ABSTRACT
Guided by the integrative theory of communication work, we present a computational and qualitative analysis of Reddit comments authored by patient-users living with chronic overlapping pain conditions. We first employed computational text analysis to compress the data corpus via structural topic modeling. The community detection algorithm Louvain was employed to identify communities of topics within the data. Then, we extracted a subset of the data and performed a computer-aided qualitative thematic co-occurrence analysis. Topic modeling revealed 35 topics and five thematic communities. The qualitative analysis revealed that these topics mapped onto and extended the typology of communication work and contributing factors and identified co-occurrence patterns both among and across research questions. Our findings offer methodological, theoretical, and practical implications.
Acknowledgements
The first author would like to acknowledge Dr. Timothy Betts, Assistant Professor at Texas Christian University, for his assistance to the first author in scraping and organizing the publicly available Reddit data used in this analysis.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Two subreddits for endometriosis were included.