Disproportional Representation of Minorities in Foster Care – A Closer Look

Native American children account for about 15% of South Dakota’s population, but over 50% of the children in foster care.  This statistic alone is a bit shocking, but cannot tell the whole story.  There are two primary types of reasons that could explain the disparity.  One is that Native Americans are treated differently than other other groups.  The other is that the Native American population has more need for intervention.  Chances are that both explanations are partially correct.

There are a number of factors not inherently related to race that  have a strong correlation with child abuse and neglect.  The following numbers are from the Fourth National Incidence Study of Child Abuse and Neglect (NIS 4) report to Congress.

  • Children with an unemployed parent are about twice as likely to suffer harm from abuse or neglect.  (Page 5-3)
  • Children from families with a low socioeconomic status (defined as families with a household income below $15,000, a parent who had not completed high school, or who participated in a poverty related public assistance program) are about five times more likely to suffer harm from abuse or neglect.  (Page 5-12)
  • Children living with married biological parents are three times less likely to suffer harm from abuse or neglect than children living in any other situation, four times less likely to suffer harm than children living with a single parent, and eight times less likely to suffer harm from abuse or neglect than children living with a single parent who has a live-in partner.  (Page 5-20)

These predictive factors are all present in at higher levels in many minority populations, including South Dakota’s Native American population.  The Kaiser Family Foundation reports that about 11% of South Dakota’s White population is living in poverty, while about 57% of South Dakota’s Native population is living in poverty.  The Annie E. Casey Foundation reports that, nationwide, about 53% of Native American children are being raised by single parents as compared to about 24% of White children.

Given the numbers above, there should be a disproportional number of Native Americans in foster care in South Dakota.  It would actually be indicative of a problem if there was not disproportional representation.  Unfortunately, the first three National Incidence Studies (NIS) have been incorrectly interpreted many times as showing that there is no difference in rates of child maltreatment between racial demographic groups.  In reality, those studies simply failed to show a racial difference that was statistically significant.  For those into statistics, the racial differences found in the NIS studies were determined not to be statistically significant if the 95% confidence intervals for the maltreatment rates among different racial groups overlapped.  So, although the NIS 2 and 3 suggested that rates of child maltreatment do vary by race, the NIS 4 was the first to show that the difference was statistically significant.  This is due primarily to the much larger sample size of the NIS 4, which provided for smaller confidence intervals.  Those interested in an analysis of these NIS findings may wish to read the Supplementary Analyses of Race Differences in Child Maltreatment Rates in the NIS–4.  I also strongly recommend the 2011 article by Brett Drake and Melissa Jonson-Reid entitled NIS interpretations: Race and the National Incidence Studies of Child Abuse and Neglect.  At the time of this writing, it was possible to find the full version of that article with a Google Search.

The facts above do not mean that there are not inappropriate race-based factors that impact the child welfare system.  Conscious and subconscious biases could have a role in how cases are reported to and handled.  Most of the research I have been able to find on this topic has been in the form of summaries in other publications.  One such source is a 2006 Synthesis of Research on Disproportionality in Child Welfare published by Casey Family Services.  Unfortunately, that publication perpetuates the falsehood that “hard evidence” shows that “parents of color are no more likely than white parents to abuse or neglect their children.”  (p. 1)  This statement is presumably drawn from the faulty analysis of the NIS 2 & 3 that were discussed above.  But, it lists a number of other studies as well that are relevant to whether bias is a factor in disproportional representation.  For example:

In conclusion, the simple fact that minorities are disproportionally represented in the child welfare system is not conclusive evidence of race-based discrimination.  There are in fact very good reasons why there should be disproportional representation, namely the correlation between minority groups and other risk factors.  However, it is probable that bias, likely at the subconscious level, is at least partially to blame for disproportional representation.  We should continue to seek to address racial bias in our child welfare system, but must be careful with the evidence we use when doing so.