This Growing Convergence Research project aims to develop algorithms and tools to better use social media data and other new forms of publicly available text data to advance understanding of human behavior and society. The research team will integrate across the social, behavioral, and computer sciences to create and adapt computer algorithms and data mining methods in ways that adhere to the design structures, measurement rigor and ethical protections of social science. While much research is emerging in this space, no established best practices exist for designing proper micro- and macro-level studies involving social media and other open-source text data. The research team, representing the breadth of behavioral/social science and computer science, will develop and test methodologies for sampling, validating, and analyzing social media data so that social scientists can easily interpret and generalize from them.

Specifically, this project will (1) develop a detailed, hybrid methodology (Iterative Method for Social Media Research - IMSMR) that integrates relevant components of existing social science methodologies with relevant components of the knowledge discovery process to enhance research practices in both social and computer science fields; (2) use IMSMR to establish guidelines for using an array of different social media data to answer questions across different social and data science disciplines; (3) test and refine the methodology and guidelines on different research exemplars that spans multiple social, behavioral, and economic disciplines; and (4) develop a shared text analytic research portal that enables social scientists to generate structured variables using state of the art natural language processing and data mining that adhere to the validity and reliability standards of social science.