Features of lexical bundles in social science research articles of native and non-native expert writers of English
Lexical bundles are combinations of three and more words which co-occur most frequently in a given register or genre (Biber et al., 1999). They have been the focus of attention of many researchers in academic written discourse in the last two decades. In spite of the fact that lexical bundles are us...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2016
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/65638/1/FPP%202016%2022%20excludes%20chapter%201.pdf http://psasir.upm.edu.my/id/eprint/65638/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Lexical bundles are combinations of three and more words which co-occur most frequently in a given register or genre (Biber et al., 1999). They have been the focus of attention of many researchers in academic written discourse in the last two decades. In spite of the fact that lexical bundles are used much frequently in academic written discourse, it appears that non-native writers have difficulty in the application of these multi-word expressions, particularly in producing writing in their specific scientific field. Having evidence of their significant role in shaping the academic texts, it is worth exploring the usage of lexical bundles in the academic texts, particularly in a specific academic discipline. The present study adopted a corpus-based approach and investigated the usage of lexical bundles in social sciences research articles written by native and non-native Malaysian expert writers in the field. Moreover, the ultimate goal of this study was to develop and offer a useful pedagogical list of lexical bundles for social sciences academic writers. In the current study, in order to examine the use of lexical bundles by native and nonnative Malaysian social scientist writers, discipline-specific corpora, namely, Social Sciences Corpus (SSC) and Malaysian Corpus of Social Sciences (MCSS) were created. The number of tokens for native and non-native Malaysian corpora is 2,007,037 and 325,164 running words, respectively. The corpora were compiled from high-impact leading and prestigious international journals in the field. Later, AntConc 3.4.3 (Anthony, 2014) software was used in order to extract target bundles from the Social Sciences Corpus (SSC). The concordancer tool was also employed on the data to help the qualitative analysis to examine lexical bundles corresponding contexts and determine the specific functions they perform. Moreover, the two modified versions of analytical frameworks, developed by Biber et al. (1999) and Hyland (2008a), were applied in the current study to classify the bundles according to their structural and functional characteristics. Furthermore, a series of data refinement criteria were set out by the researcher to create a meaningful, concise and teachable list of lexical bundles and also to limit the number of the lexical bundles for further analysis in terms of their grammatical and pragmatic features. It is worth noting that these criteria were established as a methodological decision to meet the purpose of the current study. The primary findings of the study indicated that most of the target bundles in the SSC were made up of three-word bundles, and the bundles belonged to a particular disciplinary discourse. In terms of the structural analysis, the results showed that the native social scientist writers show a greater tendency to use more noun and propositional phrases and verb or adjective phrases with „to‟ clause fragments in their academic written discourse. The functional analysis revealed that the native social scientists writers used a wide range of research-oriented bundles to describe the content and procedures of their research in detail and also to show that their interpretation is based on empirical results and that the analytical method could be reliably replicated. Overall, the structural and functional analyses led the researcher to conclude that native social scientist writers have their own distinctive characteristics to describe their methodological decisions and argue their ideas based on the existing studies and the results obtained from data. A comparative analysis between native and non-native corpora also revealed that non-native Malaysian social scientist writers overused and underused target bundles compered to their native counterparts. The results suggest that non-native writers are not able to produce a varied set of expressions. Moreover, the results of structural and functional analyses of target bundles between native and non-native corpora showed few considerable differences as far as the use of lexical bundles were concerned. However, it is interesting to note that non-native Malaysian social scientist writers used much less frequent participate-oriented bundles compared to their native counterparts. The present study had some pedagogical implications for EAP course instructors and designers to use the useful list of lexical bundles in the class and integrate them into their ESL/EFL curricula. The findings of this study could also be beneficial for writing instructors in providing the teaching materials to raise the novice and nonnatives‟ awareness of appropriate use of lexical bundles to achieve a more native-like style of academic writing. |
|---|