Abstract: Environmental Toxic Release and School Suspension: A Longitudinal Multilevel Analysis (Society for Social Work and Research 23rd Annual Conference - Ending Gender Based, Family and Community Violence)

Environmental Toxic Release and School Suspension: A Longitudinal Multilevel Analysis

Schedule:
Friday, January 18, 2019: 4:30 PM
Union Square 1 Tower 3, 4th Floor (Hilton San Francisco)
* noted as presenting author
Mi-Youn Yang, PhD, Assistant Professor, Louisiana State University at Baton Rouge, Baton Rouge, LA
Aimee Moles, PhD, GIS Analyst, Louisiana State University at Baton Rouge
Zibei Chen, MSW, Graduate Research Assistant, Louisiana State University at Baton Rouge, Baton Rouge, LA
Younghee Lim, PhD, Associate Professor, University of Mississippi, University, MS
Judith L. F. Rhodes, PhD, Associate Professor of Research, Louisiana State University at Baton Rouge
Background and Purpose

While the burgeoning body of school outcome research explored the relationship between environmental toxicity and children’ cognitive outcomes and schools’ academic performance scores, no existing studies have investigated how environmental toxin exposure affects children’s school suspensions. The purpose of this study is to determine if school proximity to Toxics Release Inventory (TRI) sites is associated with an increased likelihood of student suspension; and, if increases in the number of TRI sites within several defined proximities are associated with an increased likelihood of student suspension.  

Method

Kindergarten to 4th graders (N=7,154) from 412 Louisiana public schools in the observation school year (2007-2008) comprised the sample. By matching student and state-level education data, suspension was measured dichotomously (1= suspended once or more, 0= not suspended) in the following school year. TRI sites, obtained from the Environmental Protection Agency, and schools were geocoded using ArcGIS software; buffers at 1, 2, and 3 mile distances were drawn around schools; the number of TRI sites within each buffer were counted. Individual-level factors (e.g., students’ behavioral problems and parental school involvement) were measured by teachers' reports on validated tools during the observation year. Community-level factors (e.g., families in poverty, less than high school degree, and urban/rural school status) were measured at census tract level. This study used multivariate multilevel logistic regression to adjust for the nesting effect of schools on students’ suspension outcome.

Results

After controlling for individual- and community-level variables, elementary students who attended schools located within 1 mile radius of TRI sites were 29% more likely to be suspended (p<.01) than students who did not attend schools located within 1 mile radius of TRI sites. The analysis of the effect of distance showed that attending a school located closer to the TRI sites increased the odds of being suspended: within 1 mile radius (OR: 1.29, p<.01); within 2 mile radius (OR : 1.24, p< .05); and within 3 mile radius (OR: 1.19, not significant). In addition, the results found that increases in the number of TRI sites are associated with an increased likelihood of student suspension: one additional TRI site within the 1 mile radius increased the odds of being suspended by 11% (p<.05).

Conclusions and Implications

The results of this study indicate that environmental factors are associated with an increased risk of children’s school suspension. Moreover, closer and more abundant TRI sites relative to a school resulted in greater propensity for elementary students to be suspended from school. Implications include school system decisions on school locations. Social work and social justice initiatives should include toxic release locations when advocating for children and families. In the same way that lead-based paint negatively impacted children’s brain development and was found to adversely impact behavior, other environmental hazards must be considered when planning school location. Additional research beyond individual and family characteristics on children is needed, and integrating environmental factors with other community-level factors can better uncover mechanisms placing children at risk.