should not be assumed to work at another (Compas, 1993). Communities need to think about the full set of programs they provide and how adequately they supply opportunities for all their youth. Communities need a menu or portfolio of programs that allows adolescents and their families to select out the experiences most needed or desired.
The importance of the community as a unit of analysis for youth development is becoming increasingly popular. For example, this perspective is being encouraged by the research of the Search Institute in Minneapolis, which shows, first, that communities differ substantially in the number of psychological and social supports they have for youth and, second, that youth are healthier in communities that have more of these supports. This relation goes beyond the predictive power of the number of assets experienced by an individual youth: when a community is rich in supports, even at-risk youth seem to be doing better than in communities without such supports and opportunities (Benson, 1997; Blyth and Leffert, 1995). This perspective is also central to the work by Hawkins and Catalano (Hawkins, Catalano, and Associates, 1992; Development Research and Programs, Inc., 2000) on Communities that Care and by Public/Private Ventures on community-wide initiatives for youth (Gambone, 1997). Both of these programs are discussed in more detail in Chapter 5. Finally, the community perspective is gaining prominence in prevention work, in which more and more efforts are being made to design and evaluate community-wide interventions (see work by Biglan and colleagues as one such example [Biglan et al., 2000]).
What has not been researched systematically is how community programs fit into this picture of community supports and whether they can actually cause changes in the prevalence of both positive and risky youth outcomes. The associations undercovered by Benson and his colleagues at the Search Institute could result from characteristics of the communities that were unmeasured and influenced both the communities’ ability to generate and support multiple supports for their youth and the resilience of the youth themselves. Community-wide experiments are just beginning to be reported. For the most part, these experiments do not vary the number of supports but rather the commitment of the community to change and the provision of financial and person-power resources for a change initiative. In addition, there is not yet a model for how a community can assess whether its set of community programs provides good coverage for all youth, or whether there are major gaps. In general, we know that the availability of youth development focused institutions and programs vary with neighborhood socioeconomic status. Fewer