Family Relationships as Viewed by Parents and Adolescents: A Specification

Relational data from a representative sample of high school students matched with data from their parents (N = 3,988) are used to examine patterns of dyadic agreement and disagreement in reports of family life. The level of agreement (kappa) is uniformly low regardless of the aspect of family life being reported on, whether it is reports of concrete behaviors or evaluations of the quality of family relationships. The results replicate previous findings to an unusual degree. A new specification suggested in this analysis is that systematic response biases are limited to reports of the power dimension of family life—parents and children each systematically enlarge the degree of influence they themselves have in the dyad. Conversely, there is no systematic discrepancy between parent and child reports of the degree of closeness in the family.