Title: Computational Means, Qualitative Ends

Abstract: The majority of scholarship in the growing field of computational social science is focused on quantitative projects aimed at identifying generalizablefeatures of broad social systems. A smaller group of scholars are instead weaving computational methods into qualitative approaches to provide comprehensive, reproducible, but deeply contextual descriptions of more narrow empirical phenomena. To explore this intersection of computational and qualitative methods, we gathered over 430,000 newspaper articles describing the actions and beliefs of 530 environmental movement organizations between 1990 and 2015. Combining qualitative and interpretive methods with computational techniques, primarily natural language processing techniques and machine learning, we provide a rich, meaningful, but computational description of this movement sector andhow it has changed over time. We focus on three questions: 1) what is the full range of tactical and strategic repertoires within the environmental movement sector? 2) how have these repertoires changed over time? and 3) can we inductively but computationally identify social movement form via shared tactical and strategic repertoires? In exploring these three questions we identified a fourth question: 4) what internal movement processes led to the emergence of a new environmental form, a form that focuses almost exclusively on business sustainability? Using various clustering techniques, we explore this last question, the emergence of a new environmental movement form, to think through the implications of using computational methods for qualitative ends.