In recent years, the representation of gender across different media has been a heated topic of feminist scholarly debate (for example, see Baker, 2014; Coffey-Glover, 2015; Gupta, 2016; Jaworska and hunt, 2017). In 2014 the online movement ‘gamergate’ brought attention to the representation of gender in video games (for a full overview of the events, see Massanari, 2017). However, many of the scholarly investigations of gender in video games have relied on visual content analyses (such as Beasley and Collins-Stanley 2002; Burgess et al., 2007; Miller and Summers 2007; Kirkland, 2009). In this paper, I echo Carroll and Kowtiz’s (1994: 73) idea that “valuable as content analysis may be, it does not give insight into the significant gender differences that exist at the level of the individual linguistic item”.

I propose the use of corpus linguistics to gain a more fine-grained understanding of how gender is represented in video games. In this paper, I outline methods for creating a corpus based around the language used in video games and then take one video game series, The Witcher, as a case study. I use keyword analysis on the data set, collocational analysis on the terms man and woman, and concordance line analyses of these terms. I also report on corpus methods for examining agency of men and women (similar to Hunt’s, 2015 research).