Lexical Profiles of Pakistani Academic Writing: A Corpus-Based Study of STEM and SSH Postgraduate Research Genres

Abstract

This article is a corpus-based study of lexical profiles of the Science, Technology, Engineering and Mathematics (STEM) postgraduate research papers versus Social Sciences and Humanities (SSH) in Pakistani universities. A corpus of 240 doctoral and MPhil dissertations was aggregated in a specialised collection of 240 dissertations of 12 Pakistani Higher Education Commission (HEC)-recognised institutions and amounted to 8.2 million words. Data analysed using the AntConc 4.0 and WordSmith Tools 8.0 identified and described lexical bundle, presence of academic vocabulary and genre-specific lexical patterns in terms of Biber et al (1999)- structural taxonomy and Hyland (2008)- functional framework. Quantitative analysis showed that there is a significant difference in disciplinary variation: STEM texts are more frequent in procedural and quantifying bundles and SSS discourses are more frequent in using evaluative and positioning lexis. Coverage analysis of Academic Word List (Coxhead, 2000) indicates a difference in the distribution with STEM genres using 23 per cent more items of sublist 1 in methodological sections. Lexical specifics of the Pakistani culture appeared, such as the culture-bound formulaic patterns and institutional phraseology that represents the local academic standards. Findings reveal that postgraduate authors in both fields do not use stance adverbials as much as expected of native speakers (Hyland, 2012). The results are relevant to the English academic purpose (EAP) pedagogical models in Pakistan and curriculum design to develop postgraduate writing programme. This study fills a critical gap in the discourse analysis of South Asian academic literature and offers empirical findings regarding the use of instruction in discipline sensitive writing.

Keywords: Pakistani English, lexical bundles, academic vocabulary, STEM discourse, SSH writing, corpus linguistics, postgraduate research, genre analysis.