ABSTRACT
This chapter examines the linguistic features of Q+A forum responses through a multi-dimensional (MD) analysis methodology developed by Biber. Many corpus-based studies of cross-linguistic variation focus largely on isolated linguistic and functional features of texts. There have also been multiple studies making use of corpora in comparing and contrasting English registers and varieties of English. A complementary approach to this type of distributional research is to examine the overall linguistic profile of corpora using statistical tools without an a priori set of target features. Biber pioneered the MD analysis methodology, which classifies texts according to clusters of co-occurring linguistic features in a cross-register comparison. Biber's multi-feature, multi-dimensional analytical framework has been used in the study of a variety of spoken and written registers, including web-based texts. MD analysis data come from factor analysis (FA), which considers the sequential, partial, and observed correlations of a wide range of variables producing groups of occurring factors or dimensions.
