We evaluate the long term effects of a "girl-friendly" primary school construction program in rural Burkina Faso, funded by the Millennium Challenge Corporation, using a regression discontinuity design. The program sought to increase girls' primary school enrollment and completion rates and was implemented in 132 villages from the 10 provinces with the lowest rates of girls' primary school enrollment. Data from two follow-up surveys in 2008 and 2012 showed that the program was effective in the short and the medium-term in raising enrollment and test scores. After 6 years, the program increased enrollment by 15.3 percentage points and increased test scores by 0.28 standard deviations. Also, BRIGHT schools are effective at getting children start school on time and at keeping children in school longer. This presentation will describe the regression discontinuity design, discuss the short- and medium-term impacts, and will present the long-term impacts findings from a 2015 follow-up survey.#2015Conference