Cyberbullying Victimization and Adaptive Avoidance Behaviors at School

Abstract

Using data from the 2009 National Crime Victimization Survey, School Crime Supplement (NCVS-SCS) the current study explores the relationships between traditional bullying victimization, cyberbullying victimization, and victim adaptive avoidance behaviors. Like traditional forms of bullying, the cyberbullying literature base is developing and growing into a targeted area of emphasis in 21st century victimology. We explore the effects of these online victimization experiences, net of the impacts of traditional bullying and fear of victimization at school. Based on logistic regression analysis the results indicate that cyberbullying victimization experiences are significantly related to avoidance behaviors at school.

Keywords:

• bullying
• cyberbullying
• fear
• schools
• avoidance

Notes

1. The evolution of cyberbullying and accompanying definitional changes may have had an effect on estimates of prevalence that reflect the corresponding inclusion and exclusion of certain acts. For instance, the emergence of mobile technology as a ubiquity in the United States has undoubtedly influenced student access to cyberbullying means.

2. Data for the NCVS-SCS is collected by the United States Department of Commerce, Bureau of the Census. Households are randomly selected and all age-eligible individuals become part of the panel. The SCS was administered at the end of the NCVS interview for each eligible respondent. Weights created for the SCS are a combination of household-level and person-level adjustment factors.

3. To ensure consistency and address potential issues associated with the rarity of the outcome we also performed complimentary Log-Log models (Powers & Xie, 2000).

4. These ten items produce an alpha reliability of .76. The subset of items referring to specific locations have a corresponding alpha = .73.

5. Models presented here use the dichotomous account of cyberbullying—however, supplemental analysis using index scores did not produce substantively different results. We dichotomize cyberbullying victimization experiences for analysis purposes, as the full articulation of cyberbullying items would result in severe limitations.

6. Interestingly, while pairwise correlation between the measure of traditional bullying victimization and cyberbullying victimization was significant (r = .31), collinearity diagnostics did not suggest a problem (VIF = 1.27 and tolerance = .79).

7. Tests of the correlation between personal victimization and HH total victimization resulted in r = .55 (collinearity diagnostics fell within normal ranges).