This article attempts to identify determinants of one form of reactivity in police observational data, observer effects or when observers help police officers.
Systematic social observation (SSO) data is a key component in efforts to evaluate the “on the street” impact of policy initiatives, such as community policing or mandatory arrest policies for domestic violence. SSO police data provides a unique perspective into day-to-day police practice that goes beyond official records and police reports. Observers’ characteristics were identified as far more important in explaining whether observers became involved in encounters with suspects when compared to situational factors. Results indicated that observers with higher academic rank, lower grade point averages, and more conservative attitudes toward criminality were less likely to get involved in police work during encounters with suspects. Much knowledge about police behavior “in the field” is based on information collected by researchers who observed police work. In relation to police behavior, one relatively unexplored issue is how researchers can potentially bias observational data and undermine its reliability and validity since they are part of the context of observed behavior. The primary objective of this article is to identify determinants of one form of reactivity-observer effects (or when observers help police officers). Accounts of observer effects in the filed research literature are reviewed. The article examines the determinants of observer effects to find out when it occurs and who is most likely to trigger this form of reactivity. In addition, it examines the determinants of observer effects during non-traffic encounters with suspects using data from a large-scale observational study of police. Tables, figure, references