Should students repeat a school year? The case of grade 9 students in Colombia

This paper evaluates the relationship between grade repetition and academic performance of 9th grade students in Colombia. We use matching methods (propensity score, coarsened exact and entropy) to balance the distribution of characteristics of those students who were retained and those who were not. The results show that repetition negatively affects students’ performance in math (23% of a SD) and language (20.4% of a SD). Moreover, our findings show that the effect of grade repetition on female students’ achievement in language is positive and significant (5.6% of a SD), and in mathematics is negative and significant (19.7% of a SD).