Stability in Retrospective Self-Reports of Childhood Peer Victimization

• In Event: 58.030 - From Start to Finish: A Four-Year Study of College Student Victimization and Bystander Behavior

Abstract

Background
Bullying victimization during childhood can have a range of adverse effects that last long after the bullying has stopped. Reports of childhood bullying have been associated with increased symptoms of depression and anxiety, and decreased overall physical health in adulthood (Allison et al, 2009; Copeland et al, 2013, Holt et al, 2014). Many studies that assess the long-term impacts of bullying utilize retrospective self-report measures (Allison, et al 2009; Holt et al, 2014; Schafer, et al, 2004), in which adults reflect on peer victimization experiences during childhood and adolescence. Research examining the test-retest reliability of retrospective self-reports of adverse childhood experiences has found responses are generally stable over time (Dube et al., 2004); however, studies have yet to address the suitability of retrospective self-report methods in relation to childhood peer victimization.

Research Aims
This study evaluates the stability of reports of childhood bullying among college students, with specific attention to reports of repetition and power imbalance, two core components of bullying.

Methods
College students (n = 138, 71.3% female) completed surveys during their first and fourth years. Past bullying victimization was assessed using the California Bully Victimization Scale (CBVS; Felix et al., 2013), which measures bullying by presenting participants with specific types of victimization (e.g. teasing, rumor spreading) and asking them to specify if they have experienced each (Furlong et al, 2010; Felix et al, 2011). Based on responses, participants are prompted to report the frequency of each behavior and any power imbalance between themselves and the aggressor. Participants are classified as bullied if they endorsed repeated victimization (2-3 times per month or more) and an imbalance of power.

Results
Students reported having experienced more types of childhood victimization at the first time point (paired t-test; t = 2.004, p = .047). Preliminary analyses indicate 76.1% agreement (κ = .485, p < .001) in self-reports of repeated victimization. Reports of being hit (98.6% agreement, κ = .659, p < .001), ignored (80.4% agreement, κ = .432, p < .001), and teased (78.3% agreement, κ = .415, p < .001) showed the most stability across time points. There was 62.3% agreement (κ = .192, p = .024) in reports of power imbalance, and no significant differences in total forms of power imbalance across time points (paired t-test; t = 1.068, p = .287). Overall there was 74.6% agreement (κ = .349, p < .001) in self-reports of childhood bullying across time points.

Significance
Stability in retrospective self-reports of repeated victimization was within the range of a previous study evaluating retrospective self-reports of adverse childhood experiences (κ values ranging from .41 to .86; Dube et al., 2004). Understanding how perceptions of childhood peer victimization change throughout the lifespan can inform efforts to accurately identify and treat those at risk for negative outcomes associated with bullying. Given the prevalence of retrospective self-report methods in the existing literature, there is a clear need for research in this area.

Authors

• Rachel Oblath, Boston University

• Susan Kong

• Melissa K. Holt, Boston University

• Jennifer Greif Green, Boston University