355P
Examining County Variation in Racial Disproportionality Using GIS Mapping of National Child Maltreatment Data
A large body of research has demonstrated higher rates of child maltreatment for black children compared to white children. However, studies are often limited by aggregation at the state or national level and do not focus on specific child age groups. Recent research using national data finds that rates of family poverty parallel rates of child maltreatment, and thus risk associated with poverty is more strongly implicated in child maltreatment disproportionality than reporter bias. The current study examines the geographic variation in disproportionality at the county level, focusing on outcomes for children birth to three (the age group most likely to be maltreated) by answering the following research questions: (1) Are there differences between counties nationally and within the same state in child maltreatment racial disproportionality? (2) Is county-level maltreatment disproportionality associated with poverty disproportionality?, (3) To what extent are differences within states driven by disproportionality that exists in large urban counties? (3) Are there geographic trends that may explain variation in disproportionality within states?
Methods:
Data from the US Census and National Child Abuse and Neglect Data System (NCANDS) were used to compare rates of county racial disproportionality for poverty and child maltreatment for children age 0-3. Using ArcGIS 10.1, geographic information system (GIS) maps were created to provide visuals that convey the interactions of the racial disparity of poverty with child maltreatment report rates per county. We first geocoded the poverty and child maltreatment data from each state into shapefiles and then coded for each race category or combination thereof to create each map.
Results:
In total, there were data from 610 counties across 44 states (counties with <1000 cases were unable to be identified in the NCANDS data). Findings indicate considerable variation in county rates of maltreatment for black, white, and Hispanic children within states. The ratio of these maltreatment rates, or the disproportionality ratios, also varied considerably between and within states.
In comparing the maps of poverty racial disproportionality and maltreatment disproportionality, the geographic distributions are very similar. Notably, there is a strong within-state concentration of high disproportionality ratios for black children around urban areas. For example the highest disproportionality ratios for both black and Hispanic children were Washington, DC and New York, New York. In most states, racial disproportionality peaked around large urban centers.
Conclusions and Implications:
The finding that county maltreatment disproportionality ratios maps follow the same pattern as poverty disproportionality ratios maps provides further evidence that disproportionality in child maltreatment reports is driven by family need rather than reporter bias. The finding that there is a concentration of high disproportionality ratios around urban areas suggests that state levels of maltreatment disproportionality may be driven by large urban centers within states. Further, there may be additional aspects of parenting in urban environments that widen the gap in risk for maltreatment between black and white families.