Objective: Homework can have beneficial effects for students; however, it presents challenges, particularly for students with attention problems. Although effective homework interventions exist, intervention development and evaluation has been hampered by the lack of psychometrically sound measures. The primary purpose of this study was to evaluate the construct validity of the Homework Performance Questionnaire (HPQ), Parent and Teacher Versions, in a sample of children with ADHD. A secondary purpose was to examine variations in homework performance as a function of individual characteristics, such as academic achievement, quality of the family–school relationship, and child’s diagnostic status. Method: The sample included 91 children (34% female) with ADHD in Grades 2 to 6. Measures included parent and teacher ratings of homework performance and the quality of the parent–teacher relationship as well as direct assessment of child academic achievement and homework performance (i.e., samples of completed assignments). Correlational analyses were used to examine construct validity, and ANOVAs were used to evaluate group differences. Results: Each factor of the HPQ had a significant relationship with other measures of relevant constructs. There were no significant differences in homework performance between groups for ADHD subtype, medication status, or comorbidity, with the exception of learning disability. Children with ADHD and learning disabilities had significantly lower teacher ratings of academic competence. Conclusion: Results of the present study suggest that HPQ scores may be used to make valid inferences about the homework performance of children with attention problems. These rating scales may be helpful in progress monitoring and evaluating intervention effectiveness. ( J. of Att. Dis. 2011; XX(X) 1-XX)