Objective: To explore the significance of adding father ratings to mother and teacher ratings in the assessment of ADHD symptoms in children. Method: The ADHD Rating Scale-IV, the Child Behavior Checklist, and the Teacher Report Form were filled out by all three informants for a sample of 48 clinically referred children (79% boys) aged 6 to 15 (M = 10.1) years. Results: Correspondence between father and teacher reports on ADHD-specific symptoms (intraclass correlation coefficient [ICC] = 0.38) exceeded that between mothers and teachers (ICC = 0.23). Fathers rated their children as having fewer problems than did mothers and teachers on Total scale scores and the Inattention subscale of the ADHD Rating Scale-IV. Mother ratings were more sensitive to an ADHD diagnosis, whereas father ratings better predicted an ADHD diagnosis requiring the two-setting criterion. Conclusion: The choice of parent informant and informant combination had a considerable impact on parent-teacher concordance and estimates of ADHD symptoms and subtypes in the child.