Quantitative Text Analyses for the Sociology Decolonisation Group at Durham University
Published:
In both, R and Python, I cleaned cleaned and processed the data for text analysis to extract themes, theories, and concepts discussed across 11 modules of 20 lectures each (N=220).
Using topic modelling (LDA and NMF) I created treemap visualisations of core themes, theories and concepts across modules to show the diversity of content. This was done in Python using sklearn, see topic modelling.py file. Then for the Word clouds of theories and concepts covered in those modules to highlight the most prominent one- and two-words. Additional visualisations include the frequency of mentions of racism, colonialism, imperialism across modules. I did all those in R, see frequenct of mentions.R file. Lastly, using R and the leaflet library, I created an interactive map of countries mentioned across modules to explore the global focus of lectures in the department.
The data was collected by Maryanne Ko and Dr. Stephen Ashe from the Deconolinsation Group in the Sociology department at Durham University for the 2024-2025 Academic year.