A Corpus-based Study On The Speech Vocabulary Used By Chinese EFL Learners

The past few years saw growing attention given to vocabulary acquisition and many insightful findings on this issue. Most researchers have perceived the productive use of vocabulary in L2 writing, while the studies on vocabulary use in L2 learners'oral production are less common. The present study employs Contrastive Interlanguage Analysis (CIA) to probe into the speech vocabulary used by Chinese EFL learners with the assistance of Spoken English Corpus of Chinese Learners (SECCL). The analysis is conducted from four perspectives:vocabulary size, lexical distribution, lexical coverage and coinage, which are also discussed in relation to Chinese EFL learners' oral proficiency. The following questions are addressed:1. What is the vocabulary size of the Chinese EFL learners'speeches, and how is it related to their general oral proficiency? 2. How is the speech vocabulary distributed in the oral production of Chinese EFL learners? 3. Do lower level speakers tend to use high frequency words more exclusively? 4. Does lexical coinage phenomenon exist in the speeches of Chinese EFL learners?To answer the above questions, all the samples from 2000 to 2002 were divided into three groups, i.e. Level 1, Level 2, Level 3, according to test-takers'performance. Then from each of the three groups,10 samples in 2000,2001 and 2002 respectively were selected randomly. From Level 1 to Level 3, a hierarchy concerning speakers' general oral proficiency from low level to high level was formed. Statistics was generated after the data was processed by software Wordsmith, Range and Frequency. Basic lexical information was obtained by Wordsmith and compared among the three groups to answer the first question. LFP (Lexical Frequency Profile) generated by Range was examined to answer the second question. To answer the third question, the coverage of top high frequency words was compared based on the lists generated by Frequency. Answering the last question required identifying, categorizing the coinages in each sample.The major findings are:1. The number of tokens used by Chinese EFL learners is about 264 per person in the 3 minute speaking task. The number of both tokens and types steadily increases as speakers'proficiency levels go up.2. Over 95% of the words uttered by students are distributed in word list one and two. The difference in LFP among speakers of different proficiency levels is not so significant.3. The top 1000 high frequency words cover over 99.9% of low level speakers'speech vocabulary, but the use of high frequency words is not directly related to learners' general oral proficiency.4. Lexical coinage does exist extensively in the students' speeches but there is no necessary connection between the proficiency level and frequency of coinage.This study has illuminated many noteworthy details about the speech vocabulary used by Chinese EFL learners. The findings are expected to give some hints on classroom vocabulary instruction. For example, because of students'lack of lexical variety and sophistication as well as limited ability to use low-frequency words, language teachers may encourage students to use more advanced and newly-learned words.