A machine learning approach to police recruitment: Exploring the predictive value of social identity measurement instruments

Abstract

Existing research on police recruitment is eclectic, with examples of multiple methodologies in multiple police-related settings. These methods often resemble psychological measurement of individual traits yet neglect the potential recruits’ social resources or network-based influence. More recent research has utilised social identity and social network theory to understand the route to a police candidate’s eventual recruitment, but this is underdeveloped. This literature indicates that further research utilising social identity theory could assist with understanding what was before for police recruits and whether that matters. This study explores the use of random forest machine learning to analyse one partial and two full self-report social identity measurement instruments completed by 886 police recruit applicants. It aimed to explore whether the results of these instruments completed by potential police recruits were predictive of their success in the recruitment process. The results reveal that the combined use of these validated social identity instruments offers a reliable predictive base for successful and unsuccessful applicants, with an overall accuracy rate of 86% across the model’s performance metrics. The implications from this study highlight the significance of perceived social identity in the context of police recruitment, emphasising the potential impact of using its measurement to gain improved understanding of candidate selection. Social identity measurement instruments could be incorporated into recruitment processes, allowing police departments to enhance their ability to identify individuals who are more likely to succeed at an earlier stage via machine learning. Practically, this could reduce the need for multiple, expensive recruitment stages. Theoretically, it illustrates that a police recruit’s social identity is of importance to whether a candidate is successful or not, presenting police forces with both challenges and opportunities.