Results

In the primary measure of outcome- the overall score obtained by the students in the end line test on critical thinking skills- on average, being part of the treatment group is associated with a 17% percentage increase compared to being part of the control group. This difference is statistically significant at the 95% confidence level and multiple robustness checks indicate a causal relationship between the CT sessions and the end line test scores. Adding covariates did not affect the direction or significance but reduced the effect size slightly.

At the most conservative estimate from comparing all the models, on an average, being part of the treatment group is associated with a 14% percentage increase compared to being part of the control group. The Cohen d effect size is 0.409 standard deviations, suggesting clearly perceptible treatment effect where 0.2 standard deviations and above is the norm in education research.