School Climate and Discriminatory School Bullying, Behavioral School Victimization and Weapon Involvement on School Grounds: An Individual and School Level Analysis

Introduction: Research is limited regarding students experiencing homelessness in the US, specifically while considering their school experience. Additionally, research is scarce while considering the different subgroups of homelessness (homeless on the street/sheltered homeless/ living with a friend/relative/ living with another family) and differences in their adverse outcomes. This study utilizes a holistic approach to understand the role of school-climate with relation to school-violence, for homeless students, while using representative school-based sample from California.

Methods: Two samples from the ongoing large-scale California Healthy Kids Survey, administered biennially statewide (2011-2013), were utilized. The study includes an individual level sample of 9th and 11th grade students (total N= 389,569), and a school level sample (n=811). Bivariate analysis were used to examine differences between the different homelessness subgroups as compared to nonhomeless students (nonsheltered homeless students (2,822), sheltered homeless students (2,179), students living with relative/friend (11,917), students living with other family (10,333) and non-homeless students (358,001)). Perceived school-climate was assessed by items of several dimensions: positive relationships, school connectedness, meaningful participation and safety. The responses ranged either on a four or on a five Likert scale. To assess school victimization students were asked to report the rate of discriminatory bullying, behavioral victimization and weapon use on school grounds during the past 12 months. Three composite scales were created, then recoded into two categories (yes/no). Control variables included gender, ethnicity, gang membership, migration situation and grade. Hierarchical logistic regressions were conducted to examine the relationship between school-climate and the violence outcomes on the individual level and hierarchical linear regressions were conducted at the school level.

Results: Findings show that adding the school-climate dimensions to the demographic variables contributes significantly to each outcome. At the student level, the partial coefficients show that safety, positive relationship and connectedness were all significantly negatively associated with discriminatory bullying, with safety being the highest contributor with the lowest odds ratio (p=0.000, OR=0.84, 95% C.I. =0.83-0.85). Safety, positive relationships and connectedness were also found to be significantly negatively associated with behavioral victimization, with safety being the highest contributor with the lowest odds ratio (p=0.000, OR=0.85, 95% C.I. = 0.84-0.86). Safety, positive relationship and connectedness were all significantly negatively associated with gun involvement, with safety being the highest contributor with the lowest odds ratio (p=0.000, OR=0.77, 95% C.I. = 0.75-0.78). At the school level all models were significant (p=0.000). The partial coefficient of school-climate is negative and significantly (p=0.000) decreases the discriminatory bullying, behavioral discrimination and the gun involvement.

Discussion: The results highlight that positive school-climate serves as a protective factor for school-attending homeless youths. The study shows that homeless students, in each of the subgroups, are at high risk of adverse school related violence behaviors compared with their nonhomeless peers. The results from the school level analysis indicate as well, that when the better the school-climate is, the lower the school-violence. Homeless students deserve attention, prevention and intervention in school. Specifically, while considering school-climate, enhancing interventions related to improving school-climate at the school level, would benefit students experiencing homelessness.