How can text mining support qualitative data analysis?
Event Type
Technical Groups
Occupational Ergonomics
TimeWednesday, October 6th2:24pm - 2:42pm EDT
LocationHarborside Salon B
DescriptionQualitative research methods can be challenging as they often require a manual, time-consuming, and iterative process. Recent research suggests that text mining approaches could support qualitative data analysis. The present study explored how structural topic modeling (STM) could support qualitative data analysis and provides a guide for researchers to explore the utility of STM in their work. We used STM to analyze interview data from a study of caregivers for children with medical complexity. The STM analysis produced term frequency and inverse document frequency (tf_idf) plots, identified 15 topics, defined the topics with frequency and exclusivity (FREX) terms, and described the prevalence of topics across the dataset. These findings provide unique interpretations of the qualitative data that complement qualitative data analysis at various points in the process (e.g., data familiarization, member checking, sampling).