Assessing the Big Five personality traits with latent semantic analysis

Abstract

We tested whether the characteristics of a person's personality can be assessed by an automated analysis of the semantic content of a person's written text. Participants completed a questionnaire measuring the so-called Big Five personality traits. They also composed five short essays in which they were asked to describe what they would do and how they would feel in each of five scenarios designed to invoke the creation of narrative relevant to the Big Five personality traits. Participants' essays were processed for content by Latent Semantic Analysis (LSA; T. Landauer & S. Dumais, 1997), a model of lexical semantics. We found that LSA could assess individuals on three of the Big Five traits, and we discuss ways to improve such techniques in future work.