The present study investigated teachers' and teacher aides' (TAs) agreement in their ratings of preschoolers' externalizing behavior and their associations with observed classroom behavior for a sample of children at risk of developing a disruptive behavior disorder. One hundred twenty-two teachers rated 360 students' externalizing behavior in the first month of school, and within the next 6 weeks, children were observed in their classroom. Results indicated that teacher and TA reports were moderately correlated, teacher-rated severity of oppositional defiant disorder behaviors was related to agreement, and teacher/TA agreement did not predict observed externalizing behavior. In general, using teacher and TA ratings together versus a single informant's rating did not provide a better estimate of information gathered from observational measures. Results demonstrate the importance of gathering observations and rating scales when evaluating preschoolers' externalizing problems. Future work should explore factors that contribute to teachers', TAs', and observational methods' differential evaluations of externalizing behavior. (author abstract)