The effect of grade retention on school outcomes : an application of doubly robust estimation methods

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 Conditional Independence Assumption. I use doubly robust method, regression adjustment and inverse propensity score weighting. The results of the empirical study show that grade retention does not improve the students' educational achievement.