A Study Of Chinese EFL Learners' Use Of Filled Pause In Their Comment-making Discourse In A Testing Context

This thesis reports a study on the use of Filled Pauses by the Chinese EFL learners in their comments on a given topic in a testing situation. It seeks to describe the general characteristic of the use of Filled Pauses, and find out the difference in the use of Filled Pauses between the test-takers of different oral proficiency, and finally to discover the predicting power of the different types of Filled Pauses to L2 learners'oral performance in a testing context.The materials employed in the study were taken from SWECCL, a Chinese EFL learner corpus. The sampled 60 cases were divided into high- and low-level groups according to the scores of the comment-making the test-takers obtained in the TEM-8 oral test. In the data analysis, an independent-samples t-test and a regression analysis were conducted. The major findings yielded in the study may be summarized as follows:Firstly, according to linguistic context in which Filled Pauses occur, there are three types of Filled Pauses, namely, Repairing Pauses, False-starting Pauses, and Plain Pauses. Functionally, Repairing Pauses can be further divided into three subcategories: Repairing Pauses accompanied by repetition, self-correction and restart.Secondly, of the three major types of Filled Pauses, the type of Plain Pause was most frequently used and False-starting Pause least frequently, suggesting that the test-takers tended to use Plain Pause as a strategy of searching for a proper choice of words or the ideas to be produced in the rest of the talk. The rare occurrences of False-starting Pause are likely to be related to the task requirements according to which the test-takers were given three minutes to prepare before starting the talk.Thirdly, the independent-samples t-test analysis reveals there was significant difference in the use of Plain Pause and Repairing Pause between the low- and high-level groups, but no significant difference was found in the use of False-starting Pause between the two groups, which suggests that the excessive use of Plain Pause and Repairing Pause reflects to some degree the low-level students'insufficient oral proficiency. The micro-analysis indicates that False-starting Pause was intended by both high- and low-level learners as a means of organizing the opening of the talk.Fourthly, the multiple regression analysis indicates that the use of Plain Pause may have the strongest predicting power to the test-takers'oral performance, with Restarting Pause enjoying the second strongest predicting power. This result confirms the points made previously. The use of Plain Pause and Restarting Pause may be used as indicators to the difference in oral proficiency between the learners of different levels.Pedagogically, the use of Filled Pauses can be potentially used as one of the effective means to measure L2 learners'oral performance. Moreover, a proper or natural use of Filled Pauses may help speakers to organize their ideas into a coherent integrity though they may sound a bit broken in their oral work.