Attributes beyond Attitudes

Abstract

Models of behavior on the US Supreme Court almost universally assume that justices’ behavior depends, at least in part, on the characteristics of individual justices. However, few prior studies have attempted to assess these characteristics beyond ideological preferences. In contrast, we apply recent advances in machine learning to develop and validate measures of the Big Five personality traits for Supreme Court justices serving during the 1946 through 2015 terms based on the language in their written opinions. We then conduct an empirical application to demonstrate the importance of these Supreme Court Individual Personality Estimates and discuss their proper use.