Early vocabulary development in English, Mandarin, and Cantonese : a cross-linguistic study based on Childes

Early language development is an exciting topic in the field of child language acquisition. Only a limited amount of cross-linguistic studies has attempted to investigate the similarities and differences in child language development across different languages. In this thesis, I present a study based on English, Mandarin and Cantonese corpora extracted from the Child Language Data Exchange System (CHILDES, MacWhinney, 2000). I investigated the lexical compositions of certain lexical categories (nouns, verbs, and adjectives) in children and their caregivers’ vocabularies across eight different children age groups ranging from 13 to 60 months. ANOVA, frequency analysis, and cluster analysis were used to analyze the data. The development trajectories of lexical diversity and complexity of children’s speech were also analyzed by two novel techniques: Dmeasure and the Mean Length of Utterances. My research clearly shows that (1) in all the cultures, children’s early language development exhibits roughly similar patterns: an increasing diversity in lexicon and increasingly complicated speech patterns emerge as a function of time, and children’s vocabularies become more similar to those of their parents over time; and (2) culture variations in children’s linguistic input have strong influences on their language output, which is reflected in the noun vs. verb ratio and the varying percentages of nouns, verbs, and adjectives in the total words children are able to speak in the three cultures. I certify that I have read this thesis and find that, in scope and quality, it satisfies the requirements for the degree of Master of Arts.