The predictive validity of the parenting dimensions inventory : a replication and extension

While parent-report measures of parenting and child behavior are cost-effective, they are plagued by rater bias and poor correspondence between raters across settings. The present study examined these reliability problems and their relationship to the predictive validity of the Parenting Dimensions Inventory (PDI), a self-report measure based on a hierarchical model of parenting. Thirty-nine self-selected families from two Catholic elementary schools participated. Parents and adult friends completed the Child Behavior Checklist (CBCL) on children's home behavior while two teachers completed the CBCL on children's school behavior. Parents and adult friends also completed the PDI on mother's parenting behavior. Results indicated that home and school ratings of children's behavior, as well as home ratings of mothers' parenting, showed moderate within-setting interrater agreement, and that ratings of children's behavior were also fairly consistent across settings. Additionally, it was found that certain aspects of mother's parenting predicted a marginally significant portion of the variance of both home and school based assessments of child behavior.