Wald test between RRR of informant pairs.

Background
To fulfil the diagnostic criteria of Attention Deficit Hyperactivity Disorder in the Fifth Edition of Diagnostic and Statistical Manual of Mental Disorders (DSM-5), symptoms should be observed in two or more settings. This implies that diagnostic procedures require observations reported from informants in different settings, such as teachers in school and caregivers at home. This study examined parent-teacher agreement in reporting hyperactivity/inattention and its relationship with child’s, parent’s, and family’s characteristics.
Method
We used data from the 2004 United Kingdom Mental Health of Children and Young People survey, including 7977 children aged 4–17, to investigate cross-informant agreement between parents and teachers on the hyperactivity-inattention subscale of the Strengths and Difficulties Questionnaire. The characteristics of different patterns of informant agreement were assessed using multinomial logistic regression.
Results
Cross-informant agreement of parent and teacher was low (weighted kappa = .34, 95% C.I.: .31, .37). Some characteristics, such as male child and parental emotional distress, were associated with higher likelihood of parent-teacher discrepancy.
Conclusion
We found low informant agreement in the hyperactive/inattention subscale, as hypothesised and consistent with previous studies. The current study has found several factors that predict discrepancy, which were partly consistent with previous research. Possible explanation, implications, and further research on parent-teacher informant discrepancy in reporting hyperactivity/inattention were discussed.