Gaining insights from social media language: Methodologies and challenges.

Abstract

Language data available through social media provide opportunities to study people at an unprecedented scale. However, little guidance is available to psychologists who want to enter this area of research. Drawing on tools and techniques developed in natural language processing, we first introduce psychologists to social media language research, identifying descriptive and predictive analyses that language data allow. Second, we describe how raw language data can be accessed and quantified for inclusion in subsequent analyses, exploring personality as expressed on Facebook to illustrate. Third, we highlight challenges and issues to be considered, including accessing and processing the data, interpreting effects, and ethical issues. Social media has become a valuable part of social life, and there is much we can learn by bringing together the tools of computer science with the theories and insights of psychology. (PsycINFO Database Record

Conceptualizing and Measuring Well-Being Using Statistical Semantics and Numerical Rating Scales

“…We describe some relevant analyses related to LSA and the use of semantic representations. For a clear description of the methodology involved in performing text analyses based on LDA using large amounts of social media texts, see Kern et al (2016). All of the analyses described below may be carried out using www.semanticexcel.com (a user-friendly, online software solution; see Sikström, Kjell & Kjell, Appendix B).…”

“…For example, they found that the measures agreed with both self-reported as well as the informant's account of personality and that the measures were stable over a six-month period. Schwartz et al (2016) have also used texts from social media for predicting satisfaction with life. Lastly, Eichstaedt et al (2015) have used text from tweets in order to predict heart disease mortality on a county level, where they among other language features used LDA-derived topics.…”

“…Many of the analyses pertaining to LSA are described and possible to carry out in the online point-and-click software solution that goes by the name of Semantic Excel (Sikström, Kjell & Kjell, Appendix B). For useful descriptions of methods related to analyzing big data texts from social media, see Schwartz and Ungar (2015) and Kern et al (2016).…”

Predicting Depression From Language-Based Emotion Dynamics: Longitudinal Analysis of Facebook and Twitter Status Updates (Preprint)

“…Considering the frequent use of emotion language in status updates that relate to current experiences [13,14], for a large proportion of the population social media can provide unobtrusive access to time-sensitive and ecologically valid samples of expressed emotion [2,[41][42][43]. While observations of emotion variability and instability are yet to be applied to social media as a means of automatically screening for individuals at risk of depression, depression change has been demonstrated to be visible through changes in status update posting activity and content [2,4,10,44,45].…”

“…dictionaries were supplemented by common emoji's and internet slang that indicated positive or negative emotion (see Multimedia Appendix 1). While not definitive, these inclusions were made to better reflect the language used on social media (for further discussion see [41]). …”