Separating Polarization from Noise: Comparison and Normalization of Structural Polarization Measures

TL;DR

This paper evaluates existing structural polarization measures in social networks, revealing their limitations, and introduces normalization techniques that significantly improve their ability to distinguish polarized networks from random noise.

Contribution

It provides a systematic comparison of polarization measures and proposes normalization to enhance their accuracy in identifying true polarization.

Findings

All methods score high polarization on random networks.

Normalization improves classification accuracy by up to 220%.

Most methods outperform unnormalized scores after normalization.

Abstract

Quantifying the amount of polarization is crucial for understanding and studying political polarization in political and social systems. Several methods are used commonly to measure polarization in social networks by purely inspecting their structure. We analyse eight of such methods and show that all of them yield high polarization scores even for random networks with similar density and degree distributions to typical real-world networks. Further, some of the methods are sensitive to degree distributions and relative sizes of the polarized groups. We propose normalization to the existing scores and a minimal set of tests that a score should pass in order for it to be suitable for separating polarized networks from random noise. The performance of the scores increased by 38%-220% after normalization in a classification task of 203 networks. Further, we find that the choice of method is not as important as normalization, after which most of the methods have better performance than the best-performing method before normalization. This work opens up the possibility to critically assess and compare the features and performance of different methods for measuring structural polarization.