MULTIPLE RATERS OF DISRUPTIVE CHILD-BEHAVIOR - USING A GENETIC STRATEGY TO EXAMINE SHARED VIEWS AND BIAS

Most research on child behavior incorporates information from differen t individuals. While agreement between informants is generally only mo dest, there is little understanding of the processes underlying disagr eement. In twin studies, differential agreement among raters for MZ an d DZ twins is of particular concern. The processes underlying differen ces among mother, father, and child ratings of oppositional and conduc t disorder symptoms are explored. Evidence in favor of a shared parent al view of behavior is presented. Parental ratings give higher intrapa ir correlations, which could be due to either parents rating their twi ns more similarly or twins contrasting themselves. Rater bias and situ ational specificity are among the possible explanations of differentia l ratings. The effects of incorporating multiple raters of behavior on estimates of genetic and environmental effects are explored. These su ggest that genetic influences are greater for the shared (multiple-rat er) phenotype than for individual ratings; reduction in measurement er ror is only a partial explanation.