What Does “Research Say” on COVID-19? Data Driven Linguistic Analysis of Research Articles

Science production related to COVID-19 has increased exponentially in recent months following the pandemic outbreak, yet little has been done to investigate this huge science production from a linguistic and data-driven perspective. The research answers the following questions: What does the term “coronavirus” collocate with? and what does language tell us about the points of focus of science production on Coronavirus in general? Data for this research consisted of a large corpus of research articles that were published as part of the COVID-19 Open Research Dataset (CORD-19). Covid-19 corpus has 224,061,570 words and 50,754 documents. The analysis took a rigorous data driven approach in investigating linguistics phenomena - keyword and collocation analyses of science discourse of COVID-19. Statistic scores reported frequency of occurrences and strength of collocates. Findings showed that early science production focused on naming, describing, classifying the virus. Another point of focus is the spread of the virus. Also, findings have also noted speculation about the origin of the new virus. Science production of research investigated behavior of the virus, the life cycle of the virus and its diagnostic virology. In general, the findings are expected of science research carried out to solve a problem. As the data was collected May 2020, most research has focused on knowing more about the nature of the problem. Findings have implications for understanding in-depth points of focus in research regarding COVID-19 at the early stage of science production.