MODELING THE SOCIAL MEDIA RELATIONSHIPS OF IRISH POLITICIANS USING A GENERALIZED LATENT SPACE STOCHASTIC BLOCKMODEL

• 成果类型: Article
• 署名作者: Ng, Tin Lok James; Murphy, Thomas Brendan; Westling, Ted; McCormick, Tyler H.; Fosdick, Bailey
• 署名单位: Trinity College Dublin; University College Dublin; University of Massachusetts System; University of Massachusetts Amherst; University of Washington; University of Washington Seattle; University of Washington; University of Washington Seattle; Colorado State University System; Colorado State University Fort Collins
• 刊物名称: ANNALS OF APPLIED STATISTICS
• ISSN/ISSBN: 1932-6157
• DOI: 10.1214/21-AOAS1483
• 发表日期: 2021
• 页码: 1923-1944
• 关键词: position cluster model; bayesian-inference; selection; independents; mixture; prediction; PARTIES

摘要

Dail Eireann is the principal chamber of the Irish parliament. The 31st Dail was in session from March 11th, 2011 to February 6th, 2016. Many of the members of the Dail were active on social media, and many were Twitter users who followed other members of the Dail. The pattern of Twitter following amongst these politicians provides insights into political alignment within the Dail. We propose a new model, called the generalized latent space stochastic blockmodel, which extends and generalizes both the latent space model and the stochastic blockmodel to study social media connections between members of the Dail. The probability of an edge between two nodes in a network depends on their respective class labels, as well as sender and receiver effects and latent positions in an unobserved latent space. The proposed model is capable of representing transitivity and clustering, as well as disassortative mixing. A Bayesian method with Markov chain Monte Carlo sampling is proposed for estimation of model parameters. Model selection is performed using the WAIC criterion and models of different number of classes or dimensions of latent space are compared. We use the model to study Twitter following relationships of members of the Dail and interpret structure found in these relationships. We find that the following relationships amongst politicians is mainly driven by past and present political party membership. We also find that the modeling outputs are informative when studying voting within the Dail.