Quantifying Subjective Well-Being Using Trends in Weekend Activity

Louis Faust, David Hachen, Omar Lizardo, Nitesh V. Chawla

Abstract: The rise in popularity of physical activity trackers provides extensive opportunities for measuring personal health at scale. Coupled with renewed interest in the field of positive psychology, we aim to explore how the quantified self can inform us of well-being. In this paper, we examine Fitbit Charge HR data among a college cohort of 125 students who were measured for two academic years (56 weeks). We assess variations in step counts across a typical week through k-means clustering. From this, we observe a group of students whose step counts were highest during the weekend and a group with significantly fewer steps on weekends (t-test, p <.001). We found these trends stable: persisting across the two academic years measured. Students cluster location correlated to aspects indicating sociability and subjective well-being. We discuss the correlations between these traits and propose that weekend activity levels may serve as a measure for informing one's subjective well-being.