Abstract: Variations in voter-party congruence have been studied extensively. Citizens and their representatives are matched in broad ideological terms by conceptualizing congruence using their spatial positions on the left-right dimension. The literature has thus neglected measuring congruence on specific issues. This has an unfortunate ramification: The salience voters attach to those issues, a central theme in issue voting, is not being taking into account as an explanatory factor for differing congruence levels. The present study aims at closing this gap. Borrowing from the voting behavior and social psychology realm, it suggests that issue salience plays a vital role in voter-party congruence. Namely, higher levels of issue salience correspond to higher levels of voter-party congruence on that issue. The study draws on the new EUANDI dataset that, for the first time on an EU-wide cross-sectional basis, includes data not only on identically worded voter and party preferences for 28 issues, but also on their respective salience scores in the context of the 2014 EP-elections. Albeit a small effect magnitude, the analysis of three distinct topics, unemployment, environment and EU-austerity, reveals that indeed citizens with a high salience on these issues do have a higher congruence with their preferred party.
Keywords: Issue Salience, Voter-Party Congruence, European Union, European Elections, Voting Advice Application
What role does issue salience play in determining the level of congruence between voters and parties? The literature on representation has studied variations in congruence extensively by matching the preferences of voters and parties in terms of broad ideological similarity on the left-right dimension (Hooghe et al. 2004, McDonald et al. 2004, Mattila/Raunio 2006, McEvoy 2012, Önnudóttir 2014). Much less is known about congruence on specific issues even though research shows that political conflict is not solely constrained by the traditional left-right dimension, especially on the European level (Hooghe et al. 2004, Kriesi et al. 2006, Costello et al. 2012). The dimensional view does not allow the investigation of explanatory factors for congruence on the issue-level which has ramifications for the literature on representation: it has neglected issue salience as a potential factor explaining why some voters achieve high congruence with their favored parties while others do not. This is troubling because issue voting theory, explaining voters’ motivations and behaviors when choosing their preferred party, contends that the importance citizens attach to specific issues in fact is a strong factor for vote choice (Bélanger/Meguid 2008, DeVries 2010, Giger/Lefkofridi 2014). Within the literature on voting behavior, issue salience has become a major concept to uncover the mechanisms underlying citizens’ search for their preferred party in an election: People engage with issues important to them more frequently and hence gather more knowledge about them; this in turn helps identifying the party that matches their preferences. Thus, this paper aims at improving the existing literature by integrating the concept of salience into the analysis of congruence. Drawing on issue voting theory, it suggests that the variations in congruence levels are influenced by the importance citizens attach to specific issues. More precisely, it hypothesizes that a higher issue salience corresponds to a higher congruence between the preferences of voters and their favored party.
The analysis is conducted on the basis of three separate issues. They are consciously scrutinized to encompass one national issue (unemployment), one EU-issue (austerity) and one that can be located in between (environment). The distinction refers to the level at which the decision-making of the respective policy field is taking place. While the legislation of employment policies lies within the jurisdiction of the individual states, environmental law-making falls under the shared competences of the EU and its members. The EU’s austerity measures represent a clear example of an issue deliberated and concluded on the EU-level. The classification is chosen since this study also expects that the effect salience has on voter-party linkages should be higher for national than for EU-issues. Citizens are less engaged and know less about policies discussed and decided on the EU-level and should hence have more difficulties finding the party with matching preferences in those policy fields.
In this paper salience based-voter party congruence is examined on a cross-national basis, which was not possible in previous studies due to data availability (Giger/Lefkofridi 2014: 293/294). The use of Voting Advice Applications (VAA) as a data collection method resolves this gap. Such online applications are designed to help citizens finding their party match by comparing their stances to those of all potential parties involved in the election (Garzia/Marschall 2014). The new extensive EUANDI survey dataset (Garzia et al. forthcoming) that this study draws on makes use of VAA data collection including respondents from all EU-member states. Voters as well as the vast majority of parties answered questions regarding 28 specific issues of the 2014 EP-elections in order for citizens to find their party match.
The dataset allows for three enhancements in studying congruence: Firstly, regarding the dependent variable voter-party congruence, the everlasting problem of appropriate voter-party preference matching is solved since all questions in both the voter and the party surveys have the exact same wording. This is a major advantage to the oftentimes unreliable matching procedures of differently worded or scaled questions that previous studies employ. Secondly, the respective salience scores voters attach to each of the issues are for the first time collected on an EU-wide cross-sectional basis. They can thus be easily exploited as an explanatory factor for congruence variations. Thirdly, measuring the “Individual Citizen Congruence (ICC)” on a one-to-one basis between a citizen and her preferred party (Golder/Stramski 2010: 92f.) refines the traditional measurement of salience as a macro-level variable by directly investigating the personal importance citizens attach to issues on the micro-level. On the basis of these three points, this study enriches the representation literature by providing it with a sound theoretical and empirical basis for connecting issue-specific preferences, issue salience and the level of congruence between individual voters and their favored parties.
The investigation accounts for incorporating the concept of issue salience in the first tier of the representational chain, namely an individual’s linkage to her preferred party. Subsequent tiers that have been studied by the representation literature are for example the responsiveness of parties or party members to changes in public opinion over time or the relationship between electoral outcomes and the translation into parliamentary seats (Wlezien/Soroka 2007). Assessing levels of congruence is a previous step or a precondition of representation that tries to connect voters’ views and parties’ platforms (Walgrave/Lefevre 2013). Thus before moving up the representational chain this study aims at improving our understanding of the mechanisms underlying voter-party congruence.
To test the hypotheses, this study uses one separate OLS-regression for each of the three issues under consideration. The findings of the empirical analysis do suggest that the inclusion of salience into research on congruence is relevant, even though the effect is not large. They reveal that individuals with higher salience scores are more likely to prefer a party that shares their preferences. This positive effect holds true for all three issues unemployment, environment and austerity. Albeit its high significance, the low magnitude salience has on congruence indicates that it is only a small part of understanding issue- and individual-level factors that determine voter-party linkages. With regard to effect strengths, the results contest the expected outcome: the national issue unemployment in fact yields the lowest effect of issue salience on congruence levels.
Following, a thorough review will present existing research avenues in the literature on congruence. The theoretical section then sets the stage for the hypotheses that are underlying the analysis. The method part will introduce the research design including the data and method used in this study as well as the operationalization of voter-party congruence and issue salience. Next, a descriptive and empirical analysis will assess the validity of the hypotheses. The findings of the analysis are in the end discussed; its limitations open up promising pathways for further research.
2. Literature Review: From Dimensionality to Specific Issues
Studying congruence is part of the literature on representation. This strand of political science explores different steps of the link between public opinion and policy outcome (Wlezien/Soroka 2007). Congruence between the interests of voters and parties is pivotal for a functioning democratic representation. This is because parties competing in elections are the crucial mediators in the representational chain leading from a citizen’s preference to final policy output (Dahl 1989, Powell 2000). Voter-party congruence is thus said to be a previous step or a precondition of representation that connects citizens’ views and parties’ platforms (Walgrave/Lefevre 2013: 456). Subsequent tiers of the representational chain are focused on the outcome of elections such as the translation of ballots into parliamentary seat shares or the functioning of policy responsiveness by governments, parties and party members to changes in public opinion. Assessing congruence however means finding the sources of and measuring variations in preference linkages between voters and their favored parties in a multi-partisan political space.
Much of the research on voter-party congruence, but also more broadly on democratic representation, has been based on spatial voting (Downs 1957) and responsible party models (e.g. Ranney 1954). The former posits that preferences of voters and parties are embedded in a policy space; citizens vote for the party they deem to be closest to their own position. The latter suggests that in order for citizens to be able to make a choice, parties present distinct policy platforms which they advertize during campaigns and carry out once they hold office. In most studies on congruence, the policy space voter and party preferences are embedded in, is represented by the left-right dimension (Hooghe et al. 2004, McDonald et al. 2004, Mattila/Raunio 2006, McEvoy 2012, Önnudóttir 2014). The left-right concept summarizes all variations of political positions in a single spectrum and has been the standard to structure party competition and party systems. Congruence is thus conceptualized as the general ideological correspondence of preferences between voters and their favored party or party member, a central theme in Downs’s spatial voting theory. This approach has been useful to uncover important factors determining ideological congruence between voters and parties such as the individual’s level of education (Freire/Belchior 2011), the number of parties competing (Mattila/Raunio 2006) or the electoral system the election under consideration is taking place (Blais/Bodet 2006). The problem with the approach is that it assumes the left-right scale has the power to incorporate the great variety of issues voters are potentially concerned with. In other words, most research on congruence is “stuck” in the spatial theory of voting that solely looks at the broad and abstract ideological congruence that the left-right dimension represents. This uni-dimensional approach has been criticized arguing that it cannot account for all facets of political conflict (Hooghe et al. 2004, Kriesi et al. 2006, Costello et al. 2012).
Especially when it comes to a European-wide perspective, voters’ preferences on a variety of issues and their congruence with party platforms are not accounted for by a mere left-right conceptualization. Recent comparative as well as single-country studies consequently suggest that the move away from a dimensional to an issue-specific approach to assessing voter-party congruence is highly necessary: Thomassen (2012) explores the tenability of using the general left-right dimension for assessing voter-party linkages. Even though the ideological congruence between voters and parties in the Netherlands on the left-right scale is high, their congruence on specific issues is much lower. This is because representativeness of voters’ preferences on the left-right scale does not automatically imply representativeness on all issues; Thomassen thus argues that some policy areas cannot be subsumed under a single ideological dimension. Costello and his colleagues (2012) test the preference congruence of citizens and their preferred party for the 2009 EP-elections employing 11 attitudinal issues. Besides uncovering that party characteristics such as their size and ideological orientation affect congruence levels, they find high variation in voter-party linkages for the different issues. They conclude that a mere dimensional perspective of the policy space is not sufficient since it would miss the variation they have uncovered. The same applies to the analysis conducted by Walczak and van der Brug (2012) who test the congruence of voters and candidates in EP-elections on specific issue domains such as redistribution and immigration. They for example uncover that highly educated voters are much better represented on socioeconomic issues such as redistribution than workers. Amongst other factors like the individual’s level of education and political knowledge, the issue under consideration greatly affects the level of congruence between citizens and their preferred party. Karyotis, Rüdig and Judge (2014) use six different issues specifically related to austerity politics in Greece to assess voter-party congruence levels in that country. They find that depending on the issue at stake, Greek voter-party linkages differ greatly. The authors thus underscore the importance of distinguishing between issues when examining congruence variations. These examples show that general ideological orientations as portrayed by the left-right dimension should not be equated with substantive policy preferences (Belchior 2010).
3. Theoretical Framework: Salience-Based Congruence and Issue Voting
Issue Salience and Congruence
The move towards an issue-specific perspective of congruence is a necessary step to better understand variations in voter-party linkages. As the aforementioned works show, the preference match between citizens and their potential representatives differs across policy areas. An issue-specific perspective furthermore allows for the investigation of potential explanatory factors on the issue-level which have been largely neglected. Even those few papers that consider particular issues do not control for an issue-level explanatory variable that is widely used in other fields of political science: Issue Salience.
Issue salience has been an important factor in analyzing many aspects of politics. It is generally defined as the relative relevance or importance actors attribute to a specific issue (Leuffen 2014: 617). Depending on the context, it relates to various research objects (policies, proposals, electoral behaviors), levels of analysis (an individual’s, institution’s or country’s salience) and actors (voters, decision-makers) (Ibid). Issue salience is featuring prominently in the voting behavior realm: After the seminal article by Rabinowitz et al. (1982: 57), in which they show for the first time that “any issue singled out as personally most important plays a substantially greater role for those who so view it than it does for others”, the importance citizens attach to specific issues was given greater attention. As multiple studies since then demonstrate, public opinion and citizens’ search for their favored party is structured by levels of salience (Fournier et al. 2003, Guisinger 2009, Bromley-Trujillo et al. 2014).
Issue salience is thus on the one hand used to assess citizens’ voting behavior (demand side of voting), as well as to on the other hand explore the way parties compete in elections and how they respond to changes in public opinion (supply side of voting). In studies working on congruence, which represents the middle position between the demand and supply side, the importance citizens attach to specific issues “has thus far had a shadowy existence” (Giger/Lefkofridi 2014: 289). In fact, the authors of this quote claim to be the first to introduce issue salience into the contemporary work on congruence, though only for the single case of Switzerland. This is particularly troubling since RePass (1971), criticizing the lack of engagement with salience in the literature of representation, pointed out over four decades ago that citizens do take into account the relative importance of issues when making party choices. Another past call for introducing issue salience into the representation realm can be found in Kuklinski and McCrone (1980: 159), who state that “as efforts to develop an empirical theory of representation continue, we recommend that issue salience be given its proper place in that theory”. This paper aims at answering the call; since indeed salience has not been given its proper place in congruence research, it borrows from the voting behavior and social psychology literature in which salience is a prominent factor. In what follows, the theoretical foundations of this paper, derived from issue voting, are presented.
Theory and Hypotheses
Issue voting is based on the accessibility theory of social psychology. Here, salient stimuli are those that, through prior experiences of an individual, stand out from the rest (Vaughan/Hogg 2014: 56). People attach specific attitudinal factors such as emotions or motivations to salient stimuli, which make them more influential and coherently useable for their actions. The frequent engagement with issues that are important to us increases our disposition to recall and reflect on attitudes connected with them; they thus become more accessible from our long-term memory over time. The more important and thus the more ingrained a stimulus is in our memory, the more it affects our attitudes towards it and consequently, our behavioral patterns when reacting to it (Hogg 2006: 119). Easily accessible attitudes guide behavior in every social situation; we choose our actions according to those attitudes that are most accessible to us (Fazio 2014).
The high level of accessibility for issues that are important to us allows us to engage with topics related to the issue more often which in turn increases our knowledge about them. The frequent engagement with particular issues stabilizes our attitudes towards the topic leaving them resistant to sudden changes. The large amount of knowledge about salient issues and the attitude stability with which we encounter them affects voting behavior: On the one hand, attitudes towards a salient topic shape political preferences. Voters have a clear opinion on the way the topic should be handled politically. On the other hand citizens understand parties’ stances on issues salient to them better and are thus capable of making an informed ballot choice (Giger/Lefkofridi 2014: 292). In fact, research has repeatedly shown that while highly salient issues do shape political behavior and actions, non-salient ones have no particular effect (Krosnick 1990, Lavine et al. 1996, Fazio 2014: 168). Higher accessibility of issues, that is higher salience, enables citizens to identify the party they deem to be closest to their own preferences.
- Quote paper
- Jan Hölting (Author), 2016, Salience-Based Voter-Party Congruence in the EU, Munich, GRIN Verlag, https://www.grin.com/document/355602