PurposeThe purpose of this study is to explore factors affecting police performance, such as violent crime, fear of crime and satisfaction with police, using objectively measured policing and neighborhood characteristics.Design/methodology/approachThis study combined four data sources, including police statistics, a crime victimization survey, local government statistics and a public service satisfaction survey. Negative binomial regression and ordinary least squares models were estimated using data from 135 city governments.FindingsViolent crime was negatively associated with arrest, while fear of crime was negatively associated with patrol and positively correlated with disorder and public places. In addition, satisfaction with police had a positive association with level of patrolling and disorder policing, whereas residents receiving public assistance, ethnic heterogeneity and entertainment establishments were negatively associated with it.Research limitations/implicationsThis paper identified the different factors that impact each type of police performance, indicating its unique characteristics. Further research should be conducted to analyze data from non-urban areas to improve external validity.Practical implicationsThis study suggests that police managers need to choose the means of policing carefully depending on which performance indicators they are focusing on improving in the region.Originality/valueFew studies to date have examined the correlation between objectively measured policing and neighborhood context factors and police performance at the macro level.