Most people now use social media platforms to interact with others, get informed, or simply be entertained. During the COVID-19 pandemic, social lives moved online to a larger extent than ever before, as opportunities for face-to-face social contact in daily life were limited.
In this chapter, we focus on what can be learned about people’s emotional experiences and well-being from analyzing text data on social media. Such data is relevant for emotion research, because emotions are not only internal experiences, but often social in nature: Humans communicate their emotions in either verbal or nonverbal ways, including spoken and written language, tone of voice, facial expressions, body postures and other behaviors. Emotions are often triggered by social events: we are sad when we miss someone, happy when we meet loved ones, or angry when someone disappoints us. Emotions also provide important social signals for others, informing them of adaptive ways to interact given their own motivation and goals. Given their valuable social function, emotions are regularly shared with other people and thereby influence other people’s emotions. For instance, happiness may spread through social networks, and give rise to clusters of happy and unhappy people.
Social media continuously captures communication between millions of individuals and large groups over long periods of time. Data from these platforms provide new opportunities to trace emotions and well-being of individuals and societies at new scales and resolutions. This has motivated researchers to use social media data to investigate questions around mental health, emotional well-being, anxiety, collective emotions, or emotion regulation.
A particular strength of new computational approaches is that they can aggregate emotion data at large scales and fast temporal resolutions, often relying on text analysis. Large social media datasets that combine data from many individuals are particularly well suited to examine large group phenomena at the level of populations, especially those involving interactions between individuals. For instance, social media has made it possible to study collective emotions, which emerge from the emotional dynamics in a large group of people responding to the same situation at proximate points in time. Interaction between individuals is a key feature of collective emotions, which can change the quality, the intensity and the duration of emotional experiences.
In the following, we provide an introduction to how emotional trends in society at-large can be measured using text data from social media. We describe two studies assessing whether this social media approach in the United Kingdom (UK) and Austria agrees with surveys on short-lived emotional experiences. We also briefly illustrate their application to long-term experiences like well-being or life satisfaction. We then provide an example from the COVID-19 outbreak to illustrate how social media text analysis can be used to track emotions around the globe. Finally, we discuss the advantages and disadvantages of social media emotion measures as compared to self-report surveys.