One hundred and nine families seeking psychological help for severe interaction problems participated in a standardized parent training program. LISREL causal modeling procedures were used to evaluate the adequacy of a model relating child, parental. and socioeconomic variables measured at intake to parental perception of the child measured at intake and to parental involvement in treatment and treatment outcome measured at 12-month follow-ups. Results provided partial support for the model. They indicated that parental perception may be caused by important child characteristics and parental personal and interpersonal functioning whereas treatment involvement and outcome may be almost exclusively a function of socioeconomic variables. Limitations and implications of the results are discussed. It is suggested that this model may provide a heuristic tool for further research on causal processes in this area.