Abstract
Standardized rating scales remain the primary mechanism through which child behaviors are recorded. Despite the many advantages of such systems, the documented lack of agreement among different informants' ratings of the same child remains a pervasive problem for clinicians. This study examines the degree to which observed behavior ratings were influenced by different informants, relative to the traits these measures were designed to assess. A correlated trait-correlated method confirmatory factor analysis model is estimated on a multitrait-multimethod design in which five behavioral traits were completely crossed with three informants (mothers, fathers, and teachers). Behavioral ratings were completed on 562 first-grade children. Results indicate that ratings of child behavior are heavily influenced by the informant, that informant influences are relatively independent of one another, and that the influence of method variance is often greater than the influence of the trait being assessed.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
