Effects of Written and Spoken Discourse Variations on Receptive Vocabulary in TOEFL iBT: Vocabulary Loads and Text Coverage of the Academic Word List

Abstract

Although there is a growing body of research into the vocabulary load of international academic English tests, methodological decisions in compiling a corpus vary. Among these disparities, treatment of written and spoken discourse differs markedly. Yet, little research has been undertaken to explicitly investigate the extent to which discourse variations impact receptive vocabulary in international academic English tests. The present study examined two effects of discourse variations on vocabulary in reading and listening passages from TOEFL iBT past tests. Text coverage in Reading, Listening, and Mixed Corpora was calculated via the computer program Range, respectively. The vocabulary load coupled with text coverage by the Academic Word List were compared across the three corpora. Results demonstrated that the vocabulary load of Reading Corpus was in stark contrast to that of Listening Corpus with a difference of 3,000 word families. In addition, there was a marked difference in text coverage by the Academic Word List between the reading and listening passages. Findings of the present study indicate that reading and listening tests should be separated in exploring the vocabulary load of academic English tests and validating wordlists, as the combined reading and listening passages yielded inaccurate estimates. Pedagogical suggestions stemming from the findings are discussed in detail.