Association for Public Policy Analysis & Management

Recent study of the issues associated with differences in academic achievement consider outcome disparities to be a byproduct of poverty and racial segmentation, emphasizing the prevalence of those differences in the availability of resources needed for academic success. Observed gaps in availability of resources designed to address achievement disparities are also found to be both a significant predictor of academic outcomes, and supportive of the long-held view that addressing the observed resource disparities could be a key to reducing and eliminating differences in academic outcomes known to negatively impact those most likely to suffer as a result of these discrepancies beyond academic life—marginalized children and families in American society. Research exploring the links between social and economic characteristics of students and academic outcomes abounds, with interventions in the form of After School Programs (ASPs), among other approaches, being designed around the intense study thereof. While it has been shown that where students live relative to ASPs influences rates of participation as well as academic outcomes, existing research provides limited examinations of the impact that spatial differences has on these despite the knowledge that social and economic factors have been argued as being determinants of where members of particular social and economic groups are likely to live. The research being presented here examines the relationships between social and economic characteristics of at-risk youth and participation in after school programming, as well as with proximal outcomes known to influence long-term academic trajectory using both traditional regression modeling approach and geographically weighted regression modeling. The purpose, then is to understand how those relationships may be influenced by estimator bias that is, in reality, inherent in relationships between people and systems that are affected by differences in geospatial placement. The implications of the work are developing more robust understandings of these relationships and, if applied in the context of urban education policy analysis, improvements in the data and research that drives decision-making and resource allocation.