Cybervictimization is a social phenomenon in which a victim receives an intentional and aggressive act of harm from an aggressor/s, through technologies, from which victims cannot easily defend themselves. Victims present physical health symptoms, as well as low levels of psychological well-being and even higher levels of suicidal ideation and attempts. The alarming prevalence of cybervictimization in Primary Education in Spain ranges from 6.6% to 13.4% and has increased due to the COVID-19 confinement. For all these reasons, it is crucial to investigate the risk and protection factors that allow us to prevent them, especially those less studied in this educational stage, such as family sociodemographic variables. The objective of the study is to analyse the relationship between parental sociodemographic factors (age, gender, educational level, immigrant background, family composition, and degree of rurality of the municipality of family residence) and cybervictimization of their children in Primary Education. A sample of 1169 families with children in grades of 5th and 6th of Primary Education was selected, using a self-administered questionnaire (& alpha; = .84). Association studies were performed using binary logistic regression. The multivariate model followed a stepwise procedure, with the stepAIC function, selecting the best predictive model. The bivariate analysis identified the parental educational level as an individual risk factor for cybervictimization (p < .05). In addition, single-parent families reached almost double the risk of two-parent families. The results of the multivariate analysis showed that gender, parental educational level, and family composition are jointly significant predictors of cybervictimization in Primary Education. It is concluded that there are family sociodemographic factors that predict cybervictimization and the results are analyzed for their implications for educational practice.