ABSTRACT In this study, we compared the cross‐modality effects of three parallel vocabulary measures (including orthographic‐receptive, phonological‐receptive, and phonological‐productive levels tests) as well as the contribution of the first five 1000‐word frequency levels in each measure to second/foreign language (L2) listening comprehension. In addition, we sought to identify the vocabulary measure that most accurately discriminates L2 learners’ listening proficiency (i.e., B1 Level on the Common European Framework of Reference for Languages) on the Test of English for International Communication (TOEIC). A convenience sample of 343 university‐level L2 learners completed one of the three parallel vocabulary levels tests and an official version of the TOEIC listening subtest to assess their vocabulary knowledge and their L2 listening proficiency. Multiple regression analysis as well as Pearson correlation and chi‐square tests were conducted. The results revealed that the phonological meaning‐recognition vocabulary test had the strongest correlation with L2 listening comprehension. Although the different frequency levels of vocabulary knowledge in each measure had distinct contributions to L2 listening comprehension, vocabulary knowledge of words in Level 2 (i.e., the second 1000‐word frequency level) had a higher relative predictive value in all three vocabulary measures. In addition, the phonological meaning‐recognition test taps different types of vocabulary knowledge, rendering it a broad vocabulary measure. It could also discriminate the participants’ listening proficiency on the TOEIC more accurately. Thus, this phonological‐receptive vocabulary levels test is a more suitable and accurate aural vocabulary measure for L2 listening comprehension than the other two parallel vocabulary measures. The findings inform researchers about the central role of phonological‐receptive vocabulary in L2 listening and guide practitioners and test developers to adopt aural vocabulary measures for more accurate assessment and diagnosis of listening proficiency, especially for discriminating intermediate‐level learners.
Chenhong Li (Thu,) studied this question.