Beyond Siloed Analysis: How Multilayer Models Solve Political Network Inference Problems

Political interactions often span multiple overlapping relational contexts, an interdependence traditional network models struggle to capture. This paper introduces a multilayer approach using exponential random graph modeling (ERGM) that specifically addresses this challenge by integrating multiple relations.

Data & Methods 🎯

• Examines two political networks: policy communication and global conflict data
• Applies the new multilayer network methodology to these distinct systems

Key Findings 🔍

• Models incorporating interdependence between relational contexts fit observed data significantly better than traditional single-layer approaches.
• This multilayer method provides crucial inferential leverage for understanding complex political systems with interconnected relationships.

The approach demonstrates how accounting for multi-relational complexity enhances our analytical capabilities in political science.