Poster #88 - Student engagement and peer victimization among Brazilian children and adolescents

• In Event: 2-055 - Poster Session 06
  In Poster Session: PS 06 Section - Education, Schooling

Integrative Statement

Peer victimization and bullying have a number of negative effects on children, such as depression, anxiety, low self-esteem, and other psychosocial consequences. In educational contexts, these consequences may also reflected in maladjustment and lower achievement (Buhs, 2005). Previous studies that have demonstrating the negative relation between peer victimization and academic achievement also suggest that the association is mediated by other factors (Nakamoto & Schwartz, 2009). In this study, we investigate the association between peer victimization and aggression with student engagement, considering also the effect of gender, ethnicity and socioeconomic status (SES). Engagement here is understood based on cognitive, behavioral and emotional components (Appleton, Christenson & Furlong, 2008). The participants were 19,316 elementary and secondary students, aged between 7 and 21 years old, from five municipalities of the metropolitan area of Curitiba, in the South region of Brazil. In the current sample, 49.8% were boys, 48.2% girls and 2.0% reported themselves as a non-binary gender. Regarding ethnicity, 50.0% of the students were self-identified as white, 7.4% as black, 37.8% pardo (mixed races), 2.5% Asian and 2.4% indigenous. As part of a larger research conducted in Brazilian schools, a self-report questionnaire has been used to measure socio-demographic information, school climate, victimization, safety, and socioemotional development. Four items from a scale for peer victimization and aggression evaluation (Cunha, Weber & Steiner Neto, 2009) have been used to assess victimization (α4 = .79) and six for aggression (α6 = .72). To measure student engagement, 16 items comprising cognitive, behavioral and emotional dimensions have been used (α16 = .86). Parental educational level has been used as indicator of the SES. Gender was coded into dummy variables to understand the influences of being boy or being girl. For ethnicity, a dummy code was used to distinguish whites (1) and nonwhites (0). Table 1 shows the correlations among the study variables. All the variables were significantly correlated (p < 0.05), except ethnicity with engagement and SES with victimization. Table 2 presents the results of a multiple linear regression, using student engagement as the dependent variable. The predictors were peer victimization, aggression, gender, ethnicity and SES. The full model was a good fit to the data, F(6, 9337) = 149.62, p<0.001, R2 = 8.8%. The findings show that conforming to a binary gender norm (either as boy or as girl), and SES were significant positive predictors of engagement, suggesting that students with lower socioeconomic status and belonging to non-binary gender are more likely to be less engaged in school. An interesting finding in this study is that being white or nonwhite has not been observed as a significant predictor, thus it seems that student engagement have more relation with peer victimization than with the ethnicity itself. Indeed, both peer victimization and aggression were significant negative predictors for engagement, which indicates that being involved with aggressive situations, either as victim or as perpetrator, is linked to lower levels of engagement in school.

Authors

• Vitor Yano, Universidade Federal do Paraná
  Presenting Author

• HELLEN TSURUDA AMARAL, Universidade Federal do Paraná
  Non-Presenting Author

• Josafa Da Cunha, Universidade Federal do Paraná
  Non-Presenting Author