Voter demographics and socio-economic factors in kinetic models for opinion formation

Voter demographics and socio-economic factors like age, sex, ethnicity, education level, income, and other measurable factors like behaviour in previous elections or referenda are of key importance in modelling opinion formation dynamics. Here, we revisit the data-driven kinetic opinion formation model from Düring and Wright (2022) and compare in more detail the influence of different choices of characteristic demographic factors and initial conditions. The model is based on the kinetic opinion formation model by Toscani (2006) and the leader-follower model of Düring et al. (2009) which leads to a system of Fokker–Planck-type partial differential equations. In the run-up to the 2024 general election in the United Kingdom, we consider in our numerical experiments the situation in the so-called Red Wall in North and Midlands of England and compare our simulation results to other election forecasts.