Worker Influence on Disparity in Foster Care Entries

Background and Purpose: Efforts to address disparity in the child welfare system must contend with the question of underlying cause – is disparity generated by structural bias, individual bias, or some combination of both? Prior research has found that structural ecological factors (poverty in particular) contribute to disparity in important but complex ways. In this study, building on previous research focused on structural factors, we look for evidence of the influence of the worker on the decision to place a child into foster care. To do this, we examine the relationship between race and the frequency of placement into foster care while controlling for factors at multiple levels (child, Child Protective services (CPS) case, CPS worker, and county). We ask whether, when faced with similar allegations for similarly situated children in similar ecological contexts, do workers make similar decisions about placing children?

Methods: Using foster care entry counts by worker as the dependent variable, we considered child-level characteristics including age and gender, worker characteristics including longevity and experience with the state agency, and race-specific county-level measures of child poverty and urbanicity. We examined whether measures of child-level characteristics, case-level characteristics, worker-level field experience, and race-specific county-level poverty and socio-ecological diversity were correlated with worker-level variation in Black/White foster care placement rate disparity. We applied Poisson regression models to explore how the context at different levels affects placement rate disparity.

Results: Consistent with previous disparity research, we found that Black children are more likely to enter foster care and that entry rates are higher in counties with higher poverty rates. This relationship is strongest for infants. We found that these relationships are moderated by worker experience. Disparity rates are higher for workers who are relatively new to child welfare compared to disparity rates for seasoned child welfare workers.

Conclusions and implications: Multiple factors contribute to racial disparity in foster care entries, and the relationships between those factors and disparity are complex. Our findings suggest that sweeping, one-size-fits-all generalizations that isolate a single contributing factor do little to push the search for solutions very far. Race-specific poverty levels certainly play an important role in producing the variation we observe in disparity at the local level. The present study suggests that the knowledge workers gain through experience working with children and families also plays a moderating role in the decisions they make in the context of the many other factors that shape their encounters with children and families. This finding points toward strategies involving efforts to strengthen supervision and support for new workers as promising interventions to mitigate disparity, but these must be considered alongside strategies that focus on addressing the socio-ecological factors—race-specific poverty and social disadvantage in particular—that are strongly correlated with geographic variation in racial disparity rates.