Loading...
Japanese EFL Learners' High-Frequency Written and Spoken Receptive Word Knowledge
Martin, Eric
Martin, Eric
Citations
Altmetric:
Genre
Thesis/Dissertation
Date
2025-08
Advisor
Committee member
Group
Department
Applied Linguistics
Permanent link to this record
Collections
Research Projects
Organizational Units
Journal Issue
DOI
https://doi.org/10.34944/a9t1-g381
Abstract
Among all words in the English language, those that appear frequently are of particular importance. Research suggests that receptive knowledge of the 2,000 to 3,000 highest-frequency word families provihdes second-language (L2) English learners with a foundational basis for understanding language produced in English writing and speech (Nation & Waring, 1997). Therefore, in EFL contexts it would be beneficial to understand which of these words learners have acquired through years of study.In this study I analyzed data collected from weekly vocabulary quizzes among primarily lower-proficiency university L2 English learners in Japan. The participants were assessed on their receptive knowledge of the 1,650 highest-frequency English flemmas (i.e., groupings of a word’s dictionary form, its inflected forms, and identical forms of different parts of speech) from the New General Service List (Browne et al., 2013), in both written and spoken forms. The data were fit to the Rasch model, which produced a hierarchy of words based on their difficulty estimates. Finally, I conducted backward stepwise regressions to develop multivariable models to predict which of these words were likely to be more- or less-widely known by the participants.
The results showed a hierarchy of three-to-four levels and four-to-five levels of difficulty among written and spoken words, respectively. Additionally, comparisons of word knowledge estimates between modalities, as well as learners’ comparative abilities to identify words when read and when heard, indicated that most learners identified words more often when read, although several learners also recognized words similarly when read or heard, and a small number performed better when words were heard. There was also substantial variation among words with similar frequency profiles. Within most 100-word bands, some words were universally known while others were universally unknown, with the number of known words gradually decreasing as the bands included less frequent words.
Among single-variable predictors, age of acquisition ratings, rather than frequency, explained most of the variance in written (36%) and spoken (33%) word knowledge estimates. Frequency values based on the SUBTLEX-US corpus frequency measures were second, accounting for 28% and 29% of the variance in written and spoken word knowledge estimates, respectively. Models incorporating additional variables, such as loanword status, first-language age of acquisition, and word length measured by the number of letters, explained 44% and 45% of variance in spoken and written word difficulty, respectively. Finally, a model that included only word-intrinsic variables, such as length and concreteness, accounted for nearly as much variance in written word knowledge estimates (41%) but less in spoken word knowledge estimates (36%).
This study provides the first comprehensive high-frequency word profile of Japanese university L2 English learners and advances the development of predictive models of L2 word difficulty. The insights gained should benefit learners, teachers, and material designers in Japan by identifying high-frequency words that might warrant attention even after years of study. Because few learners had mastered the highest-frequency words, the results also highlight why Japanese EFL university learners might struggle with materials designed for L2 English learners, such as graded readers or listening exercises.
Description
Citation
Citation to related work
Has part
ADA compliance
For Americans with Disabilities Act (ADA) accommodation, including help with reading this content, please contact scholarshare@temple.edu