Friday, August 7, 2009

Academics and randomized trials

The IPA blog has an interesting response to some of the debate going around about how important randomized trials are, and what role academics should have.

Martin Rotemberg concludes that academics are needed in RCTs as they can add to the quality of what is learned from the evaluation, and thus improve a program. I agree. I think that helping people is the most important part of our work, even beyond advancing the general science.

I would like though to add one more reason for academics to be in RCTs: an evaluation requires involvement in (say it isn't so!) the real world (caveat: with enough assistants its possible to avoid such messing things, but you probably have to already know a lot about the world to even try such a thing).

One day, I hope it will be impossible for an economist to make a model, run a regression, or otherwise talk about something he/she has never seen, or a place he/she has never been. One can only dream.

No comments: