ABSTRACT
The study had two main purposes. First, to examine the psychometric properties of a measure of perceived neigh borhood context at two points in time (i.e., age three and age five). Second, to test multilevel models of linear relationships among neighborhood, family, and child development vari ables. Specifically, longitudinal (i .e . , age three to age five) and cross-sectional (i.e., age five) models were tested to examine the relationships between structural characteristics of the neighborhood (i .e., poverty rate, unemployment, percentage of female-headed households, residential mobility, and vacant housing) and perceived characteristics of the neighborhood (i .e. , social embeddedness in neighborhood networks, sense of community, satisfaction with neighborhood, and perceived crime) as they relate to parenting and development of African American children. The variables studied are depicted in Table 1 .
