Sectoral support for primary education in Zambia and Uganda

The shift by development agencies from project aid to sector and general budget support has had important implications for education in Zambia and Uganda. In addition, evaluations of educational interventions must now focus on the the sector as a whole rather than the contributions of specific development agencies. A report from the Dutch Ministry of Foreign Affairs aims to show how the impact of primary education interventions implemented through sectoral support can be analysed through the use of secondary data such as censuses, assessment tests and surveys. Donors have begun to support government priorities for sectoral development, as delineated in the national budget, rather than individual projects they conceive and execute. Rigorous evaluation methods for determining the impact of development support were designed with such projects in mind, so new methods have become necessary. It is difficult to give a statistically unbiased assessment of the effectiveness of development programmes, as many factors besides the direct interventions themselves can influence the results. The shift towards new aid modalities raises new questions, not least because at a sectoral level, there are many interventions to be evaluated. The two studies described here use secondary data in a number of ways to solve the methodological problems involved. The education sectors of both Uganda and Zambia were in a poor state until their governments focused on sectoral investment with the help of donors. In 1997, Uganda introduced Universal Primary Education (UPE), while Zambia adopted the Education Sector Investment Plan (ESIP) to improve education policy and programme coordination. Key findings for the two countries include the following: • The introduction of free primary education contributed to the rapid expansion of enrolment, indicating increased access for poor children. • Parents’ income and education has a significant positive effect on students’ test and examination results. • Learning achievement is negatively affected by high student-to-teacher ratios and high student-to-classroom ratios. • Teacher education and teacher training have a significant effect on results. Schools with a high percentage of teachers who received training in the previous year performed better. • Children perform better when they have their own books. • Improved management is the key to improving education, and is associated with more effective interventions. Both studies show that better monitoring information makes it possible to analyse the results of the sector and contributes to both functions of an evaluation: accountability and lesson learning. Secondary data are increasingly valuable sources for impact evaluations. The challenge is to improve the methods for evaluating impacts using these data. Policy implications for the education sectors of Zambia and Uganda include: • Management quality can be improved by investing in training, an effective support structure at district level and an effective inspectorate to increase effectiveness of other interventions. • Government should invest in books, teachers and classrooms. • Sectoral results should not be compared with project results, as projects focus the resources available to them on a limited number of schools.