Date of Award

1-10-2014

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Counseling and Psychological Services

First Advisor

Kris Varjas

Second Advisor

Joel Meyers

Third Advisor

Christopher Henrich

Fourth Advisor

Don Davis

Abstract

This dissertation introduces a social cognitive model of bystander behavior and examines the mediating role of self-efficacy on the relationship between bullying victimization and negative outcomes. Based on Bandura’s (1986; 2001) social cognitive theory, this model utilizes two frameworks for understanding bystander behavior in bullying: group process framework (Salmivalli, 2010) and the bystander motivation framework (Thornberg et al., 2012). A research agenda is presented based on the key elements of the proposed model, including bystander agency, bystander self-efficacy, bystander moral disengagement, and bystander collective efficacy. The research study investigated self-efficacy for coping with bullying victimization and its mediating role on the relationship between bullying victimization and the outcomes of depression, anxiety, and perceptions of school safety. The Bullying Victimization Self-Efficacy Scale (BVSES; Kim et al., 2010), the Student Survey of Bullying Behaviors – Revised 2 (SSBB-R2; Varjas et al., 2008) and the Behavior Assessment System for Children, 2nd Edition (BASC-2; Kamphaus & Reynolds, 2004) were administered to 551 elementary and middle school students in a southeastern urban school district. Using structural equation modeling, a measurement model was used to confirm the factor structure of the latent variables used in the study (i.e., victimization, the BVSES scales, depression, anxiety, and school safety). Then, the hypothesized structure model was used to determine the mediating role of self-efficacy on the relationships of bullying victimization with depression, anxiety, and school safety. An alternative model was tested where depression, anxiety, and school safety were mediators of the relationship between victimization and the BVSES scales as a comparison for the hypothesized model. The measurement model yielded a good model fit, deeming it acceptable for the structure model analysis. The hypothesized and alternative models yielded a good model fit, and significant mediation effects were found in both models. However, the low magnitude suggests that self-efficacy had a relatively weak mediation effect, which may be due to the strength of the relationship between victimization and the outcome variables. This strong direct effect suggests that self-efficacy may not be a substantial mediator influencing the relationship between victimization and the outcome variables. Theoretical implications and directions for future research are discussed.

Share

COinS