This study presents the first multifactorial corpus-linguistic analysis of lexical diversity (LD) and coverage in a dataset comprising English textbooks and exam materials used by grade 1-3 primary school children (age 6;0-8;0) in Hong Kong. Very little is known regarding how far print exposure facilitates lexical development in this key stage, and its efficacy in preparing students for high-stakes examinations. To investigate this question, we analyzed the lexical variability (MATTR and MTLD) and bigram indices (T and MI), which reflect LD at the lemma and phrasal level respectively. We predict a positive correlation between these indices and grade level as learners progress. We extracted 72 textbook chapters from major publishers and 27 exam sections from the exam authority. Using Natural Language Processing (NLP) tools, we extracted the pertinent indices from each chapter/section. Controlling for text length and syntactic complexity, we fitted the data in linear mixed-effects models. The retained model shows a significant progression for MATTR and MTLD, and pairwise comparisons indicated no significant differences between P3 and exam materials. However, we also found a descending trend in bigram indices. The result of our lexical coverage analysis shows a low correspondence between textbooks and exam papers. Together, the findings suggest that not all aspects of vocabulary knowledge were considered in textbook design. A major implication is that learners are exposed to sub-optimal print materials, which may lead to gaps in vocabulary knowledge when they reach the next stage. Based on our findings, some recommendations for educators and practitioners will be.

Authors:
Arthur Kan, University of Cambridge, United Kingdom
Chris Law, The Chinese University of Hong Kong, Hong Kong
Stephen Matthews, The University of Hong Kong, Hong Kong
Virginia Yip, The Chinese University of Hong Kong, Hong Kong

About the Presenter(s)
Arthur Kan is currently a PhD student in the Department of Theoretical and Applied Linguistics, University of Cambridge, United Kingdom. He research interests include language acquisition, bilingualism, and corpus linguistics.

See this presentation on the full schedule – Monday Schedule

Bookmark