Students' Risks for out-of-School Suspensions: Indigenous Heritage and Child Welfare System Involvement

Background

Disproportionality in out-of-school suspensions (OSS) is a persistent social and racial justice issue. Many public schools use OSS as a standard response to students' misbehaviors even though it is largely ineffective in changing behavior, and is negatively associated with physical and emotional health (Sullivan et al., 2013). Not all students are equally likely to be suspended. Indigenous students are suspended nearly five times more often than white students (Gibson et al., 2019). Despite this overrepresentation in OSS, only limited studies have investigated OSS in Indigenous students. Indigenous children are also disproportionately represented in the child welfare system and in out-of-home placements (Bussey & Lucero, 2013). Given the history of the U.S. government forcing Indigenous children from their families and tribes into carceral boarding schools, the relationships between Indigenous communities and public child welfare and education systems continue to be fraught (Johnston-Goodstar & Roholt, 2017). In this study, we examine the relationship between students’ CPS involvement, ethnicity/indigeneity, and OSS experiences following a cohort of children from 3rd through 8th grades. Our research questions are:

1. What child- and school-level factors predict OSS?
2. Does a child’s ethnicity/indigeneity predict OSS when controlling for other factors?
3. Is there a relationship between the number of CPS investigations/assessments in which a child is involved and their OSS?
4. Does this relationship vary across groups?

Methods

Education data from a cohort of 60,025 students in 3rd grade enrolled in the 2007-2008 academic year were linked to Human Services data prior to 2008. Given the high level of variance in OSS and the high percentage of students who never experienced OSS between 2008 and 2014, we employed hierarchical zero-inflated negative binomial (ZINB) regression to identify factors predicting the number of OSS incidents over a six-year period by entering each domain of variables from the child-level (Model 1), school-level (Model 2), and the interaction term of ethnicity/indigeneity and the number of CPS investigations (Model 3).

Results

Indigenous students were overrepresented in both CPS and OSS. Approximately 11% of students had at least one OSS record over 6 years with an average number of OSS of 0.18. While 2% of the students were indigenous, nearly 7% of the students involved in OSS were Indigenous with an average number of OSS of 1.25. Factors predicting risk for OSS included being Black, Indigenous, and male, socioeconomic status, emotional/behavioral disabilities, and repeat maltreatment. Of students with similar child maltreatment histories, Indigenous students are more likely to be suspended from school than their white peers. This moderating effect was not present for other BIPOC children, including Black students.

Conclusion.

This study demonstrates the importance of reducing disparities in the child welfare system by addressing underlying systemic factors. Indigenous children may experience relatively high levels of both child maltreatment allegations and OSSs, not because Indigenous families are more likely to maltreat their children, or that Indigenous children are more likely to misbehave in school, but due to systematic racism within the U.S. settler colonial state.