Researchers and policy makers worldwide are interested in measuring the subjective well-being of populations. When users post on social media, they leave behind digital traces that reflect their thoughts and feelings. Aggregation of such digital traces may make it possible to monitor well-being at large scale. However, social media-based methods need to be robust to regional effects if they are to produce reliable estimates. Using a sample of 1.53 billion geotagged English tweets, we provide a systematic evaluation of word-level and data-driven methods for text analysis for generating well-being estimates for 1,208 US counties. We compared Twitter-based county-level estimates with well-being measurements provided by the Gallup-Sharecare Well-Being Index survey through 1.73 million phone surveys. We find that word-level methods (e.g., Linguistic Inquiry and Word Count [LIWC] 2015 and Language Assessment by Mechanical Turk [LabMT]) yielded inconsistent county-level well-being measurements due to regional, cultural, and socioeconomic differences in language use. However, removing as few as three of the most frequent words led to notable improvements in well-being prediction. Data-driven methods provided robust estimates, approximating the Gallup data at up to r = 0.64. We show that the findings generalized to county socioeconomic and health outcomes and were robust when poststratifying the samples to be more representative of the general US population. Regional well-being estimation from social media data seems to be robust when supervised data-driven methods are used.

Many hoped that social networking sites would allow for the open exchange of information and a revival of the public sphere. Unfortunately, conversations on social media are often toxic and not conducive to healthy political discussions. Twitter, the most widely used social network for political discussions, doubled the limit of characters in a tweet in November 2017, which provided an opportunity to study the effect of technological affordances on political discussions using a discontinuous time series design. Using supervised and unsupervised natural language processing methods, we analyzed 358,242 tweet replies to U.S. politicians from January 2017 to March 2018. We show that doubling the permissible length of a tweet led to less uncivil, more polite, and more constructive discussions online. However, the declining trend in the empathy and respectfulness of these tweets raises concerns about the implications of the changing norms for the quality of political deliberation.

This study introduces and evaluates the robustness of different volumetric, sentiment, and social network approaches to predict the elections in three Asian countries – Malaysia, India, and Pakistan from Twitter posts. We find that predictive power of social media performs well for India and Pakistan but is not effective for Malaysia. Overall, we find that it is useful to consider the recency of Twitter posts while using it to predict a real outcome, such as an election result. Sentiment information mined using machine learning models was the most accurate predictor of election outcomes. Social network information is stable despite sudden surges in political discussions, for e.g. around elections-related news events. Methods combining sentiment and volume information, or sentiment and social network information, are effective at predicting smaller vote shares, for e.g. vote shares in the case of independent candidates and regional parties. We conclude with a detailed discussion on the caveats of social media analysis for predicting real-world outcomes and recommendations for future work.

A body of literature has demonstrated that users’ mental health conditions, such as depression and anxiety, can be predicted from their social media language. There is still a gap in the scientific understanding of how psychological stress is expressed on social media. Stress is one of the primary underlying causes and correlates of chronic physical illnesses and mental health conditions. In this paper, we explore the language of psychological stress with a dataset of 601 social media users, who answered the Perceived Stress Scale questionnaire and also consented to share their Facebook and Twitter data. Firstly, we find that stressed users post about exhaustion, losing control, increased self-focus and physical pain as compared to posts about breakfast, family-time, and travel by users who are not stressed. Secondly, we find that Facebook language is more predictive of stress than Twitter language. Thirdly, we demonstrate how the language based models thus developed can be adapted and be scaled to measure county-level trends. Since county-level language is easily available on Twitter using the Streaming API, we explore multiple domain adaptation algorithms to adapt user-level Facebook models to Twitter language. We find that domain-adapted and scaled social mediabased measurements of stress outperform sociodemographic variables (age, gender, race, education, and income), against ground-truth survey-based stress measurements, both at the user- and the county-level in the U.S. Twitter language that scores higher in stress is also predictive of poorer health, less access to facilities and lower socioeconomic status in counties. We conclude with a discussion of the implications of using social media as a new tool for monitoring stress levels of both individuals and counties.

In recent years, many studies have used social media data to make estimates of electoral outcomes and public opinion. This paper reports the findings from a meta-analysis examining the predictive power of social media data by focusing on various sources of data and different methods of prediction; i.e., (1) sentiment analysis, and (2) analysis of structural features. Our results, based on the data from 74 published studies, show significant variance in the accuracy of predictions, which were on average behind the established benchmarks in traditional survey research. In terms of the approaches used, the study shows that machine learning-based estimates are generally superior to those derived from pre-existing lexica, and that a combination of structural features and sentiment analyses provides the most accurate predictions. Furthermore, our study shows some differences in the predictive power of social media data across different levels of political democracy and different electoral systems. We also note that since the accuracy of election and public opinion forecasts varies depending on which statistical estimates are used, the scientific community should aim to adopt a more standardized approach to analyzing and reporting social media data-derived predictions in the future.