Classifying Longitudinal Changes and Examining Predictors in Abuse by Parents During Adolescence Using Growth Mixture Modeling

Objectives: This study aimed to explore distinct longitudinal trajectories of parental abuse during adolescence and to identify predictors that classify these distinctive latent class trajectories over time.Methods: Data were sourced from waves 2 to 6 of the Korean Children and Youth Panel Survey 2010, covering a five-year period from 8th grade to 12th grade. A total of 2,313 adolescents were analyzed. Growth mixture modeling was used to identify different trajectories of parental abuse, followed by multinomial logistic regression to examine predictors that distinguish these trajectories.Results: The quadratic model provided the best fit, identifying three distinct trajectories: low-worsening, high-chronic, and moderate-declining. Key predictors that distinguished the chronic group from the worsening group included gender, smartphone dependence, aggression, and delinquency. Moreover, gender, smartphone dependence, aggression, and household income were significant predictors in differentiating the worsening group from the declining group.Conclusion: The results emphasize the necessity for ongoing monitoring and intervention for adolescents who experience abuse by parents, regardless of whether it is chronic or mild, but worsening. Comprehensive interventions are essential to address the diverse factors that affect changes in parental abuse throughout adolescence. This highlights the importance of understanding both individual and familial factors in preventing and mitigating abuse.