Evidence-based intervention against bullying and cyberbullying: measurement of the constructs, evaluation of efficacy and mediation processes

The present dissertation tried to provide an answer to the need of high-quality evaluations of theoretically grounded antibullying interventions (Baldry & Farrington, 2007; Ttofi & Farrington, 2011). Literature of the last ten years shows that aggressive behaviours carried out by peers may be experienced in both face-to-face and online interactions (Tokunaga, 2010; Wingate, Minney, & Guadagno, 2013); for this reason adopting a specific focus both on bullying and cyberbullying appears to be more suitable (Menesini, 2012). The general aim of the present dissertation was to evaluate Noncadiamointrappola! program 3rd Edition by adopting an evidence-based approach in analyzing aspects related to the intervention’s efficacy. Three empirical studies are presented. They cover three main issues: 1) measurement of the cyberbullying constructs; 2) efficacy of the Noncadiamointrappola! program in reducing bullying, cyberbullying, and internalizing symptoms; 3) mediational mechanisms involved in the explanation of the efficacy of the program in reducing cybervictimization. In the first study we analyzed the psychometric properties of a revised instrument (FCBVSs; Menesini, Nocentini, & Calussi, 2011) devoted to measure cybervictimization and cyberbullying constructs. The analyses were conducted on a sample of 1142 adolescents (54.5% males) enrolled in 9th, 10th 11th grades of high schools in Tuscany. Results support a gender-invariant model based on 14 items and four factors both for cybervictimization and cyberbullying. The subscales cover four types of behaviours and describe different attacks made by peers in the cyber context (written-verbal, visual, impersonation and exclusion). The second order CFA confirmed that a “global”, second-order measure of cyberbullying and cybervictimization fits well with data. Overall the scales showed both good validity (construct, concurrent and convergent) and reliability (internal consistency and test-retest). In the second study we evaluated the effects of the Noncadiamointrappola! program in two quasi-experimental trials that involved different samples of adolescents attending the first year of Italian high schools. We found that the program in the experimental group significantly predicted a decrease in all targeted variables (victimization, bullying, cybervictimization, and 4 cyberbullying) in both quasi experimental trials. Looking at the first quasi experimental trial results (Control group, N=171; Experimental group, N=451), we found significant long-term effects of the program (six months); the outcomes did not decay over time. At the same time, we found that the program was efficacious in reducing internalizing symptoms in the experimental group through the decrease in cybervictimization above and beyond the mediational effect of the decrease in victimization. In the second independent trial (Control group, N=227; Experimental group, N=234) we tested for potential moderating effects of gender; we found that it did not have interactive effects with the efficacy of the program. In the third study, using data about the first quasi experimental trial, we analyzed the mediational mechanisms that explain the efficacy of the program in reducing cybervictimization. We found that the program predicted the increase over time in seeking support coping strategy, both on informational and instrumental aspects (distal advice) and on the more emotional way of getting help from people (close support). Noncadiamointrappola! program had significant indirect effects, through distal advice and close support, in reducing cybervictimization. Results are discussed highlighting their contributions to the literature both on evidence-based interventions and on bullying and cyberbullying phenomena. Finally, the overall strengths, limitations and implication for future studies are pointed out.