Improving data quality for the CAADP biennial review: a partnership initiative piloted in five countries

This paper presents results of a data partnership framework for strengthening evidence-based planning and implementation that was initiated in 2019 in five selected African countries (Kenya, Malawi, Mozambique, Senegal, and Togo) during the second round of the CAADP biennial review (BR) process. It analyzes the effect of the activities conducted on the data reporting rate and the quality of data reported in the five pilot countries, compared with what was achieved in like-pilot countries. The like-pilot countries are non-pilot countries that have characteristics like the pilot countries at the baseline which affect selection into the pilot or the data reporting and quality outcomes. Different methods (standard deviations, propensity score matching, and two-stage weighted regression) are used to identify the like-pilot countries, and a difference-in-difference method is used to estimate the effect of the pilot activities on the outcomes. The capacity-strengthening activities focused on working with the country Biennial Review (BR) team to: assess the inaugural or 2018 BR process and identify the data gaps; constitute and train members of data clusters to compile and check the data for the 2020 BR; and then validate and submit the data. The findings show that the activities helped the pilot countries to improve their performance in the data reporting rate and the quality of data reported in the 2020 BR. The largest improvement is observed in Togo and Senegal, followed by Kenya and Malawi, and then Mozambique. The average increase in the data reporting rate between 2018 and 2020 BRs for the pilot countries is greater than the average progress made in the like-pilot countries by about 6 to 9 % pts. This derives mostly from improvements in the data reporting rate for the indicators under theme 3 on ending hunger. Regarding the quality of data reported (measured as the percent of the data reported that have issues) too, the pilot countries on average performed better than the like-pilot countries, especially with respect to the data reported under themes 2 on investment in agriculture and 3 on ending hunger. But most of the estimated differences have low or no statistical significance. Implications for sustaining the progress made in the pilot countries, as well as for extending the activities to other countries, for the next rounds of the BR are discussed.