Fig 1.
Multidimensional correspondence analysis of national stories and vote in the US.
Fig 2.
Multidimensional correspondence analysis of national stories and vote in the United Kingdom.
Fig 3.
Multidimensional correspondence analysis of national stories and vote in Denmark.
Fig 4.
Multidimensional correspondence analysis of national stories and vote in the Netherlands.
Table 1.
The most common past (p) and future (f) story components by story type.
Fig 5.
The effect (displayed as coefficients) of respondents’ national stories on their vote choices.
Note. The boundary, survival and self-expression are three scales for the three national story types. Regression models include socio-demographic controls and attitudinal variables (political interest, political knowledge and ideological self-placement).
Fig 6.
The “us versus them” Boundary story among voters for mainstream and populist radical right parties.
Note. Fig 6 displays the dispersion of the “us versus them” Boundary story type among populist and mainstream voters in each of the four countries studied.
Fig 7.
Estimated effect of the “us versus them” Boundary story type on the vote for populist radical right parties.
Note. Marginal effect of the “us versus them” Boundary story type on the vote for populist radical right parties (vertical axis) across levels of different socio-demographic variables. Marked are 95% confidence intervals. All moderating variables (on the x axis) span from marginalized groups to more privileged voters: The social class variable spans from working to upper class; education spans from 8th grade or less to graduate and post graduate; and Urban/Rural spans from living in a rural area to living in a city/town with over 1,000,001 inhabitants. Results are based on estimation reported in S1F Appendix. The histogram below the predicted margins displays the distribution of the moderators. Figure uses the INTERFLEX package [93].