Extracting Age-Related Stereotypes from Social Media Texts

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 3183–3194 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 3183 Extracting Age-Related Stereotypes from Social Media Texts Kathleen C. Fraser, Svetlana Kiritchenko, Isar Nejadgholi National Research Council Canada Ottawa, Ontario, Canada {kathleen.fraser, svetlana.kiritchenko, isar.nejadgholi}@nrc-cnrc.gc.ca Abstract Age-related stereotypes are pervasive in our society, and yet have been under-studied in the NLP community. Here, we present a method for extracting age-related stereotypes from English language Twitter data, generating a corpus of 300,000 over-generalizations about four contemporary generations (baby boomers, generation X, millennials, and generation Z), as well as “old” and “young” people more generally. By employing word-association metrics, semi-supervised topic modelling, and density-based clustering, we uncover many common stereotypes as reported in the media and in the psychological literature, as well as some more novel findings. We also observe trends consistent with the existing literature, namely that definitions of “young” and “old” age appear to be context-dependent, stereotypes for different generations vary across different topics (e.g., work versus family life), and some age-based stereotypes are distinct from generational stereotypes. The method easily extends to other social group labels, and therefore can be used in future work to study stereotypes of different social categories. By better understanding how stereotypes are formed and spread, and by tracking emerging stereotypes, we hope to eventually develop mitigating measures against such biased statements. Keywords: stereotypes, bias, ageism, social media, computational social science 1. Introduction Stereotypes are over-generalizations about the charac- teristics of a group of people, such that an individual is assumed to have these characteristics simply based on their perceived membership in the group. Stereo- types can lead to prejudicial behaviour against mem- bers of a group, as well as psychological harm. Not all stereotypes are overtly negative; however, even stereo- types that might appear on the surface to be positive can have negative effects (for example, the stereotype that women are warm and nurturing can lead to backlash against women who violate this stereotype by acting powerful and assertive). Of particular concern is stereotyping on the basis of protected characteristics, such as race, sex, religion, or age. The latter has been particularly under-studied in natural language processing (NLP), despite the fact that ageism is widespread in North American society and can lead to age-based bias in the workplace (Perry et al., 2013), media representation (Lichtenstein, 2021), and the healthcare system (Wyman et al., 2018). Fur- thermore, such stereotypes can become a self-fulfilling prophecy when they are internalized by people who self-identify as older adults, leading to isolation and health decline (Chan et al., 2020). Unlike the traditional approach to studying stereotypes in psychology, where a small group of people is di- rectly asked to provide their own or common percep- tions of a group of interest via survey questions, in this work we analyze pervasive stereotypes from naturally- occurring data in Twitter. While we lack the benefits of a well-controlled laboratory environment and need to apply various …

Content truncated to 500 words