Big Data Inductive Theory Development: Towards Computational Grounded Theory?

摘要

It has been argued that the unprecedented availability of trace data may revolutionize the social sciences. Still, methodological knowledge is scarce as to how this abundance of data can be used to develop novel and important theory. In this essay, we inquire into how the lessons learned from Grounded Theory Method (GTM) can be used to build theory from big data. To do so, we review GTM in light of three key concepts in social analysis: the continuum of induction, the continuum of generalization, and the level of lexicon and theory. Using Habermas's concept of rational reconstruction we articulate a broader "grounded paradigm" that emphasizes the notion of emergence and provides a pragmatic epistemological foundation for different types of grounded analysis. On this basis, we propose a model that describes the process of theorizing from big data.