Predictors of AfD party success in the 2017 Predictors of AfD party - - PowerPoint PPT Presentation

▶

Nov 30, 2022 273 likes •590 views

18.3.2019 Predictors of AfD party success in the 2017 elections Predictors of AfD party success in the 2017 Predictors of AfD party success in the 2017 elections elections A Bayesian modeling approach A Bayesian modeling approach Sebastian

SLIDE 1

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 1/30

Predictors of AfD party success in the 2017 Predictors of AfD party success in the 2017 elections elections

A Bayesian modeling approach A Bayesian modeling approach

Sebastian Sauer, Oliver Gansser Sebastian Sauer, Oliver Gansser FOM FOM ECDA 2019 ECDA 2019

1 / 30 1 / 30

SLIDE 2

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 2/30

Menace to society Menace to society

Right-wing populism then and now Right-wing populism then and now 2 / 30 2 / 30

SLIDE 3

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 3/30

A model of rough populism

Cf. Kershaw, I. (2016). To hell and back: Europe 1914-1949. New York City, NW: Penguin.

Welzer, H. (2007). Täter. Wie aus ganz normalen Menschen Massenmörder werden. Frankfurt: Fischer.

3 / 30

SLIDE 4

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 4/30

AfD as a nucleus of the German right-wing movement?

Source: Decker, F. (2003). Der neue Rechtspopulismus. Wiesbaden: VS Verlag für

Sozialwissenschaften. Nicole Berbuir, Marcel Lewandowsky & Jasmin Siri (2015) The

AfD and its Sympathisers: Finally a Right-Wing Populist Movement in Germany?, German Politics, 24:2, 154-178, DOI: 10.1080/09644008.2014.982546

4 / 30

SLIDE 5

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 5/30

Popular theories on AfD success

 weak economy ("rust belt hypothesis")  high immigration ("flooding hypothesis")  cultural patterns ("Saxonia hypothesis")

Source: Franz, Christian; Fratzscher, Marcel; Kritikos, Alexander S. (2018) : German right-wing party AfD finds more support in rural areas with aging populations, DIW Weekly Report, ISSN 2568-7697, Deutsches Institut für Wirtschaftsforschung (DIW), Berlin, Vol. 8, Iss. 7/8, pp. 69-79

5 / 30

SLIDE 6

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 6/30

Behavior types model CHOUGHS

Seven behavior types according to CHOUGHS model C onformism H edonism O ut of responsibility U nderstand G ourmets H armony S elf-determined based on approx. 100k face-to-face interviews (stratified by sex and age) Multidimensional scaling was used to devise types CHOUGHS builts on Schwartz' values model

Source: Gansser, O., & Lübke, K. (2018). The development of new typologies of behaviour based on universal human values and purchasing behavior, in: Archives of Data Science, Series B, in submission. Gebauer, H., Haldimann, M., & Saul, C.J. (2017). Service innovations breaking institutionalized rules of health care. Journal of Service Management, 28(5), 972-935.

6 / 30

SLIDE 7

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 7/30

Our research model

unemployment AfD foreigners east_west personality_types

7 / 30

SLIDE 8

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 8/30

AfD votes, and socioenomic factors at the AfD votes, and socioenomic factors at the Bundestagswahl 2017 Bundestagswahl 2017

8 / 30 8 / 30

SLIDE 9

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 9/30

Unemployment and AfD votes

9 / 30

SLIDE 10

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 10/30

Foreigners and AfD votes

10 / 30

SLIDE 11

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 11/30

data analysis data analysis

11 / 30 11 / 30

SLIDE 12

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 12/30

Data preparation

Election related data were obtained from Bundeswahlleiter 2017 Behavior types data (n = 12444) were collected by the authors ´ Data and analysis are accessible at Github: https://github.com/sebastiansauer/afd_values Outcome variable: proportion of votes for AfD was log-transformed for better approximation to normality 12 / 30

SLIDE 13

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 13/30

Bayes modeling

Stan via the R package rethinking Hamiltonian Markov Chain Monte Carlo (MCMC) 2000 iterations, 2 chains, 1/2 warmup Multi level regression modeling (varying intercepts) The WAIC was used for to compare model performance: is an estimate for out-of-sample model performance based on information theory WAIC is similar to the AIC but less restrictive

Cf. McElreath, R. (2016). Statistical rethinking. New York City, NY: Apple Academic Press

Inc.

13 / 30

SLIDE 14

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 14/30

Model specication

a ∼ N (μ, σ) μ = β0e + β1f + β2u + β3t1 + β4t2 ⋯ β10t8 σ ∼ Cauchy(0, 1) f, u, t1, t2 ⋯ t8 ∼ N (1, 0) e ∼ N (0, σ2) σ2 ∼ Cauchy(0, 1)

14 / 30

SLIDE 15

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 15/30

Model specication in R

# Likelihood: afd_prop_log ~ dnorm(mu, sigma), d$ # regression: mu <- beta0[state_id] + beta1*for_prop_z + beta2*unemp_prop_z + beta3*enjoyer + beta4*harmony_seeker + beta5*self_determined beta6*appreciater + beta7*conformist + beta8*type_unknown + beta9*responsibility_denier + beta10*hedonist, # priors: sigma ~ dcauchy(0, 1), beta1 ~ dnorm(0, 1), beta2 ~ dnorm(0, 1), beta3 ~ dnorm(0, 1), beta4 ~ dnorm(0, 1), beta5 ~ dnorm(0, 1), beta6 ~ dnorm(0, 1), beta7 ~ dnorm(0, 1), beta8 ~ dnorm(0, 1), beta9 ~ dnorm(0, 1), beta10 ~ dnorm(0, 1), beta0[state_id] ~ dnorm(0, sigma2), # multi level sigma2 ~ dcauchy(0, 1)

15 / 30

SLIDE 16

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 16/30

Results: Model comparison Results: Model comparison

16 / 30 16 / 30

SLIDE 17

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 17/30

State is the strongest predictor

name predictors type WAIC SE weight m10c unemp, foreign, state Gaussian

50.97

10.74 1 m11d unemp, foreign, state, 8 consumer types Gaussian

39.02

10.31 m06d unemp, foreign, east, 8 consumer types Gaussian

6.96

12.50 m03d unemp, foreign, east Gaussian

1.24

12.44 m00d none Gaussian 54.39 16.13 m12d unemp, foreign, state, 8 consumer types Poisson 64311.15 10241.34 m09b unemp, foreign, state Poisson 64453.60 9016.30 m00e none Poisson 211670.94 51582.24 17 / 30

SLIDE 18

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 18/30

Comparing model errors

18 / 30

SLIDE 19

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 19/30

R squared estimates for each model

Beware: Unadjusted estimates, prone to overfitting

19 / 30

SLIDE 20

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 20/30

Results: Most favorable model Results: Most favorable model

20 / 30 20 / 30

SLIDE 21

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 21/30

Model specication of most favorable model

Model predictors: state (as multi level) + foreign + unemp

# Likelihood: afd_prop_log_z ~ dnorm(mu, sigma), # regression: mu <- beta0[state_id] + beta1*for_prop_z + beta2*unemp_prop_z, #priors: beta0[state_id] ~ dnorm(0, sigma2), sigma ~ dcauchy(0, 1), sigma2 ~ dcauchy(0, 1), beta1 ~ dnorm(0, 1), beta2 ~ dnorm(0, 1)

21 / 30

SLIDE 22

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 22/30

Coecients level 1

Model predictors: state (as multi level) + foreign + unemp 22 / 30

SLIDE 23

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 23/30

Coecients level 2

Model predictors: state (as multi level) + foreign + unemp 23 / 30

SLIDE 24

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 24/30

Big fat hairy catterpillars, as it should be

Model predictors: state (as multi level) + foreign + unemp 24 / 30

SLIDE 25

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 25/30

Observed vs. estimated AfD votes

Model predictors: state (as multi level) + foreign + unemp 25 / 30

SLIDE 26

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 26/30

Regional patterns of prediction errors

26 / 30

SLIDE 27

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 27/30

Conclusions Conclusions

27 / 30 27 / 30

SLIDE 28

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 28/30

Theoretical implications

Region related patterns appear to play an important role more than unemployment rate and foreigner rate not yet well understood rural? aged society? The present model is simplistic (The proposed) personality pattern didn't show strong impact Personality data representative? Let's model future elections Pathways of voter behavior remains opaque

Nicole Berbuir, Marcel Lewandowsky & Jasmin Siri (2015) The AfD and its Sympathisers: Finally a Right-Wing Populist Movement in Germany?, German Politics, 24:2, 154-178, DOI: 10.1080/09644008.2014.982546

28 / 30

SLIDE 29

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 29/30

Statistical implications

Observational research is a very limited guide for causal interpretations Overfitting (and underfitting) is to be expected Reduced sample size of electoral disctricts warrants further investigation Explorative study, no strong conclusions warranted More models are possible (but inject researchers' degree of freedom) 29 / 30

SLIDE 30

18.3.2019 Predictors of AfD party success in the 2017 elections file:///Users/sebastiansaueruser/Documents/research/polit/AfD_consumer_values/afd-modeling-ECDA-2019.html#17 30/30

Thank you Thank you

Sebastian Sauer Sebastian Sauer  sebastiansauer sebastiansauer  https://data-se.netlify.com/ https://data-se.netlify.com/  ssauer@posteo.de ssauer@posteo.de  Get slides here: Get slides here: https://data-se.netlify.com/slides/afd_ecda2019/afd- https://data-se.netlify.com/slides/afd_ecda2019/afd- modeling-ECDA-2019.pdf modeling-ECDA-2019.pdf CC-BY CC-BY 30 / 30 30 / 30