The graduate student came from a good department, so he carefully motivated his paper by talking about the policy relevance of the relationship between\u00a0D\u00a0<\/em>and\u00a0Y<\/em>, explaining that policy makers cared deeply about said relationship, and how they made a big deal of it.<\/p>\n When presenting his results, the presenter did what we commonly do in economics, which is to show a table presenting several specifications of the regression of interest, from the most\u00a0parsimonious (i.e., a simple regression of\u00a0Y\u00a0<\/em>on just\u00a0D<\/em>) to the least\u00a0parsimonious (i.e., a complex regression of\u00a0Y<\/em> on\u00a0D<\/em> and all the available controls\u00a0X<\/em>).<\/p>\n The problem, however, was that the R-squared measure–the regression’s coefficient of determination<\/a>–for the simple regression of\u00a0Y\u00a0<\/em>on just\u00a0D <\/em>(i.e., the most parsimonious specification)\u00a0was about 0.01, meaning that the treatment variable\u00a0D<\/em> explained about 1 percent of the outcome of interest.<\/p>\n I commented\u00a0that this was interesting, given that if policy makers\u00a0made a big deal about the relationship\u00a0between\u00a0D\u00a0<\/em>and\u00a0Y<\/em>, one of the points\u00a0of the paper (which the author did not already make)\u00a0should be\u00a0that policy makers should really spend their time on other things. Indeed, if\u00a0D\u00a0<\/em>explains only 1 percent of the variation in\u00a0Y<\/em>, focusing on\u00a0D<\/em> in order to stimulate\u00a0Y\u00a0<\/em>is unlikely to be cost effective. In other words, there are other factors out there that explain 99 percent of the variation in\u00a0Y<\/em>, and it is likely that among those factors, at least one or two will play a significant role–or at least, a role that is much more important than\u00a0D<\/em>.<\/p>\n The foregoing strikes me as a\u00a0useful use of R-squared, but the\u00a0measure is misused more often than not, especially by neophytes and by people outside of economics. Indeed, in his Guide to Econometrics<\/a><\/em>, which is still my favorite econometrics text, the late Peter Kennedy noted:<\/p>\n
![\"Kennedy\"](\"http:\/\/marcfbellemare.com\/wordpress\/wp-content\/uploads\/2015\/04\/Kennedy.jpg\")
<\/a><\/p>\n