Citizen-Generated Data? Participation as a Lever for Improving National Statistical Capacity

Abstract

Quality national statistics can help steer domestic policymaking, public and private investment, and civic engagement. As such, increased funding for national statistical capacity has become a key advocacy target in international development. However, there is growing concern that the political economy of data hampers capacity strengthening efforts. Could citizen participation mitigate political disincentives, and support ‘more and better’ statistics?

To answer this question, I explore whether a relationship exists between citizen participation and national statistical capacity in sub-Saharan Africa. I use a custom country-year panel dataset and compare pooled ordinary least squares and two-way fixed effects models. Independent and dependent variables of interest are sourced from the Ibrahim Index of African Governance and World Bank Statistical Capacity Index, and models control for public sector corruption, national economic development, and development assistance levels.

My findings suggest that citizen participation can influence national statistical capacity, but that this influence is intermediated through public administration capacity. This result holds only when using the Ibrahim Index of African Governance measure of statistical capacity, though I posit this is due to differences in measurement between the two dependent variables. My analysis also indicates that formalized, de jure civic checks and balances are more important than de facto civil society protections in determining the impact of citizen participation on statistical capacity.

This work aims to contribute to the modest literature examining the relationship between statistical capacity, citizen engagement, and government capacity; and seeks to provide timely inputs to the ongoing debate around what ‘more and better’ financing for data and statistics means in practice.