A Panel Data Analysis of Subjective Well-Being Based on Microblog User Information.

Big data modelling using microblogs is applied to acquire nationwide representative panel data on subjective well-being. The analysis directly validates the influence of China's Livelihood Index on subjective well-being. Using panel data on subjective well-being collected for the period from 2010 to 2021 from users of the Weibo (Sina Corporation, Beijing, China) microblogging platform, this study finds Granger causality running from China's Livelihood Index to subjective well-being and that the two are positively correlated. We also find Granger causality running from a life stress indicator to a life satisfaction indicator. The education indicator model is found to be positively correlated with life satisfaction and positive emotions, whereas the life stress indicator and life satisfaction are negatively correlated. Medical and health indicators are positively related to life satisfaction, while a negative correlation is found between the traffic indicator model and life satisfaction. The relationship between economic development and subjective well-being also displays bidirectional Granger causality and a positive correlation. However, in China's provinces and prefecture-level cities with relatively strong economic growth, the correlation between the livelihood index and economic development appears to be weaker. We suggest boosting gross domestic product per capita and absolute per capita income to increase subjective well-being in less developed western China. Bridging the gross domestic product per capita gap nationwide may also positively influence subjective well-being. To achieve this, we suggest measures that include improving medical and health services, alleviating traffic congestion, increasing the teacher-student ratio and improving the education universalisation rate. These steps would improve the equitable and balanced development of China's Livelihood Index across the country's 31 provinces.