MODELING THE HUMANITIES: DATA LESSONS FROM THE WORLD OF EDUCATION

This article will explore advances in the field of educational data modeling that have implications for modeling humanistic data. Traditional humanistic inquiry, bolstered by micro-analyses conducted by the scholar, has made way for machine-assisted methods that parse and quantify large amounts of qualitative data to reveal possible trends and focus more analogue approaches. At best, this play between human-and machine-directed approaches can lead to more profound explorations of texts. In this exploration of qualitative-quantitative methodologies that leverage human agency and machine-directed techniques, I suggest a mixed methods approach for dealing with the humanities. Specifically, this discussion will analyze the current methodological tensions related to Educational Data Mining and Learning Analytics to reveal best practices for modeling humanistic data. Principle questions of interest in this essay include: What defines literary "big data? How can we define DH modeling and where does it depart from traditional data modeling? What role does machine-based modeling have in the context of the scholarly close read? What can we learn from educational data modeling practices that are in the midst of resolving tensions between human-machine patterning?