Capturing Meaning and Impact in Real Time: Mixed-Methods and Supervised Machine Learning in Big Data Policy Analysis
Friday, November 13, 2015
Riverfront South/Central (Hyatt Regency Miami)
*Names in bold indicate Presenter
Along with quantitative and qualitative research, mixed methods is now acknowledged as a third research paradigm. The mixed methods paradigm takes a pragmatic approach by simultaneously integrating quantitative and qualitative methods, allowing the researcher to validate and generate theory at the same time. To date, big-data approaches to policy analysis have been primarily rooted in quantitive and/or predictive analytics. These provide the capacity for large-scale content analysis and hypothesis testing, but are less suitable for capturing the meaning and context of the policy process. In this paper, we outline a cost effective mixed-methods approach to big-data policy analysis that allows the researcher to quantitatively and qualitatively capture the whole textual corpus surrounding policy development in real-time: supervised machine learning. Particular emphasis is placed on using the method concurrently with practicioners and public-interest groups in a low cost and time efficient manner.