Politically Robust Randomization?

Gary King gave what looks to be a fascinating presentation on program evaluation in the context of Mexico’s  health care assistance program, Seguro Popular.

According to Gary, traditional randomization is not politically robust and he presents an alternative approach called Matched-Pair Randomization.

Here’s a couple of his slides to whet your appetite:

Image

Image

However you slice it though, from what I got out of the slides, the program so far is NOT improving health outcomes in Mexico.

2 thoughts on “Politically Robust Randomization?

  1. I guess this works best when it’s obvious the program is all or nothing — if it’s more a measure of degree I guess you’d need to have pre-specified exclusion thresholds?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s