This chapter objects to the use of surveys to monitor public confidence and police performance in any official way stating that they are based on perceptions and so abstract as to provide an uncertain guide to the service that is actually delivered. There are reasons to believe that a significant component of the general confidence in police that is measured by surveys is only tangentially related to the empirical reality of the moment. If policymakers are seeking survey data to answer the question of how they are performing it is not clear that they will always be getting a meaningful answer. The article asserts that a performance assessment survey is better off focusing on change measures rather than on level measures. The chapter also stresses the importance of ensuring that survey methodologies are employed that produce optimally accurate recall of encounters with police, and that research be conducted to illuminate problems and improve upon the ability to accurately assess encounters that respondents may be inclined to underreport. Finally, the chapter assesses whether the legitimacy of including public opinion on the list of official statistical performance indicators would be enhanced by directly addressing them proactively, rather than waiting for their inevitable appearance in the political realm. Figures, notes, references