Studying the Interaction Between Harsh Parenting and the Child’s Social Behavior Problems over Time Using Continuous Time Modeling

Parenting is typically analyzed as a unidirectional process in which the parents have an impact on the child’s behavior and development. Although this is often the main direction of influence, there are also bidirectional processes, e.g., as described in the parent-child coercive cycle (cf. Patterson GR, Coercive family process. Castalia, Eugene, 1982). These processes were mainly investigated in clinical and other studies from North America, but less in larger population samples from other countries. In the present study, data from the Erlangen-Nuremberg Development and Prevention Study in Germany (ENDPS; Lösel et al., J Exp Criminol 9(4):429–449, 2013; N = 675) were used to disentangle the bidirectional relationship between a child’s social problem behavior and harsh parenting (here corporal punishment by the father). The child’s problem behavior was rated by teachers from kindergarten and in school. The Continuous Time (CT) Framework (Voelkle et al., Psychol Methods 17:176–192, 2012) was applied to search for auto-regressive and cross-effects over time. Although the 4 and 5 measurement points, used from ENDPS, were not equally spaced and we had to deal with some missing data, the CT framework delivered a continuous model of the ongoing reciprocal processes. We started with a simple CT model with two covarying variables, and then successively, time-independent and time-dependent variables were added, resulting in a complex analysis to model the interactive dynamics between the child’s problem behavior and fathers’ harsh discipline behavior. The child’s problem behavior had a much greater influence on corporal punishment than vice versa. The highest impact of problem behavior was before the child’s school entry. It seems that father’s corporal punishment is a possible helpless reaction to the overwhelming problem behavior shown by the offspring. This idea was supported by the findings that fathers’ perceived sense of competence in parental guidance reduced harsh parenting as well as the child’s problem behavior. A CT model including two time-independent variables like (a) the child’s intelligence and (b) (difficult) temperament was also tested. Intelligence had a small but significant increasing effect on problem behavior and a lowering effect on corporal punishment. (Difficult) temperament revealed a significant increasing effect only on corporal punishment. The time-dependent variable perceived sense of parental competence suggested to significantly lower the child’s problem behavior as well as the use of harsh parental punishment. The results show the highly valuable contribution of the CT framework in the study of bidirectional reciprocal processes not only in the field of developmental psychology.