tag:blogger.com,1999:blog-38545607.post1098971049236499618..comments2024-03-09T10:26:48.789+00:00Comments on economics, psychology, policy: Analysing Social Experiments - HeckmanEmma Watsonhttp://www.blogger.com/profile/11466193733741012673noreply@blogger.comBlogger1125tag:blogger.com,1999:blog-38545607.post-57478378619205411302010-08-12T11:58:42.911+00:002010-08-12T11:58:42.911+00:00The “representativeness” issue and the extent to w...The “representativeness” issue and the extent to which you can generalise from these interventions has always troubled me, especially in light of the recent papers (including some by Prof. Heckman) which are skeptical of drawing inference from local average treatment effects derived from instrumental variable estimators. After all, Perry took place in a disadvantaged, African American community in 1960s Michigan. <br /><br />Having said that, a comparison with NLSY79 provided in this paper seems to suggest that the Perry sample was reasonably representative of disadvantaged African Americans.Mark McGovernhttps://www.blogger.com/profile/17133619200829364366noreply@blogger.com