Abstract
This chapter delves deeper into the contextual aspect of cleavage voting. The first part of the chapter focusses on the cantonal distribution of cleavage influence in order to examine the magnitude of differences and to detect possible geographical patterns. In a second step, a cluster analysis groups the cantons into three types according to contextual cleavage variables for religion, social class and rural-urban. This new typology then provides the basis for a first statistical test of different voting patterns in the three cantonal groups. The main empirical analysis using both individual and contextual variables is then presented. This part begins with a theoretical discussion about the possible effects for each cleavage, including an interplay between both the individual and contextual level. The subsequent analysis relies on multilevel modelling that allows for the inclusion of cross-level interactions. After discussion of the findings, the chapter concludes with a summary of the most important results.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The table also displays the uncertainty of the calculated lambda values in terms of the standard deviation. For smaller cantons, i.e. for those with fewer respondents, the uncertainty is especially big, which has to be kept in mind during the discussion of the results. The comparatively high general uncertainty is also one reason why I refrain from a very detailed discussion of differences between cantons, but rather group several cantons and focus more on the big picture.
- 2.
I do not present the cantonal lambda based on church attendance as the variable would need a recoding to have sufficient numbers per categories in each canton, which would weaken the explanatory power of the variable. As denomination already contains only three categories and in addition represents the empirical element of the religious cleavage, I decided to present only this religious aspect.
- 3.
At the cantonal level the eight classes according to Oesch are too demanding for the small n in several cantons. As in the longitudinal chapter, I therefore recoded the eight categories into three but this time according to the new cleavage, rather than according to the old one. The three recoded categories are (1) socio-cultural and technical specialists, (2) service/production workers, clerks and small business owners and (3) managers/administrators, liberal professions and large employers. This regrouping fits to the classification of losers and winners of globalisation, with the second category comprising the losers. A similar recoding can be found in Goldberg and Sciarini (2014).
- 4.
Actually, by using the sample mean and sample standard deviation one should rather speak of a studentization as standardisation would imply the use of the respective population parameters.
- 5.
STATA offers the command cluster with the option complete linkage, which runs exactly the described analysis.
- 6.
Measured by the proportion of persons living in a canton who do not belong to any denomination.
- 7.
Still, one has to keep in mind the restrictions of comparing logistic coefficients that may also affect the levels of significance. Alternative quantities like predicted probabilities that I calculated in Chap. 4 would help to compare effects for specific types of voters, i.e. fixed on certain values on the socio-demographic variables and controls. However, such a choice for a certain reference type may also lead to wrong or rather incomplete conclusions as some variables have a stronger/weaker effect depending on the chosen reference categories. Hence, I decided to stick to the simple comparison of significance levels with the mentioned reservations.
- 8.
The following theory and hypotheses are adapted from Goldberg (2014).
- 9.
For a general discussion of contextual influence see also Sect. 2.3
- 10.
The study by van der Brug et al. (2009) provides evidence for a similar reasoning at the European level. Religion plays a more important role in influencing party choice, the higher the degree of religious fractionalisation in a country. Although the latter measure is different to what will be used here, the argument is along the same lines - the higher importance of individual religious factors, the tighter the religious competition is.
- 11.
In fact, the labelling “rural-urban” is not ideal and might even be misleading as more adequate measures are not a mere binary measure, but include finer grained differences between citizens’ places of residence (see Sect. 3.2.3). Hence, a label such as “place of residence” would make more sense. For reasons of consistency, though, I mostly stick to the classic label of rural-urban.
- 12.
To be more precise, due to the variables and definitions chosen to measure rural-urban effects, in purely rural cantons there is absolutely no variation on the individual level. All citizens live in rural municipalities meaning that individual place of residence plays absolutely no role. In purely urban areas there are (small) variations between cities and municipalities of agglomerations. However, as the opposing group of rural residents is not present, the individual effects of residence again should be of less importance.
- 13.
In Switzerland rurality of a canton is strongly linked to the size of the canton resulting in a plurality vote in the most rural cantons (UR, OW, GL and AI). This fact leads to the withdrawal of several parties in these cantons. Interesting for this study is the fact that the typical urban parties SP and GP do not (regularly) run for election in these small rural cantons (exception SP in Glarus). In case of the GP, the decision to withdraw is mainly driven by the “mechanical” barrier for a small party, but for the SP it might also be somehow linked to its profile as a more urban party.
- 14.
The data for this calculation of vote-splitting stems from the BfS.
- 15.
An alternative approach to measuring religious competition by including also other denominations and the growing group of secular people would be the Herfindahl index as used by van der Brug et al. (2009). Testing this measure on voting behaviour, though, does not lead to significant effects as the Catholic-Protestant ratio does. This means that the main conflict at the contextual level is still between Catholics and Protestants and that the denomination of the dominating group plays a crucial role, which is irrelevant when using the Herfindahl index.
- 16.
As a reminder, Table A.4 in the appendix displays more details about the three contextual variables.
- 17.
For more details about the recoding see the operationalisation on p.42f. The corresponding regression tables in the appendix also provide information on the included variables and operationalisation used for the respective graphs.
- 18.
The separate estimation of the five models implies that the dependent variables, i.e. the voting propensities for our five parties, are independent of each other. In reality this might not be the case as the probability to vote for a given party may depend on the presence of a second very similar party. To test for such a dependence of available party choices I ran seemingly unrelated regressions (SUR) but without modelling the hierarchical structure of the data (Zellner 1962). SUR models control for the possible correlation of the residuals between the single regression models. Indeed, for some party models the residuals are correlated, unsurprisingly strongly between the SP and GP model (0.47). However, even taking this correlation into account in the estimation of the voting propensities does not result in significant changes of the coefficients. Hence, I am confident to continue the discussion using the separately estimated models.
- 19.
The non-inclusion of issue preferences is one of the main methodological criticisms of Franklin (2010) in current cleavage research. Although the focus of this book is not the distinction between the effect of group membership and value preferences, as the latter serve here more as controls, the simple comparison of model fit reveals some interesting points. As already mentioned, the inclusion of both value scale variables improves the model fit (AIC & BIC). In fact, for all parties except the CVP the best model fit results from the complete model including both socio-structural and normative variables. To be sure, though, the improvement in model fit that stems from adding the value scales is significantly bigger than from adding socio-structural factors. Hence, one may interpret this as a stronger influence of specific value preferences. The socio-structural group membership, though, still plays at least a supportive role.
- 20.
Note also that the Catholic-Protestant ratio loses in strength and becomes non-significant in the model excluding wealth inequality. The absolute change of the coefficient and standard error, though, is very minimal compared to the complete model so that no substantive change happened. In the model without urbanisation the effect of religion becomes slightly stronger.
- 21.
The finding of a systematic effect of the electoral threshold, though, would not have been unrealistic. In a similar analysis running separate models for 2007 and 2011, I could show a significant drop of CVP voting probability in bigger cantons between both elections. This drop was probably due to the now more fierce competition in the centre of the party spectrum with the additional competitors of the GLP and BDP in bigger cantons (cf. Goldberg 2014).
- 22.
In an earlier version of the analysis published in Goldberg (2014) I found a similar result using a different contextual measure, namely Catholic proportion living in a canton. The use of the ratio between Catholics and Protestants leads to even stronger results indicating that the antagonism between both denominational groups is more relevant than only considering the amount of Catholics. The strength or activation of an individual effect depends on the Catholic proportion, but also the extent to which Catholics face a strong opposing group in religiously mixed cantons.
- 23.
The appendix includes these more detailed figures also for all following marginal effect graphs.
- 24.
In this and several of the following graphs the scale for the y-axis differ in terms of the displayed values. However, the range between the lowest and highest displayed value is always the same for a given independent variable. This enables an easy comparison of the magnitude that the random slopes change by the corresponding environment.
- 25.
In fact, the effect at a level of 30 % urbanisation may already be zero indicated by the large confidence intervals for weakly urbanised cantons. The large confidence intervals are due to very few cantons with a low level of urbanisation, so the effects for low urbanisation levels strongly depend on the canton Jura and to some extent also on Graubünden and Thurgau, which also have an urbanisation of less than 50 %.
References
Ackermann, K., & Traunmüller, R. (2014). Jenseits von Schwerkraft und Höllenfeuer. Nicht-lineare Kontexteffekte auf den Zusammenhang von religiöser Gruppenzugehörigkeit und individuellem Wahlverhalten bei fünf Bundestagswahlen. Politische Vierteljahresschrift, 55(1), 33–66.
Andersen, R., & Heath, A. (2002). Class matters: The persisting effects of Contextual social class on individual voting in Britain, 1964–1997. European Sociological Review, 18(2), 125–138.
Anderson, C. J. (2007). The interaction of structures and voter behavior. In R. J. Dalton & H.-D. Klingemann (Eds.), Oxford handbook of political behavior (pp. 589–609). Oxford and New York: Oxford University Press.
Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182.
Bochsler, D., & Wasserfallen, F. (2013). Switzerland: Moving towards a nationalized party system. In R. Dandoy & A. Schakel (Eds.), Regional and national elections in Western Europe (pp. 234–253).
Botterman, S., & Hooghe, M. (2012). Religion and voting behaviour in Belgium: An analysis of the relation between religious beliefs and Christian Democratic voting. Acta Politica 47(1), 1–17.
Bovay, C., & Broquet, R. (2004). Eidgenössische Volkszählung 2000: Religionslandschaft in der Schweiz Neuchâtel: Office Fédéral de la Statistique.
Brambor, T., Clark, W. R., & Golder, M. (2006). Understanding interaction models: Improving empirical analyses. Political Analysis, 14, 63–82.
Butler, D., & Stokes, D. E. (1969). Political change in Britain: Forces shaping electoral choice. London: Macmillan.
Charnock, D. (1997). Spatial variations, contextual and social structural influences on voting for the ALP at the 1996 federal election: Conclusions from multilevel analyses. Australian Journal of Political Science, 32(2), 237–254.
Ferrer-i-Carbonell, A. (2005). Income and well-being: An empirical analysis of the comparison income effect. Journal of Public Economics, 89(5–6), 997–1019.
Franklin, M. N. (2010). Cleavage research: A critical appraisal. West European Politics, 33(3), 648–658.
Friedrich, R. J. (1982). In defense of multiplicative terms in multiple regression equations. American Journal of Political Science, 26(4), 797–833.
Geissbühler, S. (1999). Are catholics still different? Catholicism, political attitudes and party attachments in Switzerland, 1970–95. West European Politics, 22(3), 223–240.
Gelman, A., & Hill, J. (2007). Data analysis using regression and multilevel/hierarchical models. Cambridge/New York: Cambridge University Press.
Gelman, A., Kenworthy, L., & Su, Y.-S. (2010). Income inequality and partisan voting in the United States. Social Science Quarterly, 91(5), 1203–1219.
Geser, H. (1997). The persistence of confessional political cultures in Switzerland: Some puzzling empirical evidence from local parties. Swiss Political Science Review, 3(2), 1–29.
Gilg, P. (1987). Stabilität und Wandel im Spiegel des regionalen Abstimmungsverhaltens. Schweizerisches Jahrbuch für Politische Wissenschaft, 27, 121–158.
Goldberg, A. C. (2014). The impact of religion on voting behaviour – A multilevel approach for Switzerland. Swiss Political Science Review, 20(2), 305–329.
Goldberg, A. C., & Sciarini, P. (2014). Electoral competition and the new class cleavage. Swiss Political Science Review, 20(4), 573–589.
Huckfeldt, R. R. (1984). Political loyalties and social class ties: The mechanisms of contextual influence. American Journal of Political Science, 28(2), 399–417.
Jeitzinger, B., & Peters, R. (2007). Regionale Einkommens- und Vermögensverteilung in der Schweiz: Was sagen die Steuerdaten? Die Volkswirtschaft, 12, 16–21.
Johnson, M., Phillips Shively, W., & Stein, R. M. (2002). Contextual data and the study of elections and voting behavior: Connecting individuals to environments. Electoral Studies, 21(2), 219–233.
Johnston, R., Jones, K., Propper, C., & Burgess, S. (2007). Region, local context, and voting at the 1997 general election in England. American Journal of Political Science, 51(3), 640–654.
Kam, C. D., & Franzese, R. J. (2007). Modeling and interpreting interactive hypotheses in regression analysis. Ann Arbor: University of Michigan Press.
Klöti, U. (1998). Kantonale Parteiensysteme. Bedeutung des kantonalen Kontexts für die Positionierung der Parteien. In H. Kriesi, W. Linder, & U. Klöti (Eds.), Schweizer Wahlen 1995 (pp. 45–72). Bern: P Haupt.
Kotler-Berkowitz, L. A. (2001). Religion and voting behaviour in Great Britain: A reassessment. British Journal of Political Science, 31(3), 523–554.
Kriesi, H. (1998a). Einleitung. In H. Kriesi, W. Linder & U. Klöti (Eds.), Schweizer Wahlen 1995 (pp. 1–16). Bern: P Haupt.
Kriesi, H. (1998b). The transformation of cleavage politics: The 1997 Stein Rokkan lecture. European Journal of Political Research, 33 165–185.
Kriesi, H., & Trechsel, A. H. (2008). The politics of Switzerland: Continuity and change in a consensus democracy. Cambridge/New York: Cambridge University Press.
Lachat, R. (2010). Electoral competitiveness and issue voting. Paper prepared for the 2010 National Conference of the Midwest Political Science Association, April 22–25, Chicago.
Lachat, R., & Selb, P. (2005). Schweiz. In H. Kriesi, R. Lachat, P. Selb, S. Bornschier, & M. Helbling (Eds.), Der Aufstieg der SVP (pp. 41–58). Zürich: Neue Zürcher Zeitung.
Ladner, A. (2004). Typologien und Wandel: Die kantonalen Parteiensysteme im letzten Drittel des 20. Jahrhunderts. Swiss Political Science Review, 10(4), 3–32.
Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1944). The people’s choice: How the voter makes up his mind in a presidential campaign (2nd ed.). New York: Columbia University Press.
Lijphart, A. (1997). The difficult science of electoral systems: A commentary on the critique by Alberto Penadés. Electoral Studies, 16(1), 73–77.
Linder, W. (2009). Das politische System der Schweiz. In W. Ismayr (Ed.), Die politischen Systeme Westeuropas (pp. 567–606). Wiesbaden: Verlag für Sozialwissenschaften.
Magalhães, P. C., & Önnudottir, E. H. (2014). The economy and the vote: The economy as context. Paper presented at the 3rd European Conference on Comparative Electoral Research, Thessaloniki, 24–27 April 2014.
McCarty, N. M., Poole, K. T., & Rosenthal, H. (2006). Polarized America: The dance of ideology and unequal riches. Cambridge, MA: MIT Press.
Mood, C. (2010). Logistic regression: Why we cannot do what we think we can do, and what we can do about it. European Sociological Review, 26(1), 67–82.
Nef, R. (1980). Struktur, Kultur und Abstimmungsverhalten: Zur interregionalen Variation von politischen Präferenzen in der Schweiz 1950–1977. Schweizer Zeitschrift für Soziologie, 6(2), 155–190.
Oesch, D. (2006). Redrawing the class map: Stratification and institutions in Britain, Germany Sweden and Switzerland. Basingstoke & New York: Palgrave Macmillan.
Pappi, F. U. (1985). Die konfessionell-religiöse Konfliktlinie in der deutschen Wählerschaft: Entstehung, Stabilität und Wandel. In D. Oberndörfer, H. Rattinger, & K. Schmitt (Eds.), Wirtschaftlicher Wandel, religiöser Wandel und Wertwandel (pp. 263–290). Berlin: Duncker & Humblot.
Persky, J., & Tam, M.-Y. (1990). Local status and national social welfare. Journal of Regional Science, 30(2), 229–238.
Rydgren, J., & Ruth, P. (2011). Contextual explanations of radical right-wing support in Sweden: Socioeconomic marginalization, group threat, and the halo effect. Ethnic and Racial Studies, 36(4), 711–728.
Sciarini, P. (2010). Le potentiel électoral des partis de gauche. In S. Nicolet & P. Sciarini (Eds.), Le destin électoral de la gauche (pp. 87–129). Chêne-Bourg: Georg.
Stark, R., & Bainbridge, W. S. (1996). Religion, deviance and social control. New York: Routledge.
Steenbergen, M. R., & Jones, B. S. (2002). Modeling multilevel data structures. American Journal of Political Science, 46(1), 218–237.
Tillie, J. (1995). Party utility and voting behavior. Amsterdam: Spinhuis.
Tresch, A. (2005). Le soutien aux autorités lors des votations fédérales de 1971 à 1999 dans les cantons suisses. Swiss Political Science Review, 11(2), 1–25.
van den Berg, J. C., & Coffé, H. (2011). Educational and class cleavages in voting behaviour in Belgium: The effect of income, EGP class and education on party choice in Flanders and Wallonia. Acta Politica, 47(2), 151–180.
van der Brug, W., Hobolt, S. B., & de Vreese, C.H. (2009). Religion and party choice in Europe. West European Politics, 32(6), 1266–1283.
van der Brug, W., van der Eijk, C., & Franklin, M. N. (2007). The economy and the vote: Economic conditions and elections in fifteen countries. New York: Cambridge University Press.
van der Eijk, C., & Franklin, M. N. (1996). Choosing Europe? The European electorate and national politics in the face of union, Ann Arbor: University of Michigan Press.
Van der Eijk, C., van der Brug, W., Kroh, M., & Franklin, M. N. (2006). Rethinking the dependent variable in voting behavior: On the measurement and analysis of electoral utilities. Electoral Studies, 25(3), 424–447.
Vatter, A. (2002). Kantonale Demokratien im Vergleich: Entstehungsgründe Inter aktionen und Wirkungen politischer Institutionen in den Schweizer Kantonen (Vol. Bd. 159). Opladen: Leske + Budrich.
Wright, E. O. (1979). Class structure and income determination. New York: Academic Press.
Zellner, A. (1962). An efficient method of estimating seemingly unrelated regressions and tests for aggregation bias. Journal of the American Statistical Association, 57(298), 348.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Goldberg, A.C. (2017). Contextual Approach of Voting. In: The Impact of Cleavages on Swiss Voting Behaviour. Contributions to Political Science. Springer, Cham. https://doi.org/10.1007/978-3-319-46000-0_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-46000-0_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-45999-8
Online ISBN: 978-3-319-46000-0
eBook Packages: Political Science and International StudiesPolitical Science and International Studies (R0)