Developing a Comprehensive Empirical Model of Policing in the United States, 1996-1999 (ICPSR 4338)

Time Period(s) View help for Time Period(s)

Date of Collection View help for Date of Collection

Data Collection Notes View help for Data Collection Notes

Hide

Study Purpose View help for Study Purpose

The aim of this study was to provide a systematic empirical assessment of three basic premises of Community-Oriented Policing (COP): (1) a structural premise that what police departments do is shaped by their organizational structures, (2) a contextual premise that police agencies as "open systems" are constrained and influenced by their environments, and (3) a universality premise that, because the essential tasks of policing are the same everywhere, a single universal model of COP policing will apply to all sizes and types of police agencies. The project was directed by three sets of research questions corresponding to these premises: (1) How strongly is the implementation of community-oriented policing by police agencies across the U.S. predicted by the organizational structures of agencies? (2) How strongly do community contextual characteristics determine or predict the adoption of COP? (3) Does a single general organizational model universally describe the inter-relations among community contexts, police organizational structures, and implementations of community-oriented policing procedures?

Study Design View help for Study Design

This study constructed a comprehensive data set by synthesizing data available in separate national data sets on police agencies and communities. The base data source used for this synthesis was the 1999 Law Enforcement Management and Administrative Statistics (LEMAS) survey [LAW ENFORCEMENT MANAGEMENT AND ADMINISTRATIVE STATISTICS (LEMAS), 1999 (ICPSR 3079)], which contained data on police organizational characteristics and on adoption of community-oriented policing procedures. The 1999 survey was supplemented with additional organizational variables from the 1997 LEMAS survey [LAW ENFORCEMENT MANAGEMENT AND ADMINISTRATIVE STATISTICS (LEMAS), 1997 (ICPSR 2700)] and from the 1996 Directory of Law Enforcement Agencies [DIRECTORY OF LAW ENFORCEMENT AGENCIES, 1996: [UNITED STATES] (ICPSR 2260)], which provides a census of all 18,769 police agencies throughout the United States. Data on community characteristics were extracted from the 1994 County and City Data Book, from the 1996 to 1999 Uniform Crime Reports [UNIFORM CRIME REPORTING PROGRAM DATA. [UNITED STATES]: OFFENSES KNOWN AND CLEARANCES BY ARREST (1996-1997: ICPSR 9028, 1998: ICPSR 2904, 1999: ICPSR 3158)], from the 1990 and 2000 Census Gazetteer files, and from Rural-Urban Community classifications. The merging of the separate data sources was accomplished by using the Law Enforcement Agency Identifiers Crosswalk file [LAW ENFORCEMENT AGENCY IDENTIFIERS CROSSWALK [UNITED STATES], 1996 (ICPSR 2876)]. In all, 23 data files from eight separate sources collected by four different governmental agencies were used to create the merged data set. The entire merging process resulted in a combined final sample of 3,005 local general jurisdiction policing agencies (2,034 municipal-level and 971 county level) for which the researchers had matching community-demographic data on 2,449 and valid crime rate data on 2,242.

Sample View help for Sample

The base data source used for this synthesis was the 1999 Law Enforcement Management and Administrative Statistics (LEMAS) data, which contains 3,246 agencies. The editing and merging operations resulted in a combined final sample of 3,005 local general jurisdiction policing agencies. While these data are national in the scope of coverage, they do not constitute a random sample of all police departments in the United States. First, the LEMAS survey on which the merged data set was constructed uses a variable, disproportionate sampling procedure that includes all large police departments (larger than 100 full-time sworn officers) and then undersamples smaller departments. Second, the substantial amount of missing data in the final merged data set undercuts the use of probability sampling, which introduces some nonrandom selection bias into the sample. These data are broader in coverage and content than used in previous police organization research, but at the same time, are not presented as ideal or as providing final, conclusive estimates.

Universe View help for Universe

All law enforcement agencies in the United States.

Unit(s) of Observation View help for Unit(s) of Observation

Data Source View help for Data Source

Data Type(s) View help for Data Type(s)

Mode of Data Collection View help for Mode of Data Collection

Description of Variables View help for Description of Variables

Variables regarding police organizational structure include type of government (county, municipal, township), type of agency (sheriff, county, municipal, special), total number of full-time and part-time employees, total number of sworn, nonsworn, and civilian employees, total number and percent of employees in various types of units, the number of facilities operated by the agency, type of union membership of sworn employees, and whether the agency has collective bargaining for sworn and nonsworn employees. Various indices from the LEMAS surveys are provided regarding task scope, patrol types, computerization, administrative intensity, drug testing, selection/screening, educational requirements, training, unionization, and organizational height. Community-oriented policing variables are the percent of full-time sworn employees assigned to COP positions, if the agency had a COP plan (formally written, informal, no COP plan), and several indices from the 1999 LEMAS survey regarding COP training, procedures, meetings, and use of community surveys in planning and evaluation. Community context variables include population in 1994, percent of population non-White, Black, and Hispanic/Latino, percent of families and persons below poverty, employment rate, median household income, per capita income, racial heterogeneity index, percent of population 5 to 17 years of age, percent population 65 years and older, percent of adults (over 25 years of age) with high school education, percent of housing that is renter-occupied, percent of households that are single-occupancy, mean number of persons per household, population density in 1990 and 2000, population change between 1990 and 2000, rural-urban continuum (Beale) codes, urban influence codes, metropolitan vs. nonmetropolitan, total serious crime rate (average of 1996-2000), total serious crime rate (average of 1998 and 1999). Geographic variables include FIPS state, county, and place codes, and region.

Response Rates View help for Response Rates

Not applicable.

Presence of Common Scales View help for Presence of Common Scales

none.

Hide

Original Release Date View help for Original Release Date

2006-09-06

Version History View help for Version History

Hide

Weight View help for Weight

The base sampling weight factor and the final adjusted sample weighting factor from the 1999 LEMAS survey are included.

Hide