Text Analysis

One of my goals is to understand individual differences using the cues evident in the social environment. One of those cues is language.

Accordingly, some of my work entails analyzing texts in an automated way. We’ve used Jamie Pennebaker’s LIWC system for a couple of papers, including one led by myself (PDF) and another led by Angela Carey (PDF).

To go beyond simple word-counting approaches, Tal Yarkoni and I created a text analytic system called the Contrast Analysis of Semantic Similarity (CASS). You can download it for free at www.casstools.org. We’ve written a couple of papers validating CASS. You can download them here and here.