Explaining the rural-urban learning achievements gap in Ethiopian primary education: a re-centered influence function decomposition using Young Lives data

ABSTRACT

The rural-urban learning achievements gap is a persistent issue to be addressed from a different approach. This study employed re-centered influence function decomposition with Young Lives data to estimate the rural-urban education production function and decompose the rural-urban learning achievements gap in Ethiopia. Results revealed that the rural-urban education production function is different across achievements distributions. Moreover, most of the rural-urban learning achievements gap is explained by student background characteristics. Unmeasured characteristics explain much of the proportion of the gap but the importance of that proportion varies across the distribution. Policy implications of the findings were discussed.

KEYWORDS:

• Learning achievements
• rural-urban
• RIF decomposition
• Young Lives
• Blinder-Oaxaca (BO) decomposition

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Selection of the 10 sites from the two regions followed the same criteria as the core 20 sites. These regions were included in the survey because the government and education stakeholders were interested in their issues of access and use of services.

2 Anonymous referees suggested to include variables such as school ownership, school size, and community characteristics, but they were either unavailable in the dataset or could not be used from the perspective of an analysis on rural-urban learning achievements gaps because they are not measured for both types of areas.

3 One of YL priority area in the school survey was core country curricular, but also transferable common domains across participating countries. Transferable domains are a set of cognitive and non-cognitive skills that give individuals the critical tools needed to function in real-life situations like in employment or health (Rankin et al. 2015).

4 The two steps of the RIF decomposition can be estimated automatically by using the Stata command oaxaca_rif developed by Rios-Avila (2020).

5 The decomposition as presented is from the viewpoint of students in rural areas, but it can also be presented from that of students in urban areas. We chose this approach because we want to emphasize the lower achievements of rural areas, with an assumed objective of raising the performance of students from rural areas to that of those from urban areas, as opposed to lower the performance of the latter to that of the former.