Developing Textbooks to Improve Student Math Learning: Empirical Evidence from El Salvador

School enrollment has rapidly increased since 1990 in developing countries at the primary level but the quality of education has stagnated over the years. In teaching and learning practices, textbooks are an important intermediate that links curriculum, teachers, and students. Since textbooks describe the content and methodology of teaching and learning, they can improve teaching and learning practices, if they are carefully designed. This study evaluates the effectiveness of the package of interventions including the distribution of textbooks that are carefully designed to improve student learning in math through a randomized controlled trial in El Salvador. This experiment tracked same students for two years. The average one-year impact of the package on primary school 2nd grade students’ math learning is estimatedaround 0.48 standard deviation of test scores. The impact was larger on students with higher baseline scores. The average accumulated impact of the first-year interventions one year after is around 0.12 standard deviation. The package of intervention improved math learning of 2nd grade students, and the impact persisted even after schools of the control group also received the package of interventions in the following year.