The proposed project will increase our understanding of the scope of neglect among children of color. It promises to determine whether racial disproportionality can be explained by any one or more forms of bias that may arise when cases are processed through the child protective system. The central hypothesis to be tested is whether the observed racial disproportionality in reported and substantiated child neglect can be explained by (a) reporting bias, (b) substantiation laws, (c) labeling bias, (d) exposure bias, and/or (e) aggregation bias. The major contribution that the research makes to the body of knowledge is the application of a methodology called a decomposition or residual difference method that permits the assessment of the relative sizes of each of these possible biases. The methodology can be applied to data collected from other states or counties. The research project will provide other analysts with powerful tools for looking at racial inequality in other aspects of the health and welfare of children where the potential for reporting or similar biases is known to exist. Minnesota provides an excellent laboratory for the study of disproportionality; Minnesota leads the nation in racial disproportionality. If the investigators can understand racial disproportionality in Minnesota, it should be possible to understand it elsewhere. Through a unique collaboration with child protective services in three counties in Minnesota and with the Department of Human Services in Minnesota, the project will expand the body of knowledge on how large and sustained disparities in child neglect reports can arise. Four methods will be used to assess and measure potential bias in child neglect statistics in Minnesota. First, the investigators will examine statistically disaggregated statewide child neglect data collected through a new system now being implemented by the Minnesota Department of Human Services. Second, they will examine, within three counties with differing policies and approaches to delivery of services and collection of basic data on child neglect, a sample of files of neglected children. Third, the investigators will look prospectively at which cases are most likely to be screened out of the system by placing trained sentinels in the child protective services and strategically in the offices of mandated professionals. At each stage, they will specify and estimate underlying model equations and construct statistical tests measuring various forms of bias. By looking at bias from different data collection efforts and perspectives, the investigators will gain added validity in their results. The results of this research will help guide the direction that policy makers should shift in order to address the problem of persistent racial disparities in child neglect.
| Ards, Sheila D; Myers Jr, Samuel L; Ray, Patricia et al. (2012) Racialized Perceptions and Child Neglect. Child Youth Serv Rev 34:1480-1491 |
| Ards, Sheila D; Myers Jr, Samuel L; Chung, Chanjin et al. (2003) Decomposing Black-White differences in child maltreatment. Child Maltreat 8:112-21 |
