Measuring Correlation between Word Class Frequency and Simulated Early Lexical Development Based on Data Analysis from Preschool Short Stories

Abstract

This paper aimed at examining the claim that short-stories prepared for preschool children contain a theory-based system which could be reflected on early lexical development. Besides, we investigated the claim that word-class frequency and/or occurrence could be used as a predictor for a child’s early lexical development based on views from behaviourism, the frequency hypothesis and ZIPF’s law. The correlational approach was followed to measure the two identified variables, namely, simulated early lexical development (represented by preschool materials) as the dependent variable and word-class frequency and/or occurrence (with and without repetition/ frequency) as the independent variable. A series of short-story consisting of ten short-stories was analysed in terms of word-class into eight word-classes and then into different semantic categories. The use of Minitab (17th) version yet running both descriptive and inferential statistical tools indicated that the calculated percentage of the content words were clearly higher than those of the function words (43%, 5%, 6%, 26% and 6%) as compared to only (5%, 8% and 1%) respectively. Moreover, Pearson’s product-moment correlation indicated a strong yet positive correlation between word-class frequency and/or occurrence and simulated early lexical development, r=0.965, p < .0005. Similarly, independent t-test result further indicated that word-class with repetition has statistically significantly higher early lexical development rate (378 ± 438) compared to word-class without repetition (99 ± 111), t(7) = 1.72, p = 0.130. Based on our presented results, it could be concluded that short-stories prepared for preschool children contained a theory-based system which could be reflected on early lexical development-supported by the fact that there was a clear word-class classification and systematised system for the distribution of word-class in terms of content words (over majority) and function words (minor representation) yet semantic categorisation. Also, word-class frequency and/or occurrence could be used as a predictor simulating child’s early lexical development-supported with the reached strong positive correlation between the two correlated variables.