Options
Metacognition, cognition and L2 reading : a study of Chinese university EFL readers' metacognitive knowledge and strategy deployment
Loading...
Type
Thesis
Author
Zhang, Lawrence Jun
Supervisor
Hvitfeldt, Christina
Skuja-Steele, Rita
Abstract
This thesis reports on an exploratory study of the metacognitive knowledge systems of mainland Chinese university EFL readers and their links to perceived and actual reading strategy use in relation to EFL proficiency and reading comprehension across academic major and gender. The study, which involved 312 subjects, adopted a triangulated approach which used on-line think-aloud protocols, questionnaires, and retrospective interviews. The results suggest that previous studies have not addressed Chinese EFL readers adequately and furthermore , challenge the claims of some Western writers that Chinese EFL readers do not have necessary conceptual strategies and do not know how to use processing strategies.
The Chinese EFL readers in this study reveal their metacognitive knowledge on a tripartite dimension: person, task and strategy. They are aware of their roles as readers, of the demands of different tasks and of the effectiveness of good and deep strategies. This generally falls within the framework of Flavell (1979/1992) and Wenden (1991a). This awareness, however, rests on their much higher language proficiency and stronger reading ability in their first language. Evidence from the data suggests that their metacognitive knowledge is stable, statable, retrievable, utilisable and occasionally fallible, but their flexible deployment of this metacognitive knowledge is closely related to the readers' L2 proficiency levels and task requirements. While the link between metacognitive knowledge and the L2 readers' learning outcomes is found to be indirect, this metacognitive knowledge seems to have direct control over their deployment of reading strategies in actual reading situations. The readers report, however, that they have more strategies in their mental stores than they actually use finishing the two reading tasks, which suggests that having metacognitive knowledge does not always mean the deployment of this knowledge.
Chinese EFL readers seem to have a unitary view of reading in EFL and reading in their first language; that is, the goal of reading is for meaning-construction and reading in L1 and L2 shares certain similarities in terms of the processes involved and the deployment of strategies in processing the written input, though requirements vary according to text types and contents. This metacognitive knowledge helps higher-proficiency EFL readers to shape and modify their conceptual processing strategies used for reading comprehension in EFL. The poor readers, restricted by their lower L2 proficiency, find it difficult in real reading situations to activate the metacognitive knowledge that can direct the use of strategies generally regarded as effective. An EFL reader ability continuum exists and individual differences are salient features in these readers who are distinguished from one another by many person-variables. In general, global (good) readers and local (poor) readers appear to be different not only in their flexible choice of strategies but also in their concentration spans. The frequency with which the readers use local strategies decreases as their proficiency level rises. Global readers use strategies more effectively and generally, their use of strategies is less frequent, which is in contrast with the low-proficiency readers whose use of these strategies is more frequent, especially with regard to strategies that are cognitively less demanding. One-way ANOVA results from the EFL-RSI also suggest that proficiency levels are a determinant of Chinese EFL readers' effective strategy deployment for meaning-construction.
Results show that seven of the thirteen clusters of strategies have statistically significant correlations with the subjects' EFL learning achievement (p<0.5). Multiple regression analyses suggest that two clusters of strategies, "vocabulary detailing" and "guessing at unknown words", and one reader background variable, "self-evaluation of their reading ability in L2", significantly predict their EFL achievement (R2=.219, F=4.079, p<.001). Further analyses of the interviews and the think-aloud protocols reveal that the degree of flexible use of specific strategies by the high-proficiency and low-proficiency groups in accomplishing the reading tasks is different in the two groups. Only one strategy is found to be typical of poor readers, that of repeating words and sentences over and again. Reader schemata are found to have varying degrees of impact on the readers' levels of comprehension, but there is a statistically significant correlation for the linguistically less challenging passage when the correlations between the readers' background knowledge scores and their recall task performance are calculated (r=.3912, p<.05). Variations in recall performance also separate good/global readers from poor/local readers in that the former's recalls are more integrative and the latter's more form-based and non-integrative in nature.
Although t-test results show that male readers reported more frequent use of four clusters of strategies than their female counterparts, h2 results show that the associations of strategy choice with gender are weak. Nevertheless, the think-aloud and the interview data suggest that male readers appear to have more schemata or background knowledge than do females in the two specific reading tasks used in the study, which coincides with research findings in second language reading research (e.g., Young & Oxford, 1997). The common belief that students in arts fields are better language learners than those in science fields does not seem to be endorsed by this study. The thesis also addresses the pedagogical implications of the findings and makes recommendations for further research.
The Chinese EFL readers in this study reveal their metacognitive knowledge on a tripartite dimension: person, task and strategy. They are aware of their roles as readers, of the demands of different tasks and of the effectiveness of good and deep strategies. This generally falls within the framework of Flavell (1979/1992) and Wenden (1991a). This awareness, however, rests on their much higher language proficiency and stronger reading ability in their first language. Evidence from the data suggests that their metacognitive knowledge is stable, statable, retrievable, utilisable and occasionally fallible, but their flexible deployment of this metacognitive knowledge is closely related to the readers' L2 proficiency levels and task requirements. While the link between metacognitive knowledge and the L2 readers' learning outcomes is found to be indirect, this metacognitive knowledge seems to have direct control over their deployment of reading strategies in actual reading situations. The readers report, however, that they have more strategies in their mental stores than they actually use finishing the two reading tasks, which suggests that having metacognitive knowledge does not always mean the deployment of this knowledge.
Chinese EFL readers seem to have a unitary view of reading in EFL and reading in their first language; that is, the goal of reading is for meaning-construction and reading in L1 and L2 shares certain similarities in terms of the processes involved and the deployment of strategies in processing the written input, though requirements vary according to text types and contents. This metacognitive knowledge helps higher-proficiency EFL readers to shape and modify their conceptual processing strategies used for reading comprehension in EFL. The poor readers, restricted by their lower L2 proficiency, find it difficult in real reading situations to activate the metacognitive knowledge that can direct the use of strategies generally regarded as effective. An EFL reader ability continuum exists and individual differences are salient features in these readers who are distinguished from one another by many person-variables. In general, global (good) readers and local (poor) readers appear to be different not only in their flexible choice of strategies but also in their concentration spans. The frequency with which the readers use local strategies decreases as their proficiency level rises. Global readers use strategies more effectively and generally, their use of strategies is less frequent, which is in contrast with the low-proficiency readers whose use of these strategies is more frequent, especially with regard to strategies that are cognitively less demanding. One-way ANOVA results from the EFL-RSI also suggest that proficiency levels are a determinant of Chinese EFL readers' effective strategy deployment for meaning-construction.
Results show that seven of the thirteen clusters of strategies have statistically significant correlations with the subjects' EFL learning achievement (p<0.5). Multiple regression analyses suggest that two clusters of strategies, "vocabulary detailing" and "guessing at unknown words", and one reader background variable, "self-evaluation of their reading ability in L2", significantly predict their EFL achievement (R2=.219, F=4.079, p<.001). Further analyses of the interviews and the think-aloud protocols reveal that the degree of flexible use of specific strategies by the high-proficiency and low-proficiency groups in accomplishing the reading tasks is different in the two groups. Only one strategy is found to be typical of poor readers, that of repeating words and sentences over and again. Reader schemata are found to have varying degrees of impact on the readers' levels of comprehension, but there is a statistically significant correlation for the linguistically less challenging passage when the correlations between the readers' background knowledge scores and their recall task performance are calculated (r=.3912, p<.05). Variations in recall performance also separate good/global readers from poor/local readers in that the former's recalls are more integrative and the latter's more form-based and non-integrative in nature.
Although t-test results show that male readers reported more frequent use of four clusters of strategies than their female counterparts, h2 results show that the associations of strategy choice with gender are weak. Nevertheless, the think-aloud and the interview data suggest that male readers appear to have more schemata or background knowledge than do females in the two specific reading tasks used in the study, which coincides with research findings in second language reading research (e.g., Young & Oxford, 1997). The common belief that students in arts fields are better language learners than those in science fields does not seem to be endorsed by this study. The thesis also addresses the pedagogical implications of the findings and makes recommendations for further research.
Date Issued
1999
Call Number
PE1128 Zha
Date Submitted
1999