Abstract

Classic approaches to measurement of psychological traits and many health outcomes have relied on the tried and true methods of self-report. However, self-report by its definition requires people to actively respond to surveys, which imposes a big time and monetary cost for recruitment of large samples. Furthermore, its exceedingly difficult to ever examine how people’s social environments—especially micro-level social environments—affect them since measuring the traits and values of the people in one’s social network has been near impossible. Recent advances in big data analytics, however, are changing this. In this talk, I will discuss new data from my lab on 30 million Americans, how we can make predictions about their health, and then study the way their social environment impacts these health outcomes. The talk will also include an overview of how big data analytics can be done by behavioural and social scientists interested in applying these methods.