This study tests the extent to which giving credit for African American English (AAE) responses on a General American English sentence imitation test mitigates dialect effects. Forty-eight AAE-speaking second graders completed the Recalling Sentences subtest of the Clinical Evaluation of Language Fundamentals-Third Edition (1995). A Bayesian Markov Chain Monte Carlo method was used to determine the relationship between the students' scores and the presence of third person singular "-s," a feature largely absent from AAE morphosyntax, in the subtest sentences. Even when given credit for AAE responses, the estimated effect of third person singular "-s" was significant, high relative to those of negation and counterfactual conditional "if" + "ed," and correlated with an independent measure of the students' rootedness in AAE syntax. It is argued that these results reveal a receptive language effect not addressed by crediting dialect productions.