Conflicting Signals and Bias in Public Performance Evaluation: Leniency and Centrality in the Public Sector

ABSTRACT Performance evaluation in public administration plays a critical role in improving public sector accountability and effectiveness, yet it is often vulnerable to systematic biases. This study examines how leniency and centrality biases affect program evaluation, particularly when evaluators face incongruent performance signals. Drawing on 1414 cases from Korean central government program evaluations and using heteroskedastic regression techniques, we find that when evaluators face conflicting information about a program's performance, they tend to assign higher and less variable scores—suggesting both inflation and centrality effects. However, these biases are notably reduced when programs receive broader public and political attention, especially when designated as presidential priorities. These findings highlight the importance of transparency and external oversight in mitigating bias and improving the credibility of performance assessments. Korea's performance evaluation system serves as a vital model for low‐ and middle‐income countries. Our findings directly inform the design choices facing these modern budgeting reformers.