Key points are not available for this paper at this time.
In the past decade, we have witnessed a confluence of two previously independent modes of research in the area of urban service delivery. The first mode of analysis involves measuring service delivery performance characteristics using data from official archives of public agencies. Sometimes called measures, these indicators are used to document such performance criteria as effectiveness, efficiency, and equity of policy inputs, outputs, and impacts. The second mode involves measuring a sample of citizens' and clients' attitudes about service delivery through some form of survey research. Citizens or clients are asked to evaluate the quality and / or quantity of a given service in general, or with respect to some specific aspect of the service. Most such research employing indicators of service delivery has focused on effectiveness. More recently, policy analysts have begun to raise questions about the extent to which the two analytical modes produce consistent or contradictory results. What variables account for variance in citizens' subjective evaluations of urban services in their neighborhoods? To what extent do citizens' subjective evaluations reflect objective service conditions in their neighborhoods? Answers to these questions are particularly important, since presumably one of the main purposes of citizen satisfaction surveys is to provide information for city officials to use in making decisions about service delivery. This paper addresses these and similar questions, and presents a variety of tests of the empirical relationship between objective and subjective measures of police service delivery. First, as a background to this study, some of the relevant findings of previous research are summarized. Second, the methodology, hypotheses, and data employed in this study are described in detail. Third, we present the results of our analysis and compare our findings with those of previous studies. Finally, we draw some conclusions about the two modes of analysis.
Brown et al. (Sat,) studied this question.