During the past few decades, the use of community policing has proliferated throughout American policing. Despite numerous case studies there is very little empirical evidence as to the determinants of the variation in community policing implementation. Using open-systems theories (i.e., contingency and institutional theories), 1 develop and test a structural model that not only assesses the substantive and relative effects of the organizational context, organizational structure, and organizational commitment on community policing implementation, but also explores the relationships among these various determinants. I assess this model via structural equation modeling. This research also utilizes confirmatory factor analyses to improve the measurement of many of the constructs of interest, including community policing implementation.
To form the measurement models and test the structural model, I compiled data from the 1997 Law Enforcement Management and Administrative Statistics survey, 1990 U.S. Decennial Census, organizational surveys conducted Edward Maguire and William King, and the Office of Community-Oriented Policing Services. Merging these data created a sample of 401 large, municipal police organizations that comprises approximately 83 percent of the total number of such agencies in the U.S.
The findings reveal that a second-order measurement model of community policing implementation fits the data rather well. The main purpose of this research is to explain variation in this construct. The standardized effects of the structural model illustrate that organizational commitment has the largest influence on community policing implementation, followed by formalization, geographic region, income heterogeneity of the community, the number of functional units in the police organization, police chief turnover, population mobility, centralization within the police organization, and funding incentive (several other elements of the organizational context and organizational structure had no direct statistical relationship with the implementation of community policing). These determinants explain about 28 percent of the variation in community policing implementation. Furthermore, I show that organizational structure does not maintain a pervasive role in COP implementation, but the relationships that do exist indicate that mechanistic-type structures coincide with greater COP implementation. The findings also reveal that many of the COP implementation determinants have associations with each other, resulting in these variables exhibiting indirect effects on COP implementation. Notably, I discern that task scope, socio-economic status, organization size, environmental dispersion, environmental capacity, organization age, and spatial, occupational, hierarchical, and functional unit differentiation have no direct association with COP implementation but nonetheless exhibit an indirect effect on COP implementation through the various relationships among the determinants. In addition to illustrating that hierarchical and spatial differentiation only maintain an indirect relationship with COP implementation, further examination of the relationships among the determinants discloses that the organizational context explains a fairly large proportion of the variance in spatial (about half) and hierarchical (about 61 percent) differentiation. Despite popular belief, these findings coupled together indicate that making arbitrary changes in spatial and hierarchical differentiation to facilitate COP implementation will not only have little effect on COP implementation, but may make the organization less able to respond to the organizational context. In short, the existence of these various relationships implies that the specification of direct and total effects only in a model of community policing implementation oversimplifies the complexity of the determinants in terms of their interrelatedness.
The formal examination of the relationships among the determinants of COP implementation also offers some evidence suggesting that more structurally complex police organizations tend to exhibit greater structural control mechanisms. Occupational differentiation enhances formalization but also reduces the proportion of employees in an administrative capacity. Greater functional differentiation in terms of number of functional units and differentiation among functional units coincides with more formalization, administrative weight, and employment screens. Yet, neither spatial nor hierarchical differentiation determine any measure of structural control. These findings provide moderate support for Durkheim (1933) and Rushing's (1967) hypothesis that the more complex a social unit becomes the greater is its need for internal control.
The results of this study indicate that both the task and institutional environment are important for understanding the form and function of large, municipal police organizations. This conclusion reconciles the perception that contingency and institutional theory are competing theories of organization. Contrarily, I demonstrate that these theories are complementary. While doing so, I also address many of the deficiencies noted by implementation scholars as impeding the development of implementation as a science.