ステレオタイプ情報はよりリツイートされるか？

ツイートの言語表現に注目した検討

抄録

Previous studies have revealed that stereotypic (vs. non-stereotypic) information about individuals spreads more easily among communicators in face-to-face communication. It is also known that more dispositional terms (i.e., more "abstract" words such as nouns and adjectives according to the Linguistic Category Model) are used to describe behaviors that are consistent with stereotypes of the actor's group, whereas more context-specific and less dispositional terms (e.g., verbs) are used to describe counter-stereotypic behaviors. In the present study, we investigated to what extent the stereotypicality and linguistic abstractness influence the spread of information in social media which is currently growing into a major arena for prejudiced communication in place of face-to-face discourse. Specifically, we examined whether tweets with higher stereotypicality and linguistic abstractness receive a greater number of retweets. Five hundred tweets posted by Japanese users containing "men" or "women" in Japanese were sampled, and were rated by independent coders with respect to their perceived gender-steretypyicality. The number of nouns, adjectives, and verbs contained in each tweet was respectively counted to assess the linguistic abstractness. Consistent with our predictions, gender-stereotypic tweets contained a greater number of abstract terms (i.e., nominal adjectives) than did less stereotypic tweets. Furthermore, perceived stereotypicality significantly predicted whether or not the tweet was retweeted, as well as how many times the tweet was retweeted. Implications of the results are discussed with regard to the influence of linguistic abstractness on stereotypic generalization in social media. Theoretical significance concerning communication and collectively shared cognition is also discussed.