Large Learning Gains in Pockets of Extreme Poverty: Experimental Evidence from Guinea Bissau

Abstract

Fazzio and Eble share "co-first author" status, as they contributed equally to the work (order of co-first author names was randomized according to Ray and Robson 2018).We are grateful to our administrative, research, and implementation teams in Bissau for their tireless work. Ana Forjaz oversaw the fieldwork at endline. Mark Fisher designed and maintained the database. Gilda Piaggio conducted the power calculation and randomization at the beginning of the study. Yixun Zeng provided research assistance in the preliminary stages of the data analysis. Clive Belfield and Sharon Wolf gave helpful comments on the manuscript. We are grateful to the National Bureau of Economic Research (NBER) for funding the exploratory research on learning levels in Guinea Bissau that motivated this study. We thank the editor and two anonymous referees for generous input which greatly improved the work. This study was funded by Effective Intervention, a UK registered charity. The study passed ethical/ institutional review by the Ministry of Education of Guinea Bissau on 30 August, 2012, and the NBER ethics committee on 1 July, 2014 (ref: IRB Ref#14_06). The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

How much can we remedy very low learning levels in rural parts of low-income countries? Impact and generalizability of a multi-pronged para-teacher intervention from a cluster-randomized trial in the Gambia

“…This combination of multiple known best practices, offered through a contracted provider, parallels solutions from studies of the provision of healthcare in similarly remote and disadvantaged areas(Salehi et al 2018) and of "graduation" programs helping individuals out of extreme poverty(Banerjee et al 2015).3 These learning levels are substantially worse than those observed in similar assessments of children's learning in other developing countries, such as India and Tanzania (Pratham 2010; Rajani 2010) and similar to what some of us have observed in recently completed work in rural Guinea Bissau(Fazzio et al 2020). This pattern also appears in regular national assessments of child learning using EGRA-and EGMA-style tests administered by the Gambian government and by third parties (e.g., Sprenger-Charolles 2008).4 Burde and Linden (2013) find that a supply-side intervention yields learning gains of 0.4-0.65 SD in rural Afghanistan, a setting with comparably low baseline literacy and numeracy.…”

“…In a separate project, some of us took this model to Guinea Bissau to try to raise learning levels in remote rural areas there (Fazzio et al 2020). In that context, there is essentially no reliable state implementation of education.…”

“…As a result, the core research question of that study diverged, and instead aimed to see whether the main thrust of the model could be delivered via a dramatically different method. The intervention studied in Fazzio et al (2020) replaced the early grades of government primary schools with privately operated schools which use only two of the core insights -scripted lessons and frequent monitoring. Because of the extremely low levels of human capital in Guinea Bissau, implementing the "para teacher" component -hiring literate and numerate individuals locally and training them to serve as teachers -was impossible, and trained teachers had to be hired instead.…”