An exploratory study of clinician real-time morpho-syntactic judgements with pre-school children

The purpose of this study was to explore the feasibility of employing real-time morpho-syntactic judgements. A system of capturing real-time judgements of children's utterances was implemented using a modified talk-time sampling procedure with 13 monolingual, English-speaking US pre-school children who had standard scores above 85 on a norm-referenced measure of language ability, four of whom had histories of diagnosed language deficits and were receiving language therapy. The remaining participants (n = 9) were typically-developing. The primary research questions were whether the linguistic data from abbreviated talk-time samples were comparable to traditional 100 utterance samples, and whether utterance level judgements made in real-time were in concordance with the linguistic data available in talk-time samples. Results of this investigation revealed positive and significant correlations in linguistic data across both forms of language sampling, as well as statistically significant inverse correlation between real-time judgement of utterance errors and mean length of utterances in morphemes (MLU-m), indicating that clinician decisions about children's utterances were closely associated with morpho-syntactic accuracy. Further analyses revealed that the percentage of conversational utterances with errors was different for children with and without histories of language impairment. Implications for potential clinical application and directions for future research are discussed.