Figure 1 in Cotton et al. 2024, explaining the process
Visiting PhD student Isabel Cotton joined my team and the ResNet project for 3 months last year, from her home unit at the Tyndall Centre at the University of East Anglia, and recently published a methods-oriented research note on that work in Journal of Environmental Psychology. The paper, Comparing thematic and search term-based coding in understanding sense of place in survey research, shows the result of exploratory work we did with Bangor University linguistic scholar Thora Tenbrink that took a qualitative approach by contrast with most survey-based assessments of sense of place. We compared the results of inductive and search-term-based coding of a free-text survey question on a survey of Minas Basin house residents asking them to, “Please describe [their] local area, in terms of what it means to [them] personally and how [they] use it”. The two sets of categories varied in their correlation, with more tangible themes like recreation/experience, relational/family and cultural heritage/history the strongest across the methods, compared with less comparable themes about restorative environments and small-town identity. What was particularly interesting, however, was that it was possible using word clouds to identify new terms to allow the search-term approach to improve its performance significantly (for instance recreation went from 0.75 to 0.96 correlation by adding only one term, exercise; see above).
