Notes: This table summarizes the results of our balancing tests. To test random assignment, we regress treatment status on baseline characteristics. We run a separate linear probability model for each baseline characteristic. Table 3 shows a detailed list of all baseline characteristics and individual point estimates. All regressions include strata fixed effects for the level of randomization. Standard errors are clustered at the school level. Column (1) reports the total number of balancing tests and the number of statistically significant tests for different levels of significance. Column (2) reports the number of coefficients we would expect to be statistically significant due to chance under random assignment.