Purpose: The purpose of this paper is to investigate the impact of a police foot patrol considering micro-geographic units of analysis.

Design/methodology/approach: Six years of monthly crime counts for eight violent and property crime types are analyzed. Negative binomial and binary logistic regressions were used to evaluate the impact of the police foot patrol.

Findings: The impact of police foot patrol is in a small number of micro-geographic areas. Specifically, only 5 percent of the spatial units of analysis exhibit a statistically significant impact from the foot patrol.

Originality/value: These analyses show the importance of undertaking evaluations at the micro-scale in order to identify the impact of police patrol initiative because a small number of places are driving the overall result. Moreover, care must be taken with how small the units of analysis are because as the units of analysis become smaller and smaller, criminal events become rarer and, potentially, identifying statistically significant change becomes more difficult.