SES != socio-economic status

I was in DC on Monday to talk about an evaluation of the SES program. SES stands for supplemental educational services. It is a program I had not even heard of before but it turns out to be quite an interesting one. Under No Child Left Behind (NCLB) schools that fail to meet their performance standards (AYP in the jargon, for adequate yearly performance as I recall) for three years running must spend some of their Title I federal funds (or an equivalent amount from other sources) on what are essentially vouchers for after-school tutoring. Districts can re-capture the vouchers by attracting students to their own after-school programs but they must offer parents a choice among all state-approved providers, which includes both non-profits (both faith-based and whatever the opposite of faith-based is) and for-profits such as Sylvan learning centers. The money is pretty big here - the providers can make around $40 per hour of tutoring provided.

One important feature of this program is that spillovers are built in, as Title I funds devoted to SES are not spent on other things. This paper, which one of our ace graduate students pointed me to, is by a student of Caroline Hoxby's at Harvard and emphasizes the spillover issue.

The program raises other interesting issues as well. If vouchers for tutoring are okay, why not vouchers for the school day itself? Is it a good idea to create a new set of private sector actors who depend on federal funding? There is, apparently, a K street lobbying organization for providers of SES services, who will now fight tooth-and-nail to keep the program in place regardless of any evaluation results that might come forth. How do parents choose among alternative tutoring providers? My sense is that the information they have to go on is modest, though it may include recommendations from other parents with students in the various programs. Much of the competition may be on convenience in terms of timing and location. What determines the "dosage" of tutoring that students who do take up the services receive? That is, some students go only once, others go all year and others somewhere in between. What factors lead to this variation? What is the optimal dosage and how does it vary among students? As always, there are more interesting research questions than time to address them.

Hat tip: Alex Resch