Discovering higher-level Patterns (E2) - Nelson 2015 & Nelson 2017

[HyunkuKwon]

Post questions about the following exemplary reading here:

Nelson, Laura K. 2015. “Persistent Political Logics: Geographical Differences and Temporal Continuities within the Women’s Moevements in Chicago and New York City.” Working paper. (But you must also skim the framework she draws from her analysis in Nelson, Laura K. 2017. “Computational Grounded Theory: A Methodological Framework.” Sociological Methods & Research

[chiayunc]

This is probably a pretty trivial question. The first piece used pattern detection to come up with prevalent words used by Chicago and New York women movements. The author then, based on her judgment, assigned qualities to assess these two different groups of words - concreteness and abstractness. Although it is definitely in alignment with what the author provided in her paper, I wonder if this ‘axis’ of concrete-abstract is chosen too arbitrarily? This will go on to affect her later endeavor to use crowd-sourcing to confirm how concrete or how abstract these words are. I am not sure how this solidifies the validity of her characterization or is this at all important? (I can easily argue that Chicago is more about solutions or actors, while New York focuses on ideology or awakening. Putting this to crowd-sourcing, I feel like the crowd would also agree.)

[xxicheng]

I am interested in the computational grounded theory in this article. The author claims that this method can "discard previous assumptions about the nature of these organizations and the categories through which they are different," but it is still not clear at this point to me. Could you please elaborate more about how this method deals with the pre-existing differences in writing styles in two waves?

[MOTOKU666]

I'm also interested in this computational grounded theory. By combining with "the substantive expertise and ability to interpret brought by human readers", I'm wondering how does this theory being applied in a broader context. In addition, would there be any county-level difference that can be detected in the analysis? What is a suitable level we shall look into when facing similar problems?

[sabinahartnett]

When reading the computational grounded theory method - I am most concerned about what is missed when the first step is only computational - especially when, as in the authors analysis, the relevance is text-frequency based. This runs the risk of 'missing' the significance of certain documents (i.e. circulation frequency/readership, etc) that may be understood by a historian and thus differently weighted