Early and effective treatment for children with developmental language disorder (DLD) is important. Although a growing body of research shows the effects of interventions at the group level, clinicians observe large individual differences in language growth, and differences in outcomes across language domains. A systematic understanding of how child characteristics contribute to changes in language skills is still lacking. To assess changes in the language domains: expressive morphosyntax; receptive and expressive vocabulary; and comprehension, in children in special needs education for DLD. To explore if differences in language gains between children are related to child characteristics: language profile; severity of the disorder; being raised mono- or multilingually; and cognitive ability. We extracted data from school records of 154 children (4-6 years old) in special needs education offering a language and communication-stimulating educational environment, including speech and language therapy. Changes in language were measured by comparing the scores on standardized language tests at the beginning and the end of a school year. Next, we related language change to language profile (receptive-expressive versus expressive-only disorders), severity (initial scores), growing up mono- and multilingually, and children's reported non-verbal IQ scores. Overall, the children showed significant improvements in expressive morphosyntax, expressive vocabulary and language comprehension. Baseline scores and gains were lowest for expressive morphosyntax. Differences in language gains between children with receptive-expressive disorders and expressive-only disorders were not significant. There was more improvement in children with lower initial scores. There were no differences between mono- and multilingual children, except for expressive vocabulary. There was no evidence of a relation between non-verbal IQ scores and language growth. Children with DLD in special needs education showed gains in language performance during one school year. There was, however, little change in morphosyntactic scores, which supports previous studies concluding that poor morphosyntax is a persistent characteristic of DLD. Our results indicate that it is important to include all children with DLD in intervention: children with receptive-expressive and expressive disorders; mono- and multilingual children, and children with high, average and low non-verbal IQ scores. We did not find negative relations between these child factors and changes in language skills. What is already known on the subject Intervention studies indicate that intervention can be effective, but not for all children with DLD, and not in all language domains. Longitudinal studies on language development show stable growth patterns in children with DLD at the group level. A systematic understanding of how child characteristics contribute to changes in language skills is still lacking. What this paper adds to existing knowledge In this study, we report on the language gains of a cohort of 154 children with DLD (4-6 years old), in a special education setting for children with language disorders. Our sample includes children with receptive-expressive disorders and expressive-only disorders, and monolingual as well as multilingual children. Our results show that children's language skills improved. The co-normed tests we used revealed that the children had much lower growth in morphosyntax than in the other language domains. Language gains between children with receptive-expressive and expressive-only language disorders did not differ, children with lower initial test scores showed more improvement than children with higher initial scores, multilingual children showed more gains in expressive vocabulary than monolingual children, and there was no effect of non-verbal IQ on change in language scores. What are the potential or actual clinical implications of this work? The results suggest that catching up on language is possible for children with DLD. It is important to include all children with DLD in intervention: mono- and multilingual children; children with receptive-expressive and expressive disorders; and children with high, average and low non-verbal IQ scores. We did not find negative relations between these child factors and changes in language skills. The limited growth in morphosyntax compared with other linguistic areas warrants the attention of both practitioners and researchers, with a particular focus on the implementation of research findings in clinical practice.