Comparing language characteristics of 3-6 years old children with specific language disorders to those of normally developing children

ABSTRACTIt is a difficult process to select appropriate measures in determining speech and language disorders. In the literature, researchers mentioned various measures used so far but they are still searching for a more effective, practical and distinguishing one. Speech and language therapists have preferred two main methods; one is standardized assessment tests and the other is criterion based tools. Standardized assessment tests are sometimes too structured to determine language disorders. On the other hand, criterion based measures used the data elicited by spontaneous language sampling of the children and therefore provides a more detailed analysis of the outcome.Today, computer based analysis programs such as SALT (Systematic Analysis of Language Transcripts) in assessing children? expressive language. Quantitative measures were calculated by SALT by eliciting spontaneous speech samples. It is said that mean length of utterance for morphemes and words, number of different words, number of total words and type token ratios provide data to compare children with specific language impairment (SLI) and the children with typical development. The aim of the study is therefore to find out whether quantitative measurements were discriminative enough to diagnose children with SLI and to differentiate those children from the normally developing children in terms of question words, conjunctions, pronouns, adverbs, and the like..The participants of the study were 18 children with SLI between the ages of 3 and 6. The age- matched typically developing children (ranged between 33 and 47 in number) were selected from SALT data base. Spontaneous language samplings were considered in two parameters: duration/time (15 minutes) and length (50 utterances and 50 words)The results of study pointed out that children with language disorders demonstrated lower performance (in mean length of utterance in morphemes, mean length of utterance in word, number of different words, number of total words measurements) than their age?matched typically developing children in time and length parameters. There was no difference between the group of children with SLI and the group of normally developing children in type token ratio (TTR) measurements limited by time. On the other hand, the group of children with SLI showed lower TTR performance than typically developing children by when the data limited by word length.It was found that the percentages of question-response and intelligibility variables were lower than those of normally developing children; but the maze words produced by children with SLI was higher than those of the age-matched typically developing children. The children with SLI produced question words, conjunctions, pronouns and noun and verb morphemes less than the typically developing children.Key words: Specific language impairment, spontenous language lampling, quantitative measures, SALT analysis