Education without formal schooling through tablets and tutors: Evidence from out-of-school children in Bangladesh during the COVID-19 pandemic

This paper estimates the impact on children's learning of one specific education technology (EdTech) intervention in Bangladesh: providing tablets with educational software, combined with private tutoring, to out-of-school students using a randomized control trial. The provision of tablets and tutors led to positive impacts on both the math and the Bangla language scores of out-of-school children, increasing math scores by approximately 0.25 standard deviations (SDs) of the distribution of test scores, and Bangla scores by approximately 0.17 SDs. The effects of the intervention were especially strong for girls compared to boys. Rural out-of-school children, but not urban out-ofschool children or out-of-school children in urban slums, benefited greatly from the program. The program has little effect on noncognitive traits such as competence, selfesteem, and grit. These findings have broader implications for implementing programs targeted to out-of-school children and distance education during school closures, such as those caused by pandemics.