Abstract
<p>Peer victimization in adolescence has been consistently associated with increases in internalizing symptoms, such as depression and social anxiety. Recently, both empirical and public attention has focused on cyber victimization, or negative peer experiences that occur in an on-line context. Adolescent Internet use has reached nearly universal rates (recent estimates report over 93% of adolescents have access to the Internet, with the majority accessing the Internet daily), and the rise of social networking and media tools has created a diverse and complicated environment for adolescents to interact within in cyberspace. Research has made great strides in identifying the potential negative effects of cyber victimization, which include poor academic functioning, frustration, sadness, low self-esteem, low peer support, feelings of depression and social anxiety, and suicidal ideation. However, existing measures of cyber victimization: a) are often limited by their focus on older communication tools or limited item content, which may miss important information about a variety of diverse aversive experiences adolescents face on-line; b) lack prospective data on the effects of cyber victimization on psychological outcomes, which limit the ability to make inferences about causality or the directionality of effects; and c) have not examined potential moderators of associations between cyber victimization and internalizing distress, which limits of understanding of these pathways to intelligently inform targeted and meaningful clinical interventions. Finally, research is needed that examines the interplay between traditional, face-to-face, peer victimization and cyber victimization, particularly with a consideration of types of both peer and cyber victimization. The current study sought to (a) examine the psychometric properties and validate a new measure of cyber victimization, the Cyber Victimization Scale for Adolescents (CVS-A), (b) explore the concurrent and prospective associations between cyber victimization and internalizing symptoms (depression, social anxiety and general anxiety), and (c) identify potential moderators (peer social support, gender, Internet use) of the relationship between cyber victimization and internalizing symptoms, both concurrently and prospectively. Participants were 1,045 public high school students (58% girls), ages 13-19 years, 73% Hispanic, 11% White non-Hispanic, 12% Black (African-American, Caribbean American and Hispanic Black) and 4% Asian. Participants were recruited from two public high schools in an urban area of the Southeastern United States. Participants completed questionnaires during class periods at two time points during the spring of the academic year, approximately six weeks apart. These questionnaires included information on demographics and electronic media usage, the CVS-A, the Revised Peer Experiences Questionnaire, the Center for Epidemiological Studies – Depression scale, the Social Anxiety Scale for Adolescents, the short form of the Revised Children’s Manifest Anxiety Scale – 2nd Edition, and the Perceived Social Support scale. Data were collected as part of a larger study that sampled adolescents at three time points throughout the academic year. Data were analyzed using prudent statistical techniques. To evaluate the psychometric properties of the CVS-A, confirmatory factor analytic strategies were utilized as appropriate based on theoretically derived models. For other planned analyses, hierarchical linear regression techniques, controlling for demographic variables and traditional peer victimization to test for the unique contributions of cyber victimization on three separate measures of internalizing symptoms, focusing specifically on symptoms of depression, social anxiety and general anxiety, were utilized. For prospective analyses, symptomology at baseline was also controlled. In regards to the first study aim that sought to evaluate the psychometric properties of the CVS-A, results provided good evidence for psychometric support of a single factor structure utilizing 13 items of the CVS-A (a = .80). Additionally, there was some support for subscales of the CVS-A based on a theoretical model that parallel known subscales of traditional peer victimization, including Aggressive, Relational and Reputational forms of cyber victimization. Regarding the second study aim, controlling for traditional peer victimization, cyber victimization was found to predict significant incremental variance in adolescents’ symptoms of depression both concurrently (ß = .09, p < .05) and prospectively, (ß = .13, p < .01). Relational peer victimization, in particular, was significantly and strongly associated with social anxiety, both concurrently (ß = .36, p < .001) and prospectively, (ß = .08, p < .01). Regarding the third study aim, perceived social support from peers and gender moderated the relationship between cyber victimization and depressive symptoms over time. Specifically, counter to expectations, cyber victimization was more strongly associated with depressive symptoms over time among adolescents who reported higher levels of perceived social support from friends (ß = .18, p < .001) compared to those who reported lower levels of peer social support from friends (ß = .10, p < .001). Additionally, when utilizing CVS-A subscales, reputational cyber victimization predicted increases in depressive symptoms over time for boys (ß = .09, p < .05), but not for girls (ß = .06, p = .08), and aggressive cyber victimization predicted depressive symptoms over time for girls (ß = .20, p < .001), but not for boys (ß = .05, p = .16). These findings provide initial support for the CVS-A as a reliable and valid instrument to examine cyber victimization among adolescents. The CVS-A offers several advantages over existing measures of cyber victimization, including the incorporation of newer communication methods, inclusion of a variety of on-line experiences, and the ability to delineate between potential subtypes of cyber victimization. Additionally, these results provide evidence that cyber victimization may have a unique and negative effect on adolescents’ symptoms of depression, above and beyond the effects of traditional, face-to-face peer victimization. Results also suggest the possibility that cyber victimization may play a unique role in depressogenic pathways of internalizing distress, as opposed to anxious pathways. Furthermore, it appears that adolescents with higher levels of peer support may be at increased risk of reporting depressive symptoms after experiencing cyber victimization, contrary to expectations that adolescents with high peer social support are buffered from psychological distress. Differential findings among boys and girls suggest potential gender differences in the interpretation and meaning of certain cyber victimization and its role in emotional functioning, which may impact our treatment of youth experiencing cyber victimization. This study was limited by several factors (e.g., lack of comparison measure of cyber victimization, reliance on self-report). Future clinical and empirical work would benefit from a consideration of these findings to enhance targeted interventions for adolescents’ experiencing peer and cyber victimization and inform areas of future research by further examining potential gender differences as well as the existence of subtypes of cyber victimization.</p>