This document studies the effect of preschool education on cognitive outcome of young people at the end of high school in Colombia. Using the Instrumental Variables Methodology, negative gains are found from preschool attendance. An additional year of preschool education is associated with a reduction of 0.121 and 0.079 standard deviations in language, and mathematics scores of Saber 11 test, respectively. Nevertheless, there are heterogeneous effects when the sample is divided by socioeconomic variables like mother’s education level, household income, and sector to which the school belongs.
