A multilevel meta-analysis of cyberstalking: Domains (and subdomains) of contributing factors

Abstract

PurposeThe prevalence of cyberstalking, with its substantial harm to victims, highlights the importance of identifying the factors contributing to its occurrence. This study assesses, through an advanced three-level meta-analytic approach, the relative predictive validity of sociodemographic, background, risk, and protective domains of predictors associated with cyberstalking perpetration and victimization. MethodsIn each domain, multiple subdomains were measured for dozens of individual contributing factors. An extensive systematic search yielded 57 eligible studies, describing 54 independent samples between 2002 and 2022 with 515 effect sizes. ResultsMultilevel meta-analysis results revealed that the background domain had the largest composite effect sizes for both cyberstalking perpetration and victimization, followed by the risk domain, whereas sociodemographic and protective domains had no significant effect. According to the moderator analyses, the effect sizes of background and risk domains varied depending on the age of the sample and the country of the data, highlighting the need for future synthesis studies to identify the unique contributing factors to cyberstalking in adult samples from various countries. ConclusionsFindings from this meta-analysis provide valuable evidence for the development of primary prevention and risk assessment-management, as well as cyberstalking security and guardianship.