Qualitative content analysis was used to assess the presence of racial bias in completed family assessments (n = 49) in an urban county child welfare agency. Objectivity was assessed using five mutually exclusive categories: factual/objective, strengths-based, negative bias/undertone, report/hearsay, or generic. Sentence-level coding found negativity or bias present in 92% of the assessments. Section-level coding illuminated racial and ethnic differences related to six of the 13 sections including family perception, family strengths, culture/religion/ethnicity, medical/mental health, substance use, and legal history. These findings support the need for ongoing education and training to illuminate and address individual bias within the child welfare workforce.
For decades, academics and government agencies alike have documented the disproportionate racial representation and disparities within the child welfare system. The National Incidence Study of 2010 (NIS-4) (Sedlak et al., 2010) found that children who were identified as Black or African American experienced higher rates of maltreatment compared to children who were identified as White or Latinx. Publicly available data illustrates that the problem continues to persist (Statista Research Department, 2021). Bias in the child welfare system has been found to exist at all levels, including reporting, data collection, and interpretation of the reported information--i.e., understanding the context surrounding the circumstances that lead to a family being reported to the child welfare system (Dettlaff, 2011). Disentangling the impacts of socioeconomic circumstances such as exposure to poverty and decreased access to critical services from worker bias on these disparities has been a long-debated issue among child welfare researchers and practitioners. This study adds to the discourse and supports the need for ongoing attention to worker bias to improve outcomes for Black and Brown children and families involved in the child welfare system.
Types of Bias
Implicit bias-"a preference (positive or negative) for a social category that operates outside of awareness" (Marsh, 2009, p. 17)-has been proven to be pervasive and impact decision-making across several social systems including child welfare (Benbenishty et al., 2015; Dehon et al., 2017; Levinson et al., 2017; Payne et al., 2017; Staats et al., 2016). More specifically, racial bias has been found to have significant negative impact on judgment toward, treatment of, and services for those who are Black, Brown, Indigenous, and other people of color (BIPOC) (Arnold et al., 2018; Capers et al., 2017; Nance, 2016; Obermeyer et al., 2019; Wilson et al., 2017).
Wells and colleagues (2009) identified two types of biases: cognitive (implicit) and motivational (explicit). Cognitive biases reflect those errors in the thought process that can be influenced or changed by the introduction of new information. Motivational biases are more difficult to change as they are a form of maintaining a specific way of thinking. Both biases influence decision-making processes at both the individual and systemic levels, either consciously (explicitly) or unconsciously (implicitly). In other words, how one feels (attitude) impacts how we make decisions (think) and how we operationalize those decisions through policy or action (behavior). Sources of bias may include past familial and personal experiences, social pressure, societal or cultural...