Multiple informant agreement of child, parent, and teacher ratings within community samples.


Objective: Extant research concerning the degree of multiple informant (that is, parent, clinician, teacher, and child) agreement for child anxiety ratings generally uses clinical samples, and results have been mixed. Method: Our study used a community sample of public school children (n = 1039) to investigate child (self), parent, and teacher reports of child anxiety across 3 time points (pretreatement, posttreatment, and follow-up) in 3 independent school prevention and intervention trials. Results: Results showed that parents and teachers had high informant agreement for ratings on anxiety across the 3 time points (r = 0 95 to 0.96, P < 0.001); agreement between parent and child (self) reports and between teacher and child (self) reports consistently showed lower agreement across the 3 time points ( r = 0.14 and 0.28, respectively, P < 0.001) Group differences were also significant for sex and grade, whereby females more commonly self-reported higher anxiety and children in grades 3 and 4 self-reported higher anxiety, compared with students in grades 5 to 7 Conclusion: Correlations between parent and teacher with child ratings were poor over 3 time points, and significant differences were found for sex and grade. Research is needed to understand reasons for poor concordance between parent, child, and teacher ratings of anxiety for all children.