The second of our text analysis sessions will offer a brief overview of some methods used in this area, along with a deeper dive into topic modeling. This particular approach aims to find (latent) groups of words belonging to a larger body of text or collection of documents (similar to a latent class analysis).
Additionally, we will begin to analyze a healthcare data set with a sociological/psychological hypothesis to simulate real-world anlaysis.
At the end of the session: participants will leave with code to undertake basic topic modelling.
Optional Reading on topic modelling (13 pages):
Sirkka, O., Walo, S., Purhonen, S., Verboord, M., Janssen, S., & Bonnet, P. (2024). Understandings of Culture: A European Bottom-up Study Using Structural Topic Modeling. Socius, 10.