Abstract

Excerpted From: Ian Ayres, Sonia Qin and Pranjal Drall, Racial and Gender Bias in Child Maltreatment Reporting Decisions: Results of a Randomized Vignette Experiment, 21 UC Law Journal of Race and Economic Justice 183 (June, 2024) (181 Footnotes) (Full Document)

The goal of the child welfare system is to promote the safety, well-being, and permanency of children, and prevent incidents of child maltreatment. In the United States, the child welfare system is comprised of a vast network of public and private agencies with annual spending of approximately $33 billion, and which employ tens of thousands of workers for the purposes of intake, screening, and investigating reports of child maltreatment.

In 2019, around 7.9 million children were screened following allegations of abuse or neglect, and 3.5 million children were the subject of a child maltreatment investigation; of those, over 650,000 children were determined to have been maltreated. Public child protection agencies provided follow-up services in the homes of an estimated 1.1 million children.

1,840 children died from abuse or neglect in 2019. Beyond this risk of death, developmental neuroscientific findings reveal that various forms of child maltreatment can have serious, long-term, and potentially devastating effects on the brain, the nervous system, and overall health. The economic consequences of child maltreatment are also substantial, involving greater use of child welfare services, higher health and mental health utilization rates, worse academic outcomes and work productivity, increased risk of violence for both the perpetrator and the victim, and reduced quality of life and life expectancy.

Once a report of child abuse or neglect is made to authorities, child protective services (“CPS”) must investigate the situation to determine whether the report is substantiated. In substantiated cases where a child is found to have been abused or neglected, the state's response depends on multiple factors, including: the severity of the maltreatment, an assessment of the child's immediate safety, the risk of continued maltreatment, the services available to address the family's needs, and whether the child has already been removed from the home. Possible actions in substantiated cases range from closing the case with just a warning that substantiated maltreatment was found, to removal and placement of the child in foster care. In cases considered low risk, children may remain at home with their family, while their family receives support and services such as parenting education, counseling, and therapy. Parents typically have little choice in the types of services they are required to engage in to have their children returned. For high-risk cases, children will be removed from their home and placed either with relatives or foster families, with parents allowed only supervised visits.

Recent federal data suggests that more than 670,000 children were served by the foster care system in 2019, with 251,000 children entering foster care that year alone. This translates to the separation of hundreds of thousands of families. Youth in foster care who are neither reunited nor placed with a new permanent family are typically emancipated at their state's legal age of majority; this is referred to as “aging out” of foster care. More than half of the 249,000 children who formally left foster care in 2019 returned to their parents or went to live informally with a relative, while 37% left through adoption or guardianship, and 8% aged out of care.

The consequences of being entangled in the child welfare system as a potential abuser are significant and often destructive. In substantiated cases, regardless of the final outcome of the case, the name of the person accused of committing the abuse or neglect will be recorded in a statewide central registry that maintains child maltreatment investigation records. In some states, all investigated reports of abuse and neglect are kept in the central registry. These registries are used in background checks for certain professions; for example, information from substantiated reports is made available to employers in the child care, education, and health care industries. Information from central registries is also used in background checks for foster or adoptive parents. The length of time the information is held and the conditions for expunction vary by state.

The entry point to the child welfare system is the call that is made to authorities to report a suspected incident of abuse or neglect. It is when someone--a relative, neighbor, teacher, doctor, or even a mere stranger-- chooses to call authorities and report a parent or caregiver that the door to the child welfare system is opened. Certain individuals, by virtue of their professions, are categorized as “mandatory” or “mandated” reporters, meaning that they are mandated by law to report suspected maltreatment. While each state has slightly different designations of professionals who are mandated to report, common categories of mandatory reporters are health care providers, people who interact with minors in a school or daycare setting, law enforcement officials, and members of the clergy. Some states designate all individuals, regardless of profession, as mandatory reporters. Those who are not mandatory reporters are considered “permissive reporters,” meaning they are not required by law to report, but are nonetheless encouraged to report any suspected child abuse and neglect.

This paper is an empirical study of the implicit biases present in the reporting of child maltreatment. More specifically, this study surveyed 4,000 nationally representative respondents to test for mandatory and permissive reporters' gender and racialbiases when they encounter an interaction between a parent and child that may appear to involve child maltreatment. We used short vignettes taken from actual events to illustrate parent-child interactions and asked participants to rate how likely they would be to report each interaction as child maltreatment. In each vignette, we randomly varied the gender and race of the parent and child using gendered white-, Black-, and Hispanic-sounding names. In each vignette, the race-valence of names randomly assigned to the parent and child were always the same, but the gender-valences for names randomly assigned to the parent and child in a vignette could be the same or different.

We find strong statistical evidence that different naming cues caused respondents to change their reporting likelihood. Respondents were less likely to report potential child maltreatment when the vignette used nonwhite names to describe the family participants. Respondents were less likely to report when a male child was involved, and more likely to report when a male parent was involved. Women respondents were more likely than men respondents to discriminate on the gender of the parent in deciding their reporting likelihood. The uncovered racial and gender biases were more pronounced in vignettes that were of intermediate severity.

The body of this Article is divided into five Parts. Part I discusses the legal definition of child maltreatment and explains the mechanics of reporting child maltreatment, introducing the role of mandatory reporters. Part II reviews the role of race and gender in the history of the child welfare system in the United States. Part III describes the design of this study and presents alternative hypotheses with regard to race and gender discrimination. Part IV presents the core results--highlighting the racial and gender determinants of child maltreatment reporting. Part V discusses normative implications of the findings and presents policy proposals in response to the findings.

[. . .]

In 2019, more than 250,000 children were separated from their families and placed into foster care, and more than 420,000 children were in the system in total. Around 44% of the children in foster care are Black or Hispanic. If the goal of the child welfare system is truly to protect the welfare of children, then we must more closely examine the dynamics at play at the point of a family's entry into the system. If the reporting of child maltreatment is plagued by racial and gender bias, then that translates to racial and gender disproportionality in the makeup of families in the child welfare system, subsequently leading to disproportionality in family separation.

This study relied on vignettes with random assignment of name cues to test for racial and gender bias in child maltreatment reporting. Our results show that mandatory reporters are more likely to report instances of suspected child maltreatment. Moreover, people are more likely to report fathers as well as situations involving children who are girls. Non-white parents were less likely to be reported than white parents. Hispanics, women, non-Republicans, religious people, lower income individuals, and less-educated individuals are more likely to report. Discrimination was the strongest in vignettes with the most “moderate” depictions of potential maltreatment. In conclusion, this study proves that there are disparities in reporting decisions that are tied to implicit biases.

While our results provide strong causal evidence of discrimination, disparate treatment with regard to naming cues on an online survey may not translate into racial or gender discrimination with regard to real-world reporting decisions. Even if our results were indicative of analogous real-world bias, our findings do not tell us whether we should reduce bias by increasing the reporting of the low-likelihood group or by decreasing the reporting of the high-likelihood group. For example, if the public is less likely, ceteris paribus, to report female than male adults, should this bias be remedied by lower the reporting rates of males or by increasing the reporting rates of females (or some combination of both). Vignette studies like ours cannot answer important policy questions. Nonetheless, documenting several forms of reporting bias can play a role in spurring additional research and provide an input to policy deliberation, as designers of the child welfare system strive to shield children from trauma, without inadvertently introducing the new trauma of overly aggressive state intervention into the lives of the families and the very children it is designed to protect.

Oscar M. Ruebhausen Professor, Yale Law School.

Class of 2022 J.D. Graduate, Yale Law School.

Law Student, Yale Law School.