Categorical Data in the Evaluation of School-Based Cyberbullying Prevention Programs: A Review of the Literature

Categorical data analysis offers valuable tools for evaluating school-based prevention programs, yet these methods remain rarely applied in cyberbullying research. This literature review examined how categorical approaches, including contingency tables and related techniques, have been used in studies evaluating school-based cyberbullying prevention. A comprehensive search was conducted in Web of Science covering publications from 2020 to 2025, yielding 100 articles. After applying predefined inclusion and exclusion criteria, 24 studies were reviewed in full, of which 8 met all requirements for final analysis. The results revealed a predominant reliance on linear statistical techniques, such as <i>t</i>-tests, ANOVA, and regression models, applied mainly to continuous variables. By contrast, categorical analyses were seldom employed. The chi-square test appeared as the most frequent approach, but its use was generally restricted to descriptive purposes, with little application of complementary methods such as standardized residuals, effect size measures, or logistic models. This restricted application reduced the ability to capture response patterns, subgroup differences, and categorical associations essential for evaluating program outcomes. The findings highlight a methodological gap in cyberbullying prevention research and emphasize the potential of categorical data analysis to enrich interpretation. Incorporating these methods could increase methodological rigor, reveal nuanced behavioral patterns, and provide actionable evidence for educators, policymakers, and program designers seeking to strengthen school-based prevention strategies.