Longitudinal associations between features of toxic masculinity and bystander willingness to intervene in bullying among middle school boys
Katherine M Ingram,Jordan P Davis,Dorothy L Espelage,Tyler Hatchel,Gabriel J Merrin,Alberto Valido,Cagil Torgal
DOI: https://doi.org/10.1016/j.jsp.2019.10.007
Abstract:Bystander intervention (i.e., a third party decides to defend a victim when witnessing a conflict) has been identified as an effective strategy to resolve bullying incidents (O'Connell, Pepler, & Craig, 1999). Researchers suggest that student willingness to intervene (WTI) is a robust predictor of bystander intervention (Nickerson, Aloe, Livingston, & Feeley, 2014). Toxic masculinity has been defined as "the constellation of socially regressive [masculine] traits that serve to foster domination, the devaluation of women, homophobia, and wanton violence" (Kupers, 2005, p. 71). Though some aspects of toxic masculinity (e.g., low empathy) have received some empirical attention regarding their role in determining prosocial behavior, many aspects of toxic masculinity have not. Little research has examined how constructs such as attitudes surrounding bullying and sexual harassment, social dominance orientation, and homophobic bullying are related to longitudinal changes in WTI across adolescence. The present study uses growth mixture modeling (GMM) to examine the heterogeneity of WTI among middle school boys in the Midwest (N = 805). Students were classified into three profiles of WTI over time: a "stable high" class (70.9%), a "decreasing" class (22%), and a "stable low" class (7.1%). When compared with the "stable high" class, students with higher levels of dominance and pro-bullying attitudes were associated with an 11% (AOR = 1.11, 95% CI [1.01-1.21] and a 55% (AOR = 1.55, 95% CI [1.05-2.31] increase in the odds of being in the "decreasing" class, respectively. Youth who reported higher rates of homophobic name calling perpetration had a 16% (AOR = 1.16, 95% CI [1.02-1.34] increase in the odds of being in the stable low class compared to the stable high class. Additionally, both homophobic name calling victimization and empathy were associated with a 17% (AOR = 0.83, 95% CI [0.70-0.98] and 18% (AOR = 0.82, 95% CI [0.69-0.98] lower odds of being in the stable low class. The findings support the theoretical framework which posits that features of toxic masculinity are associated with less WTI and thus carry implications for intervention design (Carlson, 2008; Leone et al., 2016).