Purpose: The purpose of this study was to investigate the extent to which a parent-implemented language intervention improves language skills in toddlers at risk for persistent language impairment (LI) as compared with a group of typically developing toddlers. Method: Thirty-four children with LI between 24 and 42 months of age were randomly assigned to a treatment or nontreatment experimental condition. Participants in the treatment group received 24 biweekly 1-hr sessions for 3 months. An additional sample of 28 age- and gender-matched children with typically developing language (TL) was also included. Norm-referenced child assessments and observational measures were used to assess changes in children's language growth. Results: Results from multilevel modeling indicate that children in the treatment group made greater gains than children in the control group on most language measures. Whereas children in the treatment group had lower language scores than children with TL at the end of intervention, the rate of language growth was not significantly different between groups. Child receptive language and parent use of matched turns predicted expressive language growth in both children with and without LI. Conclusion: The results of this preliminary study indicate that parent-implemented interventions may be an effective treatment for children with expressive and receptive LI.