Abstract
In recent years, the sentence diversity method of language development assessment has been proposed overseas as an approach for differentiating whether “language delays” in the 2-year-old age range are within the scope of individual differences or are a sign of later language developmental disorders. To apply this method in Japan, we developed a Japanese version of the sentence diversity method for early language development, collected data from 30- and 36-month-old children with typical development and children at risk of language development disorders, and examined its validity. On the one hand, a strong positive correlation was found between the sentence diversity produced by children with typical development, the number of sentences, and the mean length of utterance in morphemes（30M: r=.827, r=.701; 36M: r=860, r=568），which are conventional measures of grammatical development. On the other hand, children at risk of language developmental disorders showed less sentence diversity than did children with typical development. The sentence diversity method seemed to be valid for assessing grammatical development in the 2-year-olds, but 36-month-old children showed large variability, suggesting the need for further collection of data by age group in the future.
