Reciprocal Relationships between Parental Psychological Aggression and Adolescent Behavioral Problems

Yoonyoung Kwak, Purdue University


An extensive body of research has shown that parents’ harsh parenting and children’s behavioral problems are bidirectionally associated. These findings are in line with the transactional process theoretical model. Parental psychological aggression (PPA) is a non-physical type of harsh parenting that is more prevalent among adolescents than corporal punishment and has rarely been tested for bidirectional relationships with diverse behavioral problems. This dissertation includes an investigation of the bidirectional relationship between PPA and four adolescent behavioral problems (delinquency, alcohol use, smoking, and running away) separately using samples from two different cultural contexts. The first sample is 591 adolescents who had been involved with Child Protective Services (CPS) investigations for alleged maltreatment in the U.S. (Study 1). The second sample is 2,938 adolescents living in South Korea (Study 2). A cross-lagged panel model controlling for covariates is estimated to assess bidirectional relationships. The covariates include adolescent sex, race/ethnicity, and age, primary caregivers’ age and education, annual family income, and physical abuse measured at T1. Each bidirectional model is tested for adolescent sex and age moderation. The results of the cross-lagged panel models differ by the sample and behavioral problem. In Study 1, a bidirectional relationship is found for delinquency only and only parent-driven effects are found to be significant for the other behavioral problems. The parent-driven effects are stronger for older adolescents’ alcohol use vs. younger adolescents and male adolescents’ delinquency and alcohol use vs. female adolescents. In addition, female adolescents’ running away is also bidirectionally associated with PPA. In Study 2, only child-driven effects are found for smoking and running away. The moderation effects of sex and age are either non-significant or marginal, indicating that the effects across sex and age are similar in this study. Overall, the transactional process hypothesis between PPA and behavioral problems is not supported by the results of the studies. In general, in the US sample, parental behaviors are found to drive adolescent behaviors and not vice versa. However, there are reciprocal effects with adolescent delinquency and female adolescent running away indicating that these types of behavior from adolescents may escalate together with PPA. This may be because delinquent behaviors and running away are more visible to and involve parents more compared to adolescent substance use, which may be clandestine. Parent’s lower use of PPA would benefit adolescents and prevention of PPA as a reaction to adolescent behavior problems would help reduce behavior problems. For the Korean sample, the levels of behavioral problems are very low overall, indicating that Korean adolescents do not tend to externalize problems. Therefore, PPA in this sample does not seem to have the same effect on adolescent externalizing behaviors. Due to cultural constraints on externalizing behaviors, it may be that Korean adolescents are more likely to suffer from internalizing problems as a result of PPA. Related to this, the externalizing problems that are exhibited by adolescents, namely smoking and running away, engender rebuke in the form of PPA from parents, perhaps because it is so far outside the norm in this culture. In sum, the findings of this dissertation reveal that the transactional process is conditional on the type of behaviors and cultural differences in parent-child relationships.




Christ, Purdue University.

Subject Area

Developmental psychology|Individual & family studies

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server