Location Map

XFI Conference Room 1

enlarge mapshrink map

Bipartite attitude networks: a Method for Exploring Opinion-Based Groups, Group-Based Opinions, and Social Polarization by Professor Michael Quayle at University of Limerick

An Institute for Data Science and Artificial Intelligence seminar
Date17 November 2022
Time14:00 to 15:00
PlaceXFI Conference Room 1

Hybrid delivery by Zoom.

I introduce a network-based approach to modelling attitude-based groups as bipartite networks, where people are linked by the attitudes they jointly hold, and attitudes become socially connected when jointly held by people. This structure simultaneously links people into groups and attitudes into meaningful clusters. With examples from opinion-surveys, social-media, and simulations, I show how group structure emerges in this "attitude space", and how people are located in it.

I show how projections of the bipartite network allow us to visualize (1) the opinion-based groups (of people linked by jointly held attitudes) and (2) group-based opinions (linked by people who jointly hold them). I demonstrate how analysis of this bipartite network structure allows us to quantify multidimensional polarization in ANES samples, and visualize the increase in  multidimensional polarization in US society over this time period (on these selected dimensions). Using toy data, I explore some implications of this model for conceptualizing polarization: 1) A group can absorb logically inconsistent attitudes. 2) Group-members can hold substantively different positions, so long as there is sufficient cross-cutting ingroup cohesion. 3) Polarization does not require extremism, and polarized configurations are possible that are not bipolar. 4) Social opinion systems are dynamic, and polarization can result from within- or between-group dynamics. 5) Individuals are located in the opinion space by the combination of attitudes they hold. Because the system is dynamic, individuals' positions in the identity-space can change even if their attitudes do not. 6) Specific attitudes can become progressively more identity-relevant as the system polarizes. I discuss the value of the framework for understanding identity ecosystems in which social group structure and attitudes are co-constituted; and the theoretical consequence that attitude change is usually also identity change.

Delivery and Registration:

To be delivered hybrid. To manage the registrations, we ask participants to complete a simple form, which closes on the morning of the seminar, but please don't let that put you off. If you do miss the registration cut-off, then please email IDSAI. To register, please click here.

If you have any queries, please contact idsai@exeter.ac.uk.

The seminar forms part of the IDSAI Research Seminar Series for 2022-2023. Click here to find out more.


Add this event to your calendar
Bookmark and Share