Tate Watkins, referring to an old Chris Blattman post, asks me such a question. One might raise a similar query about Sri Lanka. In both cases there are some obvious answers, having to do with ease of access and languages and also levels of violence (in the past, if not always today).
But I would like to raise a more general issue and I hope the development economists and social scientists reading this can add their expertise in the comments section. To what extent is the choice of venue for study due to what I will call “social science infrastructure”?
I don’t mean roads and bridges. I mean having trained armies of local assistants, data gathering and processing facilities, populations which are used to signing informed consent forms, medical clinics which understand how to work with social scientists and register data, and other less visible assets. I once visited a Poverty Action Lab evaluation in Hyderabad and it struck me immediately just how much local assistance they needed to get their study of micro-credit off the ground. As far as I could tell, the local assistance seemed really quite able but of course that cannot be taken for granted in all locales.
It has struck me for a while just how many RCT papers appear to be set in a relatively small number of places in Kenya. Presumably these parts of Kenya have a very good social science infrastructure.
So what is going on here? Who can shed light on this? And to what extent does having a good social science infrastructure correlate with other features which will bias the results of these studies?