Golshan Madraki, University of New England – Polarized Social Media Networks

On University of New England College of Business Week: Everything is polarized these days, including social media networks.

Golshan Madraki, associate professor of supply chain management in the college of business and director of the Supply Chain Management Center, looks to quantify this.

Dr. Golshan Madraki is currently a tenured Associate Professor of Supply Chain Management in the College of Business at the University of New England (UNE), Maine, USA. Before joining UNE, she was an Associate Professor of Engineering and Management in the Reh School of Business and the Associate Director of the Honors Program at Clarkson University. She received her Ph.D. in Industrial and Systems Engineering from Ohio University in 2017. Her research focuses on various applications of algorithms and data analytics, graph theory, operations research, and mathematical modeling in manufacturing systems, supply chain networks, social media networks, and related societal issues. Dr. Madraki has taught a variety of courses, including Business Analytics, Systems Engineering and Management, Project Management, and Engineering Economics, at the University of New England, Clarkson University, and Ohio University. She has received several internal and external research grants, as well as multiple teaching and research awards throughout her career.

Polarized Social Media Networks

A key aspect of research into social media polarization is the development of methods to quantify individual users’ polarization. This approach focuses on the connections within a network, modeling it as a directed graph and computing a new metric called “Connectivity Score” for each individual user based on their follower/following relationships. This method relies on objective, public, and memory-efficient data.

We applied our method to the Twitter network of the 116th U.S. Congressmembers and compared with two measures : 1)content-based measures, 2) real-world political behavior-based polarization  measure such as voting patterns and co-sponsoring bills in congress. The results highlight that users’ social media connections reflect their polarization behavior. Furthermore, a significant correlation was observed between Connectivity Scores and real-world political behavior measures like Roll Call and Ideology scores.

Interestingly, Democrats’ behavior on social media showed stronger alignment with their legislative actions compared to Republicans. These findings emphasize the potential of network-based approaches to provide deeper insights into polarization dynamics.

By leveraging structural data, this research advances the understanding of political polarization on social media. Moreover, the results underscore the importance of whom we follow on social media.