This study examined effects of negative versus positive symptom formats on the assessment and subsequent classification of ADHD in children in public schools. Symptoms associated with the disorder based on the Diagnostic and Statistical Manual of Mental Disorders Fourth Edition (DSM-IV) were presented to parents and teachers of referred children according to both types of formats. Informants were then asked to indicate whether the symptoms described the child's behavior over the last six months. Overall, the negatively phrased symptoms appeared to generate response bias which resulted in an increased percentage of children meeting the DSM-IV criteria for ADHD. Additionally, the decision reliability or classification agreement between the two formats was low suggesting the possibility of confirmation bias in the assessment process. Implications for practice and future research are discussed.