Sara Binzer Hobolt, Department of Politics and International Relations, University of Oxford, Manor Road, Oxford OX1 3UQ, UK. Tel.: +44 (0)1865 278830; E-mail: email@example.com
Abstract. One of the criticisms often levelled against direct democracy is that citizens lack sufficient knowledge to vote directly on policy issues. The ‘No’ votes in the French and Dutch referendums on the Constitutional Treaty have highlighted the importance of examining voter competence in referendums. This article proposes a theoretical framework for evaluating competence in EU referendums. It suggests that competent voting in EU referendums is based on issue-specific preferences and requires political information. Since most voters have little detailed knowledge of European integration, they rely on heuristics and cues when deciding how to vote. The important question is how much and which type of information voters require to make competent choices. This article examines whether and under what conditions the use of party endorsements as information cues can enhance competent voting in EU referendums. These theoretical questions are examined in an analysis of the 1994 Norwegian referendum on EU membership.
The question of whether these [competency] conditions are fulfilled to the extent required in order to make democracy work should not be answered by reckless assertion or equally reckless denial. It can be answered only by laborious appraisal of a maze of conflicting evidence. (Schumpeter 1942: 254)
The rejection of the Constitutional Treaty in referendums in France and the Netherlands in 2005 brought about a political crisis in the European Union (EU). Paradoxically, the constitutional process that was initiated to ‘bring the Union closer to its citizens’ may thus ultimately have fallen at the hands of the people. These referendum outcomes raise the question: Are voters sufficiently competent to vote on complex matters, such as the future of European integration? Decisions taken in referendums are by definition beyond the scope of government, and it has often been questioned whether we ‘should endorse the use of an institutional device, which allows the citizens to undercut their elected representatives’ (Qvortrup 2002: 2). Rather than assessing the fundamental desirability of direct democracy vis-à-vis representative democracy, this article examines a crucial aspect of this debate – namely whether and under what conditions voters act competently when deciding in EU referendums.
One of the most common criticisms of direct democracy concerns the lack of voter competence. Since voters are struggling to make informed choices in normal elections, scholars have questioned whether they will be able to make competent decisions when asked to vote directly on unfamiliar and complex issues (Bowler & Donovan 1998; Christin et al. 2002). What is missing from this debate, however, is a systematic exposition of what exactly we mean by ‘competent voting’ in these referendums and how we can measure this. This article proposes a framework for evaluating voter competence in EU referendums. First, it expounds on the normative expectation of issue proximity voting, proposing that voters are expected to make decisions on the basis of preferences relevant to the task at hand (e.g., the future of European integration). Second, the article considers the type of information required for voters to accomplish this task. Recent studies on information and voting have shown that voters can make competent choices even with limited information by relying on informational shortcuts such as elite cues (see Lupia 1994; Bowler & Donovan 1998; Lau & Redlawsk 2001). Yet, very few studies have examined the use of heuristics in EU referendums (see Hobolt 2006 for an overview). Following the debate in the voter information literature, this article examines how different types and degrees of information impact behaviour and, in particular, how the use of heuristics may enhance voter competence. This analysis focuses on the highly visible ‘endorsement heuristics’ provided by political parties in referendum campaigns and examines whether these cues can function as substitutes for detailed knowledge of the ballot proposal. The main findings suggest that partisan endorsements can aid competent behaviour, but only to the extent that voters are sufficiently knowledgeable about party positions on the European integration dimension.
The article proceeds as follows. First, it considers the information requirements in referendums and outlines a conceptual framework for understanding and measuring voter competence in EU referendums. Second, it discusses the extent to which partisan endorsements can be expected to provide reliable substitutes for detailed information. Finally, it evaluates the theoretical propositions empirically in a detailed analysis of the 1994 Norwegian referendum on membership.
Information, knowledge and voter competence
‘If most American voters lack factual information, a constraining ideology and an ability to deal with issues in highly publicised candidate races, how can they be expected to sort through the complex policy choices they face in the low information setting of direct democracy?’ As Bowler and Donovan (1998: 24) indicate in this quote, it seems reasonable to question the ability of voters to make reasoned decision under direct democracy, given the findings on voter ignorance in representative processes of democracy. Despite the democratic ideal of an electorate that chooses among candidates and parties on the basis of an informed understanding of policy issues and party platforms, numerous studies of American voters have shown that most people have limited ability to think in the abstract about candidates and issues and they lack factual information about politics (see, e.g., Campbell et al. 1960; Converse 1964; Delli Carpini & Keeter 1991; Alvarez & Brehm 2002). Voter ignorance is not a uniquely American phenomenon. Only shortly after the new European Constitution was signed by European leaders in a high-profile ceremony in Rome in 2004, a third of the European public according to Eurobarometer 214 (see Appendix) declared that they had never heard of the document. Respondents were also asked a number of factual questions about the new Constitution and, as the results in Table 1 indicate, less than half of respondents were able to answer correctly on most questions. Furthermore in post-referendum surveys, a third of Dutch voters and a quarter of Spanish voters mentioned ‘lack of information’ as the reason for their negative vote in the referendum on the Constitutional Treaty (European Commission 2005a, 2005b). A set of questions in a 1993 Eurobarometer survey allows us to compare citizens' knowledge of European and national political affairs.
Table 1. Public knowledge of national and European politics
Correct answer (%)
Notes: * Eurobarometer 39.0 (1993); N: 14,039 (respondents from 12 Member States and Norway). ** Special Eurobarometer 214 (2004); N: 24,786 (respondents from 25 Member States).
The figures in Table 1 suggest that while voters generally have little knowledge of political affairs, they are more ignorant about political affairs at the European level than at the national level. Europeans are unable to answer basic questions about the Union's political system, such as the final legislative authority (only 19 per cent answered correctly) or identify members of the Union's executive (only 7 per cent could identify a member, while 42 per cent could name the president). These findings of public ignorance have been used as a justification for avoiding plebiscites altogether. In the words of Belgium's former foreign minister, Erik Derycke (quoted in Humo, 2 September 1997): ‘I'm glad that we have no referendums. How for God's sake are you going to explain a complicated thing like the Euro in a yes-and-no question to voters?’ Just how much detailed information do voters need to make competent choices in referendums? Are voters required to have a detailed understanding of the intricacies of the treaties or to know the names of all of the members of the European Commission in order to make a reasoned decision in an EU referendum? In recent years, a number of studies of information and heuristics have moved away from the very pessimistic view of voter competence and rallied behind Key's (1966) dictum that ‘voters are not fools’. In essence, these studies have changed the focus from information levels to information processing.
Heuristics may also help voters in EU referendums to arrive at competent decisions despite their lack of factual knowledge. In particular, party endorsements may offer a type of judgement shortcut that allow citizens to infer their own position on a ballot issue without detailed information about the proposal itself (Sniderman 2000; Lau & Redlawsk 2001). However, as the critics of the heuristics literature have pointed out, it is not reasonable to simply assume such heuristics can work as substitutes for domain-specific knowledge (Bartels 1996; Kuklinksi & Quirk 2000). In his empirical analysis of information effects in American presidential elections, Bartels (1996) finds that fully informed voters act very differently from less informed voters. This suggests that information differentials are not entirely overcome by the use of heuristics. However, these findings are not surprising given that some contextual knowledge is needed to use heuristics effectively. Bartels examines differences across levels of political sophistication, but he does not examine whether different types of information may have similar effect on patterns of behaviour. Yet, if we want to examine the effectiveness of cues in aiding competent decision making, the latter is as important as the former. Hence this article examines whether certain types of information – in particular party endorsements – can work as efficiently as detailed information about the question at hand. As Lupia and Johnston (2001:196) have pointed out, ‘short cuts are no panacea. If used incorrectly, reliance on short cuts can lead to grave errors.’ Thus, while the heuristic school paints a more optimistic picture of the possibilities of voter competence, it cannot be assumed that the mere presence of cues will lead voters to act competently. Assessing the impact of information on behaviour is thus an essential component of examining voter competence in referendums. Yet first we need to consider what we mean by voter competence in the specific context of EU referendums.
Conceptualising and measuring voter competence in EU referendums
When we study voting behaviour, we make inferences about what causes people to act in a certain way: how do voters decide between candidates, parties or issues? Examining voter competence is somewhat different because we are not only interested in understanding how people decide how to vote, but we also have certain normative expectations about how they ought to behave. Yet how do we decide which normative criteria to apply? This question is seldom answered explicitly in empirical studies of voter competence. Part of the reason lays in the inherent difficulty in establishing competence measures in politics where there are no objective standards for better and worse, right and wrong. Yet, while we cannot establish a universally agreed upon conceptualisation and measurement of voter competence, we can make an attempt to engage in a systematic debate about these issues by setting out explicit theoretical propositions and empirical measures that can be evaluated.
Kuklinski and Quirk (2001: 287) suggest that the first step in evaluating civic competence should be to ‘identify the task that the actor is asked to undertake’. Competence, then, concerns the ability of individuals to accomplish this specific task. The task of a voter in a referendum is to choose between two policy alternatives, the ballot proposition and the status quo. In referendums on European integration, the question on the ballot falls into one of three categories: either accession to the EU, the ratification of a new EU Treaty or a decision on a specific policy issue related to the integration process. Yet would it be regarded as normatively desirable for voters to answer the question in any way they see fit? Would we consider it competent behaviour if voters simply pick the first option on the ballot, or if they, for example, answer on the basis of the number of red lights that they crossed on the way to the polling station, their feelings about the state of the national economy or their evaluation the performance of the government of the day? While the latter two reasons, at least, cannot be dismissed as entirely irrational behaviour, they may be seen as normatively undesirable. Our normative expectation is that there is a clear link between the question asked (the ballot proposal) and the answer provided by the individual voter (the vote choice). Hence, while a rational choice may simply imply ‘one that is based on reasons, irrespective of what these reasons may be’ (Lupia et al. 2000: 7), a competent choice is related to the accomplishment of a specific task and it should thus be based on preferences pertaining to that task.
As Druckman (2001a: 232) notes, democratic competence ‘concerns the expression of [citizen] preferences to which governors can and should respond’, but governors can only respond appropriately to the expression of voter preferences if these expressions relate to the question posed. This article thus proposes a narrow conceptualisation of competence that, unlike the broader notions of reasoning or rationality, refers to the accomplishment of a specific task. A competent vote in referendums can thus be defined as one that is based on preferences specific to the issue on the ballot and that would be the same if full information were available. Note that this definition differs from the more generic version presented in the heuristics literature since voters are not merely expected to vote ‘as if’ they were fully informed, they are also expected to rely on ‘proximity considerations’ related to the ballot proposal. Issue voting is normatively desirable because of the specific objective of the task posed to citizens. This will be discussed in more detail below.
Issue-specific versus second-order voting
To understand how citizens can competently accomplish a political task, we thus need to consider the specific objective of that task. To be precise: what is the purpose of asking people to decide on a proposition concerning European integration? From the perspective of democratic theory, the broad objective may be to ensure ‘government by the people’. In reality, of course, modern democracies are mostly governed by the principle of ‘government by the representatives of the people’. More specifically, the purpose of holding referendums on aspects of European integration, at least from a normative perspective, is the public legitimisation or veto of an EU-related policy proposal (Hug 2002). In the context of European integration, public veto power can be seen as particularly important since these proposals involve the surrender of national sovereignty to a supranational body – that is, the transfer of powers from national representatives to European (elected and non-elected) officials. Yet to achieve the politically desirable outcome of legitimising a transfer of sovereignty, it is vital that voters decide on the issue posed to them. As Van der Eijk and Franklin have noted in their study of elections to the European Parliament:
The logic of democratic elections presupposes that the political verdict of electorates can be construed as emanating from the political preferences of voters, preferences that are relevant to the decision-making arena concerned. (Van der Eijk & Franklin 1996: 6)
The same is true for EU referendums, where we expect voters to base their choice on preferences related to the European question so the outcome of the referendum can be seen as democratic mandate for (or against) a particular policy option. Some may regard it as self-evident that individuals vote on the basis of their preferences towards the issue at stake, but in fact several empirical studies of voting in EU referendums and elections have shown that voters tend not to vote on the basis of European issues and instead vote on the basis of ‘second-order’ factors such as the satisfaction with the national government (see Franklin et al. 1994, 1995; Van der Eijk & Franklin 1996; Marsh 1998). Reif and Schmitt (1980) pioneered the idea of European Parliament elections as ‘second-order national elections’ in which considerations about ‘first-order’ national politics drive political behaviour. In particular, voters are expected to use second-order contests to signal their satisfaction, or dissatisfaction, with their domestic political parties and governments. This framework has also been used to explain behaviour in EU referendums (see Franklin et al. 1994, 1995; Franklin 2002), yet it has been implied that such patterns of voting behaviour are undesirable from a democratic perspective (see, e.g., Van der Eijk & Franklin 1996; Norris 1997). Garry et al. (2005: 205) have reasoned that while issue voting ‘assumes a rational and reasoned calculation by the voter based on his or her views on EU matters . . . (t)he second-order approach sees voting in EU referendums simply as a chance for voters to express their domestic political preferences’. A common interpretation thus suggests that second-order voting is less rational and reasoned than issue-specific voting. Yet, ‘second-order’ voting is may in some cases be rational – that is, based on strategic and reasoned considerations. For example, it may be rational for a voter to vote against the EU Constitution because he or she wants to punish Tony Blair for going to war in Iraq. However, this vote choice is not competent because it does not accomplish the specific task – namely to choose the better of the two alternatives: an EU Constitution or the status quo. Hence, to conceptualise voter competence we must also consider the objective of that task voters are asked perform. This can be summarised graphically as shown in Figure 1.
Figure 1 illustrates that voter competence, as it has been conceptualised here, is related to the specific task facing the voters and the objective of that task. Following this, our normative expectation concerning competent voting in EU referendums involves the ability of voters to decide on a ballot proposition on the basis of their (broadly defined) preferences on the issue of European integration. This can be represented in a simple spatial framework, drawing on the proximity theory of voting. The basic idea of this spatial theory of voting, originally popularised by Downs, is that voters vote for the candidate holding policy positions most similar to their own positions (Downs 1957; Davis et al. 1970; Enelow & Hinich 1981). If competence implies voting on the basis of your own issue preferences, then this could equally be interpreted as voting for the alternative with the greatest proximity to your own ideal point (for a criticism of the ‘proximity model’, see Rabinowitz & Macdonald 1989). In a referendum on European integration voters are asked to choose between two alternatives; the proposal described on the ballot, x, or status quo, q. The utility of voting for the ballot can this be written as:
where each voter's evaluation of the ballot proposition's x utility depends upon the squared distance between voter i's own preferred position Ii and the voter's perception of the position of the ballot proposal Pix, on a single issue dimension. The term cix represents all the considerations involved in the voter's overall evaluation of x other than those pertaining to the specific issue dimension included in the model. Following the logic of the Downsian proximity theory what matters is which of the alternative policy outcomes, x or q, is closer to the voter's ideal point I. Competent voting thus implies deciding on how to vote by comparing the expected utility of voting in favour of the proposal x with the expected utility of status quo, q: E(Uix) ≥ E(Uiq). We can refer to this as the normative expectation of issue proximity voting.
Maximising utility requires voters to have a certain knowledge of how their issue preferences related to the specific ballot proposal and the reversion point – that is to identify (Pix − Ii). To vote on the basis of one's own preferences, voters must figure out whether the policy described is better or worse with respect to his or her interest than the policy entailed by the rejecting the proposal, and this requires information. We may refer to this as the information requirement of voter competence. The important question is how much and which kind of information is required for competent voting, and how to assess this empirically.
Measuring voter competence
According to Kuklinski and Quirk (2001), the conceptualisation of civic competence (i.e., the identification of the particular task and the performance criterion) should be followed by the selection of empirical indicators that allow us to measure competence empirically and standards that enable us to categorise performance. As mentioned above, it has been common practice in studies of civic competence to focus on information levels, and this type of measurement of competence levels tends to involve some analysis of people's factual knowledge of political affairs (Luskin 1987; Delli Carpini & Keeter 1991). As the heuristic school shifted the focus away from information levels to an emphasis on information processing, they also adopted a different approach to the measurement of competence levels. Rather than setting certain standards for factual knowledge levels, many researchers have focused on whether voters can act ‘as if’ they were fully informed (Lupia 1994; Lau & Redlawsk 1997; see also Bartels 1996). Competent behaviour can thus be examined by comparing voters across different levels and types of information; or what Kuklinski and Quirk (2001:296) have referred to as the ‘well-informed-proxy group comparison’. Applying this method involves selecting a group of well-informed citizens and comparing the behaviour of voters that are less informed with the behaviour of this group. If the less-informed approximate the behaviour of the well-informed then their behaviour is seen to be competent, and the presumption is that competency is achieved by relying on informative heuristics (see, e.g., Lupia 1994; Christin et al. 2002). Hence it is assumed in the literature that the more informed voters are the competent voters. Following the conceptualisation of voter competence presented above, we would also expect that more detailed knowledge allows voters to choose the alternative that is most compatible with their own preferences. However, unlike most studies on information effects that only examine variations across general levels of political information,1 we are also interested in analysing whether the use of heuristic information may lead to the same patterns of behaviour as high levels of domain-specific information. Hence, in the empirical examination of voter competence, we compare voter competence across both levels and types of information to establish what type of heuristic information is sufficient for competent performances (Lupia n.d.: 10). Before proceeding to the empirical analysis, the next section considers the use of party endorsements as heuristics in referendums.
Party endorsements as cues in EU referendums
As Lau and Redlawsk have pointed out, elite endorsements have an obvious heuristic value, since ‘all that is necessary is to learn the candidate endorsed by a group and one's own attitude toward the group, and an obvious cognitively-efficient inference can be made’ (Lau & Redlawsk 2001: 953; see also Sniderman et al. 1991; Sniderman 2000). In European referendums, the endorsements of political parties are arguably the most visible elite cues. Kriesi (2005: 139) has referred to the partisan heuristic as ‘the quintessential shortcut in direct democratic votes’. Political contestation in Europe is essentially framed by political parties, and we would thus expect party labels to carry extraordinary weight also in the context of referendums. Moreover, an analysis of news coverage in the Norwegian accession referendum campaign shows that party endorsements received far more attention than the endorsements provided by interest organisations, nongovernmental organisations and public personas.2 These findings illustrate that the referendum was mainly framed as a contest between political parties: almost three-quarters of the news that referred to particular elite actors mentioned political parties. Moreover, as the setter of the referendum, the governing party or parties have considerable signalling power in terms of deciding the specific timing and wording of the referendum proposal. Since political parties are likely to provide the most important – albeit not the only – informational cues to voters about how to vote in referendums, it is important to understand the nature of these cues and how they are processed.
Partisan endorsements can enhance competence by signalling to voters where they are located in relation to the two alternatives on the ballot – that is, by informing the voter which choice is better for them. Relying on the party endorsements of the preferred party seems an obvious choice for most voters since people tend to listen to the elites sharing their ideological predispositions (Zaller 1992). Hence, by structuring the ‘choice set’ for citizens, parties may allow them to make choices approximately coherently, despite their informational shortfalls (Sniderman 2000). Experiments on framing have shown that party cues can help citizens overcome ‘framing effects’ since these cues provide information about the credibility of the information provider (Druckman 2001b, 2001c). Despite these potential advantages of endorsements, we need to consider two important caveats. First, while heuristics can help voters overcome information shortfalls, the effective use of endorsements also requires information (Kuklinski & Quirk 2000). Second, under certain circumstances, party endorsements may in fact mislead citizens. Even advocates of the heuristics approach acknowledge that some information is needed to take advantage of heuristic shortcuts. As Sniderman (2000: 72) has pointed out: ‘[I]t takes smarts to take advantage of smart moves.’ Not only may citizens lack the necessary information about the endorser to use the cues effectively, they may also be misinformed (Kuklinski et al. 2000). It is reasonable to assume that a partisan cue will only work effectively if it gives reliable information about the specific task at hand (i.e., the referendum vote). As Lau and Redlawsk (2001) have observed, endorsements can work against voters when the question at hand is not structured in the typical manner, and this may be the case in most EU referendums.
The main dimension structuring political contestation in Europe is the left-right dimension. In national elections most voters choose a preferred party based on where they are located on the left-right dimension. Voters can therefore rely on partisan cues to give them information on how their preferences relate to a specific policy proposal based on this knowledge. However, in EU referendums voters are asked to vote on issues related to European integration, and the question is thus whether the EU positions of political parties can be read from their position on the left-right dimension. If these two dimensions are orthogonal, then partisan endorsements may be misleading and fail to enhance competence in EU referendums. Several studies have shown that the European integration and left-right dimensions are in fact independent of each other (see, e.g., Hix 1999). Whereas European integration engages national sovereignty and mobilises territorial groups, the left-right contestation involves the allocation of values among functional interests. If that is the case, voters relying on partisan cues due to proximity on the left-right dimension will not acquire accurate information about positions on the dimension of interest – namely European integration.3Van der Eijk and Franklin (1996: 381) have even argued that ‘the democratic deficit in European elections is due to the lack of correspondence between the dimensions of competition among national parties and the dimensions of conflict within the European arena’. A party expert survey carried out by Marks and Steenbergen (1999) shows that there is a non-linear relationship between the two dimensions since centrist parties tend to be more pro-integration and peripheral parties more anti-integration (see Hooghe et al. 2002).
Figure 2 depicts the positions of parties in the EU-15 Member States and illustrates a nonlinear relationship between the left-right position and the European integration position of parties, resembling an inverted U-curve. Parties on the extreme of the political spectrum tend to be more anti-European than parties located around the centre. However, this relationship disappears when we only examine the mainstream parties (i.e., the Socialist/Social Democratic, Liberal, Christian Democratic and Conservative party families) that dominate national governments and represent approximately 80 per cent of electoral votes across the EU. This is illustrated in Figure 3.
As we can see from Figure 3, there is no apparent relationship between the left-right and the European positions of mainstream parties. Apart from the British Conservative Party, which is the outlier in the lower right-hand quadrant, all of the European mainstream parties are broadly pro-European. It is not possible to discern a party's more exact location on this dimension from any knowledge of its left-right position. We can therefore hypothesise that, since partisanship mainly relates to proximity on the left-right dimension, party endorsements can enhance competent voting in EU referendums only when voters have basic knowledge about the positions of parties on the EU dimension. The next section examines these propositions empirically using a post-referendum study from the 1994 Norwegian accession referendum.
The Norwegian accession referendum 1994
The 1994 referendum was the second Norwegian referendum on accession to the EU. In 1972, a small majority of voters (53.5 per cent) rejected membership. In 1994, 89 per cent of Norwegian voters turned out to reject the accession proposal for a second time by an even smaller majority of 52.2 per cent. The Norwegian 1994 referendum is an apposite case for examining voter competence in EU referendums because it provides a critical test of our examination of voter competence and cue-taking. European integration has been more politicised in Norway than in most other countries and all political parties have taken a clear stance on the issue (Aardal & Valen 1997; Midtbø & Hines 1998; Saglie 2000; Hobolt 2005). This high salience of the European issue implies that we would expect voters to be more certain of their own issue preferences and better equipped to judge how the ballot proposal and the reversion point relate to these preferences. Moreover, party competition on the issue ought to make party endorsements more reliable cues in the referendum. In other words, if we find that voter competence is low and that partisan endorsements provide inadequate information in this referendum, we would expect even greater problems in other EU referendums held in low-salience and low-information environments. On the other hand, if we find that voter competence is high among certain voters, this can teach us important lessons about the circumstances under which voters can act competently and the type of information necessary to achieve this. In addition to providing a ‘critical test’ of voter competence, the post-referendum survey from the 1994 referendum contains a number of questions on knowledge and information as well as the respondent's perception of party positions that allow us to examine the processing of cues. This analysis thus contributes to the (mostly American) debate on information and voting behaviour by examining the question in a European setting and by explicitly analysing how different types of information and heuristics affect voter competence.
Party endorsements and voter preferences in the Norwegian referendum
The Norwegian debate on membership had begun already in November 1992 when the Labour government applied to join the EU, and it intensified after membership negotiations began in April 1993. The proposal of membership was initiated by the governing party, Arbeiderpartiet (Labour), which with 41 per cent of seats was by far the largest party in the Norwegian parliament (Stortinget) after the 1993 election. In addition to this, the main opposition party, the conservative Høyre, also supported membership, as did the populist right-wing party, Fremskrittspartiet (Progress Party). Altogether these pro-EU parties made up 65 per cent of seats in Stortinget. Hence on the basis of this information alone, we would predict that the membership proposal would pass, but only 60 per cent of the voters who identified with the pro-EU parties actually followed their recommendation. In comparison, 76 per cent of voters who identify with the anti-EU parties voted ‘No’. EU membership was opposed by Senterpartiet, an agrarian centre party, as well as two small parties on the far left and two small centre-right parties. Hence, while Norwegian party politics is structured along the left-right dimension, the European integration issue has split the traditional blocs and created new political alliances (Aardal 1995; Saglie 2000). Figure 4 shows the location of the parties on the European and the left-right dimension, using data Ray's party expert survey.4 This figure also includes the mean position of the voters on these dimensions.
As Figure 4 illustrates, there is no unambiguous relationship between the left-right dimension and the EU dimension. The extreme left tends to be very anti-EU, but not more so than the Centre Party. Another important finding is that voters tend to be closer to their parties on the left-right dimension than on the European dimension. This is particularly true for voters of the two largest parties, Arbeiderpartiet and Høyre, who are relatively closely aligned to their party on the left-right dimension, but far more eurosceptical than their preferred parties. As argued above, these orthogonal dimensions may potentially reduce the ability of voters to act competently since adhering to the endorsements of preferred parties based on proximities on the left-right dimension provides inaccurate information on where they are located in relation to the proposal on the European dimension. The important question is therefore not only the distance between voter and preferred party positions on the EU dimension, but also whether voters are aware of this distance. Voters are known to project their own views onto parties they support, thereby enhancing the appearance of consistency in their positions. Since the Norwegian post-referendum survey contains questions on both self-placement and the perceived position of parties on different policy dimensions, we can compare the ‘perceived’ distance between voters and parties with the ‘actual’ distance, based on party expert survey data. Table 2 shows the results for both dimensions.
Table 2. Distances between voters and parties
Low political knowledge
Medium political knowledge
High political knowledge
Notes: S.D. = Standard deviation. Absolute values (in parentheses) indicate the numerical value of the distance on the scale without regard to its sign.
Sources: EU-avstemningen 1994 and Ray's party expert survey. See Appendix.
EU distance (ideal – actual party position)
EU distance (ideal – perceived party position)
Projection: difference between actual and perceived proximity
Left-Right distance (ideal – actual party position)
Left-Right distance (ideal – perceived party position)
Projection: difference between actual and perceived proximity
This table illustrates that the distance between parties and voters on the EU dimension is greater than on the left-right dimension. Using party expert survey data, voters are on average 2 points more eurosceptical than their preferred party and the distance is 2.8 points in absolute values. This is a considerable distance on a 10-point scale. In comparison, voters are on average 0.7 point more left-wing than their preferred party, with an absolute (mean) distance of 1.6. There is, however, no difference in the extent to which voters project their own views onto their preferred party. On both dimensions, voters on average perceive themselves to be 0.8 points closer to their party than the expert evaluation. It is interesting to note that if we divide these results by the level of political knowledge, voters with very limited knowledge of politics are far more likely to ‘project’ (report a smaller distance than the experts) on the EU dimension than voters with higher level of political awareness, but these differences are not found on the left-right dimension. This indicates that a higher level of knowledge is required to interpret partisan cues on the European dimension correctly. Following this, we can conjecture that party endorsements may be potentially misleading if people have a very limited knowledge of party politics (see Lau & Redlawsk 2001). On the other hand, the clear positions of parties on this issue (Figure 4 shows that parties are located at each end of the spectrum) may provide useful cues to voters given that they are aware of these positions and how they differ from their own position. These propositions will be tested below.
The impact of information on voter competence
In this article, voter competence in EU referendums is conceptualised as the ability of voters to choose the alternative (the proposal or status quo) that is closest to their preferences on the issue of European integration. It has been argued that information is required for voters to perform this task, but it is inherently difficult to establish absolute standards for competent voting in a specific referendum. In order to classify voters into categories of ‘competent’ and ‘not competent’ voters, we would need to observe not only their true preferences on the EU dimension, but also the location of the ballot proposal and of status quo. While we can reasonably argue that accession to the EU is more ‘pro-integration’ than staying outside the Union, we cannot precisely locate these alternatives on a scale, nor can we be certain that the reported answers on EU attitude questions represent true preferences for all voters. Yet since the value we want to observe is how voters would have voted if they were themselves fully informed about these positions, we can get an approximate idea of competent voting by comparing groups according to their level of information. By using the method of ‘well-informed proxy group comparison’ we can compare voting behaviour across types and levels of information. The assumption is that if you are very knowledgeable about European politics, you will also better equipped to choose the best alternative according to your own preferences, and you will not be mislead by elite cues that are incompatible with your own preferences (Zaller 1992; Kriesi 2005). Hence, we want to examine the extent to which the vote choice is compatible with the voter's EU preferences and how this compares across groups with different levels of information. Moreover, we are also interested in finding out to what extent endorsement heuristics can aid competent voting.
To examine the effects of information on competent voting, we distinguish between three types of information: cue reception, cue knowledge and EU knowledge. The ‘cue reception’ indicator simply measures individual exposure to partisan endorsements. This scale thus measures the extent to which people have been exposed to and can identify the endorsements of parties, but it does not measure whether individuals identify party positions correctly. The ‘cue knowledge’ indicator measures whether individuals can correctly identify the positions of parties. This scale is created by calculating the distance between the voters' perception of party positions and their actual position (according to the expert surveys): the smaller the distance, the greater the knowledge of party positions and the higher the score on the scale. The last scale measures detailed knowledge of the EU. This EU knowledge scale is created as a summated ‘political awareness’ scale by calculating correct answers to factual knowledge questions on the EU, supplemented by subjective knowledge questions (Zaller 1992; Converse 2000).
By examining how these types of information interact with vote considerations, we can arrive at a better understanding of how information and heuristics influence voter competence. However, two potential concerns should be addressed. The first concern is that the three measures may capture the same thing and are so highly correlated that any distinction is meaningless. While this is a valid concern, the conceptual distinction between the three measures is very important when examining competence. Moreover, although the indicators are positively correlated – as we would expect – this correlation is very low (below 0.2) in all cases,5 and the inter-item reliability of each scale is very high.6 A closer look at the three indicators also reveals that the ‘EU knowledge’ group seems to capture what is generally referred to as political ‘sophistication’ or ‘awareness’ (Zaller 1992). A second potential problem is that these measures – particularly the EU knowledge measure – are endogenous to our dependent variable. In other words, one could argue that the more people support the integration project, the more likely they are to acquire information about the issue. While this is a valid argument, the empirical evidence shows no correlation between EU knowledge and support for European integration, and only a very low correlation between cue-taking and support. Hence, we have grounds to believe that these indicators are capturing three distinct types of information processing that could have markedly different effects on competent voting.
To test the impact of different types of information on voter competence, we can thus compare the behaviour of voters who differ only in the amount and types of information they possess. Our expectation – as outlined above – is that both detailed knowledge of the EU politics and knowledge of party positions will make issue preferences more important in determining the vote choice. However, we do not expect that simply receiving party cues will have the same effect since such party endorsements may be potentially misleading without a basic level of information. To test these propositions, I have specified three logit models including interaction terms for each specific information type and the variables that are expected to be most significant in determining the vote: EU preferences, preferred party endorsement and (perceived) distance to the voter's preferred party. The dependent variable is ‘Yes’/‘No’ vote in the referendum.7 The variable on individual voters' EU preferences – or ‘ideal points’ on the European integration dimension – is based on a question where respondents are asked to place themselves on a 10-point EU-attitude scale. They are also asked to place each of the parties on the same scale, and this allows us to calculate the perceived distance between the respondent and his or her preferred party. The preferred party was determined on the basis of the respondent's vote in the last Storting election (in 1993).8 Party endorsements are coded as a dummy variable depending on whether the respondents' preferred party recommended a ‘Yes’ or a ‘No’ vote. In addition, the model includes controls for age, income, region, left-right ideology and government satisfaction.9 To test the hypothesis of how the voters' utility calculus varies depending on their knowledge, the model includes two interaction terms for the effect of knowledge and party endorsement and knowledge and preferences.
Table 3 reports the logit results. These results suggest that the type and the amount of information that people have affect the way in which they decide in the referendum. We can observe that the interactions between EU preferences and cue knowledge and factual EU knowledge are significant, but that this is not the case for cue reception. In other words, both factual knowledge of the EU and knowledge of party position will make issue preferences a more important determinant of the vote, whereas simply receiving cues makes no difference. In addition, the data illustrate that knowledge of partisan positions mediates the extent to which party recommendations affect behaviour (and the sign is negative) and makes it less likely that voters follow the recommendation of their preferred party (again, the sign is different from the main effect). The results thus seem to indicate that, as expected, both ‘encyclopaedic’ information and knowledge of partisan cues can enhance voter competence (since they augment the effect of issue preferences), whereas exposure to endorsements has no effect. The fit of each of the models is very good with a pseudo R2 of approximately 0.60 and over 87 per cent of cases correctly predicted by the models (with a distribution of the dependent variable of 46 per cent ‘Yes’ voters, according to the survey data). However, logit coefficients do not provide intuitive information about the absolute magnitude of these differences across types of information. Also, the logit models do not give us a straightforward way of comparing between different levels of information. Hence, to aid the interpretation and understanding of these information effects, I calculated the predicted probability of voting ‘Yes’, given 0.5 standard deviation change in each of the significant predictor variables, for each of the information types (cue reception, cue knowledge, EU knowledge) across three information levels (low, medium, high10). The results are reported in Table 4.
Table 3. The effect of information on voting behaviour
Model 1: Reception of cues
Model 2: Knowledge of cues
Model 3: Knowledge of the EU
Notes: The dependent variable is the vote in the referendum (‘Yes’/‘No’). All models are estimated using logistic regression and also include controls for age, income, region, left-right ideology and government satisfaction. None of these control variables are statistically significant. ** Significant at 0.01. * Significant at 0.05.
Table 4. Probability of a ‘Yes’ vote by information categories and levels
Model 1: Reception of cues
Model 2: Knowledge of cues
Model 3: Knowledge of the EU
Impact of change in variable (%)
Resulting change in ‘Yes’ vote (95% C.I.)
Impact of change in variable (%)
Resulting change in ‘Yes’ vote (95% C.I.)
Impact of change in variable (%)
Resulting change in ‘Yes’ vote (95% C.I.)
Notes: All models are estimated using logistic regression. Each of the independent variables were increased half a standard deviation, keeping all the other variables at their mean, and the change in the probability of a ‘Yes’ vote was calculated. Model 1: Vote (yes) = α + β1EU preferences + β2Partisanship + β3EU knowledge + β4Cue knowledge + β5Distance to party + β6Left-right placement + β7Government satisfaction + β8Sex + β9Age + β10Income + ε (sample divided according to level of cue reception). Model 2: Vote (yes) = α + β1EU preferences + β2Partisanship + β3EU knowledge + β4Cue reception + β5Distance to party + β6Left-right placement + β7Government satisfaction + β8Sex + β9Age + β10Income + ε (sample divided according to level of cue knowledge). Model 3: Vote (yes) = α + β1EU preferences + β2Partisanship + β3Cue knowledge + β4Cue reception + β5Distance to party + β6Left-right placement + β7Government satisfaction + β8Sex + β9Age + β10Income + ε (sample divided according to level of EU knowledge).
In Table 4, the EU preference variables were increased half a standard deviation, ceteris paribus, which is roughly equivalent to a 11/2 point increase in EU positive attitudes on a 10-point scale. The results show that for all groups of voters, this change in EU preferences has a greater effect on the probability of voting ‘Yes’ than party endorsements (and also greater than the effect of any other predictor included in the model). This suggests that voter competence may be relatively high, given that our normative expectation is that voters should vote on the basis of their issue preferences. However, voting on the basis of issue preferences in a competent manner requires information. Since our assumption is that high levels of detailed knowledge of the EU enable voters to identify correctly the choice that maximises utility, our ‘proxy group’ for competence is the group of voters with high levels of knowledge of the EU. The easiest way to compare across groups is to illustrate these predicted probabilities graphically.
Figure 5 shows that EU preferences have the most significant impact on voting for the ‘proxy group’ of voters with high levels of factual information; yet information about party cues produces similar patterns of voter behaviour. If a voter is well-informed about party positions on the EU, then issue preferences will matter more and party endorsements less (partisanship is insignificant for the high information group in model 2). This similarity of patterns of voting behaviour across levels for these two types of information suggests that knowledge of party cues can be used as a reliable substitute for detailed knowledge of the EU. On the other hand, we do not detect this variation across levels when we look at ‘cue reception’, suggesting that simply being exposed to cues is not going to make one better equipped to vote competently. This confirms our expectation that a certain level of knowledge about party positions is required in order to use party endorsements as effective heuristics.
The models presented in Tables 3 and 4 estimate the effect of different information types separately, but, in reality, some people have knowledge of party cues as well as detailed information about the EU, whereas others will lack both detailed knowledge and an understanding of party positions. To analyse how this combination of information effects influences voting behaviour, we can simulate the impact of a change in EU preferences on the probability of voting ‘Yes’ (as in Table 4) for a combination of the two information types at each information level (low, medium, high). Table 5 reports the predicted impact of a change in EU preferences for each category, including the 95 per cent confidence intervals (in brackets) and the number of respondents in each cell.
Table 5. Impact of EU attitude change on the probability of voting ‘Yes’, across information categories (percentages)
Knowledge of party cues
Factual EU knowledge
Notes: Cells include the simulated percentage impact of 1/2 S.D. change in EU preference variable on the probability of voting ‘Yes’ with number of respondents in brackets. * Failed to converge because EU preferences > 5 predicts data (‘Yes’ vote) perfectly.
As expected, Table 5 illustrates that the people who lack knowledge of both cues and the EU are least likely to rely on their EU preferences, but this group only constitutes about 10 per cent of people who voted. The impact of EU preferences increases the higher people score in the factual knowledge category, whereas the evidence is more ambiguous for the ‘cue knowledge’ category. These findings suggest that extensive EU knowledge makes attitudes more relevant than cue knowledge. While there appears to be little difference between the medium and high ‘cue knowledge’ groups when combined with factual knowledge, there is a considerable difference between the medium and high ‘factual knowledge groups’ when combined with cue knowledge. Hence, as suggested in other studies on the use of heuristics, political sophisticates are also better equipped to use cues effectively. Yet, both types of information appear to enhance competent issue voting: the people who rely most on their issue preferences are those people with high knowledge of the EU and medium/high knowledge of party positions (for the group of voters with the highest scores in both information categories, their votes are perfectly predicted by their ‘ideal points’).
In sum, these comparisons of voting behaviour across types and levels of information provide important insights into our understanding of voter competence in EU referendums. The findings suggest that while issue preferences matter to all voters, they are more important to the well-informed voters. Moreover, we find that voters can act competently without detailed knowledge of the EU by relying on party endorsements, if they have a basic knowledge of party positions on the EU. However, the argument that the mere presence of elite cues enhances reasoned voting is not supported by the data. The results in Tables 3 and 4 indicate that high exposure to the position of elite cues is not going to make voters better equipped to act on the basis of their preferences.
John Kenneth Galbraith noted that ‘there is certainly no absolute standard of beauty. That precisely is what makes its pursuit so interesting’. Equally, there are no absolute standards for what makes a political decision right or wrong, competent or incompetent, but this has not prevented countless scholars from judging political actions against normative standards. While we cannot establish absolute standards for competency in politics, it is still important to address the issue of competent voting and how it can be advanced. Some behavioural patterns are clearly regarded as better than others. For example, we would consider it better if citizens base their vote choices on their preferences towards the issue at stake than if they treat the democratic processes as a lottery and vote randomly. At the very heart of the idea of direct democracy is the notion that people have meaningful attitudes towards the issues at stake and that they vote on the basis of these attitudes. The aim of this article has been to provide a framework for conceptualising and measuring voter competence in EU referendums. According to this framework, a competent vote is based on issue proximity voting. The important question is how much and which type of information is required for voters to be able to match their own preferences with the choice set presented to them. This question has been debated extensively in the American literature on heuristics, but has rarely been examined in the European context.
The empirical analysis in this study has examined voter competence by comparing across types and levels of information. Assuming that people who are very well informed about the details of EU politics will vote relatively competently, we can infer how other types of information influence voter competence. The findings from the Norwegian accession referendum illustrate that knowledge of party endorsements has the same effect on voting behaviour as detailed knowledge of EU politics. This supports the main argument of the heuristics school – namely that elite cues (and other heuristics) can be used as reliable substitutes for more detailed knowledge. However, the results also show that simply being exposed to the cues of political parties had no ‘competence enhancing’ effect on behaviour. These findings therefore speak directly to the debate on whether heuristics can enhance civic competence by showing that endorsements may structure the choice set for voters, but only to the extent that voters have sufficient knowledge about the position of the endorser. Overall, the evidence from the Norwegian referendum gives reasons for optimism about voter competence in EU referendums: vote choices are guided mainly by issue preferences and party endorsements provide reliable information shortcuts, provided that voters have basic knowledge of party positions.
The Norwegian case was presented as a ‘critical test’ of voter competence since the EU issue has been very salient in the Norwegian political debate, and we would thus expect Norwegian voters to be better informed about the issue itself as well as the position of parties. Hence, we might suppose that voters in other EU referendums act less competently. Moreover, this article has highlighted a potentially ‘competence reducing’ aspect of the political context of EU referendums: party endorsements may be misleading since the EU dimension does not constitute an integrated part of the main political dimension(s) of contestation in European politics. Given that findings show that mere exposure to elite cues has no competence enhancing effect, we can hypothesise that in countries where the issue of European integration has remained low-salience and parties have avoided to compete on the issue, party endorsements may not provide reliable cues for voters and competence levels will be lower. These issues can only be properly addressed in a comparative empirical analysis of competence and cue-taking in EU referendums. Yet, this article has drawn attention to the role and responsibility of political parties in providing clear cues and policy alternatives on the issue of European integration. As Van der Eijk and Franklin noted in their study of European Parliament elections:
Elections conducted without proper debate of the issues relevant to that arena concerned cannot function as real elections with consequences for the exercise of power in that arena . . . only by allowing voters to choose alternative visions of the future of Europe can a crisis of democratic legitimacy be averted. (Van der Eijk & Franklin 1996: 388)
Referendums on European integration can serve to enhance the legitimacy of the integration process, but only if voters are capable of expressing their preferences concerning European integration in these referendums and if politicians, in turn, are responsive to these preferences. The former condition – competent voting – crucially depends on the information and cues provided to citizens by political elites. The study of the Norwegian referendum has shown that when voters are given such information they respond in a competent manner, although this may not have been the response that the European elites were hoping for.
Previous versions of this article have been presented at the Comparative Research Workshop, University of Michigan, February 2005; the International Studies Association annual conference, Hawaii, March 2005; and the European Union Studies Association biannual conference, Austin, April 2005. I am grateful for the support of the American National Election Studies and the Center for Political Studies at the University of Michigan. For insightful comments and suggestions, I would like to thank Orit Kedar, Arthur Lupia, Michael Marsh, Keiko Ono, Jae-Jae Spoon and Pieter van Houten, as well as the anonymous reviewers.
Appendix: Data Sources
The analyses presented in this article are based on data from the surveys referenced below. I am grateful to the Norwegian Social Science Data Service (NSD) for giving me access to the Norwegian referendum surveys. I am also thankful that Leonard Ray, Gary Marks and Marco Steenbergen make their party expert survey data publicly available. The responsibility for the analyses and interpretations presented in this article rests solely with the author.
Public opinion surveys:
EU-avstemningen 1994. Survey data were made available by the Norwegian Social Science Data Service (NSD). The dataset is owned by Statistisk Sentralbyrå Seksjon for intervjuundersøkelser (SSB).
Eurobarometer 39.0. Fieldwork conducted in 1993 for the European Commission.
Special Eurobarometer 214, wave 62.1. Fieldwork conducted in November 2004 by TNS Opinion Social and EOS Gallup Europe for the European Commission.
Leonard Ray expert survey data on party positions about European Integration, 1984, 1988, 1992, 1996 (described in Ray 1999).
Leonard Ray expert survey of Norwegian party positions, 1998 (described in Ray & Narud 2000).
Bartels (1996), for example, examines differences in patterns of voting behaviour across levels of political sophistication using the interviewers' classification of the respondent's information level as a measure. He therefore does not model explicitly the effect of different types of information.
To examine elite cues in the Norwegian context, I content analysed the Norwegian newspapers Aftenposten and Dagbladet in the three months leading up to the referendum. These are the best-selling quality newspapers in Norway. In each daily issue, all articles related to the EU were coded in terms of frequency (how many), salience (front page or not), content (20 categories including ‘referendum’) and main actors portrayed in the article (who?). The latter part of the analysis showed that members of political parties were the most frequently cited actors.
Some studies have found that the left-right dimension is related to European integration on a subset of issues – namely those concerned with redistribution and regulating capitalism (see, e.g., Hooghe et al. 2002). Yet EU referendums tend to concern the broader issues of integration and sovereignty rather than redistributional policies.
Unfortunately, there was no party expert survey conducted in Norway in 1994, so for these estimates I used the Leonard Ray's party expert survey from 1998 (see Ray & Narud 2000). To ensure that the positions of the Norwegian parties did not shift significantly in the period between the referendum and the survey, I cross-checked the estimation with the party expert estimations from 1992 and 1996 and the Marks and Steenbergen survey from 1999, and they show that the Norwegian party positions remained very stable in this period.
A factor analysis also shows that the three variables extract three factors with each of the variables loading more heavily on each of the three factors, thus suggesting that the variables can be considered orthogonal.
The information scales all have alpha reliabilities in the 0.80 to 0.90 range, and thus have high internal consistency.
Only respondents who actually voted are included in the analysis since the aim is to examine competent voting. The question of turnout is an equally important issue, yet it is not addressed in this article.
I ran the same model using a ‘party sympathy’ question instead of the vote question, and the results were almost identical.
Missing data pose a serious problem in multivariate models applied to survey data. Using listwise deletion would mean losing a third of the data, and thereby losing valuable information and risking selection bias. To alleviate this problem, the analyses presented in Tables 3 to 5 have been run with imputed values for missing data using the multiple imputation software programme AMELIA (see King et al. 2001).
These groups constitute a large middle group of half of the voters and two smaller top and bottom groups with approximately 25 per cent of the voters in each group.