Abstract

In this paper, two word association (WA) studies are presented in support of recent arguments against the use of native-speaker (NS) norms in WA research. In Study 1, first-language (L1) and second-language (L2) WA norms lists were developed and compared to learner responses as a means of measuring L2 proficiency. The results showed that L2 norms provided a more sensitive measure of L2 lexical development than did traditional NS norms. Study 2 was designed to test the utility of native norms databases in predicting the primary WA responses of Japanese learners to high-frequency English cues. With the exception of only extremely frequent cues, it was shown that native norms were not successful in predicting learner responses. The results of both studies are discussed in terms of cultural and linguistic differences, geographic distance, and dissimilarities in word knowledge between respondent populations. Finally, a proposal is made for the construction of a Japanese WA database of English responses (J-WADE). The methods by which it will be developed, key features, and employment in future research are outlined.

Suggested citation