The different ways to write publishable research articles: An exploration using automated language processing tools

The exploration of linguistic profiles of research articles (rAs) has been underrepresented in the existing literature. This study utilizes two automated language processing tools and a cluster analysis approach to explore linguistic features and variation of published research articles (N=360) in two hard science disciplines (i.e., Biology and Medicine). Findings show five different profiles characterized by their use of distinct combinations of linguistic features. The identified profiles not only vary between the two disciplines, but also within each discipline. The distinct profiles within each discipline represent the potentially different ways for researchers to write publishable research articles. The study fills the research gap and contributes to a new understanding of linguistic features and variation of RAS.