The present study estimates the longitudinal effects of in utero cocaine exposure on language functioning at 3, 5 and 7 years of age in an urban sample of 443 full-term children (236 cocaine-exposed and 207 noncocaine-exposed) participating in the Miami Prenatal Cocaine Study. The sample was enrolled prospectively at birth, with documentation of prenatal drug exposure status through maternal interview and urine and meconium toxicology assays. Language functioning was measured at ages 3 and 5 years using the Clinical Evaluation of Language Fundamentals-Preschool (CELF-P) and at age 7 years using the Core Language Domain of the NEPSY: A Developmental Neuropsychological Assessment. Longitudinal Generalized Linear Model and Generalized Estimating Equations (GLM/GEE) analyses revealed an association between prenatal cocaine exposure and deficits in total language functioning after statistically controlling for child sex, visit age, prenatal exposure to alcohol, marijuana and tobacco and over 20 additional medical and sociodemographic covariates drawn from potentially confounding influences assessed at birth and follow-up visits (D=-0.17; 95% CI=-0.32, -0.03; P=.019). The link from prenatal cocaine exposure to later language deficits does not appear to be mediated by cocaine-associated deficits in birth weight, length or head circumference. Overall, the evidence tends to support an inference of a stable cocaine-specific effect on indicators of language functioning during early childhood through age 7 years.