This paper uses household, school, and test score data from Bangladesh to compare and contrast the effectiveness of NGO-run and state-run schools in the provision of primary education. I study how the entry of NGOs in primary education has affected educational outcomes of girls and examine the mechanisms which account for the relative performance of NGO versus state schools in improving female educational outcomes. The results show that the entry of NGO schools has significantly increased girls’ enrollment as compared to boys. Constructing cohorts from cross-sectional data using year of birth and year of NGO school establishment, I show that cohorts which were exposed to NGO schools have higher probability of enrollment and the effect operates mainly through girls. The two most prominent characteristics of NGO schools that encourage girls’ enrollment are the high percentage of female teachers and having Parent-Teacher Associations (PTAs). NGO schools show strong effects in improving children’s test scores.