Does information improve service delivery? A randomized trial in education in India

Abstract

From a cluster randomized control trial in 610 villages, the study evaluates the impact of a community-based information campaign on school outcomes in three Indian states. The campaign consisted of eleven to fourteen public meetings over two rounds in treatment villages to disseminate information to the community about its state-mandated roles and responsibilities in school management. No intervention took place in control villages. The paper reports on the final follow up survey two and half years after the campaign. Providing information improved teacher effort and learning outcomes in schools. Bigger gains were seen in the two states, Madhya Pradesh (MP) and Uttar Pradesh (UP), where baseline level of outcomes was lower than in the third state, Karnataka. The impact on teacher effort, primarily for civil-service teachers with permanent jobs and therefore lower accountability, was between 16%-43% in MP and UP. The shares of children able to do basic mathematics competencies improved. Due to low baseline learning levels, the magnitudes of the percentage increases were much larger compared to the absolute increases in shares. Fewer improvements occurred in language. This can be because of low teaching effectiveness as well as more time needed for larger and wider impacts on learning. School councils became more active after the campaign. Focus group discussions indicated discussions within communities and communities actively bringing up issues with teachers and school councils. Impacts were generally larger or broader than those at midline survey 2–4 months after one round of intervention. Overall providing information holds promise in improving public services via worker accountability.