My colleague Tom Toch makes the point that regardless of whose estimates you believe in the ongoing graduation rate debate they still point to a substantial problem and many of the preventative steps would be the same in either case. But Tom then goes on to say that the research dispute about methods doesn’t really matter that much at all. There are some arcane disputes in this business, but I’m not sure that latter point is right in this instance.
I do think the grad rate debate is pretty one-sided and mostly political at its core but it is really about different datasets rather than methods (in other words the methods conform to the competing datasets). I don’t buy that the EPIers data is superior and while the various cumulative methods have some shortcomings I think they’re the best at hand right now for information and current estimates.
But, that said, if the EPIers are right about the datasets it still has pretty big implications for public policy in ways other than the general issue of combating the problem. That’s because the National Governors Association and almost all the nation’s governors have embraced a compact (pdf) essentially based on the Greene method, Ed Week is about to launch a new project to report grad rates based on Chris Swanson’s method (he now works at Ed Week) so that’s going to become part of the conversational grammar, and these numbers have real consequences for state accountability systems and No Child Left Behind going forward. So while the general policy prescriptions might remain the same in terms of dropout prevention, these numbers do matter in terms of holding schools, school districts, and states accountable in practice and with the public. Consequently, seems to me that whether they are right, or as right as can be, does matter does matter at that more micro level. It’s a distinction with a difference.