Gender and Language (English vs. Russian) as Predictors of Toxic Statements Perseption in Online Multiplayer Games

This study investigates the relationship of perceived toxicity in online multiplayer games with gender and language. We aim to close a gap in the research literature, as this issue has previously been studied only in the English-speaking community. This paper answers two main research questions: (1) How does gender influence the perception of toxic communication in online multiplayer games? and (2) How do perceptions of toxicity differ between English-speaking and Russian-speaking players? The research design of this paper involves an online experiment involving English-speaking and Russian-speaking Dota 2 players. Participants are placed in a simulated situation where they are asked to rate the toxicity of 25 phrases on a five-point scale. Next, they take a post-experimental questionnaire containing information on demographic and self-reported gaming performance. Criteria for inclusion in the sample included Dota 2 gaming experience, Russian or English language proficiency, and binary (male/female) heder individuation. The main channels for recruiting the sample were profile chats in various social networks, as well as online forms dedicated to the game Dota 2. To analyze the data obtained as a result of the experiment, we used quantitative methods. We use such methods as: textual analysis, multilevel modeling, various statistical tests, as well as linear regression models. As a result, we found that English-speaking players have a higher level of perceived toxicity than Russian-speaking players. Also, in the English-speaking language community there is a difference in perceived toxicity depending on the gender of the respondent. English-speaking women have a stronger perception of toxic communication than English-speaking men. A similar correlation could not be found in the Russian-speaking gaming community. No correlation could be established between how the subject who reproduces toxic communication gendered himself and how the subject who receives toxic communication gendered himself. As a result, we were able to discover an important relationship between language play community and perceived toxicity, thus partially closing an existing gap in the research literature. The next step in this study will be to increase the language sample and modify the research design to examine the relationship between the gender of the sender of a toxic message and the gender of the receiver of a toxic message.