| Following a bioecological framework, this dissertation closely examines the prevalence, predictors, and associated outcomes of self care (unsupervised time outside of school). The dissertation utilizes the nationally representative Panel Study of Income Dynamics Child Development Supplement (N=2,570, ages 5--18) and the longitudinal NICHD Study of Early Child Care (N = 959, ages 8--10), two large-scale national datasets that offer detailed time-diary information as to child and adolescent activities and supervision. Both variable- and pattern-based approaches were used to create self-care types based on three contextual features (i.e., supervision, location, and the presence of peers during self care) in order to elucidate: (A) factors that predict the use of child and adolescent self care, and (B) the circumstances under which child and adolescent self care are associated with study outcomes. Overall results suggest that self care is (1) a common context for children and adolescents, (2) that the type of self care utilized differs according to several characteristics (e.g., amount of time in self care, child age, family income-to-needs, and neighborhood poverty), and (3) that associated outcomes of self care differ by self care contextual features. More specifically, the location of and peer association during self care are associated with several study outcomes, including higher externalizing behavior (adolescents but not for younger groups), lower well-being, and higher body mass index. Relations between self care and academic outcomes were not statistically significant. |
