Improving High School Students’ Performance on “Discourse Structure” Tests Through Instruction of Text Structure and Think-aloud Modeling

• Main Authors: Lting Chen, 陳俐婷
• Other Authors: Vincent W. Chang
• Format: Others
• Language: en_US
• Published: 2003
• Online Access: http://ndltd.ncl.edu.tw/handle/10569640690954696785

碩士 === 國立臺灣師範大學 === 英語研究所 === 91 === This thesis explores the nature of the “Discourse Structure” (DS) test and the role awareness of text structure plays in taking the DS test. An integrated treatment of instruction of text structure and think-aloud modeling followed by think-aloud practice was administered to see its effect on students’ performance on the DS test and on their metacognitive awareness. Sixty-nine female senior high school students participated in this study and were grouped as High, Mid, and Low according to their English proficiency levels. They received an instruction of text structure, which was based on Halliday and Hasan’s (1976) model of cohesion and Chang’s (1997) list of grammatical cohesive ties. Then, a recorded think-aloud modeling designed by the researcher was played to them both to make concrete to them the invisible thinking process in taking the DS test, especially strategies involving the use of cohesive ties as clues, and to familiarize them with the think-aloud technique. Afterwards, the participants practiced with four DS tests through think-alouds. Meanwhile, four participants with different English proficiency had their think-aloud protocols recorded for analysis. Besides the treatment, the participants took pretests and posttests in the DS test mode, answered a metacognitive awareness questionnaire both before and after the treatment, and gave responses to questions concerning the effect and application of the treatment. Several findings are revealed from the analysis of the results. First, it is found that there is a significant difference between scores of pretests and posttests for each group, indicating that all the participants have benefited from the treatment in taking the DS test. The test results also show that the treatment has a more facilitative effect on the performance of higher-proficiency participants on the DS test. Second, the treatment has enhanced the participants’ metacognitive awareness of text structure in taking the DS test, especially the use of cohesive ties such as pronouns, demonstratives, lexical ties and conjunctions as clues. Third, the analysis of think-aloud protocols suggests that proficient readers are more sensitive to prominent cohesive ties in text and are more capable of correcting misjudgment of cohesive ties as clues. Finally, most of the participants have displayed their recognition of the value of the treatment in promoting their confidence in taking the DS test, ranking the instruction session as the most effective. The results suggest that the integrated treatment administered in this study can be a viable approach to help improve students’ performance on the DS test, and that passages in the DS test mode, in turn, can serve as excellent teaching material to enhance text structure awareness. It should be noted, however, that in implementing the think-aloud technique, students’ individual thinking mode and learning approach should be taken into consideration to avoid potential rejection.