Attention-deficit/hyperactivity disorder (ADHD) is characterised by developmentally inappropriate levels of hyperactivity, impulsivity, and/or inattention, leading to significant impairment across multiple domains of functioning. To receive a diagnosis, symptoms must be present across two or more settings; thus requiring information to be collected from multiple informants, typically parents and teachers. Research consistently shows low to moderate agreement between parent and teacher reports; however, mechanisms underlying these discrepancies remain unclear. The present study aimed to: (1) describe patterns of reporting children's ADHD symptoms by parents and teachers in New Zealand; (2) replicate previous research examining agreement between parents and teachers; and (3) investigate the role of several potential mechanisms for informant discrepancies; those being contextual influences (i.e. situational specificity) or biases in informants' reports (i.e. source biases). Fifty five children and their parents/guardians and teachers participated. Parent and teacher ratings were measured using standardised questionnaires (i.e. ADHD-RS-IV, BASC-2) and children's off-task behaviour using a clinician-rated classroom observation scheme (i.e. BOSS). T-tests showed that overall parents rated their children as more symptomatic of ADHD than teachers. Diagnostically, the choice of informant and the rule for combining information from multiple informants dramatically altered the ADHD subtype assigned to the child.Correlations and kappa statistics showed that parent-teacher agreement was slightly higher than previous studies; however, it remained within the moderate range. The level of agreement differed across the type of symptoms reported, with hyperactive/impulsive symptoms being rated more concordantly than inattentive. Finally, results concerning the mechanisms underlying discrepancy suggest a combination of factors. Implications for the assessment of ADHD in clinical practice are discussed. ACKNOWLEDGEMENTS