Finding relevance in the news: The scale of self-reference.

According to both professional journalists and news users, news should be relevant. While a great deal of research that treats relevance as co-constructed starts from the text of news stories, this paper asks how news users explicitly construct the (ir)relevance of particular news reports, taking a language-centered lens to open-ended survey responses. This paper makes a methodological argument in favor of a language-centered approach to open-ended survey data. Given the ubiquity of online surveys in many social science disciplines, the present paper provides an example of how this approach can deepen our understanding of survey responses. We find that news users construct relevance at varying scales, using a number of linguistic strategies of self-reference. Those who said they found the story they saw relevant used pronouns with a different distribution than those who did not, and these differences exceeded chance. In general, those who referred to themselves as members of larger collectivities were more likely to say they found a news story relevant, suggesting that relevance is discursively constructed in part through practices of self-reference.