Paper detail

Predicting Emotional Volatility Using 41,000 Participants in the United Kingdom

Emotional volatility is a human universal. Yet there has been no large-scale scientific study of predictors of that phenomenon. Building from previous works, which had been ad hoc and based on tiny samples, this paper reports the first large-scale estimation of volatility in human emotional experiences. Our study draws from a large sample of intrapersonal variation in moment-to-moment happiness from over three million observations by 41,023 UK individuals. Holding other things constant, we show that emotional volatility is highest among women with children, the separated, the poor, and the young. Women without children report substantially greater emotional volatility than men with and without children. For any given rate of volatility, women with children also experience more frequent extreme emotional lows than any other socio-demographic group. Our results, which are robust to different specification tests, enable researchers and policymakers to quantify and prioritise different determinants of intrapersonal variability in human emotions.