The intervention involved organizing residents of selected neighborhoods through door-to-door canvassing, block meetings, neighborhood meetings, and related strategies, with emphasis on the block watch model. A quasi-experimental research design was used to evaluate the programs. Measurement was done before and after program implementation in the experimental neighborhoods, and identical measurement occurred simultaneously in control comparison neighborhoods. Extensive telephone surveys were conducted with residents in panel samples and independent samples. Although residents' crime-prevention awareness and participation increased in the experimental neighborhoods, there was no significant differential change over time between the experimental and control areas. In fact, the three neighborhoods with the strongest evidence of program implementation showed significant increases in a number of problem areas, such as fear of crime, perceptions of the crime problem, vicarious victimization, and concern about the neighborhood's future. Possible explanations for these findings are explored. 60 tables, 3 footnotes, 32 references, survey instrument. (Author abstract modified)