ABSTRACT
This chapter explores patterns of micro-level linguistic variation in a corpus of Facebook and Twitter posts using the dimensions of web registers from Titak and Roberson (2013). We apply a multi-dimensional analysis to a web registers corpus of topic-coded, public Facebook and Twitter posts collected from 2011 to 2016 (N = 5,331,120 words). The texts within each domain are grouped into six categories: Personal, Politics, Business, Weather, Entertainment, and Sports. Results show notable structural differences between Facebook and Twitter postings on the whole (as two linguistically distinct registers) and also across topics within and between these two registers.
