Abstract/Summary

The thesis focuses on the individual differences which can explain the variability in listening comprehension among adult Chinese learners of English. Although there is now a large number of studies on listening comprehension among second language (L2) learners, individual differences in listening have received less attention than individual differences in reading. The thesis aims to fill some of the gaps in our knowledge in this field of research. Listening comprehension was measured with two different tests (the College English Test Band 4 listening section and the Cambridge Preliminary English Test listening section). Four groups of explanatory variables were included in the study: linguistic knowledge, sentence processing speed, cognitive factors and learners’ use of English in daily life. Structural equation models of listening comprehension were built based on Andringa, Olsthoorn, Van Beuningen, Schoonen and Hulstijn (2012). This model was tested among 187 Chinese learners of English (one group in China, N = 147; the other group in the UK, N = 40). The results indicate that the listening comprehension of learners in China was significantly lower than that of Chinese learners in the UK. Linguistic knowledge, frequency of English use in daily life and aural sentence processing speed were key predictors of listening in these groups. Phonological knowledge was the most important predictor of listening comprehension among the variables measuring linguistic knowledge and word recognition from speech explained variance in listening comprehension over and above the contribution of word segmentation from speech. When the two groups of learners were considered separately word recognition from speech was the most important predictor for learners in China whilst for learners in the UK, learners’ grammar knowledge and the reasoning ability were key. Finally a comparison of both listening tests revealed that the listening section of the CET4 only measures learners’ ability to comprehend information explicitly available in the text and not inferencing skills. Pedagogical implications for teachers and learners of English and test developers in China and in the UK are provided based on these findings.