Quantitative text analysis in Geography: facilitating access and fostering collaboration

Abstract

Quantitative text analysis can support researchers working with a large number of documents. Corpus linguistic methods are already employed by geographers in the context of discourse studies, and recent discussions about big data and digital geographies point to a potential broadening of their application. However, building a corpus and adapting to existing data analysis tools can be challenging. In this paper, we outline possible steps towards collaborative quantitative text analysis through the use of computational methods and corpora that can be incorporated into a variety of research settings. We summarise key steps for creating annotated corpora from text sources using state of the art methods and tools. Using the open source software Corpus Workbench (Evert and Hardie 2011) and CQPweb (Hardie 2012), we provide a platform to access corpora and corpus analysis functionality via a web interface. We invite researchers to use this existing infrastructure for corpus linguistic methods in their teaching and research, and to collaborate in making interesting material available to the geographic research community.