The driving force behind establishing the community policing movement in the United States was to improve the relationship between the police and the public. However, since this movement, there has been little effort to link research on confidence in the police with literature on community policing. This study attempts to provide a theoretical basis for the measurement of police performance. It provided police practitioners with information on the sources of confidence in the police, with the potential of using it to measure officers’ effectiveness in community policing. Data were obtained from two evaluation studies in a mid-size northwestern city in 1997. Results indicated that among all significant predictors of confidence in the police, the effect of volunteers was the strongest one and the number of contacts with the police increased confidence in the police. In addition, the effect of informal collective security was the second strongest predictor of confidence in the police, and victimization experiences and traffic tickets tended to reduce confidence in the police, while voluntary contacts with the police increased confidence in the police. In summary, the results indicated that various community policing programs tended to increase confidence in the police. Appendix and references