Language Sample Length Effect on Mean Length of Utterance in Morpheme in Normal Persian Children

Language sample analysis (LSA) is one of the best methods of language evaluation and mean length of utterance (MLU), and also it is one of the most popular methods for analyzing language samples in children. Spontaneous utterance sampling determining MLU requires a lot of time and now it is not clear how many utterance in spontaneous speech sample of Persian-speaking children can represent their ability in morphology and syntax. Determining the relationship between the length of language sample and the MLU can help further research in determining the reliable sample to evaluate MLU in Persian speaking children. Therefore, this study aimed to determine the effect of language sample length on MLUm (Mean Length of Utterance in Morpheme) in Persian-speaking children aged 4-5 years. 30 children (15 girls and 15 boys) were selected from kindergartens in Babol. Language sampling was performed in a 30-minute free play between the child and researcher. Four sets of speech samples in different lengths of 50, 100, 150, and 200 utterances were transcripted sequentially. The mean MLUm in different lengths was compared by one-way ANOVA and Bonferroni's post hoc test. The mean MLUm was significantly different in samples with 50 (mean = 4.40) and 150 (mean = 4.71), and samples with 50 (mean = 4.40) and 200 (mean = 4.79) utterances. No significant difference was found between MLUm in other lengths (p<0/05). In spontaneous speech samples of Persian-language children, the length of the language sample was effective on MLUm, therefore, that an increase in sample length of more than 100 utterance can improve the MLUm. This effect is not significant in language samples with more than 100 utterances.