Text mining and IRT for psychiatric and psychological assessment

The information age has made it easy to store and process large amounts of data, including both structured data (e.g., responses to questionnaires) and unstructured data (e.g., natural language or prose). As an additional source of information in assessments, textual data has been increasingly used by cognitive, personality, clinical, and social psychologists in attempt to understand human beings. The questions how to handle these textual data and how to combine them with structured data in psychiatric and psychological assessments are the major themes in this thesis. The research initiates a new intake procedure for psychiatric screening for posttraumatic stress disorder (PTSD), which combines the utilization of advanced text mining techniques and item response modeling in one framework. The project consists of three parts: (1) computerized text classification on patients’ self narratives to screen for PTSD. (2) exploring the generalizability of DSM-IV diagnostic criteria for PTSD using item response modeling and identifying the differential symptom functioning related to various background variables such as gender, marital status and educational level. And (3) combining textual assessment on patients’ self-narratives and item-based assessments in one systematic framework to further improve the screening efficiency. Further, the automated intake framework is also validated in two psychological assessments. One focused on the application of text mining techniques in learning patterns among undergraduate students’ personality adaption with their life stories. The other engaged in an interesting analysis of Internet data that was collected from Facebook. This study investigated the relationship between people’s self-monitoring skills and verbal features in their textual posts on the Facebook wall. To sum up, the combination of the text mining technique with item response modeling is an innovation in psychiatric and psychological assessments. It is promising to be applied for a broader scope within the field of education, psychology, sociology in both research settings and practical use.