Modeling and Visualizing Locus of Control with Facebook Language

Abstract

A body of literature has demonstrated that users' psychological traits such as personality can be predicted from their posts on social media. However, there is still a gap between the computational and descriptive analyses of the language features associated with different psychological traits, and their use by social scientists and psychologists to make deeper behavioral inferences. In this study, we aim to bridge this gap with a visualization that situates the language associated with one psychological trait in the context of other psychological dimensions. We predict Locus of Control (LoC), an individual's perception of personal control over events in their lives, from their Facebook language (F1=0.82). We then look at how language explains the relationship of LoC with consciousness and emotional stability.