The research used a pretest-posttest design that was part of a quasi-experimental evaluation in communities that took part in the Eisenhower Foundation Anti-Crime Self Help Program. The data came from 841 participants in block watch or seminars in Baltimore, Washington, DC, New York City, Brooklyn, Philadelphia, Boston, Minneapolis, and Cleveland. Results revealed that those who participated in block watch programs experienced an increased sense of community crime prevention efficacy, and perceptions of informal social control and reduced fear of crime. These effects did not occur in participants in seminars and workshops. Results indicated that workshops and seminars should not be the sole type of crime prevention activity. In addition, the results of previous research reporting a lack of effect or a negative effect of crime prevention are due to the unit of analysis used, the type of participation studied, and self- selection bias. Tables, notes, and 37 references