A corpus-based analysis of lexical bundles in english introductions of chinese and international students’ theses

1Citations
Citations of this article
11Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Phraseology is attracting an increasing number of linguistics researchers’ attention. Based on the data from our self-made corpus, this paper aimed to make a comparative analysis of the lexical bundles in the English introductions of Chinese graduate students’ theses and those written by the graduate students in the international prestigious universities. The most frequently used four-word lexical bundles in the corpus were identified and classified structurally and functionally and corresponding analysis was made. Results indicated that lexical bundles in the English introductions of Chinese students’ theses were structurally incomplete in comparison with those in the international students’ papers, and that lexical bundles were less varied in form, meaning and function in Chinese students’ English thesis introductions despite the larger number both in type and token. It is necessary to strengthen the richness, diversity and fluency of lexical bundles in them.

Author supplied keywords

Cite

CITATION STYLE

APA

Guiling, N. (2015). A corpus-based analysis of lexical bundles in english introductions of chinese and international students’ theses. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9332, pp. 486–493). Springer Verlag. https://doi.org/10.1007/978-3-319-27194-1_49

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free