The research design is based in one theoretical model of the ecological predictors of organizational integration and another model of the organizational predictors of official delinquency. Based on these models, the following hypotheses are proposed for the research: (1) the more community-based an organization in terms of budget source, source of volunteers and staff, and responsiveness to community needs, the greater its openness to police referrals and its willingness to coordinate with other local agencies; (2) the less bureaucratized a police department and the more it is staffed by locals the greater the likelihood of informal handling of juveniles through dismissals and referrals; (3) a police department with a specially trained youth police unit is expected to make more referrals to youth services than a police department without such a unit; and (4) communities characterized by a high degree of coordination of their youth service organizations and a police department simultaneously low in bureaucratization and high in youth police professionalization will have lower degrees of official delinquency than communities with the opposite characteristics. The research design provides for access to multiple communities so that the merits of different service delivery systems can be comparatively assessed as well as access to measures of youth needs within each community, so that the efficiency of different systems can be evaluated. The project selected four Michigan communities in an area currently characterized by high unemployment and increasing economic hardship. The project is expected to yield definitive guidelines about the design and implementation of community programs that will be of interest to a broad range of policymakers. Community characteristics are appended, and 17 references are provided.