everyone got an A - this was apparently a legendary class among undergrads, everyone wants in. Sigh...] So the increase in success rates from spending more on outside the classroom services could be because these services are so useful or it could be because the drop in allocation for the classroom (read more instructors and fewer professors) have made passing classes and graduating easier.
In other words an easy way to improve 'success' rates in any endeavor is to make it easier, but is this really benefitting students? I am all for increasing student services, but we have already ventured too far along the instructor track in my view. It is not, by the way, that instructors are bad teachers, they are generally exceptional. But incentives matter, so we should be making these instructors professors.