Document Type
Capstone Experience
Graduation Date
5-2025
Degree Name
Master of Public Health
Department
Biostatistics
First Committee Member
Cheng Zheng, PhD
Second Committee Member
Lynette M. Smith, PhD
Third Committee Member
Christopher S. Wichman, PhD
Abstract
The replicated study is titled “Investigating the Association Between Bullying Victimization and Suicidal Ideation Among Adolescents: Evidence from the 2017 Youth Risk Behavior Survey” by Philip Baiden and Savarra K. Tadeo, published in Child Abuse & Neglect in 2020. The study employed 2017 youth risk behavior survey data publicly available via Center for Disease Control and Prevention (CDC). Descriptive, binary, and multivariate logistic regression was employed to examine the association between bullying victimization and suicidal ideation. The study analysis included 14,603 adolescents aged 14 to 18, of whom 52% were female. Approximately 23% of the sample experienced bullying, and 18% reported suicidal ideation. Bivariate and multivariate logistic regression indicated a significant association between bullying and suicide ideation. The odds of suicidal ideation were 3.26 times greater among adolescents who experienced both school and cyberbullying (AOR = 3.26, p < 0.001, 95 CI = 3.10–3.43), 2.15 times higher among those who experienced only school bullying (AOR = 2.15, p < 0.001, 95 CI = 2.04–2.27), and 2.00 times greater among those who experienced only cyberbullying (AOR = 2.00, p < 0.001, 95 CI = 1.87–2.14). The study analysis also identified other significant risk factors for suicide ideation such as substance use, depression, and forced sexual intercourse.
The primary objective of this replication study was to conduct a pure replication to determine whether original findings can be replicated using original data and methods. While most of the replicated results aligned with the original study showing significant associations between study variables and suicidal ideation, discrepancies were observed in the magnitude of these associations. The pure replication successfully reproduced the original study’s findings in terms of the direction and statistical significance of most associations, with a few exceptions. While the majority of variables showed consistent results, discrepancies emerged for age, sex, alcohol use, and marijuana use, where the replication differed in either significance or magnitude. Measurement and estimated analysis indicated robust results. Linearity assumption for the continuous variable age satisfied through the Box-Tidwell test. Variance inflation factor results showed no multicollinearity among the study variables. A theory of change analysis extended the study by applying Latent Class Analysis (LCA) to identify adolescent subgroups based on bullying victimization, mental health indicators, forced sexual intercourse, and substance use. The analysis revealed that 62.6% were in the low-risk group, 19.9% in the high-risk group, and 17.6% in the moderate-risk group.
Recommended Citation
Kwizerimana, Jean D'Amour, "Association between Bullying Victimization and Suicidal Ideation among Adolescents: A Replication Study" (2025). Capstone Experience. 387.
https://digitalcommons.unmc.edu/coph_slce/387