In a comparison of generalised linear mixed-effects models, generalised linear mixed-effects model trees and random forests, the author applies the three methodologies to a binary variable from the field of interactional pragmatics, the choice between filled and unfilled pauses across varieties of English represented by components of the International Corpus of English. Based on a large number of examples annotated for linguistic and extralinguistic factors the steps and decisions involved in the analyses are demonstrated. Though different in essence, the three resulting models share central trends. A more fine-grained evaluation of results and interpretations shows, however, that the three approaches differ in their systematicity of handling multiple observations from the same source, in that only the mixed-effects models explicitly account for and systematically partial out the relatedness of data points contributed by the same speaker. As to the way the approaches balance researcher involvement and control of the outcome, the approaches also differ substantially. A modelling choice can thus lead to notably different perspectives on an identical set of data and variables.

Setting the agenda for the volume, this introduction amalgamates the so far relatively isolated strands of research into genderlectal variation and World Englishes, relying on state-of-the-art empirical approaches. As they apply to the vast majority of speakers of English around the world, the notions of English as a second language and English as a foreign language are introduced and – in this light – recent attempts at bridging this paradigm gap between these two speaker groups as well as the models employed in these attempts are briefly discussed. For the study of gender and language, the central pillars of its most prominent theoretical waves – the dominance, the difference and the social construct framework – are presented and the corresponding methodological approaches critically appraised. Against this background, it is concluded that responsible explorations of genderlectal variation in World Englishes need to be based on transparent empirical foundations – both in terms of datasets and statistical modelling. For this reason, the tenets of corpus linguistics are explored and the benefits of multifactorial statistical techniques as consistently applied in this volume are illustrated. After previews of the individual chapters in the volume, the introduction ends with summarising remarks including the moderator function of gender in World Englishes.

Women have often been profiled as prototypical users of hedges, i.e. linguistic devices such as I believe lowering the pragmatic force of a statement to potentially save interlocutors’ faces. Still, empirical investigations of gender-preferential hedging as employed by learners – specifically in postcolonial territories – are not available. This study establishes corpus-linguistically a) whether men or women use more hedges in native-speaker and postcolonial learner contexts, b) what factors determine hedge choice and c) on a theoretical level, the relation between learners and the evolutionary progress of their postcolonial habitat. A total of 1,530 hedges are extracted from texts by British native speakers and by learners (maximally level B1 in the Common European Framework of Reference for Languages) from Hong Kong, the Philippines and Singapore. Males use more hedges in Britain and Singapore, while female learners employ more hedges in Hong Kong and the Philippines, but the concrete hedge chosen is determined by region – with Singapore being notably different from other territories – mode and gender. More generally, the findings suggest that speaker status differences, i.e. whether speakers are second-language or foreign-language users, may be less important in explaining linguistic choices than the evolutionary status of their sociolinguistic habitat.