This presentation will report on the long-term impact of an education project called IMAGINE (IMprove the educAtion of GIrls in NigEr) in Niger, financed by the U.S. government (through the Millennium Challenge Corporation). The IMAGINE project was completed in 2009 and included the building of 68 “girl friendly” schools in rural areas of Niger, with amenities such as separate latrines for boys and girls, a water source, and housing for female teachers. The project intended to incorporate the use of complementary interventions after the completion of school construction, such as conducting a societal awareness campaign, training to increase the literacy of mothers, additional tutoring and training, and providing attendance incentives for school-age girls and female teachers; however these were not implemented because of suspension of the threshold program.

Mathematica completed an impact assessment of the school construction activities in 2011 and found small positive impacts on school enrollment but no impacts on attendance or test scores. These impacts are driven entirely by effects of the program on girls. These findings are smaller than expected given that an evaluation of a similar intervention in neighboring Burkina Faso found large impacts of the program on enrollment, attendance, and test scores. However, there are several factors that might help to understand the results from the initial study, including the presence of schools in nearly all recipient villages prior to the program, villages were selected by the central ministry to receive schools and did not go through an application process, not all intervention activities were fully implemented, and outcomes were measured only one to two years after completion of the project. Given the short period of exposure, stakeholders requested a study evaluating the longer term effects of the program. 

This follow-up evaluation of the IMAGINE program four years after its completion will address key questions regarding the sustainability of educational development programs that focus on school infrastructure.  We utilize a random assignment design to answer the following research questions:

1. Have the previously completed IMAGINE investments been sustainable?
   1. What is the current level of functionality and use of the infrastructure constructed under the IMAGINE program?
   2. Did the IMAGINE program have any lasting impacts on school enrollment, attendance, and test scores?
   3. Are the impacts different for girls than for boys?
   4. Are the impacts different for children from households with different asset levels?

Preliminary results suggest that longer-term impacts are larger than previously observed and remain larger for girls than for boys.