Friday, April 17, 2009

Measuring participation in government programs

I want to give a shout out to the paper "The Under-Reporting of Transfers in Household Surveys: Its Nature and Consequences" by Bruce Meyer, Wallace Mok and James Sullivan.

The amount of under-reporting in the most widely-used programs should give pause to anyone who produces or consumes research on participation in transfer programs or on the effects of transfer programs. It underscores the need to do a better job of linking up widely-used survey data sets with administrative records from the transfer programs themselves. This in turn will require a seriousness about evidence-based policymaking heretofore largely absent among the political class.

Bruce has been engaged in this sort of "meat and potatoes" research on measurement-related topics for some time now. I think this work is of fundamental importance. It is, at the same time, under-produced and, I think, undervalued in the profession. The spillovers from this type of work are huge, as it affects every study using these data.

One thing that any researcher can do to bring attention to this issue is to ask about measurement error in seminars. That will at least encourage the reading of what literature there is on measurement and perhaps, indirectly, the production of more research and the improvement of the data on which our knowledge relies.