Effects of speech modification, prior knowledge, and listening proficiency on the lecture listening comprehension of Chinese EFL (English as a Foreign Language) students

This thesis reports the results of an empirical study to investigate the effects of speech modification, prior knowledge, and listening proficiency on the comprehension of lecture information by Chinese EFL (English as a Foreign Language) students.;In the study, intermediate-level EFL students were divided into two listening proficiency levels on the basis of scores achieved on the CELT (Comprehensive English Language Test) with the median serving as the cut-off score for high-intermediate-listening-proficient level (HILP) and low-intermediate-listening-proficient level (LILP). Both HILP and LILP students were randomly assigned to one of four groups and asked to listen to a lecture the discourse of which was (1) familiar-unmodified, (2) familiar-modified, (3) unfamiliar-unmodified, or (4) unfamiliar-modified. Five minutes after the lecture presentation, subjects took a multiple-choice comprehension test on the information presented in the lectures.;The test data were subjected to analyses of variance with repeated measure (ANOVR). The results showed that both the HILP and LILP EFL students' scores were higher when listening to the familiar-topic lecture than when listening to the unfamiliar-topic lecture. A significant interaction between speech modification and listening proficiency indicated that the HILP students comprehended better than the LILP students in the modified and unmodified lectures, whether the lecture-topic was familiar or unfamiliar. However, the redundancy of information helped improve the lecture comprehension of the HILP listeners, but not the LILP listeners.;The study also examined the effect of prior knowledge on two different types of test items: (1) passage-independent and (2) passage-dependent items. The results suggest that a test consisting of both types of items may be a test which better measures the integration of a listener's prior knowledge and the new information contained in the lecture discourse than a test consisting only of one test type. Implications of the findings for the development of curriculum materials and tests for English for academic purposes in an EFL environment are reported, and limitations of the study and suggestions for further research are also presented.