The purpose of this study was to assess validity and reliability by comparing agency-level data reported in the 2003 and 2007 waves of the Law Enforcement Management and Administrative Statistics (LEMAS) surveys with available external sources.
The study was divided in four phases. The first phase consisted of the merging of the 2003 and 2007 Law Enforcement Management and Administrative Statistics (LEMAS) survey data (collected in 2002 and 2006, respectively) for large municipal police departments and conducting basic descriptive analyses of those data.
The second phase of the study aimed to assess the reliability of the complaints data by conducting a within-agency comparison of the two years of the LEMAS survey data. A distributional rationale was used for identifying relatively large changes from 2002 to 2006 in both the volume of complaints and sustained complaints and rates of such complaints (per 100 officers). Where large changes were observed, the researchers used publically available reports, any publically accessible records and/or data, and direct contacts with agency personnel to explain these changes.
The third phase of the study aimed to assess the validity of the complaints data using a simple random subsample (n=100) of agencies. Researchers used the 2003 and 2007 LEMAS survey data and publically available reports and records and/or data and direct contact with agency personnel to ensure the survey data corresponded with the external criterion within an acceptable range.
The fourth phase of the study intended to replicate an analysis of administrative, organizational, and environmental correlations of complaints about police use of force. The results from the second and third phase led researchers to conclude that the fourth phase was unnecessary.
The dataset contains data from the 2003 and 2007 Law Enforcement Management and Administrative Statistics (LEMAS) surveys. The agencies selected for validation and reliability analysis (n=100) were selected at random and if they were present in both waves of the LEMAS surveys.
Large municipal police departments in the United States.
administrative records data
The dataset (n=496) contains 24 variables.
Researchers used two variables (PERF_ID and SURVEYID) in order to link this dataset to the 2003 (ICPSR 4411) and 2007 (ICPSR 31161) Law Enforcement Management and Administrative Statistics (LEMAS) survey data on large police municipalities.
The dataset contains variables related to the number of formal citizen complaints received during 2002 regarding police use of force. This includes total number of complaints as well as unfounded, exonerated, not sustained, sustained, pending and other disposition complaints.
The dataset contains variables related to the number of formal citizen complaints received during 2006 including total complaints as well as pending, sustained, and other disposition complaints. Notably missing are the exonerated and not sustained complaints. These were not asked in the 2007 LEMAS survey.
Researchers also included variables that identified the name of the police agency, identified whether or not the agency was in both the 2003 and 2007 LEMAS survey data, whether or not the agency was selected at random for validation, and how an agency, if selected at random for validation, responded to researcher contact.
Finally, researchers included variables that served as verification and validation of the total use of force complaints in both 2002 and 2006 and the sustained use of force complaints in both 2002 and 2006.