Doing sociolinguistic research on computer-mediated data: A review of four methodological issues

• No single methodological approach to CMC data is per se better than any other.
• Methodological challenges arise as a result of the diversity of online language use.
• Ethical-decision making no longer relies on the dichotomy of public and private.
• Scholars researching language use online increasingly mix methodologies.

This article focuses on four methodological issues which raise challenges for sociolinguists working with online data: (1) ethics; (2) multimodality; (3) mixed methodologies and the relationship between online and offline settings; and (4) web corpora and annotation. While there are currently numerous publications dealing with questions of ethics, data and methodology from within communication studies and social scientific research more generally, there are only a handful of publications which specifically focus on empirical linguistic research. In addition to delineating the diversity of computer-mediated data, in the course of the article we review each of these methodological issues in turn, thereby discussing key terminology and reviewing relevant literature.