Text Analysis Workshop, April 30th & May 1st

A Text Analysis Workshop

James Pennebaker (University of Texas at Austin)

    Monday, April 30th, 12:00pm and
    Tuesday, May 1st, 12:00pm
    Location to be determined

This hands-on workshop will describe the logic of computer-based text analytic methods, with special emphasis on word counting approaches.  Participants will be encouraged to bring their own computer (either PC or Mac) and will be given a trial version of the LIWC text analysis program. Over the two half-day sessions, the following topics will be covered:
  • Overview of various text analytic approaches commonly used in the social sciences
  • The psychometrics of words and word aggregation methods
  • Popular text analysis strategies: Google Books, Google Trends. Thinking about "corpus psychology." Twitter, Facebook, Wikipedia, Reddit, CNN, IMDb, Project Gutenberg, and other places where words exist.
  • LIWC -- how it works, its development, input and out
  • Analyzing and interpreting LIWC output: personality, cognition, emotion
  • Approaches to tracking social processes and dynamics in dyads, groups, and communities
  • Topic modeling, with particular attention to the meaning extraction method (MEM)
  • Playing with data: machine learning, author profile matching (unmasking), identifying narrative
  • Applications in medicine, business, accounting, history, marketing, professional sports, and much more
In the workshop, participants are encouraged to bring their own data sets to play with. In addition, some shared data sets will be available.