Crime Analysis Report

Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=gppr20

Police Practice and Research An International Journal

ISSN: 1561-4263 (Print) 1477-271X (Online) Journal homepage: https://www.tandfonline.com/loi/gppr20

Community hierarchy of needs and policing models: toward a new theory of police organizational behavior

Melchor C. de Guzman & MoonSun Kim

To cite this article: Melchor C. de Guzman & MoonSun Kim (2017) Community hierarchy of needs and policing models: toward a new theory of police organizational behavior, Police Practice and Research, 18:4, 352-365, DOI: 10.1080/15614263.2016.1242425

To link to this article: https://doi.org/10.1080/15614263.2016.1242425

Published online: 10 Oct 2016.

Submit your article to this journal

Article views: 897

View related articles

View Crossmark data

Citing articles: 1 View citing articles

Police Practice and research, 2017 Vol. 18, no. 4, 352–365 http://dx.doi.org/10.1080/15614263.2016.1242425

Community hierarchy of needs and policing models: toward a new theory of police organizational behavior

Melchor C. de Guzmana and MoonSun Kimb

acriminal Justice and criminology, school of liberal arts, Georgia Gwinnett college, lawrenceville, Georgia, Usa; bcriminal Justice, school of education and human services, the college at Brockport, state University of new York, Brockport, new York, Usa

ABSTRACT This study explains the influences of environmental variables on the emergence of varying policing models. It empirically tests a new perspective on the influences of community variations to police organizational behaviours and practices among local police departments in the U.S. Using the U.S. Census, the Uniform Crime Reports and the Law Enforcement Management and Administrative Statistics data, the study tests a perspective suggesting that community hierarchy of needs influences the degree level of the implementation of different models of policing, particularly community policing. The study presents constructs to operationalize hierarchy of needs. Multiple regression analyses were used to examine the relationships of hierarchy of needs, organizational factors, and other control variables to police departments’ pursuit of different policing models. Findings suggest that community hierarchy of needs and organizational factors significantly constrain the departments’ implementation of policing models. Lower hierarchy of needs tends to be associated with pursuit of traditional policing while a community with higher level needs tend to pursue community policing. Similarly, organizational complexities influence the implementation of different policing models.

Introduction

Responsiveness of the police to the needs of the community has been recently highlighted in the headlines of daily news in the United States. The history of the police in the United States seems to be driven by the dynamic changes in the needs of the community (Travis & Langworthy, 2008). This research adds to the existing empirical research that investigates the relationships of community needs and policing models. The historical accounts of police organizational developments in the United States is interspersed with the confluence of multiple factors that are predominantly at the macro-level such as technological developments, ideological shifts, political and civil rights struggles, economics and urbanization, among others. As police departments are primarily organized at the city or municipal level, it is important to examine the impact of their immediate operational environments on police practices.

In his analysis of the implementation of community policing in the United States, Wilson (2006) found that departments have varying levels of the implementation of community policing. This finding

© 2016 informa UK limited, trading as taylor & Francis Group

KEYWORDS community hierarchy of need; community policing; social disorganization; U.s. local police departments; leMas; social order repair

ARTICLE HISTORY received 1 June 2016 accepted 24 september 2016

CONTACT Melchor c. de Guzman [email protected]

PoLICE PRACTICE And RESEARCh 353

should not be surprising given that community policing should be implemented based on the needs of the community that they serve. See Travis and Langworthy’s (2008) extensive discourse on the variety of communities and the fit of certain police models. Wilson’s (2006) argument begs the question as to what brings about these disparities in the implementation of community policing. Moreover, it raises the spectre that different models of policing are pursued in place of community policing depending on the community’s circumstances or needs.

This current research suggests that variations in Community Oriented Policing (COP) implemen- tation could be explained not only by departmental variables as suggested by Wilson (2006) but also, most importantly, by community structural variables. This paper proposes that community character- istics as conceptualized through social disorganization variables could be significant predictors of the level of community policing implementation. The paper empirically studies the relationships between geodemographics and policing styles by testing Lambertus and Yakimchuck’s (2007) proposition that COP implementation is predominantly constrained by the levels of the community’s hierarchy of need. Lambertus and Yakimchuck suggest a linear path for COP to prosper by arguing that COP could only be implemented after the application of the traditional model of police enforcement in order to maintain order in society. Once order is established, only then COP implementation becomes possible and order enhancement activities that community policing desire to achieve could be implemented. Using secondary data analysis, this proposition could be tested cross-sectionally by examining the relationships that exists between the status of community’s needs and the level of COP implementation.

Importance of the study

This study advances knowledge in several respects. First, it adds to the existing literature examining the use of social disorganization as a predictor of social control (Kubrin & Weitzer, 2003; Warner, 2007). In the social disorganization literature, it has been contended that formal social control mechanisms are more pronounced in socially disorganized communities (Renauer, 2007). Hence, since COP has elements of informal social control with community members being able to direct and even do actual policing (See Travis and Langworthy (2008) as well as Bayley and Shearing (2001) descriptions of these new models of policing), its implementation in socially disorganized communities would be lesser.

Second, the study provides a new framework to explain the variations in COP adoption given contingency and organizational factors (Grinc, 1994; Kerley & Benson, 2000). Lastly, the study could serve as a replication of several research examining the constraining effects of environmental factors on community policing (See Renauer’s (2007) study on this topic).

Aside from the paper’s contribution to the body of knowledge, the results from this study could inform practitioners and policy makers about police interventions and their alignments with their community’s social needs. The findings could serve as guides to police departments enabling them to align their police practices with the community they serve. Using the information on this study, police departments could achieve greater efficiency by aligning their modes of interventions to produce the most benefits with greater certainties of success. More importantly, community interventions could be devised that could alter social order and forcing the police models on communities that are not best served by that particular model. Additionally, police departments could explain the constraints exerted by their working environments on their police efforts and provide an empirical foundation for their choice of models. Finally, although the data are derived from the United States’ context, some of these structural and organizational conditions are universal in nature that the theory could be tested internationally with some adjustments on the measures of certain concepts.

Review of related literature

COP has been found having a differential impact on communities (See Reisig & Parks, 2000; Renauer, 2007; Skogan, 1986, 1990; Skogan & Hartnett, 1997). Skogan and Hartnett (1997) found that disad- vantaged communities come out worse and that better-off communities benefit most from COP. Two

354 M. C. dE GUzMAn And M. KIM

possible explanations could be deduced from this effect disparity – either police departments are not uniformly implementing COP or that the necessary community settings for COP to prosper should be present. Wilson’s (2006) study on the implementation of COP among police departments in the United States (US) provided evidence to the argument that police departments implement COP in a disparate manner. This disparity is not surprising and in fact desirable as police departments must align their policing models with their communities (Perkins, 2016; Travis & Langworthy, 2008). Wilson’s (2006) analyses accounted for these COP implementation variations using mostly organizational factors but hardly examined the impact of external environmental factors. This gap in the study needs to be considered especially since police contingency theorists (e.g. Caldero & Crank, 2004; Cordner & Scarborough, 2010) suggest that police departments have to operate within their environmental constraint. See Travis and Langworthy’s (2008) lengthy discussion on the effects of environmental factors to police organizational behaviors.

It is our contention that alleged failures of COP to make a difference in communities is due to the lack of fit that happens between the type of policing model that is applied to the community more than the inherently inability for COP to work in disadvantaged communities. For example, in an elaborate study by Perkins (2016), he found that social and structural characteristics (geodemographics) can affect the perceptions of the community about their police. He concluded that police styles should fit the level of perceptions that community have of their police. By extension, given the same contexts, community cooperation and action, the ingredients necessary for COP, may be hindered by social and structural characteristics and prevent the police from fully implementing community policing.

Community features and social controls

Social disorganization has always been utilized to explain crime in society. In reality, social disorganiza- tion should primarily predict the form of social control and the extent of the use of this form of control in society (Kubrin & Weitzer, 2003). Social disorganization theories have contended that structural stresses weaken informal social controls (Merton, 1968; Sampson & Groves, 1989). Although policing has always been conceptualized as formal form of social control, its latest permutation of community policing is really a manifestation of increased informal social control as it increases community par- ticipation in policing (See Bayley & Shearing, 2001; Travis & Langworthy, 2008).

In their book, Neighborhood and Crime, Bursik and Grasmick (1993) seem to suggest a separation between the formal and informal models of social control. However, Bayley and Shearing (2001) have argued that such distinctions and delineations are no longer as apparent in the modern policing structures. Bayley and Shearing (2001) contend that communities are now empowered to define their own safety needs and are allowed to implement policing measures that they can do on their own or in cooperation with the police or, perhaps leave the police alone in the provision of these safety needs. Lambertus and Yakimchuck (2007) tend to lend support to these configurations by arguing that communities that are plagued by levels of crime and disorders would seek recourse to a more aggressive formal means of social control rather than use their local resources to control these threats to achieve community stability. Hence it is expected that the police would be pressured to adopt a more traditional style of law enforcement – one that is more aggressive and formal (i.e., use of arrests as opposed to less formal means such as warnings, suggestions, or reprimands) as they are being mandated by the community to increase their law enforcement efforts. See arguments on this topic by Travis and Langworthy (2008); de Guzman and Kumar (2011) as well as by Wilson (1968). Although Bursik and Grasmick (1993, p. 17) contend that the ability to exert informal controls at a parochial level are mediated by the ability of the community to solicit services from public agencies, such community capability might be undermined by their perceptions that public control agencies are not responsive to their needs. Indeed, strings of research suggests that police organizations tend to neglect the needs of their constituents (Crank, 2003). Thus, police models that are not viewed as responsive to parochial needs are dismissed and, are not successful in bringing communities to have a perception of collective efficacy (Boessen & Hipp, 2016; Hipp, 2016). When this disconnect happens, it would ultimately limit the implementation of COP.

PoLICE PRACTICE And RESEARCh 355

The most current evidence on the relationship of COP and community characteristics may be found in Renauer’s (2007) study. He found that informal social controls could hardly exist among disadvantaged communities. Furthermore, he found that the application of community policing strat- egies in disadvantaged communities led to community distrust with the police. Lastly, he found social cohesiveness and government responsiveness are best predictors of the emergence of informal social controls. Renauer’s (2007) study showed that as community participation in COP became less, reliance on formal social controls (e.g. police) increased. In short, the lack of community participation resulted in lesser adherence to the basic tenets of community policing including the formation of collaborative partnerships between the community and the police.

Similarly, such debilitating effects of social disorganization on the development of informal social control were evident in Warner’s (2007) study. She found that communities beset by greater social disorganization would tend to seek the formal intervention by the police. She argued that among disadvantaged communities, community members could not directly perform their roles as sources of informal controls (Warner, 2007). Hence, community members would tend to rely on the police to restore social order through traditional law enforcement methods rather than enhancing order through informal interventions.

These two recent studies on social disorganization highlight the constraining effects of community conditions on implementation of informal social control (i.e., operationalized as COP). In Renauer’s (2007) study, the police efforts of community oriented policing failed to flourish and enhance order in the community. Worse, community policing strategies increased community dissatisfaction with the police. Similarly, Warner’s study (2007) indicated that community exercise of informal control (i.e., participation in COP) was predicated on the establishment of order by the police. These two studies lend credence to the views of Lambertus and Yakimchuck’s (2007) perspective on the relationships of community hierarchy of needs and policing practices.

Methods

Theoretical perspectives

The perspective tested in this paper was derived from Lambertus and Yakimchuk (2007) and grounded on the contingency perspective on policing (Cordner & Scarborough, 2010). Lambertus and Yakimchuck (2007) have suggested that communities should be treated as living organisms with differing levels of needs, more specifically, safety needs. They posited that their perspective about community needs may be analogous to Maslow’s (1954) theory on worker’s hierarchy of needs.

Lambertus and Yakimchuck (2007) presented three hierarchical levels of community safety needs. They state that the basic need of communities is social order. They believe that it is the very foundation for community’s continued existence and a catalyst for its members to socially participate. Therefore, communities that are plagued by crimes and disorder should aim for social order recovery/repair. Once social order is restored and repaired, communities would could aspire for the next higher social order which is to maintain that state of social order. Lambertus and Yakimchuck called this as order maintenance level of need. The maintenance of this desired state would inspire citizens to reach for the next higher level which is called social order enhancement. This level is synonymous to a self-ac- tualized community that are empowered and have the ability to reach the highest form of order that they could aspire.

They postulated that these social order needs are hierarchical, that is, lower order needs must be satisfied prior to being able to move the next higher level needs (Lambertus & Yakimchuk, 2007). They also suggested that most urban communities would be plagued with lower level needs as indicated by the broader base of the triangle. This level of disorder in urban areas being was earlier predicted by Lundman (1980) in his theory about the development of the police. Later his thesis was empirically demonstrated by Travis and Langworthy (2008) and was confirmed in another study by de Guzman and Kumar (2011).

356 M. C. dE GUzMAn And M. KIM

Figure 1 depicts these social order needs. Furthermore, Lambertus and Yakimchuk (2007) suggested that the most basic need of the community is to establish order in society. They argued that societies beset by massive social disorders could not elevate themselves to maintain and enhance order in soci- ety. They suggested that formal social controls should have more instrumental roles in the base level and that informal social controls would only emerge at a gradual level after the recovery of order in society. Thus, they suggested that that communities that are beset by physiological and security needs could not make community oriented policing as viable. They expected that greater implementation of community policing could only be achieved in communities that have achieved a level synonymous to self-actualization or an order enhancement phase.

Lambertus and Yakimchuck’s perspective appear to be shared by Travis and Langworthy (2008) on their propositions on the relationship of community structures and models of policing. Travis and Langworthy (2008) suggested that COP could primarily emerge only in solidary communities that are independent from the larger society in the provisions of locality relevant functions (e.g. security) and more importantly, the members have high consensus on the provision of these locality relevant functions. Absent these characteristics, they suggested that fragmented or disorganized communities emerge where police departments would tend to employ the traditional form of policing.

This current research empirically investigates Lambertus and Yakimchuk’s (2007) propositions by examining the tendencies of different police departments to adopt and implement community oriented policing in urban settings. In addition, it also provides an empirical validation of Wilson and Kelling’s (1982) idea that police could implement COP as a response to disorders in society. Our prediction is that communities that have greater lower social needs will tend towards a more traditional type of policing (i.e., lesser tendencies to implement community policing) and that communities that have achieved greater social needs will tend towards the adoption and more extensive implementation of community oriented policing. The theoretical perspective is illustrated in Figure 2.

Figure 1. lambertus & Yakimchuk’s (2007, p. 31) hierarchy of social needs.

Figure 2. theoretical diagram of the relationship between community hierarchy of needs and police practices.

PoLICE PRACTICE And RESEARCh 357

The upright triangle represents the community hierarchy of needs with the base as a situation where communities have the lowest level of needs (i.e., highly disorganized) and the apex representing community self-actualization (i.e., highly organized). The inverted triangle-shaped figure represents the level of implementation of community oriented policing. At the base of that inverted triangle is the fullest implementation of community oriented policing and the apex of the inverted triangle is the least implementation of community policing.

Thus, this study posits that levels community needs will have varying effects on the emergence of informal means of social control or community policing. Therefore, police departments in commu- nities that have greater lower level needs would tend to have lower levels of COP implementation. Alternatively, police departments that operate within communities with higher level needs satisfied would be able to greatly implement community oriented policing.

Conceptualization

The concept of policing models was operationalized by using Wilson’s (2006) measurements on the level of implementation of community policing. Police departments could either pursue COP more or less vigorously. Wilson’s (2006) measures and analyses provided an empirical way of validating the normative claims by most police departments of pursuing community policing. Thus, we can now have empirical measures of COP and traditional policing models. A lower level of COP implementation suggests that a more traditional and aggressive type of policing is being pursued. Hence, the dependent variable will be measured by indicators for implementation of COP.

With respect to the independent variable, that is community hierarchy of need, the concept is more challenging to measure as the hierarchy of needs has always been conceptualized on the individual level (Maslow, 1954). The challenge is to find a data-set at the community level to represent a community’s level of order needs. Social disorganization presents a logical proxy for community hierarchy of needs. Despite the direct measures of social disorganization presented by Sampson and Groves (1989), most researchers still consider structural variables such as poverty, population heterogeneity, or mobility among others (See Cantillon et al., 2003; Lowenkamp, Cullen, & Pratt, 2003; Veysey & Messner, 1999) as the most appropriate constructs to measure social disorganization. For this empirical study, there- fore, structural variables were chosen as indicators for the hierarchical levels of community needs. In order to have parallel levels of units of analysis, city or municipal levels of disorganization data are derived. This is important as police organizations and policies are structured on the department rather than at the precinct level. In other words, police policies and practices are determined at the central offices and are assumed to apply across their operational jurisdictions. It is also important to note that policy makers and executives formulate and implement policies based on a city/municipal level perspective. Thus, the variables are conceptualized as city or municipal level indicators rather than as the neighbourhood level indicators.

Data

This paper utilized four data sources: U.S. 2000 Census Summary File 3 (SF3) ‘place’ level data including city, town, and village demographic information; Law Enforcement Management and Administrative Statistics 2003 (LEMAS 2003, ICPSR# 4411); and FBI’s Uniform Crime Reports 2000 (UCR 2000, ICPSR# 3447). Law Enforcement Agency Identifier Crosswalk 2000 (ICPSR# 4082) was used to merge census data and police/crime data with FIPS (Federal Information Processing Standard) and agency ORI (UCR Originating Agency Identifier) code. Considering the availability of updated existing data, the temporal orders among census variables, crime variable, and community policing implementation variables, and also potential causal orders, we adopted the databases mentioned above. Likewise, the succeeding LEMAS data starting in 2003 has been revised that dropped some measures of COP implementation that the use of more recent database is not warranted by the research. LEMAS sur- vey instruments after 2003 have been changed to focus more on anti-terrorism efforts rather than

358 M. C. dE GUzMAn And M. KIM

on community policing. Likewise, the 2000 U.S. Census data was used to derive measures of social disorganization in order to closely reflect the LEMAS survey respondent’s contextual references with respect to the responsiveness of the police to their operational environments. A larger gap between the two datasets will not make the findings less credible with respect to potential causal inferences and respondents’ contexts for providing those information about implementation of COP.

In order to maintain the consistent units of analysis, say municipality level, all data sets were care- fully merged using unique IDs for each place. Merging these datasets and selecting only municipal police agencies (i.e., villages, towns, and cities), a total 1248 police agencies were included for analysis. Among the agencies in the sample, 32.4% (n = 404) of agencies covered small communities with less than 10,000 population; and 15.4% (n = 192) of agencies covered bigger communities over 100,000 population. The rest may be considered as medium-sized areas.

Dependent variables

Community oriented policing implementation index (COPI)1 The variable, COPI, captures the levels of community oriented policing implementation in which higher index values indicate more imple- mentation of community oriented policing in the police department. The index is a factor weighted composite variable constructed from fourteen (14) observed community policing measures in LEMAS 2003 (Wells & Falcone, 2005 and Wilson, 2006). That is, COPI =

∑n

i=1 x i w

i where x is the value of a

given variable, w is the factor loading weight for a given variable, and i indicates different community oriented policing measures. The measures of community orientated policing we used in the construc- tion of COPI are explained below.

LEMAS contained three (3) measures of community policing training within the agency. Police departments reported the proportions of recruit, in-service sworn, and non-sworn personnel who received at least 8 h of training within the 12-month period. For each of the personnel categories, organizations had the option to respond with none, less than half, more than half, or all. Original responses to the LEMAS survey were reported on a four point scale with 1 (all) to 4 (none). This research reverse-coded the original responses to represent a four-point scale ranging from 0 (none) to 3 (all).

LEMAS also requested the organization to indicate if the department included community policing components in its mission statement; if it maintained or created a formal, written community policing plan; and if the organization had a dedicated person to address community policing. These three (3) measures were coded as 0 and 1.

Four (4) measures were included to capture problem solving efforts in the police department: SARA project, assigned beats, problem solving, and upgraded technology. LEMAS asked organizations to respond whether they actively encouraged their officers to engage in the SARA model (scanning, analysis, response, and assessment) to address community problems within the last twelve months; if they gave patrol officers responsibility for specific geographic areas/beats within the last twelve months; if the organization included collaborative problem-solving projects in the evaluation criteria of patrol officers; if the organization used an upgraded technology to support the analysis of community prob- lems. The responses for each of the four measures were coded as 0 (no) and 1 (yes).

Four (4) dichotomous measures were also included in the LEMAS survey to estimate the degree of citizen involvement in policing: if the organization conducted a citizen police academy or not; if the organization trained citizens in community policing such as community mobilization and problem solving; if the organization had a problem-solving partnership or written agreement with community groups; if the organization sponsored or surveyed their citizens on co-production of safety and police services in the community.

All measures included in COPI are reliable to secure internal consistency to represent the features of community oriented policing (Cronbach’s alpha = .80). COPI is also stable from the error due to the items with different scales by employing the factor score weights which indicate the predicted value for COPI when the measured indicator goes up by 1 unit.

PoLICE PRACTICE And RESEARCh 359

Independent variables

Social Disorganization Index (SDI). The variable, SDI, is a proxy measure of community hierarchy of needs. Higher index values indicate more disorganized characteristics of the communities which need to satisfy lower level of community needs in hierarchy first. The SDI is a factor weighted composite variable which utilized several structural variables in the U.S. Census 2000. The index value was cal- culated as, SDI =

∑n

i=1 x i w

i where x is the value of a given structural variable, w is the factor loading

weight for a given variable, and i indicates different structural measure. The measures to calculate the SDI include population having less education, population having

non-professional occupations, racial heterogeneity, family disruption, household income, and the lack of home ownership.2 For each measure, higher value indicates more disorganized feature. For the operationalization of each Census variable, lower education was measured by the percent population who received lower than BA degree among 25 + years old. The percent population who worked for non-professional and non-managerial occupations among 16 + years old was captured as non-profes- sional occupations measure. We measured ethnic heterogeneity based on the composition of different racial groups. Originated by Blau (1977), and widely used in many social science studies (Fitzpatrick & Hwang, 1992; Sampson, 1984), the racial heterogeneity index was calculated with, D = 1 −

∑N

i=1 p2 i ,

where p is the proportion of racial groups in a given category, and i is the number of different categories of the feature across all groups. The scores could range from 0 (i.e., perfect homogenous population) to 1 (i.e., perfect heterogeneous population). Family disruption was measured with the percentage of married people who were separated or divorced (Lowenkamp et al., 2003). Household income was estimated with the natural logarithm of an inverse median household income, Ln(1/income), so that higher income could get lower value and lower income could get higher value. The lack of home own- ership was estimated by the percentage of renter occupied housing units.

All measures included here are acceptable to secure internal consistency of SDI (Cronbach’s alpha = .62). The use of the factor score weights for each item further secures SDI from the issue of unreliable fluctuations caused by some items with different scales.

Control variables

To estimate the net effect of community hierarchy of needs on the implementation of community oriented policing, we included urbanity, geographical region, crime rate, and organizational features as control variables. Since previous research (Wilson, 2006) suggested that the features of police department should be addressed to understand the implementation of community policing, we con- trolled for five organizational variables: Scope of department tasks, spatial differentiation, occupational differentiation, functional differentiation, and formalization. The number of sworn officers measuring the department size used in other literature, however, was not directly included here due to the mul- ticollinearity concern with ‘scope of department tasks’ in our analytical model.

Using LEMAS 2003, and guided by Wilson’s work (2006), the following organizational variables were constructed. Scope of department tasks was calculated with the number of tasks the police depart- ment maintained as a primary responsibility. Total 37 tasks were included in LEMAS 2003.3 Spatial differentiation measures the total number of facilities or stations in addition to headquarters, which includes district of precinct stations, fixed and mobile neighbourhood or community substations, and other facilities. Occupational differentiation measures the heterogeneity of the department com- position in term of sworn and non-sworn officers. It was calculated with the mathematical formula, D = 1 −

∑

i=1 p2 i , where p is the proportion of occupational groups in a given category (Maguire,

2003; Wilson, 2006). The scores could range from 0 (i.e., all sworn or all non-sworn) to .5 (i.e., equal number of sworn and non-sworn officers). Functional differentiation measures the number of special units in which at least one full-time employees is assigned. Total 22 specialized units were included in LEMAS 2003.4 Finally, formalization represents the total number of written policies for fifteen different topics in the department.5

360 M. C. dE GUzMAn And M. KIM

In addition to the characteristics of the department, urbanity was included because it influences the policing styles and police-community relations (Crank, 1990; Flanagan, 1985; Meagher, 1985). Urbanity was measured by the percentage of population who lived in urban area including urban cluster. The geographical region was included as a dichotomous variable to identify agencies in the West or other regions based on U.S. Census definition. Previous research suggested that the departments in West region seemed to apply more innovative approaches in policing than those in other regions (Wilson, 2006; Zhao, 1996). Each police agency was assigned into either West or non-West. Crime rate of the community is another variable which affects police resources and practices. (Kovandzic & Sloan, 2002) Therefore, crime rate was measured by the number of UCR Part 1 index crimes per 100 residents. Table 1 shows a summary of each variable and the descriptive statistics in our analyses.

Analytical model

A multivariate analysis with the ordinary least square (OLS) regression model was used to iden- tify the relationship between SDI and COPI controlling for other variables. COPI was regressed on SDI, department variables, and other control variables. Following equation shows our full model. COPI  =  a  +  b1SDI  +  b2Scope  +  b3Spatial  +  b4Occupational  +  b5Functional  +  b6Formalization  +  b7Region  +  b8Urbanity  +  b9Crime  +  e, where a indicates the constant term; b1 is the coefficient of SDI; b2, b3, b4, b5, and b6 show the coefficient of each departmental variables (Scope of department tasks, spatial differentiation, occupational differentiation, functional differentiation and formalization, respectively); b7, b8, and b9 represent the coefficient of geographic region, urbanity, and crime rate respectively; e represents random errors. For all models in our analysis, we did not find any indication of the violation of the OLS assumptions. The threat of multicollinearity among variables in the right hand side of the equation was checked with VIF and no serious threat was found (maximum 1.445 which is far below than common practice of 10.0 or even 4.0 for stricter rule).

Table 1. Variable descriptions and descriptive statistics (N = 1248).

Variables description (Measure) Mean Sd Min Max Dependent variables coPi coPi as an overall measure (Factor weighted com-

posite variable with fourteen indicators) 1.177 .738 .00 2.83

Independent variables sdi sdi as an overall measure (Factor weighted com-

posite variable with six indicators) 69.380 13.323 16.57 96.83

Control variables department scope of department tasks the number of tasks the police department

maintained as a primary responsibility 19.999 4.056 3 33

spatial differentiation the total number of facilities or stations 2.254 6.020 0 100 occupational differentiation the heterogeneity of the department composition

in term of sworn and non-sworn officers .314 .141 0 .5

Functional differentiation the number of special units in which at least one full-time employees is assigned

3.242 5.210 0 22

Formalization the total number of written policies for fifteen different topics in the department

12.543 2.497 0 15

others Geographic location (West = 1) a dummy variable – Geographic locations in

census (West vs. non-West) .208 .406 0 1

Urbanity % population living in census urban area and urban cluster

91.1 25.2 .00 100.0

crime rate Ucr index crime per 100 residents 6.333 5.959 .00 168.64

PoLICE PRACTICE And RESEARCh 361

Results

Estimates of the covariate of SDI with the implementation of community oriented policing (COPI) were derived. The results in Table 2 show that the SDI is negatively and significantly associated with COPI. Police departments within socially disorganized areas are tend to have lower levels of COP implementation. The negative effects of SDI on COPI is consistent throughout the different models we have constructed. All the organization variables except one (i.e., spatial differentiation) are signifi- cantly associated with the variations in COPI. Thus, departments that have greater task scopes, greater occupational differentiation, more functionally differentiated, and more formalized rules implement COP to a significantly higher extent. The effects of other environmental control variables seem to follow a consistent pattern. Regional location, urbanity, and crime rates significantly contribute to COPI. However, it should be noted that urbanity does not reach significance in the full model. This means that urbanity is not factor in the presence of other explanatory variables.

Although organizational variables are significant predictors of COP implementation levels as expected, the level of COP implementation is also clearly influenced by the community features out- side of the department. Thus, community hierarchy of needs and organizational capabilities constrain the implementation of COP.

Discussion

Although community hierarchy of needs have modest effects, the findings lend credence to Lambertus and Yakimchuck’s (2007) main hypothesis that community hierarchy of needs influences the imple- mentation of community policing. When community needs are made compete with other explanatory variables, organizational dynamics seem to exert much greater weight. Community hierarchy of needs seems to have less influence than what Lambertus and Yakimchuck (2007) have proposed. Despite these relatively minimal effects, the findings in the study lend support to several studies on community oriented policing where COP was considered the policing strategy of choice in socially disorganized neighbourhood (Skogan, 1986; Skogan & Hartnett, 1997). The findings also lend support to Wilson and Kelling’s (1982) advice on the importance of fixing broken windows. This finding suggests that police organizations might consider COP as an appropriate response to socially disorganized communities. This tendency is probably driven by the promises of COP as a remedy to re-establish order in the community. However, the police department’s desire to vigorously implement community oriented policing in communities could be greatly constrained by department resources. Thus, findings in the study could be interpreted to mean that the level of community hierarchy of need is an impetus for the pursuit of COP but organizational resources limits the vigorous pursuit of COP.

Table 2. ols regression models explaining the implementation of community oriented policing.

**p < .05; ***p < .01.

Variables

(1) (2) (3)

B SE (B) Β B SE (B) β B SE (B) β SDI −.003** .001 −.047 −.006*** .002 −.117 −.003** .001 −.050 task scopes .014*** .004 .079 .015*** .004 .083 spatial differentiation .002 .003 .016 .002 .003 .017 occupational differentiation .925*** .122 .177 .683*** .128 .131 Functional differentiation .049*** .004 .348 .048*** .004 .336 Formalization .089*** .007 .302 .083*** .007 .280 location (West = 1) .393*** .047 .216 .233*** .041 .128 Urbanity .007*** .001 .246 .001 .001 .040 crime rates .019*** .003 .154 .005 .003 .044 (constant) −.500** .150 – .770*** .137 – −.535*** .152 –

R2 = .406 R2 = .158 R2 = .423

362 M. C. dE GUzMAn And M. KIM

Interestingly, crime rate has no significant effects on COP implementation after controlling for department and social structural variables. In fact, the data seem to suggest that department concern about crime rate dissipates with the presence of organization resources. Among the organizational variables, departments with more functional units and task units seem to have greater levels of COP implementation while having more precincts does not seem to have a significant influence on COP implementation. This finding negates some of the traditional belief that widespread structural police deployment was a key component of COP. Perhaps, what police actually do in the field is more impor- tant than structural spatial coverage.

Conclusion

Community policing has been the current philosophy that has engulfed the police agencies in the last 30 years (Wilson, 2006). Community policing has been considered by police departments as an effective response to the disorders in society (Reisig & Parks, 2004; Skogan, 1996; Skogan & Hartnett, 1997; Wilson & Kelling, 1982). In response, police departments seemed to have embraced this prom- ise. However, as revealed by the analyses, community policing initiatives seem to be primarily driven by organizational characteristics than community needs. This finding indicates that police might be recognizing the resource-driven demands of COP. Findings in this study suggest that environmental contingencies seem to have modest impacts than organizational contingencies. Despite the robust impact of organizational variables, the findings suggest that police departments seem convinced that COP is an appropriate response among socially disorganized communities. However, faced with organizational constraints, police departments will perhaps tend to lessen its community oriented policing programs.

As stated in previous research, the character and behaviour of the police are products of its soci- ety (Black, 1980; Caldero & Crank, 2004; de Guzman & Kumar, 2011; Lundman, 1980; Travis & Langworthy, 2008). COP was touted as a response to the demands of the environment (Reisig & Parks, 2004; Skogan, 1986; Wilson & Kelling, 1982). This present research adds to the strings of research in line with these assertions. However, certain theoretical perspectives have asserted different directions on how the police react to their environment. Research has shown the important influence of the envi- ronment on police behaviour (Cordner & Scarborough, 2010; de Guzman & Jones, 2012; de Guzman & Kumar, 2011). Other researchers assert that police organizations are closed and that they cling on to long held beliefs and traditions as the bases for their practices and behaviours (Caldero & Crank, 2004; Crank & Langworthy, 1992; Kraska, 2001, 2007). This paper looks at contingency theory concepts (crime and social disorganization), as well as organizational level factors (occupational differentiation, formalization. We acknowledge that institutional level factors may have significant impact on police models adoption. Burruss and Giblin (2014), for example, specifically looked at institutional-level factors, such as the impact of journals, meetings, and networks on the implementation of COP. They found that institutional factors were strong predictors of COP implementation above organizational and contingency factors. Thus, future research should integrate some of these competing explanations for police behaviors. Police departments implement policing in part because they react to pressures to do something about crime (contingency theory), but they also implement COP because their peers are doing the same and the administration wishes to mimic peer institutions (institutional theory). Thus, it may be possible that social disorganization may no longer be a significant predictor if the institutional level predictors are added. However, this could not be addressed in the current research. Despite these plausible counter-arguments, this current research presents a refinement of our under- standing of the influences of environment on the police and has shown additional complexities on the factors that shape police organizational behaviour. Police respond to community needs but are sometimes hampered by the realities within their organizations.

PoLICE PRACTICE And RESEARCh 363

Limitations and future directions

Despite initial support for Lambertus and Yakimchuck’s model, a more extensive research is needed. Due to the limitations of current data, there should be a collection of datasets that would address the other concepts involved in social disorganization theory. For instance primary data that validly captures collective efficacy or perceived risks instead of disorders are necessary. To capture the features of police departments, this study used five organizational variables in the equations. If we could operationalize other factors such as institutional variable discussed by Burruss and Giblin (2014) in the analysis, we would enhance our understanding of the relationships between police organizational behaviors and the operational environments.

Another caution should be noted about applicability of the findings to international police organ- izations. Although most countries will have an appreciation of the universality of some of the social disorganization measures, policing organizational cultures and constraints must be considered. For example, a lot of countries have a national or centralized form of policing that a city or municipal level of analysis might need to be more carefully structured. Despite these limitations, the lesson in this paper is that COP may not be appropriate for all circumstances and jurisdictions. Instead, the more appropriate approach would be to have a more appropriate adoption of police models and its features that correspond to the needs of its constituents. Future research is needed to identify those circum- stances where the community and the police could find a common ground to be able to enhance order.

Notes 1. For COPI calculation, weights were obtained from a confirmatory factor analysis: training for new officers = .072,

training for sworn officers = .064, training for non-sworn = .079, citizen survey = .125, citizen partner = .145, citizen training  =  .252, citizen academy  =  .192, technology use  =  .191, inclusion of problem solving in evaluation = .232, fixed beat = .211, use of SARA = .282, mission statement = .156, written plan = .234, and dedicated person = .169.

2. Since many structural variables are correlated with each other, we only included these variables after addressing multicollinearity issue using Variance Inflation Factor (VIF) less than 4.0.

3. LEMAS 2003 included 37 different functions: 5 law enforcement functions, 5 traffic and vehicle-related functions, 4 criminal investigations functions, 6 court-related functions, 6 special public safety functions, 4 special operations, 4 detention-related functions and 3 other functions.

4. Twenty-two special units in LEMAS 2003 are bias/hate crime, bomb/explosive disposal, child abuse/ endangerment, community crime prevention, community policing, crime analysis, cyber crime, domestic violence, drug education in schools, gangs, impaired drivers, internal affairs, juvenile crime, methamphetamine labs, missing children, prosecutor relations, repeat offenders, research and planning, school safety, terrorism/ homeland security, victim assistance, and youth outreach.

5. Written policy directives on the following fifteen areas were included in the calculation: Use of deadly force/ firearm discharge, use of less-than-lethal force, code of conduct and appearance, off-duty employment of officers, maximum work hours allowed for officers, dealing with the mentally ill, dealing with the homeless, dealing with domestic disputes, dealing with juveniles, strip searches, racial profiling, citizen complaints, off-duty conduct, interacting with the media, and employee counselling assistance.

Disclosure statement No potential conflict of interest was reported by the authors.

Notes on contributors Melchor C. de Guzman is professor of Criminal Justice and Criminology at Georgia Gwinnett College. He earned his doctorate degree in criminal justice from the University of Cincinnati. Dr de Guzman’s research includes the examination of citizen participation in the control of the police. He also investigates organizational and environmental factors that influence police behavior. His most recent research includes the utilization of websites by police departments in the delivery of police services and the role of the police in homeland security. Dr de Guzman could be contacted through email at [email protected] or by phone at 470-217-5512.

364 M. C. dE GUzMAn And M. KIM

MoonSun Kim is an associate professor of criminal justice at The College at Brockport, State University of New York. He earned his PhD from the School of Criminal Justice, University at Albany, SUNY. He has been involved in various research projects for local and state law enforcement agencies by utilizing advanced statistics and geo-spatial data anal- yses. His current research interest centers on police intervention research, community policing, and terrorism. Dr Kim could be contacted through email at [email protected] or by phone at 585-395-2915.

References Bayley, D., & Shearing, C. (2001). The new structure of policing: Description, conceptualization, and research agenda.

Washington, DC: National Institute of Justice, Office of Justice Programs. Black, D. (1980). The manners and customs of the police. New York, NY: Academic Press. Blau, P. (1977). Inequality and heterogeneity. New York, NY: The Free Press. Boessen, A., & Hipp, J. R. (2016). Close-ups and the scale of ecology: Land uses and the geography of social contexts

and crime. Criminology, 53, 399–426. Burruss, G. W., & Giblin, M. J. (2014). Modeling isomorphism on policing innovation: The role of institutional pressures

in adopting community-oriented policing. Crime & Delinquency, 60, 331–355. Bursik, R. J., & Grasmick, H. G. (1993). Neighborhood and crime: The dimensions of effective community control. New

York, NY: Lexington Books. Caldero, M. A., & Crank, J. P. (2004). Police ethics: The corruption of a noble cause (2nd ed.). Cincinnati, OH: Anderson

Publishing. Cantillon, D., Davidson, W. S., & Schweitzer, J. H. (2003). Measuring community social organization: Sense of community

as a mediator in social disorganization theory. Journal of Criminal Justice, 31, 321–339. Cordner, G. W., & Scarborough, K. (2010). Police administration (7th ed.). New Providence, NJ: Lexis-Nexis. Crank, J. P. (1990). The influence of environmental and organizational factors on police style in urban and rural

environments. Journal of Research in Crime and Delinquency, 27, 166–189. Crank, J. P. (2003). Institutional theory of police: A review of the state of the art. Policing: An International Journal of

Police Strategies and Management, 26, 186–207. Crank, J. P., & Langworthy, R. H. (1992). An institutional perspective of policing. The Journal of Criminal Law and

Criminology, 83, 338–363. de Guzman, M. C., & Jones, M. A. (2012). E-policing. International Journal of Electronic Government Research, 8, 64–82. de Guzman, M. C., & Kumar, S. K. (2011). Extending Lundman’s theory on policing: the evidence from the literature

focusing on India. Policing: An International Journal of Police Strategies & Management, 34, 403–418. Fitzpatrick, K. M., & Hwang, S. (1992). The effects of community structure on opportunities for interracial contact.

Extending Blau’s macrostructural theory. The Sociological Quarterly, 33, 51–61. Flanagan, T. J. (1985). Consumer perspectives on police operational strategy. Journal of Police Science and Administration,

13, 10–21. Grinc, R. (1994). ‘Angels in marble’: Problems in stimulating community involvement in community policing. Crime

and Delinquency, 40, 437–468. Hipp, J. R. (2016). Collective efficacy: How is it conceptualized, how is it measured, and does it really matter for

understanding perceived neighborhood crime and disorder? Journal of Criminal Justice, 46, 32–44. Kerley, K. R., & Benson, M. L. (2000). Does community-oriented policing help build stronger communities? Police

Quarterly, 3, 46–69. Kovandzic, T. V., & Sloan, J. J. (2002). Police levels and crime rates revisited. Journal of Criminal Justice, 30, 65–76. Kraska, P. (2007). Militarization and policing – Its relevance to 21st century police. Policing, 1, 501–513. Kraska, P. B. (2001). Militarizing the American criminal justice system: The changing roles of the armed forces and the

police. Boston, MA: Northeastern University Press. Kubrin, Charis E., & Weitzer, R. (2003). New directions in social disorganization theory. Journal of Research in Crime

and Delinquency, 40, 374–402. Lambertus, S., & Yakimchuk, R. (2007). Future of policing in Alberta: International trends and case studies (A discussion

paper). Alberta: Alberta Solicitor General and Public Security Office Research Unit. Lowenkamp, C. T., Cullen, F. T., & Pratt, T. C. (2003). Replicating Sampson and Groves’s test of social disorganization

theory: Revisiting a criminological classic. Journal of Research in Crime and Delinquency, 40, 351–373. Lundman, R. J. (1980). Police and policing: An introduction. New York, NY: Holt, Rinehart, & Winston. Maguire, E. R. (2003). Organizational structure in American police agencies: Context, complexity, and control. Albany,

NY: SUNY Press. Maslow, A. (1954). Motivation and personality. New York, NY: Harper Row. Meagher, M. S. (1985). Police patrol styles: How pervasive is community variation? Journal of Police Science and

Administration, 13, 36–45. Merton, R. K. (1968). Social theory and social structure. New York, NY: Free Press.

PoLICE PRACTICE And RESEARCh 365

Perkins, M. (2016). Modelling public confidence of the police: How perceptions of the police differ between neighborhoods in a city. Police Practice and Research, 17, 113–125.

Reisig, M., & Parks, R. (2000). Experience, quality of life, and neighborhood context: A hierarchical analysis of satisfaction with police. Justice Quarterly, 17, 607–630.

Reisig, M. D., & Parks, R. B. (2004). Can community policing help the truly disadvantaged? Crime & Delinquency, 50, 139–167.

Renauer, B. C. (2007). Is neighborhood policing related to informal social control? Policing: An International Journal of Police Strategies & Management, 30, 61–81.

Sampson, R. J. (1984). Group size, heterogeneity, and intergroup conflict: A test of Blau's inequality and heterogeneity. Social Forces, 62, 618–639.

Sampson, R. J., & Groves, W. B. (1989). Community structure and crime: Testing social-disorganization theory. American Journal of Sociology, 94, 774–802.

Skogan, W. (1986). Fear of crime and neighbourhood change. In A. Reiss & M. Tonry (Eds.), Crime and justice (Vol. 8, pp. 203–229). Chicago, IL: University of Chicago Press.

Skogan, W. (1990). Disorder and decline. Berkeley, CA: University of California Press. Skogan, W., & Hartnett, S. (1997). Community policing, Chicago style. Chicago, IL: University of Chicago Press. Travis, L. F., & Langworthy, R. H. (2008). Policing in America: A balance of forces (4th ed.). Cincinnati, OH: Prentice Hall. Veysey, B. M., & Messner, S. F. (1999). Further testing of social disorganization theory: An elaboration of Sampson and

Groves’s ‘community structure and crime’. Journal of Research in Crime and Delinquency, 36, 156–174. Warner, B. D. (2007). Directly intervene or call the authorities? A study of the forms of neighbourhood social control

within a social disorganization framework. Criminology, 45, 99–129. Wells, L. E., & Falcone, D. (2005, April). Policing in the United States: Developing a comprehensive empirical model (Final

report to the National Institute of Justice. NCJ 210830). Illinois State University [publisher], National Institute of Justice [distributor].

Wilson, J. (2006). Community policing in America. New York, NY: Taylor & Francis. Wison, J. Q. (1968). Varieties of police behaviour. Boston, MA: Cambridge University Press. Wilson, J. Q., & Kelling, G. L. (1982). Broken windows. Atlantic Monthly, 211, 29–38. Zhao, J. (1996). Why police organizations change: A study of community-oriented policing. Washington, DC: Police

Executive Research Forum.