Methods: The sample was 704 children in 2nd-5thgrades in 74 afterschool programs. To measure the setting-level quality of afterschool programs, several observational instruments were used: Arnett’s (1989) Caregiver Interaction Scale (CIS), Vandell et al.’s (2004) Promising Practices Rating Scale (PPRS) and the Academic Climate Assessment (ACA). For our dependent variables, we used 3 measures of child-reported behavioral outcomes: problem behavior and prosocial behavior (Strengths and Difficulties Questionnaire, SDQ, Goodman et al., 1994) and a measure of substance use (Russo et al., 1993). We employed hierarchical linear modeling (HLM) to take into account the nested design of the data. For each of three behavioral outcome measures, we first estimated an unconditional model with no predictors and then added individual-level covariates, including gender, race, and age, and setting-level measures of afterschool quality.
Results and Conclusion: The results of preliminary analyses indicate that some of our measures of afterschool quality significantly affect the level of problem behaviors. Specifically, the CIS-Sensitivity/Detachment measure was significantly, negatively associated with the SDQ-PB. This suggests that children show less problem behaviors in afterschool programs where staff have more positive interactions with children. The ACA, the overall measure of the after-school site’s behavioral management strategies, was also significantly related to decreased problem behavior (SDQ-PB) and substance use. These results together suggest that quality and sensitivity of afterschool programming matters for children’s behavioral development. More complete findings will be presented and discussed during the meeting presentation.