The Social Contagion on Police Misconduct

The discourse around police misconduct often pivots on whether such behaviors are driven by a few “bad apples” or larger systematic or institutional problems. The truth, as it is in so many instances, is likely somewhere in the middle. Whether criminal offending or corporate malfeasance, deviant behavior is a learned behavior and is influenced by the structure and content of one’s social networks. Police misconduct is no exception and is most likely learned through informal relationships and interactions between officers. Indeed, on the first day “on the job,” police officers often hear some iteration of the idea: “Forget what they taught you in the police academy, I’m going to show you how real policing works.”

The Social Contagion of Police Misconduct project emerged from a collaboration between Professors Andrew Papachristos (Yale) and Daria Roithmayr (USC), members of “Deconstructing Ferguson” working group co-sponsored by The Justice Collaboratory and the ISPS Center for the Study of Inequality, with objective of applying developments in network science to the study of police misconduct. Led by Papachristos, the Collaboratory team has begun a pilot study examining the structure and composition of social networks among more than 3,000 police officers in the Chicago Police Department. Exploratory data analyses are looking for structural differences in the networks of officers involved in misconduct and those who are not, testing such differences against both real and simulated network conditions and constraints. The overarching hypothesis of this project is that the tendency towards specific behaviors (e.g. the use of deadly force or, conversely, behaviors that lead to positive interactions with individuals) may be transmitted between officers in a way that is statistically predictable. Understanding how networks might influence misconduct might also provide a way to proactively identify officers who will be associated with more serious allegations, including shootings, and design interventions and prevention efforts accordingly.

Andrew Papachristos
Daria Roithmayr
Joscha Legewie

Research Team
Jennifer Wu
William Cai
Claire Ewing-Nelson
Isabel Cruz
Keniel Yao