LAUSR

In a world increasingly driven by digital technology, cyber-safety is becoming a pressing concern for Internet users. This article contributes to sociological, psychological, and criminological literature on digital risks by investigating the determinants of individuals’ cyber-safety behaviors. Our study adds new insights to digital inequalities studies, cognitive health behavior models, and fear of crime literature by developing and testing a comprehensive theoretical model in a developing economy (Uruguay). To validate our model, we fit structural equations with data from the 2017 WIP+DiSTO Uy survey, a representative sample of Uruguayans (N=653). We found that operational digital skills were the strongest predictor of cyber-safety, while also providing support for the sequentiality of the digital divide hypothesis, as education and age-based disparities affect cyber-safety through their effect on Internet use, which in turn affects digital skills. Additionally, findings contribute both to cognitive behaviors and fear of crime literature, by attesting that gender and age-based disparities only have an indirect effect on cyber-safety, which is mediated by user beliefs regarding the severity of cyber-victimization. The study also provides evidence for the generalization of cyber-safety behavior theories — originally formulated based on data from developed economies — to developing ones, by stressing the role of digital skills and perceptions of victimization severity as the main direct antecedents of cyber-safety.