Grammatical Morpheme Effects on MLU

Studies of children with specific language impairment (SLI) often include 2 comparison groups of typically developing children--a group matched according to age and a group matched according to mean length of utterance (MLU). In these studies, both groups of typically developing children often perform better than the SLI group. For many of these investigations, grammatical morpheme use constitutes the dependent measure. The use of grammatical morphemes requires longer utterances than the failure to use these morphemes. If children with SLI show less use of grammatical morphemes than typically developing children matched for MLU, shouldn't they produce some other detail of language more frequently than the MLU-matched group? In this article, the authors report 2 studies showing that such offsetting effects are not necessary in principle, given the nature of MLU. The implications of these findings are discussed.