英语论文网

.4 Data coding and analysis .........91-96
Chapter 4 Results and discussion: Use of epistemic devices.........96-135
    4.1 Overall distribution......... 96-101
        4.1.1 Overall frequency and variety of devices......... 96-98
        4.1.2 Grammatical distributions .........98-99
        4.1.3 Categories of epistemic devices .........99-101
    4.2 Distribution of boosting epistemic devices......... 101-109
    4.3 distribution of hedging epistemic devices .........109-128
    4.4 Discussion .........128-135
Chapter 5 Results and discussion: Development.........135-175
    5.1 Variations in the use of epistemic devices.........135-155
    5.2 Variations of performance between hig.........155-168
5.3 Discussion .........168-175

Conclusion

While the present study provides some theoretical, methodological and pedagogicalimplications for ILP research in particular and SLA in general, there are still somelimitations. These limitations are represented in the comparability of the two corpora, therepresentativeness of the data and the subjects, and the identification of epistemic devices.One limitation is from the inherent nature of contrastive interlanguage analysis.Because the great majority of learner corpora are monolingual in nature, researchers requirean external NS control corpus in order to conduct contrastive learner corpus analysis. Thisprocedure is problematic, however, because it means that the data to which the learnerproductions are compared are produced at a different point in time, under differentcircumstances, and in different contexts (Cobb, 2003; Granger, 1998a; Granger & Tribble,1998). The contextual disparity between a learner corpus and an external NS control corpusis especially prejudicial with regard to pragmatic competence because such competencegenerally is defined as language use in social context where context shapes use (Kasper &Rose, 2002). Although we have tried to minimize the disparities, such a condition issomewhat inescapable in the present learner corpus-based studies.
Generalizations cannot be made about how Chinese students modify claims withEnglish epistemic devices since the participants do not represent the whole populations.Because the collection of the data took place in a university setting, it lacked randomselection and random assignment of participants to groups. In addition, the participantswere university students majoring in EFL, and the findings may vary if applied to othergroups of participants. Considering the limitation in the representativeness of the data andsubjects, the findings yielded from this study are open to future confirmation or refutation.Another concern has to do with the identification of epistemic devices in the twocorpora. Since the study involves completely manual tagging of the devices identified, theprocess of which is to some extent subjective. Sometimes, it is hard to judge whether anitem express epistemic meanings, particularly the modal verbs, therefore, such anidentification is inevitably somewhat arbitrary in some cases. Besides, all the tagging workis done by the author and a co-worker, and no native speaker expert interpr本论文由英语论文网提供整理，提供论文代写，英语论文代写，代写论文，代写英语论文，代写留学生论文，代写英文论文，留学生论文代写相关核心关键词搜索。