Inferring the dynamics of rising radical right-wing party support using Gaussian processes.

Area of Science:

• Social Sciences
• Political Science
• Statistical Modeling

Background:

• Classical regression techniques in social science may limit the discovery of complex, nonlinear mechanisms.
• Existing methods often depend heavily on analyst expertise and prior expectations.
• There is a need for advanced methodologies to better understand social phenomena.

Purpose of the Study:

• To introduce a novel regression methodology combining interpretability with Bayesian statistical flexibility.
• To identify and explain the mechanisms driving the rise of Radical Right-wing Populist parties (RRPs).
• To demonstrate the superiority of the new approach over traditional methods for social systems analysis.

Main Methods:

• Development of a hybrid regression methodology integrating traditional statistical interpretability with Bayesian predictability.
• Application of the methodology to Swedish municipality-level data from 2002 to 2018.
• Analysis of factors influencing support for Radical Right-wing Populist parties (RRPs).

Main Results:

• No evidence found that the proportion of foreign-born residents predicts increases in RRP support.
• Education levels and population density identified as significant predictors of RRP support.
• Spatial and temporal control variables also influenced RRP support dynamics.

Conclusions:

• The proposed methodology offers superior model fit for complex, nonlinear social systems.
• This approach enhances hypothesis testing and exploration of theories concerning RRPs and other social movements.
• Findings challenge common assumptions about immigration's role in the rise of right-wing populism.