Keywords, time and multiple-correspondence analysis: exploring press discourse over time
In corpus linguistic research, keywords can be understood as words which occur with a significantly higher frequency in the corpus under analysis in comparison with another corpus. As a technique, keywords is still relatively new, with work using it only beginning on any scale with the development of Wordsmith Tools in the late 1990s (Scott, 1996). While there is little doubt that the technique has proved helpful in corpus studies of discourse, there is still scope for its development.
Accordingly this talk is centrally concerned with one aspect of keyword analysis that has been relatively neglected in this development work – change over time. I will explore what value the analysis of keywords can bring to the study of temporal variation in discourses by focusing on a topic that has received sustained attention in CADS: the representation of Islam and Muslims in the UK press. I will review a new technique used to derive keywords, based on Multiple Correspondence Analysis (MCA) and will show the potential of the new technique to control, guide and provide a more fine-grained, keyword-driven analysis of discourse through time.
Tony McEnery is Distinguished Professor of English Language and Linguistics at Lancaster University and Changjiang Chair at Xi’an Jiaotong University, China. Tony was previously Director of Research and Interim Chief Executive at the UK’s Economic and Social Research Council (ESRC), Director of Research at the UK's Arts and Humanities Research Council, Dean of the Faculty of Arts and Social Sciences at Lancaster University and Director of the ESRC Centre for Corpus Approaches to Social Science (CASS) at Lancaster. He has published extensively on corpus linguistics.