Objective: Short-forms of Intelligence Quotient (IQ) tests are used to estimate IQ because they take less time to administer and are particularly useful when there is no interest in profile analysis of subtest or Index scores. Individual differences occur between short-form IQ estimates and actual IQ. For clinical populations, where specific abilities are differentially impaired, even more variability in short-form predicted IQ scores are apparent. This study examined the performance of 10 short-forms of the Wechsler Intelligence Scale for Children- Fourth Edition (WISC-IV) in children with Attention-Deficit/Hyperactivity Disorder (ADHD) to determine which provided the most accurate estimate of IQ. Method: The sample included 391 children diagnosed with ADHD. They were 10.48 years old, 70.08% were male, and referred for evaluation because of academic concerns. The WISC-IV was administered as part of this evaluation. A senior neuropsychologist diagnosed ADHD based on test results, parent interviews, behavioral ratings and review of pertinent records. Results: A short-form consisting of the Block Design, Similarities, Comprehension, Vocabulary, and Letter-Number Sequencing subtests provided the most accurate estimate of FSIQ with an absolute mean difference of 3.57 IQ points and a correlation with FSIQ of 0.93. The least accurate short-form included the Similarities and Matrix Reasoning subtests (absolute mean difference =7.48; r = 0.80). Conclusion: Results are discussed in the context of ADHD assessment and broader issues regarding utility of short-forms. Future research might consider demographic or other factors that characterize children whose predicted IQ's are most discrepant from FSIQ's in order to provide guidance for short form administration in clinical and research settings.