Institute of Sociology
of the Federal Center of Theoretical and Applied Sociology
of the Russian Academy of Sciences

Kuchenkova, A.V. (2020), “Measuring subjective well-being based on social media texts. Overview of modern practices”, RSUH



Kuchenkova, A.V. (2020), “Measuring subjective well-being based on social media texts. Overview of modern practices”, RSUH/RGGU Bulletin. “Philosophy. Sociology. Art Studies” Series, no. 4, pp. 92–101, DOI:10.28995/2073-6401-2020-4-92-101
ISSN 2073-6401
DOI 10.28995/2073-6401-2020-4-92-101
ÐÈÍÖ: https://www.elibrary.ru/item.asp?id=45609491

Posted on site: 27.04.21

Òåêñò ñòàòüè íà ñàéòå æóðíàëà URL: https://philosophy.rsuh.ru/jour/article/view/343 (äàòà îáðàùåíèÿ 27.04.2021)


Abstract

Along with numerous studies of subjective well-being through sociological methods (first of all, surveys), attempts are being made to use Big Data, “digital footprints” (social media texts, social network profile informa-tion, search query statistics, personal electronic device data) as an additional source of information. Based on a review of foreign literature, the author reveals major practices of the social media texts analysis to measure subjective well-being. Including the experience of constructing the Gross National Happiness index for Facebook and the Hedonometer for Twitter based on the analysis of emotive vocabulary and the tone of publications of network users. Possibilities of searching for “digital traces” of life satisfaction in the social media texts are revealed.Methodological difficulties and limitations in that area of research, which have not yet been overcome, are highlighted: the issue of the obtained conclu-sions generalization and the validity of the constructed indices of “happiness” in their correlation with the “real” subjective well-being measured through surveys. Difficulties in measuring subjective well-being are associated with the effects of self-presentation in social media, varying degrees and strategies of users’ publication activity, imperfection of the analysis algorithms that are still inferior to “manual coding”.