Date of Conferral
2020
Degree
Ph.D.
School
Psychology
Advisor
Sandra Caramela-Miller
Abstract
U.S. law enforcement agencies are facing a legitimacy crisis. Incidents of police misconduct are the subject of widespread media coverage. Officer conduct continues to be a problem despite effectiveness of candidate screening. Underlying causes of ethical drift must be understood to reduce police misconduct. The purpose of this nonexperimental quantitative study was to examine the relationship between police ethical drift and agency size, officer age, officer gender, and officer education level. Ethical drift was the conceptual framework. Archival secondary data from local law enforcement agencies and the Florida Department of Law Enforcement Criminal Justice Standards and Training Commission were obtained via public records. Personnel records for 143 law enforcement officers were analyzed for information regarding officer age, gender, and education and number of officers employed. A multiple linear regression machine learning algorithm was developed and applied. A post hoc analysis involving multinomial logistic regression resulted in a moderately predictive model for ethical drift as a function of agency size. Law enforcement agency leadership may apply the results to identify officers at risk for ethical drift. Findings may also be used to promote positive social change through stronger police relations with communities and improved police legitimacy
Recommended Citation
Mann, Ryan, "Use of Machine Learning to Predict Ethical Drift in Law Enforcement" (2020). Walden Dissertations and Doctoral Studies. 9598.
https://scholarworks.waldenu.edu/dissertations/9598
