Learning losses during the COVID-19 pandemic in Ethiopia: comparing student achievement in early primary grades before school closures, and after they reopened

This Insight Note provides a new perspective on numeracy achievements of Grade 1 and Grade 4 pupils by comparing learning at the start of each academic year and the gains over the course of the year across two academic years: 2018-19 and 2020-21. During the 2018-19 academic year, the Research on Improving Systems of Education (RISE) Ethiopia programme collected data on students’ numeracy achievement from 168 schools. After schools reopened in October 2020, data on students’ numeracy achievements were collected for two new cohorts of pupils in Grades 1 and 4 in the same schools using the same instruments. This has enabled us to compare learning patterns between two cohorts in the same grades and schools before and during the pandemic. In comparison with the 2018-19 cohort, students entering Grade 1 after schools reopened began the year with lower foundational numeracy achievements and made slower progress in foundational numeracy during the year. As a consequence, they ended the year with an even wider difference in foundational numeracy compared to the pre-COVID cohort. Similarly, Grade 4 students in the 2020-21 cohort started the year with lower numeracy achievements than those in Grade 4 in 2018-19. Their learning progress over the school year was also lower by about 12 points (SD=0.12), widening the difference in numeracy achievement between the two cohorts by 22 points (SD=0.22) at the end of the school year. The differences in educational achievement remain evident in both grades even when adjusting for the duration of instructional time and are estimated to be equivalent to more than one-third of an academic year lost. Pupils from rural areas and those from the poorest households remain within the lowest numeracy achievement levels. There are also variations by gender for Grade 4 pupils, where boys showed relatively higher progress than girls by the end of the school year. Overall, our results continue to suggest the need for targeted support for low-performing students, particularly in rural areas and for those from the poorest backgrounds, so that a further widening in learning inequalities can be prevented.