Model analyses of factors affecting the neighboring activities of youth who reside in a revitalizing community

This study examined the attitudes and behavior of 119 youth living in a small community located adjacent to a major midwestern university which began a community revitalization project in 1995. Survey data were collected in 1999. A model of prosocial community behavior was created based on research literature from sociology, psychology, and social work on community change, as well as literature on adult and adolescent community attitudes and behavior. The analysis examined a model of four latent variables (residential stability, community attitudes, social integration, and neighboring activities). Residential stability was hypothesized to have a direct negative effect on community attitudes and a direct positive effect social integration. Community attitudes and social integration were hypothesized to directly affect neighboring activities and mediate the effect of residential stability on neighboring activities.; A specified model with adequate fit indices resulted from structural equation model analyses. The latent constructs "community attitudes" and "neighboring activities" were supported. Paths between length of time in the community and community attitudes, length of time and social network, community attitudes and neighboring activities, and social network and neighboring activities were significant (p ≤ .05). Those who resided in the community for longer periods had less favorable community attitudes and a more extensive social network that included university affiliates. Persons with more favorable attitudes about the community and a social network that included more university affiliates participated more often in neighboring activities. The model created by these paths resulted in adequate goodness of fit indices but explained only 17% (R square standardized solution) of the variance in neighboring activities indicating that the model could benefit from further development, specifically in regard to a possible feedback loop between attitudes and neighboring activities. Further research is also needed to test the model with gender, minority status, and socioeconomic status. These analyses were not possible due to the small sample size.