Definitions of specific language impairment (SLI), for both research and clinical purposes, often state that nonverbal IQ scores must be within normal limits. This use of nonverbal IQ has been criticized on several grounds, including lack of equivalence between tests. In the current study, a sample of 204 adolescents with and without language impairment took two nonverbal IQ tests. Scores on the tests were compared, as was the classification of individuals into diagnostic categories according to the two tests. Results indicated that there were significant differences between test scores for clinical subgroups, and classification of individuals as having SLI varied substantially depending on the test used. The findings are interpreted as providing further evidence that nonverbal IQ should be used only cautiously, if at all, in defining SLI. Learning outcomes: Readers will (1) understand problems with the use of nonverbal IQ criteria in diagnosis of SLI and (2) be able to cite data showing that different nonverbal IQ tests can lead to different diagnoses.