Causal Effects of Corruption Perceptions Measurement Reform on Institutional Accountability in Uganda: A Difference-in-Differences Design

Measurement is not a neutral mirror of corruption; it is an institutional intervention that reshapes the incentives of the actors it observes. We argue that the mechanism linking measurement reform to accountability is not the accuracy of the new information per se, but the shift in the locus of monitoring from central principals to distributed citizens, which alters the reputational and electoral incentives facing public officials. The article specifies a difference-in-differences design that exploits the staggered rollout of citizen-reporting platforms across Ugandan districts, addressing the critical threats of non-parallel trends and spillover effects. We formalise the identification assumptions, propose a measurement model for the accountability outcome, and discuss the interpretation of the resulting estimates. The framework contributes to the psychology of corruption measurement by conceptualising perception indices not as passive thermometers but as active components of the governance system they purport to describe.