Effectiveness of interactive satellite-transmitted instruction: Experimental evidence from Ghanaian primary schools

Rural students in lower- and middle-income countries dramatically underperform their urban peers. Through a cluster randomized controlled trial, we explore live, interactive instruction via satellite as a potential solution. The studied program equipped 70 randomly selected schools in two areas of Ghana with technology to connect to a studio in Ghana's capital Accra, where instructors provided math and English lessons simultaneously to 10-12 schools at a time, supported by in-class facilitators. An additional 77 schools served as control. We estimate substantial gains in rural students’ numeracy and literacy skills. After one year, we observe an immediate gain of 0.21 standard deviations in numeracy. After two years, we observe a gain of 0.26 standard deviations in numeracy and a gain of 0.33 standard deviations in literacy. These effects are remarkably consistent across different satellite instructors and geographic areas. Much larger gains in foundational literacy and numeracy skills and evidence from quantile regressions suggest that the program was effective in tailoring instruction at the appropriate level and in closing substantial pre-existing gaps in basic skills. The program demonstrates the feasibility of establishing a successful distance instruction model in remote and resource-constrained contexts; however, given large infrastructure and implementation expenses, the program is not (yet) cost-effective.