Corpus-based Investigation of the Listening Input in Young Beginners' English Textbooks

The quality of listening materials for young beginners in EFL settings is an area in need of research. This paper investigates language representation in the listening components of two young beginners' textbook series. The listening scripts were primarily examined in terms of the usefulness of lexico-grammatical items which is inferred from a native speaker corpus of child and carer spoken English, Childes-En. The listening scripts of the two textbook series were made into two separate corpora (the KTLC and the OTLC), and lexico-grammatical under- and over- representations were inspected in comparison to Childes-En 0-12 using WordSmith Tools 5.0. It was found that features of spoken register are similarly restricted in the scripted listening of the young beginners' textbooks. Different under-represented items showed that frequency data from authentic texts are not an overriding principle in inventing listening scripts for young beginners. Over-represented items indicated that language representation is greatly influenced by the underlying viewpoints on language acquisition or the organizing principles. The findings derived from the analysis of lexico-grammatical representation in young beginners' textbooks were discussed in relation to language acquisition and pedagogy.