Comparison of LIWC and open-vocabulary features within predictive models of gender, age, and personality.

accuracy: percent predicted correctly (for discrete binary outcomes). R: Square-root of the coefficient of determination (for sequential/continuous outcomes). LIWC: A priori word-categories from Linguistic Inquiry and Word Count. Topics: Automatically created LDA topic clusters. WordPhrases: words and phrases (n-grams of size 1 to 3 passing a collocation filter). Bold indicates significant (pLIWC alone).