11
IDEOLOGY AND ELECTORAL CHOICE

Martin Elff

 

 

 

Introduction

Few topics in Political Science appear as burdened with difficulties as the term “ideology.” On the one hand, the term “ideological” is often used in political discussions to criticize or vilify opponents either as irrational and intransigent – as sticking to principles beyond reason – or as insincere – as masquerading special interests as a common good. On the other hand, the meaning of the term “ideology” in Political Science seems to have been “thoroughly muddied by diverse uses” (Converse 1964: 207). Yet the diversity of uses does not indicate that it is impossible to associate a meaning with the term. While “ideology” is used to refer to different phenomena in different areas of scholarship, this does rarely lead to disagreements about its denotation within a particular field. Furthermore, despite this diversity, the usages of the term are not unrelated. Usually they refer to a set of abstract or general ideas, distinct from parties’, candidates’ or citizens’ positions on particular issues or from specific policy plans. Disregarding the more polemic uses of the term, “ideology” has been used to refer to (see also Sartori 1969a; Jost 2006; Knight 2006):

 

1 a set of ideas that justify a social or political state of affairs

2 a set of ideas that give the (usually economic) interests of a certain social group a moral or (more generally) normative appeal

3 a particular line or tradition of political thought, such as Liberalism, Conservatism, Socialism, Fascism, etc.

4 the set of ideas that determine the political aims and policy positions of a political party

5 factors that structure citizens’ attitudes and values.

 

While the first four of these uses are coherent with Downs’ definition of ideology as “a verbal image of the good society and of the chief means of constructing such a society” (Downs 1957: 96), it is the fifth of these that appears to be immediately relevant for voting behavior. But the third and the fourth uses of the term “ideology” are also relevant for voting behavior and its analysis, though in a less obvious way. Their relevance comes from the consideration that it would be surprising to find persistent and coherent patterns of voting if not for persistent and coherent differences between parties in terms of the polices they announce or promise to voters. These differences between parties may be related to their membership in a party family – for example, of the liberal, conservative or social democratic parties – which are each rooted in one of the grand traditions of political thought that emerged in the nineteenth century – that is, Liberalism, Conservatism, Socialism, etc. (see Heywood 2003). Yet while categorizing parties in party families may facilitate the comparative description of class voting or voting along religious-secular lines, it masks out the variability of parties’ policy positions. If parties that compete for voters in a country become more similar in terms of their policy positions, announcements and promises, they give fewer incentives to voters from different social groups to diverge in terms of voting behavior. Thus the political convergence of parties of the left and the center/right has been suggested as an explanation of the decline in class voting in Western Europe (see, for example, Elff 2009 and Evans and Northmore-Ball in this volume).

The spatial theory of party competition in the tradition of Downs (1957) views parties’ ideologies as positions in an abstract unidimensional space, the principal directions of which are typically identified with the labels “left” and “right” (when applied to European politics) or “liberal” and “conservative” (when applied to American politics). The idea of such an overarching political dimension that lies behind the various policy or issue positions is however not confined to the spatial theory of party competition, but also common in the empirical analysis of patterns in citizens’ political opinions and attitudes. Therefore, the question of whether such “citizen ideologies” (as distinguished from party ideologies) are adequately described by a single dimension or by multiple dimensions will be a central one in this chapter as well as the question about the origins of such dimensions. Another question addressed in this chapter is whether ideologies can be the foundation of new lines of cleavage, especially if traditional social cleavages based on class and religion have faded away. Yet the first question to be addressed is to what degree citizens have ideologies at all, since it was argued early on (Converse 1964) that ideologically coherent patterns of citizens’ attitudes and values are a rare phenomenon and that – at least in the United States of the 1950s – most citizens are (or were) “innocent of ideology” (but see Jost 2006; Bølstad in this volume).

The question of coherence in belief systems

In his seminal contribution “The Nature of Belief Systems in Mass Publics,” Philip Converse suggested substituting the “muddied” concept of ideology with the concept of a belief system, which he defines as “a configuration of attitudes and ideas in which the elements are bound together by some form of constraint or functional interdependence” (Converse 1964: 207). While such a “constraint” is understood as an intra-individual connection between these beliefs and ideas, so that the change of one idea or attitude would “psychologically” require the change of another idea or attitude, in actual empirical research such constraint is usually operationalized as a correlation between attitudes or ideas. Based on the low correlation among attitudes toward political issues that are related in content or in terms of their ideological significance, Converse argues that such belief systems in the (then) contemporary American public are quite rare, restricted to a thin elite of highly involved political activists and experts.

Intra-individual connections between beliefs and ideas are however not sufficient for the existence of correlations between ideas or attitudes in the general public, which poses a problem for the operationalization of constraint in terms of correlations. That is, if there is a high variety of different belief systems – each with a different pattern of constraint – the correlation among measurements of ideas and attitudes from different people may be quite low. For strong correlations to exist there must also be a high level of polarization between a small number of types of individual belief systems. Indeed, Converse’s pessimistic assessment has been questioned and criticized repeatedly, if only for slightly different reasons. The first type of criticism was methodological. One argument, which goes back to Achen (1975), is that the low correlation among issue attitudes that Converse found were not indicative of a lack of constraint, but of a lack of reliability in attitude measures. The idea of inter-individually differing belief systems has led other authors to posit the existence of “hierarchical” or “vertical” patterns of constraint. These are constraints between abstract principles and attitudes toward the issues in a particular policy area that citizens might be particularly interested or engaged in (Peffley and Hurwitz 1985). The other type of criticism questioned the general relevance of the finding of low constraint. Thus it has been suggested that the apparent low level of constraint was a phenomenon of the halcyon American politics of the 1950s and that attitude constraint and polarization was much higher in the more turbulent 1970s (see, for example, Nie and Andersen 1974) and in the highly polarized politics of the current decade (see, for example, Jewitt and Goren 2016). Yet there are also disagreeing voices that state that partisan polarization and issue polarization between parties has increased, but not issue constraint (see, for example, Mason 2015). A result that still stands, however, is that ideological constraint is related to political sophistication and to education and indeed the gains in constraint appear to be restricted to the educated and politically engaged (see, for example, Jewitt and Goren 2016); but ideological constraint varies not only systematically across individuals but also across countries. For example, a high visibility of the welfare state also increases the constraint among attitudes toward the welfare state (Gingrich 2014).

As just discussed, there is evidence of systematic differences between individuals in terms of belief system constraint, at least if measured in terms of correlations among attitudes. Those correlations are however inter-individual patterns, while constraint really is a concept that concerns intra-individual patterns of attitudes. Thus the findings of variations over time and across countries may be the artifacts of variations in the polarization of attitudes, which in turn may be affected by the polarization at the level of political elites. On the other hand, it is quite plausible that the variations across individuals, in so far as they are related to education and political involvement, may indeed reflect variations in intra-individual constraint. To uncover these effects of political context and individuals’ political sophistication, it is necessary to find ways to measure intra-individual constraint of political attitudes independent from correlations across individuals, which is still an unmet challenge (but see Erikson this volume).

Ideology as a means to facilitate electoral choice

A fundamental idea of the spatial theory of ideologies, which goes back to Downs’ Economic Theory of Democracy (1957), is that they can be represented by locations on a single dimension, usually identified with a left–right or liberal–conservative axis: The more similar two ideologies are, the closer to each other they are on this axis. Extreme ideologies are more distant from the middle of the axis than centrist ideologies. In so far as ideology is relevant for citizens’ voting decisions, they choose the party or candidate that has the closest to their own ideological position.

In Downs’ version of the theory (1957), neither parties nor voters are interested in ideologies themselves. For parties and candidates, ideologies are just means to win elections. For voters, ideologies are means for the reduction of information costs. They simplify the choice between parties, because voters do not need to acquire detailed information about parties’ potential government activities and to evaluate these in terms of their own well-being. Instead, voters can identify a party’s notion of the good society and what place they themselves would have in it. That is, parties’ or candidates’ ideological positions are some sort of heuristics.

It is not immediately clear how this spatial notion of ideologies can be reconciled with the idea that they are descriptions of “good societies” and the way toward them. In order to be able to compare parties’ ideological positions with their own position, they need to be able to have such ideological positions, and that means to have their own notion of a good society. This seems to require a certain degree of abstract thinking, which, as we saw previously, not many citizens are capable of or engaged in. Alternatively, one could assume that citizens do not care about the ideologies in themselves. Instead they look at ideologies only in terms of those aspects they are affected by. These aspects then do not need to lie on the same dimension as the ideologies. This idea was formalized by Hinich and Pollard (1981) and Enelow and Hinich (1982), who posit that voters care about parties’ or candidates’ positions on issues, but find it costly to learn about these positions directly. Instead, they use linear mappings to predict these positions from parties’ or candidates’ ideological positions. While this idea has nice mathematical properties, it actually begs the question of how citizens construct these mappings. Irrespective of whether these mappings are the product of logical inference or of learning from experience, they would constitute a considerable cognitive achievement. In fact, this idea of linear mappings between ideological positions and particular issue positions seem to fit together with the idea of attitude constraint manifested in correlations among attitudes: Like in the general factor analysis model, the less “noisy” individuals’ mappings between ideological positions and issue positions are, the higher the absolute correlation among issue attitudes will be.

The idea of ideological distinctions as inferential devices is not restricted to research motivated by the spatial theory of voting. From cognitive psychology comes the idea of the left–right distinction as a schema (Conover and Feldman 1984). A schema is a cognitive structure that helps individuals to organize diverse experience. Schemas can have different levels of abstraction and are typically domain-specific, that is, used (only) in a particular area of experience. But this means that if the notions of “left” and “right” each refer to a schema or if the left–right axis is a schema, as it appears to be in West European politics (Fuchs and Klingemann 1990), it is not necessarily the only one. There may be other, general or domain-specific, schemata that are used as an alternative or as a supplement to the left–right schema to make sense of political information and to take positions on particular issues (see, for example, Medina 2015).

If voters do not care for each and every implication that an ideology has, then a complete and coherent belief system may not be necessary for ideological cues to be useful for them. But at least they will need a basic understanding of what it means for a party or candidate to be either “left” or “right,” “liberal” or “conservative.” Again it was Converse who brought dismal tidings: Only a small section of the citizenry – at least at the time of his writing – actively used ideological categories explicitly in their evaluation of parties and an only somewhat broader section appeared able to make sense of ideological labels when presented with them (Converse 1964). While results for West European countries are somewhat more favorable, there are still differences between educational groups in terms of recognition and understanding of “left” and “right,” even though they appear small in comparison to the differences between the West European countries and the US (Klingemann 1972). Yet respondents in the study by Fuchs and Klingemann (1990) often use dichotomies of “progressive” vs. “conservative” and “communism” vs. “fascism” as interpretations of “left” and “right” and thus use categories no less abstract than the original terms, while they rarely use more concrete meanings such as “worker” vs. “entrepreneur” or “poor” vs. “rich.” But if citizens are only able to paraphrase abstract terms with different abstract terms, there is room for doubt that they are able to make any specific inferences with regards to parties’ or candidates’ policy positions.

Even if a citizen is able to associate substantial issue content with labels such as “left” and “right” in a wide range of policy areas, and place themselves and parties correctly on a left–right scale, this will not prove that he or she uses left–right positions to infer positions on particular issues. While such a finding would be consistent with such a use, it is also possible that he or she has just merely learned, after picking up a certain set of positions, that certain positions and the use of “left” and “right” go together. To prove his or her inferential use of left–right positions, one will need to provide him or her with left–right positions of (fictitious) parties and candidates and record whether he or she makes correct predictions about these parties’ issue positions. It appears that no research in this direction has been undertaken yet.

The content and dimensionality of “left” and “right”

It would be a fallacy to conclude from the ubiquity of the use of “left” and “right” as political categories that there exists a consensus about their issue content. This insight has led many scholars to look into the correlates of citizens’ left–right self-placement. If these correlates are indicative of the content of “left” and “right,” then there is ample evidence that it varies considerably across space (i.e., countries) and time (see, for example, van Elsas and van der Brug 2015).

Further, there is evidence that ideological positions cannot or can no longer be adequately described by positions on a single axis. A growing literature (see, for example, Hooghe, Marks and Wilson 2002; Feldman and Johnston 2014) suggests that there are instead at least two ideological dimensions:

 

an economic left–right or “materialist” dimension that contrasts

demands for redistribution of assets and income toward more equality, public provision of welfare benefits and an active role of the state in controlling the economy, with

an affirmation of the freedom from interventions into property rights and demands for the state to let markets run their course

 

a “non-materialist,” “post-materialist,” “social,” “authoritarian-libertarian” or “GAL/ TAN” dimension that contrasts

the affirmation of citizens’ political rights and individuals’ liberties to choose their way of life, with

an emphasis of authority of the state to constrain or coerce citizens in the name of public security and an affirmation of traditional norms of piety and modesty.

 

The notion of two ideological dimensions has become popular in the literature that postulates that a value change from materialist to post-materialist priorities has occurred. This value change is supposed to have led to a confrontation between “old politics” and “new politics,” which cuts across traditional left–right alignments based on economic interests and class divisions (see, for example, Inglehart 1984). But this second dimension is arguably a quite old one, which derives from religious-secular divisions and conflicts about the contrast between individual freedom and the authority of the national state, conflicts that go back to the late nineteenth and early twentieth century and that have become manifest in the divergent ideological movements of Liberalism and Conservatism. Considering that in some cases the authority of the nation-state could get into conflict with Catholic church religiosity, it might be reasonable to distinguish two non-economic ideological dimensions (see, for example, Elff 2009), one contrasting individual wayof-life liberties with traditional-religious norms and one contrasting citizens’ rights with the authority of the nation-state (Lipset and Rokkan 1967; Elff and Rossteutscher 2017).

So how many dimensions of ideology exist? The significance of the answer to this question depends on whether ideological dimensions are genuine (latent) factors or epiphenomenal to clusters of ideas. In the first case, if ideologies are genuine latent factors, they will restrict the degrees of freedom of positional changes by parties – for example, by restricting them to movements on a single left–right axis. The number of ideological dimensions could then be determined by a combination of theoretical reasoning and sophisticated analysis of data on parties’ electoral platforms and/or voters’ issue preferences. In the second case, if ideologies are primarily sets or clusters of ideas, each cluster being held together by a common origin and/or logical or factual coherence, then ideological dimensions are a mere means for the description of differences among these idea clusters. Sophisticated data analysis may still lead to results that suggest a representation of ideologies by a small number of dimensions, but the implications of such findings will be limited and likely to be dependent on context and time. But the different clusters/ dimensions may also be ephemeral side-effects of an evolution of the meaning of left–right itself as new issues arise that will eventually be absorbed into the left–right dimension (van der Brug and Franklin in this volume), perhaps as a pre-condition for their having effects on party support.

Operational and symbolic ideology

It is tempting to equate citizens’ self-placement on a scale with ideological labels such as “left” and “right” or “liberal” and “conservative,” etc. with actually having corresponding ideological orientations. Yet if someone reports a “moderately leftist” score of 3 on a 10-point left–right scale, for example, then this does not imply that he or she will support all the issue stances that a party or candidate with such an ideological position will assume. Instead, respondents’ left– right or liberal–conservative self-placements and their issue positions may empirically diverge and in some instances they do. Ellis and Stimson (2009) therefore distinguish between an “operational ideology” and a “symbolic ideology.” An operational ideology is a coherent set of attitudes and beliefs similar to Converse’s concept of a belief system. An individual’s symbolic ideology is his or her self-identification with ideological labels, such as “left” or “right,” or with groups denoted by such labels, such as “liberals” or “conservatives.” The phenomenon of “conflicted conservatives,” who identify themselves with an ideological symbol without actually supporting any of the policies that members of the political elite would ascribe to it (Ellis and Stimson 2012), are an instructive example for such a divergence.

While the concept of symbolic ideology suggests that it is causally prior to vote decisions and to some degree also to operational ideology, other interpretations of ideological self-identification view it as a consequence of partisanship and/or party preference. Thus when Inglehart and Klingemann (1976) examine the relative influence of value orientations and partisanship on left–right self-placement, they find that the latter dominates. Thus the question may arise about the relative priority of party identification and partisanship on the one hand and ideological self-placement or symbolic ideological identification on the other. Both causal pathways appear equally plausible. If partisanship is a product of socialization (by parents and perhaps also by peer groups) and the meanings of ideological “labels” are learned from their application to the policy positions of the party one identifies with, then partisanship is causally prior. This is likely to be the case in the US where the ideological polarization between parties used to be low (even if it has increased recently). In contrast, if political socialization does not lead so much to the identification with parties than with ideological labels or ideologically labeled groups, then ideological self-identification is likely to condition individuals’ adoption of issue positions and electoral choices, which may then coagulate into partisan attachments. This is more likely to be the case in countries with several small to medium-size parties that sort themselves into ideological “camps,” such as in pre-1990 Italy. The causal relation between partisanship and ideological identification may however be reciprocal if the party system is sufficiently concentrated so that certain parties can, at least in public perception, be the exclusive “owners” of particular ideological brands. For example, if there is a major party calling itself the “Socialist Party” or similar and if this party describes itself as “moderate leftist” – as in many West European countries before 1990 – then partisan attachments and ideological identifications are likely to develop in tandem. It is of course difficult to disentangle these causal pathways empirically, because this requires panel survey studies with the right instruments. These are quite rare, with the various studies conducted by Michael Lewis-Beck and his co-workers on France as exceptions (e.g., Fleury and Lewis-Beck 1993). They find ideology to be a cause of partisanship rather than the other way around – though the findings are consistent with a view that party choice is a cause of both (see Dinas in this volume).

Psychological underpinnings

Much of the discussion about the concept of ideology and the causes and consequences of ideological thought and conceptualization considered so far can be characterized by a “top-down” perspective, where ideologies are originally systems of thought adopted by parties and political elites, which then color the perceptions of those who identify with those parties (see Heath in this volume). A more psychologically oriented “bottom-up” perspective can be contrasted with this (Jost, Federico and Napier 2009). In this perspective, ideologies are patterns of sentiments and cognition, rooted in individuals’ personality, their particular situation or experience, or even their genetic makeup.

Much of the research tradition on the psychological underpinnings of ideology goes back to Adorno et al.’s (1950) Authoritarian Personality and focuses on personality traits that dispose individuals toward right-wing authoritarian attitudes (see, for example, Altemeyer 1981) or hierarchical relations between groups (see, for example, Pratto et al. 1994). This research thus favors a uni-polar conception of ideologies or is restricted to a particular ideological content. Later authors focus on more general attitudes such as (in)egalitarianism and resistance to change and their relation to liberal–conservative or left–right self-placement (see, for example, Thorisdottir et al. 2007). Explanations of these phenomena appear to draw from the full inventory of psychological factors. They include situational factors such as perceived threat (see, for example, Jost 2009), psychological needs (see, for example, Thorisdottir et al. 2007), experiences during youth and childhood (see, for example, Block and Block 2006), personality factors such as the “Big Five” (Jost 2006), or even physiological and genetic factors (Smith et al. 2011; Funk et al. 2013).

Another area of psychological research is less focused on the content of ideological thinking than on the psychological foundation of what Sartori (1969b) calls the ideological mind-set. The central concept of this research is “motivated reasoning” (Redlawsk 2002), which involves not so much the deductive search for implications and consequences from certain premises as it tries to find reasons to justify a given and predetermined political decision – that is, to rationalize them. In particular, such research focuses on to what degree and under what conditions individuals retain certain political positions or preferences, despite being faced with factual information that contradicts the reasons that appear to justify these positions or preferences (see, for example, Kahan 2012).

While these lines of research contribute important insights into phenomena that are related to ideologies in the various meanings discussed at the beginning of this chapter, care must be taken that phenomena such as resistance to change are not confused with ideologies themselves. Furthermore, it seems worthwhile to explore the relation of psychological factors to politics in more refined ways than with reference to left–right or liberal–conservative self-placements. Finally, it appears that psychological factors are perhaps rather moderators than primary factors since, for example, the need for security varies between Eastern and Western Europe in terms of the relation with left–right self-placements (Thorisdottir et al. 2007).

Outlook: an ideology-based cleavage structure?

Apart from giving an overview of the role played by ideology in shaping electoral choices, a main purpose of this chapter is to examine whether ideologies can be the basis of new political cleavages after the traditional ones based on social structure have faded away, as often is claimed (see Evans and Northmore-Ball in this volume). As it turns out, the answer is (as ever so often) “it depends.” In the present case, it depends on the meaning of the terms “cleavage” and “ideology.”

While the concept of cleavage is no more clear than that of ideology, for reasons of space the following “minimal” definition will have to suffice: A cleavage is a persistent division of groups with systematically divergent patterns in political behavior, including patterns of voting (this minimal definition is inspired by Rae and Taylor 1970). A “social cleavage” is a group division where members of different groups differ in their position in the system of social stratification or other major aspects of their social life – such as religiosity or church attendance or religious non-affiliation. An “ideological cleavage” then would be a division between groups that differ in their ideologies, irrespective of their social position. Since one can distinguish between symbolic ideology and operational ideology, one can analogously distinguish between symbolic ideological cleavages and operational ideological cleavages.

While stable programmatic divisions between parties may be important for the existence of social cleavages, they are obviously even more important for cleavages based on operational ideology. It will be hard to understand why voters with different operational ideologies vote for different parties, if not for differences between parties in terms of ideological messages or ideologically relevant policy proposals. But the existence of such cleavages also requires a sufficient amount of constraint among voters’ political opinions and attitudes to produce such coherent responses to parties’ messages. Clearly, the dependence of such cleavages on party agency makes it unlikely that such cleavages will be uniform across countries and over time. If most citizens lack the necessary structuration of opinions and attitudes, it will be unlikely that such cleavages will range deep into society, if they occur at all. The matter may be different in the case of cleavages based on symbolic ideology, at least at first glance. If there are groups defined by the identification with ideological symbols and if certain parties can be associated with these symbols, then a more or less stable pattern of voter alignments may result. But it is possible that symbolic and operational ideology do not match, and it is plausible that such a mismatch undermines such ideological cleavages. Whether this is empirically the case is still an open question.

Another question is whether the psychological dispositions that are at the center of the “bottom-up” perspective could form the basis of new political cleavages. On the one hand, being affective dispositions, they are less cognitively demanding than operational ideologies and thus potentially more widespread. On the other hand, the psychological dispositions on which the “bottom-up” approach focuses are not really ideologies in any of the senses discussed at the beginning of this chapter. It is however plausible that these dispositions contribute to the psychological underpinnings of the value conflicts that lead to the tension between liberal and conservative ideologies.

The discussion so far focused on the possibility that ideological cleavages emerge at all. This leaves open the question about the content of potential ideological cleavages, provided that the conditions are met for their manifestation: If ideological cleavages substituted social cleavages, what would they be about? While inventive scholars will always try to demonstrate otherwise, it does not seem plausible that ideological cleavages will result in the emergence of something radically new. First, ideologies with mass appeal cannot be invented out of the blue, but take time to evolve, so that socialism, liberalism, conservatism, Christian democracy, etc. are likely here to stay. Second, the major issues of the day, inequality and immigration, are far from alien to the historical currents of ideology. Inequality has always been a foundational concern of socialism, and ethnic divisions, which are created or highlighted by immigration, have been exploited by right-wing nationalism since the beginning on the twentieth century. The new populist movements virulent in certain European countries and elsewhere are thus not much more than old wine in new bottles, while socialism may again become resurgent should the parties of the moderate left cease or undo their attempt to re-define themselves in terms of a “New Center” or a “Third Way.” The only genuinely new ideological current that has emerged in the last few decades is ecologism. Yet its electoral impact has remained limited, even though demands for fighting pollution and wildlife protection have often been picked up by the “traditional” political left or even addressed by supranational regulations. Whether further climate change lends new fervor to this movement remains a matter of future research.

References

Achen, C. H. (1975) “Mass Political Attitudes and the Survey Response,” American Political Science Review, vol. 69, no. 4, December, 1218–1231.

Adorno, T. W., Frenkel-Brunswik, E., Levinson, D. J. and Sanford, R. N. (1950) The Authoritarian Personality, New York: Harper and Row.

Altemeyer, B. (1981) Right-Wing Authoritarianism, Winnipeg: University of Manitoba Press.

Block, J. and Block, J. H. (2006) “Nursery School Personality and Political Orientation Two Decades Later,” Journal of Research in Personality, vol. 40, no. 5, October, 734–749.

Conover, P. J. and Feldman, S. (1984) “How People Organize the Political World: A Schematic Model,” American Journal of Political Science, vol. 28, no. 1, February, 95–126.

Converse, P. E. (1964) “The Nature of Belief Systems in Mass Publics,” in Apter, D. E. (ed.) Ideology and Discontent, New York: The Free Press: 206–261.

Downs, A. (1957) An Economic Theory of Democracy, New York: Harper and Row.

Elff, M. (2009) “Social Divisions, Party Positions, and Electoral Behaviour,” Electoral Studies, vol. 28, no. 2, June, 297–308.

Elff, M. and Rossteutscher, S. (2017) “Religion,” in Arzheimer, K., Evans, J. and Lewis-Beck, M. S. (eds.) The SAGE Handbook of Electoral Behaviour, London: Sage: 199–219.

Ellis, C. and Stimson, J. A. (2009) “Symbolic Ideology in the American Electorate,” Electoral Studies, vol. 28, no. 3, September, 388–402.

Ellis, C. and Stimson, J. A. (2012) Ideology in America, New York: Cambridge University Press.

Elsas, E. van and Brug, W. van der (2015) “The Changing Relationship Between Left–Right Ideology and Euroscepticism, 1973–2010,” European Union Politics, vol. 16, no. 2, June, 194–215.

Enelow, J. M. and Hinich, M. J. (1982) “Ideology, Issues, and the Spatial Theory of Elections,” The American Political Science Review, vol. 76, no. 3, September, 493–501.

Feldman, S. and Johnston, C. (2014) “Understanding the Determinants of Political Ideology: Implications of Structural Complexity,” Political Psychology, vol. 35, no. 3, June, 337–358.

Fleury, C. J. and Lewis-Beck, M. S. (1993) “Anchoring the French Voter: Ideology Versus Party,” Journal of Politics, vol. 55, no. 4, November, 1100–1109.

Fuchs, D. and Klingemann, H. D. (1990) “The Left-Right Schema,” in Jennings, M. K. and Van Deth, J. (eds.) Continuities in Political Action: A Longitudinal Study of Political Orientations in Three Western Democracies, Berlin: Walter de Gruyter: 203–234.

Funk, C. L., Smith, K. B., Alford, J. R., Hibbing, M. V., Eaton, N. R., Krueger, R. F., Eaves, L. J. and Hibbing, J. R. (2013) “Genetic and Environmental Transmission of Political Orientations,” Political Psychology, vol. 34, no. 6, October, 805–819.

Gingrich, J. (2014) “Visibility, Values, and Voters: The Informational Role of the Welfare State,” Journal of Politics, vol. 76, no. 2, February, 565–580.

Heywood, A. (2003) Political Ideologies: An Introduction, 3rd Edition, Basingstoke: Palgrave Macmillan.

Hinich, M. J. and Pollard, W. (1981) “A New Approach to the Spatial Theory of Electoral Competition,” American Journal of Political Science, vol. 25, no. 2, May, 323–341.

Hooghe, L., Marks, G. and Wilson, C. J. (2002) “Does Left/Right Structure Party Positions on European Integration?,” Comparative Political Studies, vol. 35, no. 8, October, 965–989.

Inglehart, R. (1984) “The Changing Structure of Political Cleavages in Western Society,” in Dalton, R. J., Flanagan, S. C. and Beck, P. A. (eds.) Electoral Change: Realignment and Dealignment in Advanced Industrial Democracies, Princeton, NJ: Princeton University Press: 25–69.

Inglehart, R. and Klingemann, H. D. (1976) “Party Identification, Ideological Preference and the Left-Right Dimension Among Western Mass Publics,” in Castiglione, D. and Hoffmann-Martinot, V. (eds.) Party Identification and Beyond: Representations of Voting and Party Competition, London: John Wiley & Sons: 243–273.

Jewitt, C. E. and Goren, P. (2016) “Ideological Structure and Consistency in the Age of Polarization,” American Politics Research, vol. 44, no. 1, January, 81–105.

Jost, J. T. (2006) “The End of the End of Ideology,” American Psychologist, vol. 61, no. 7, October, 651–670.

Jost, J. T. (2009) “ ‘Elective Affinities’: On the Psychological Bases of Left–Right Differences,” Psychological Inquiry, vol. 20, no. 2–3, April, 129–141.

Jost, J. T., Federico, C. M. and Napier, J. L. (2009) “Political Ideology: Its Structure, Functions, and Elective Affinities,” Annual Review of Psychology, vol. 60, no. 1, January, 307–337.

Kahan, D. M. (2012) “Ideology, Motivated Reasoning, and Cognitive Reflection: An Experimental Study,” Judgment and Decision Making, vol. 8, no. 4, July, 407–424.

Klingemann, H. D. (1972) “Testing the Left-Right Continuum on a Sample of German Voters,” Comparative Political Studies, vol. 5, no. 1, April, 93–106.

Knight, K. (2006) “Transformations of the Concept of Ideology in the Twentieth Century,” American Political Science Review, vol. 100, no. 4, November, 619–626.

Lipset, S. M. and Rokkan, S. (1967) “Cleavage Structures, Party Systems, and Voter Alignments: An Introduction,” in Lipset, S. M. and Rokkan, S. (eds.) Party Systems and Voter Alignments: Cross-National Perspectives, New York: Free Press: 1–64.

Mason, L. (2015) “ ‘I Disrespectfully Agree’: The Differential Effects of Partisan Sorting on Social and Issue Polarization,” American Journal of Political Science, vol. 59, no. 1, January, 128–145.

Medina, L. (2015) “Partisan Supply and Voters’ Positioning on the Left–Right Scale in Europe,” Party Politics, vol. 21, no. 5, September, 775–790.

Nie, N. H. and Andersen, K. (1974) “Mass Belief Systems Revisited: Political Change and Attitude Structure,” Journal of Politics, vol. 36, no. 3, August, 540–591.

Peffley, M. A. and Hurwitz, J. (1985) “A Hierarchical Model of Attitude Constraint,” American Journal of Political Science, vol. 29, no. 4, November, 871–890.

Pratto, F., Sidanius, J., Stallworth, L. M. and Malle, B. F. (1994) “Social Dominance Orientation: A Personality Variable Predicting Social and Political Attitudes,” Journal of Personality and Social Psychology, vol. 67, no. 4, October, 741–763.

Rae, D. W., and Taylor, M. (1970) The Analysis of Political Cleavages, New Haven: Yale University Press.

Redlawsk, D. P. (2002) “Hot Cognition or Cool Consideration? Testing the Effects of Motivated Reasoning on Political Decision Making,” Journal of Politics, vol. 64, no. 4, November, 1021–1044.

Sartori, G. (1969a) “From the Sociology of Politics to Political Sociology,” in Lipset, S. M. (ed.) Politics and the Social Sciences, Oxford: Oxford University Press: 195–214.

Sartori, G. (1969b) “Politics, Ideology, and Belief Systems,” American Political Science Review, vol. 63, no. 2, June, 398–411.

Smith, K. B., Oxley, D., Hibbing, M. V., Alford, J. R. and Hibbing, J. R. (2011) “Disgust Sensitivity and the Neurophysiology of Left-Right Political Orientations,” PLoS ONE, vol. 6, no. 10, October, 1–9.

Thorisdottir, H., Jost, J. T., Liviatan, I. and Shrout, P. E. (2007) “Psychological Needs and Values Underlying Left-Right Political Orientation: Cross-National Evidence from Eastern and Western Europe,” Public Opinion Quarterly, vol. 71, no. 2, Spring, 175–203.