Psychology, Department of


Date of this Version



Published in the Journal of Speech, Language and Hearing Research (August 2006) 49: 783-808. Copyright 2006, American Speech-Language-Hearing Association. Used by permission.


Purpose: Although mean length of utterance (MLU) is a useful benchmark in studies of children with specific language impairment (SLI), some empirical and interpretive issues are unresolved. The authors report on 2 studies examining, respectively, the concurrent validity and temporal stability of MLU equivalency between children with SLI and typically developing children. Method: Study 1 used 124 archival conversational samples consisting of 39 children with SLI (age 5;0 [years;months]), 40 MLU-equivalent typically developing children (age 3;0), and 45 age-equivalent controls. Concurrent validity of MLU matches was examined by considering the correspondence between MLU and developmental sentence scoring (DSS), index of productive syntax (IPSyn), and MLU in words. Study 2 used 205 archival conversational samples, representing 5 years of longitudinal data collected on 20 children with SLI (from age 5;0) and 18 MLU matches (from age 3;0). Evaluation of growth dimensions within and across groups was carried out via growth-curve modeling. Results: In Study 1, high levels of correlation among the MLU, DSS, and IPSyn measures were observed. Differences between groups were not significant. In Study 2, temporal stability of MLU matches was robust over a 5 year period. Conclusions: MLU appears to be a reliable and valid index of general language development and an appropriate grouping variable from age 3 to 10. The developmental stability of MLU matches is indicative of shared underlying growth mechanisms.