Empirical studies on representation have been based mainly on the descriptive analysis of levels of political or ideological congruence between MPs and voters. Very few studies focus on explaining congruence and those which have done so do not explore all the explanatory dimensions. This article contributes to filling this gap by testing whether three theoretical models can explain left-right congruence among European parties. These models explore causality at the micro or individual level (the characteristics of voters and MPs), the meso level (party characteristics), and the macro or system level (party system and institutional characteristics). Based on data from the PIREDEU project, the study examines the party systems of the 27 countries of the European Union with reference to the 2009 European Parliament elections. The findings reveal that MP-voter congruence is best explained at the party level and by key MP (candidates) characteristics.

The article addresses the crucial but still relatively understudied issue of political perceptions, specifically the perceptions of members of parliament (MPs) The findings show that MP perceptions are shaped by wishful thinking, reiterating previous research, and that party integration is an important new factor explaining perceptual accuracy.

Most research on political congruence has failed to address a fundamental aspect: the explanation of its causality. Party characteristics are acknowledged to be relevant, but there is no consensus as to their effective importance. Nor has there thus far been an analysis of the implications for results of using different methods to assess congruence.The present article seeks to respond to these goals by testing the importance of party type (catch-all versus ideological) to an explanation of levels of intra-party congruence, comparing the main methods of assessing congruence. The study focuses on the Portuguese party system and makes use of a set of 14 policy issues across two dimensions: socioeconomic left-right and libertarian-authoritarian (plus left-right selfplacement).Findings support the notion that Portuguese catch-all parties tend to display slightly higher levels of MP-voter congruence than more ideological parties, although that is essentially dependent on the issue in question. These results were supported by the different methods used (albeit to a lesser extent by centrism).

Keywords: political parties, congruence, left-right, MPs, voters. Some significant gaps remain in the empirical study of MP-voter representation; this is due largely to data restrictions as these studies usually require an equivalent data base for citizens and parliamentarians. In an attempt to overcome this weakness, we assess representation through measures of MP-voter congruence within political parties.Although the study of representation implies varied levels of analysis ranging from the formation of citizens' preferences to policy outcomes, we disregard the diachronic analysis underlying that process and focus on the synchronic analysis of citizens and their ideological congruence towards the parliamentarians within European political parties. 3Political parties are not only fundamental actors in the democratic process, but also the key official channels that connect the will of the people and its representation in parliaments in Western Europe parliamentary democracies. Furthermore, the link provided by elections serves as a base on which to assess congruence between citizens and elites (Dalton, 1985: 278; Powell, 2000: 5; McDonald & Budge, 2005). Not only has this role been the focus of little attention to date, but no comparative research has yet been conducted on ideological representation measured through congruence that includes the set of European countries considered here. This is the main purpose of this paper.We begin with a discussion on representation through congruence. Given the many ways of conceptualizing and operationalizing representation and the implications of its study through congruence, we start by presenting how representation should be considered in this research and how it is measured through congruence. This is followed by the analysis of left-right congruence. The use of this variable raises some comparability problems between MPs and voters that must be duly addressed. The hypotheses and the methodological notes are then presented, after which MP-voter congruence is assessed in each country in order to test if European MPs are ideologically more extreme than their supporters as mainstream research on this subject suggests. In the next section, a study is made of MPs' perceptions of parties' and voters' ideological positioning. Finally, the last section estimates MP-voter congruence within individual European political parties. Representation through CongruenceRepresentation is often said to be a complex concept that goes beyond the assessment of MP-voter congruence. However, congruence can be seen as a valid instrument to assess 4 representation albeit limited to measuring the distances between the positions of the representatives and the represented. To this extent, representation occurs if the elected MPs share the ideological positioning and policy preferences of those citizens who voted for them. This straightforward approach of representation provides helpful information about the levels of representatives' responsiveness. We will return to this issue later.The Responsible...

In the literature elite-mass congruence is usually measured either in terms of issue preferences or in terms of left-right self-placement. The two forms of congruence are only occasionally used together and seldom were they used with the purpose of seeing if the levels of congruence were similar. With this intention we use Portugal as an extreme case (due to low level of clarity of party policy alternatives, and to relatively low levels of education, media exposure, and political interest). We found that in many situations the results for MPs-mass congruence were different depending on the use of left-right self-placement or substantive issue preferences.Keywords: political representation; left-right; ideology; Portugal. 3 IntroductionThe empirical study of representation in democratic regimes has relied extensively upon the "responsible party model" and its normative implications (Miller and Stokes, 1963;Powell, 2000 and2004;Kitschelt et al, 1999;Miller et al, 1999;Thomassen, 1994;Schmitt and Thomassen, 1999a;Esaiasson and Holmberg, 1996; Wessels, 2007;Belchior, 2010; about the concept of representation, see Pitkin, 1976). This model points to some features about voters-parties' relations: 1) electors share various packages of issue positions; 2) electors compare their issue positions with the ones presented by the political competitors in each election; 3) electors vote for the party which presented the issue package which is closest to their own; finally, 4) once elected, the party officials in parliament, etc., remain united and trying harder to enact the issue package (they presented to the electorate) into public policy (Pierce, 1999: 9; Thomassen and Schmitt, 1999: 13-19).Although some voters in at least some occasions do in fact compare their issue positions with the ones presented by the parties, in order to decide how to vote, the truth is that this task is very demanding in terms of data collection and processing, and the benefits of the act of voting are not very high.1 Thus, most electors most of the time use shortcuts like the left-right divide to evaluate where the parties stand in terms of packages of issue positions, and also to evaluate how close the parties are to their (packages of) issue positions (Downs, 1957; Popkin, 1992). Then they tend to vote in the party that is closest to their issue positions, although this is usually evaluated in terms of the appropriate short-cut: usually the left-right divide, in Europe, and the 1 Of course, this is only one theory about voting behaviour, which explains only a part of the variance in the vote and a part of the variance from country to country and from election to election. This is more likely to happen to more sophisticated voters and especially in occasions when the clarity of policy alternatives presented to voters by parties is higher. 4liberal-conservative schema, in the US. Moreover, to communicate with the voters, parties often use the language of left and right: to inform where they stand in terms of issue positions (Fuchs and Kl...