An Examination Of Exercise Effects On Self-Reported Well Being Using Social Media Data

Official URL: https://risc01.sabanciuniv.edu/record=b3205821

The increasing usage of social media has opened up new avenues for studying the relationship between physical exercise and well-being. While exercise has long been recognized for its positive impact on physical health, its effects on mental and emotional well-being have gained significant attention. In this context, Twitter, as a popular social media platform, provides a valuable source of real-time data that reflects individuals’ thoughts, emotions, and behaviors in their daily lives. Furthermore, the rise of wearable devices has allowed researchers to collect detailed activity reports, including exercise data. By leveraging Twitter data and sentiment analysis, we explored the effects of exercise on daily life well-being. Through the analysis of sentiment scores of tweets posted before, during, and after exercise periods, we compared the average sentiment scores of exercise and non-exercise periods to gain insights into the impact of exercise on overall sentiment. Our methodology involved applying causal inference models to time series data by using propensity score matching methods, revealing how exercise periods influenced people’s sentiment status. The results of our study highlight the constructive influence of regular physical activity on mental well-being. We have identified the positive effect of exercise on the daily posted content in terms of sentiment during the exercise periods. This research contributes to the growing body of knowledge on the relationship between exercise and daily life well-being