In this paper I demonstrate, both theoretically and empirically, that the interpretation of regression estimates of between-group differences in economic outcomes depends on the relative sizes of subpopulations under study. When the disadvantaged group is small, regression estimates are similar...

In this study, I estimate the average causal treatment effect of grade retention on several educational outcome variables, such as completion of upper secondary school, graduation grades in math and German, as well as average final grade using a data set from Germany. The analysis relies on...