3. Clustering & Topic Modeling to Discover Higher-Order Patterns of Meaning - [E2] Nelson, Laura K. 2021.

[lkcao]

Post questions here for this week's exemplary readings:

2. Nelson, Laura K. 2021. “Cycles of Conflict, a Century of Continuity: The Impact of Persistent Place-Based Political Logics on Social Movement Strategy.” American Journal of Sociology 127 (1): 1-59. You might also peek at Laura’s approach to content discovery: Nelson, Laura K. 2017. “Computational Grounded Theory: A Methodological Framework.” Sociological Methods & Research: 1-40.

[bucketteOfIvy]

Geographer's often talk about the power of place, with spatial interactions being viewed as a key piece in understanding how the social dynamics, institutions, and systems. Nelson (2021) exemplifies place's power, demonstrating the ability of place as an explanatory force and relevant factor in our analysis. However, visualizing and explaining the intersection of text and place seems complex. For instance, in an analysis of ideas expressed about Chicago's community areas over time, one might be interested in thinking about commonly used words to describe each of the areas. Word clouds might be a way to analyze that data, but that would result in 72 word clouds, which is hardly legible. What data visualization and communication best practices exist when using place-based content analysis?

As a similar followup: geography often considers questions of autocorrelation, hot spots, and spatial clustering. How can we work content analysis methods and insights into methods from quantitative geography?