Evidence from Sub-Saharan Africa

Researchers in this paper partnered with the Ghanaian government to simultaneously test four methods of increasing achievement—assistant-led remedial pull-out lessons, remedial after-school lessons, smaller class sizes and teacher-implemented partial day tracking—in schools with low and heterogeneous student achievement. The interventions increased student learning by about 0.1 standard deviations, rising to 0.4 standard deviations when adjusting for imperfect implementation, with no effects on attendance, grade repetition or drop-out. Test score increases were larger for girls. Test score gains persisted after the program ended. Assistants implemented the program with higher fidelity than teachers, although their fidelity decreased over time while teacher fidelity marginally improved. Explore the findings in detail here.

This working paper uses a quasi-experimental design to investigate the continued impact of a scaled-up intervention undertaken by the Ministry of Education in Madagascar. It studies a package consisting of TaRL training to primary teachers and community volunteers, capacity strengthening for school management committees to lead extra-curricular remedial activities, and a public education forum in the region to develop collaboration among different actors, such as educational administration and local governments,  to improve foundational skills. The study finds that TaRL training plus the public education forum improved learning outcomes in numbers and four basic operations by 0.37 standard deviations. The collaboration of different actors to improve foundational skills in Madagascar shows an example of collective impact to address the learning crisis in a low-income country. For more detailed insights click here. 

This working paper uses a quasi-experimental design to investigate the impact of a scaled-up intervention undertaken by the Ministry of Education in Madagascar. After COVID-19, the MOE rolled out and progressively scaled an education package consisting of Teaching at the Right Level training for teachers and community volunteers, plus interventions to strengthen the capacity of school management committees to lead extracurricular remedial activities using TaRL. With the foundation of the improved capacity of the SMCs, the TaRL training increased the proportion of grade 2 through 4 students who could read words written in the local language by 15.9 percentage points and those who could read a story by 3.1 percentage points. Further, analysis suggests that these impacts were realised widely across the targeted region. The case of Madagascar indicates that it is possible to improve learning at scale through extra-curricular remedial activities organised by SMCs in support of learning recovery. Read more about the findings here.

Targeted instruction is one of the most effective educational interventions in low- and middle-income countries, yet reported impacts vary by an order of magnitude. This working paper studies this variation by aggregating evidence from prior randomized trials across five contexts, using the results to inform a new randomized trial. The authors find two factors explain most of the heterogeneity in effects across contexts: the degree of implementation (intention-to-treat or treatment-on-the-treated) and program delivery model (teachers or volunteers). Accounting for these implementation factors yields high generalizability, with similar effect sizes across studies. Thus, reporting treatment-on-the-treated effects, a practice which remains limited, can enhance external validity. A new Bayesian framework is introduced to formally incorporate implementation metrics into evidence aggregation. Results show targeted instruction delivers average learning gains of 0.42 SD when taken up and 0.85 SD when implemented with high fidelity. To investigate how implementation can be improved in future settings, we run a new randomized trial of a targeted instruction program in Botswana. Results demonstrate that implementation can be improved in the context of a scaling program with large causal effects on learning. While research on implementation has been limited to date, our findings and framework reveal its importance for impact evaluation and generalizability.

This study evaluates the impact of tracking (grouping students by academic ability) in 61 Kenyan schools, where students were randomly assigned to first-grade classes, compared to 60 schools that assigned students based on initial achievement. The study found that in non-tracking schools, students benefited from being with academically stronger peers. However, in tracking schools, all students, regardless of their academic level, showed improved performance, scoring 0.14 standard deviations higher after 18 months compared to those in non-tracking schools, with the effect persisting even a year after the program ended. The study suggests that while there is a direct positive effect of being with high-achieving peers, tracking benefits lower-achieving students indirectly by allowing teachers to tailor their instruction to a more appropriate level for each group. This indicates that tracking can be beneficial across the entire spectrum of student abilities, not just for higher-achieving students. For more detailed insights click here.