BackHome
Refresh: OffView Personal Schedule
Individual Submission Summary
Share...

Direct link:

Poster #225 - Longitudinal Relationships between Victimization and Bullying/Cyberbullying: The Role of Self-Esteem

Thu, March 21, 4:00 to 5:15pm, Baltimore Convention Center, Floor: Level 1, Exhibit Hall B

Integrative Statement

Background: A host of negative developmental outcomes, including mental, physical and socioeconomic problems resulting from being bullied during childhood and adolescence develop over the course of the lifespan (Arseneault, 2018). One such outcome that is not well researched is that being bullied might make the victim a bully him/herself, thus a vicious bully-victim cycle. Previous research has shown that being a bully-victim has poorer outcomes as compared with pure bullies, pure victims, or adolescents not involved in bullying (Kowalski & Limber, 2013). Therefore, understanding why victims develop into bullies seems paramount. One potential explanatory mechanism includes self-esteem in the victim-bully link. On the one hand, being bullied may lead to lower self-esteem (Cammack-Barry, 2005); on the other hand, low self-esteem is itself predictive of bullying (Kokkinos & Panayiotou, 2004). Thus, the current longitudinal study had the following two main research goals: First, to test whether victimization predicted bullying over time; second, to test whether self-esteem mediated the victimization-bullying link. Parallel hypothesized models were specified and tested, one focused on traditional bullying, the second focused on cyberbullying and victimization. Method: Data over the course of two years, four assessments, were collected as part of an ongoing longitudinal study among 569 adolescents in Czech Republic (58.6% female; Mage=12.42 years, SD=0.66 at Wave 1). Bullying and Cyberbullying as well as victimization were measured at each assessment by scales developed by Gradinger, Strohmeier, and Spiel (2010). Self-esteem was measured at all four assessments using the low self-esteem subscale from the Weinberger Adjustment Inventory (WAI): Short Form (Weinberger & Schwartz, 1990). Wave 1 control variables included adolescent age, sex, family structure, and socioeconomic status. Results: Descriptive statistics and bivariate correlations were computed in an initial step, followed by two mediation path models ( one for the traditional victimization-bullying link, the other for cyber-victimization-bullying link; see Figure 1). Correlations showed that victimization and bullying had a weak to moderate positive relationship across assessments; the cyber-victimization-bullying link held across most time points. Self-esteem had a weak negative relationship with bullying/cyberbullying victimization across assessments, but the correlations between self-esteem and bullying and cyberbullying perpetration were mostly non-significant. Results of mediation analysis indicated that victimization at W1 predicted bullying at W3 (ß = .13, p< .01), and victimization at W2 predicted bullying at W4 (ß = .13, p< .01); self-esteem was not a significant mediator for either line. For the victimization-cyberbullying link, victimization at W1 predicated self-esteem at W2 (ß =-.13, p< .001), but not cyberbullying at W3, and self-esteem at W2 predicated cyberbullying at W3 (ß =-.13, p< .05); victimization at W2 predicted cyberbullying at W4 (ß =.24, p< .001), but not self-esteem at W3. Discussion: Study findings support the hypothesized traditional victimization-bullying as well as cyberbullying links. Self-esteem did not mediate the victimization-bullying link but showed some effects between cyberbullying victimization and perpetration. They imply that future interventions should take into account the nature of the relationship between victimization experiences and both traditional bullying and cyberbullying perpetration as well as the effects of self-esteem on the link.

Authors