Navigation – Plan du site

AccueilNuméros23Géographie et sociologie électora...Does Associational Membership Aff...

Géographie et sociologie électorales : duel ou duo ?

Does Associational Membership Affect Extreme Right Voting and Anti-Immigrant Attitudes? An Individual and Contextual Analysis in Belgium

L'appartenance à une association affecte-t-elle le vote pour l'extrême droite et les attitudes anti-immigrés ? Une analyse individuelle et contextuelle en Belgique
Quentin David, Jean-Benoit Pilet et Gilles Van Hamme


L’article explore le rôle de l’appartenance directe à des associations et de leur présence dans le contexte de l’individu sur ses attitudes vis-à-vis des immigrés et de l’immigration ainsi que sur le vote d’extrême droite. Par rapport à la littérature existante sur le sujet, nos analyses montrent le poids très limité du capital social pour expliquer ces comportements. A partir des enquêtes post-électorales PIOP-ISPO de 1999 et 2003, nos résultats indiquent que, dans le contexte belge contemporain, l’intensité de la vie associative dans l’environnement n’a pas d’influence significative sur les attitudes et le comportement électoral, et que le capital social attaché à l’individu dépend essentiellement de la nature des associations : en particulier, l’appartenance à des associations politiques diminue les attitudes négatives vis-à-vis de l’immigration (et des immigrés) ainsi que le vote d’extrême droite. En fin de compte, l’influence du capital social au sens large semble avoir une influence bien moins directe que ce que la théorie et certaines études empiriques ne supposent.

Haut de page

Texte intégral


1In the last fifteen to twenty years, extreme right parties have increased their electoral appeal in many Western European countries. The most obvious examples are the Front National in France, the FPÖ in Austria, the SVP in Switzerland, the FolkeParti in Denmark, the Lijst Pim Fortuyn in the Netherlands, the BNP in Britain, and the Vlaams Blok/Belang in Belgium. Although these parties are heterogeneous (see Mudde, 2007; Hainsworth, 2008), they share common features that link them to the extreme right.

  • 1 Mayer, 1987; Betz and Immerfall, 1998; Kitschelt and Mc Gann, 1997; Chapin, 1997; Eatwell, 2003; Ea (...)

2In order to understand the factors of these electoral successes, a vast body of literature has focused on the determinants of voter support for extreme right parties1. Using the metaphor of the electoral market, we can divide these theories into supply-side and demand-side explanations. Political supply-side factors relate to the configuration of the electoral arena, the number of parties, and their respective ideological positions. In particular, factors such as multipartism, the absence of big mainstream right-wing parties, and the presence of an anti-immigration party led by a charismatic leader, combine to explain the high electoral performance of the extreme right (van der Brug et al., 2005). As for the demand-side explanations, the literature emphasized the impact of the individuals’ social background and their economic and social environment on extreme right voting. This article adopts the demand-side perspective.

3More specifically, we are interested in the effects of social capital and social embeddedness on the propensity to vote for the extreme right and to adhere to what constitutes the core of their policy platform: anti-immigration positions (Fennema, 1997). Scholars have noted how social isolation and anomy make citizens more vulnerable to the message of extreme right parties (Ignazi, 1999). However, systematic accounts of the effect of this variable remain rare and are not fully satisfactory (Coffé et al., 2007). The existing studies do not combine the influence of individual and contextual social capital, although the direct and indirect (rainmaking) effects of associational participation are central in Putnam’s theory of social capital (Rossteutscher, 2008).

4In order to fill this gap, this article proposes a systematic assessment of the impact of individual embeddedness in social networks on extreme right voting and on the adhesion of citizens to the anti-immigrant discourse of extreme right parties. In this sense, we try to address one of the issues noted by van der Brug and Fennema in their review of existing works on extreme right parties (2007): “The theory that lack of civic community leads voters to vote for radical right is one of the oldest in the history of electoral research on the radical-right parties (Arendt 1951; Bendix 1952). Yet, it has never been convincingly been proven or discarded. New databases and new methods may well do better here”. We believe that our approach combining contextual and individual measures of social capital and the use of original data can provide appropriate though partial answers to this issue.

5The article is organized as follows. We begin in section 1 with an overview of the literature on extreme right voting. Particular attention is given to the role of social capital. In section 2, our analytical model is specified for the case of Belgium. Section 3 presents our data and the general approach used in section 4. Results are detailed and discussed in section 5.

Theoretical framework: the role of social capital on extreme right voting and anti-immigrant attitudes

  • 2 Arzheimer, 2009; Betz, 1994; Givens, 2002; Golder 2003; Rink et al., 2009, Jackman and Volpert, 199 (...)

6Many scholars have tried to provide explanations for the growing support of extreme right parties in several Western European countries. Among demand-side explanations, the factors that are most frequently mentioned are the socio-economic situation of voters2 or the adherence to some attitudes like nationalism or authoritarianism (Lubbers and Scheepers, 2000). Another characteristic of voters adhering to the ideas of extreme right parties that is often mentioned is social isolation.

  • 3 Ignazi, 1999; Lubbers and Scheepers, 2000; Billiet and De Witte 1995; Lubbers et al., 2002.

7The general idea is that the processes of individualization and de-massification (Franklin et al., 1992; Pakulski, 2005; Flanagan, 1987) have made some voters more vulnerable to the discourse of extreme right parties. This de-massification is presented as the result of the decline of mass organizational memberships, notably to parties and trade unions (Swyngedouw, 1998). Combined with higher levels of education, these evolutions supposedly allow individuals to be more autonomous in their access to information and in their political decisions. It would result in growing electoral volatility and in the possibility for new political forces to emerge (Dalton, 1984). However, this is only one side of the story. The other side is that the destruction of social links (associations, communities, families, etc.) results in growing social isolation and anomy. A growing share of the population might suffer from this situation and favor the search for order and security, which potentially leads to extreme right support3.

8Starting from this idea, a few authors have mentioned a link between social capital and extreme right voting. Theoretically, the idea is that higher levels of social capital would reduce the likelihood to vote for extreme right parties (van der Brug and Fennema, 2007; Ignazi, 1999) as well as adherence to ethnic-centrist discourses (Hooghe, 2008).

9Empirically, most pieces of research trying to assess the relation between social capital and either extreme right voting or anti-immigrant attitudes focus on one specific aspect of social capital: membership of associations. The concept of social capital goes obviously beyond being member of associations. First, social capital contains both a structural component, that refers to the ‘social connections between people and groups’ like those developed via associational membership, but also a cultural component that refers more to the social norms, obligations and values like trust (van Deth, 2008). Moreover, even for the structural component of social capital only, social connections cannot be reduced to membership of associations but also include other forms of social inclusion via family, friends, neighbors or colleagues (Mutz, 2002).

10The most frequent operationalization of social capital in political science, and especially in studies of extreme right voting, focuses on membership of voluntary association (Rydgren, 2009; Billiet and De Witte, 2001; Coffé, 2002, Mayer and Perrineau, 1992). In many respects, this choice goes back to Putnam’s work on the effect of associational participation on democratic attitudes and behaviors (Putnam, 1993; 2000). According to Putnam, associational participation has a double influence on the relation of individuals to society. First, the participation of citizens in associational life increases their social trust, that is, their trust in other members of the community (Verba et al., 1978; Newton and Delhey, 2003; Stolle, 1998). Second, through voluntary association, citizens develop democratic values. As argued long ago by Alexis de Tocqueville, associations function as schools of democracy. Consequently, members of associations show stronger levels of political trust and stronger feelings of political efficacy (Newton, 2007). On the contrary, people who are socially isolated and have a low level of social capital show low levels of social and political trust. These lower levels of trust are often associated to a greater likelihood of adhering to populist and extreme right discourses.

11In addition, the presence of voluntary associations is perceived to have a similar effect on non members of voluntary associations through a rainmaking effect (Putnam, Pharr and Dalton, 2000): they impact the level of social and political trust of both members and non-members who belong to the same civic community. In that sense, Coffé et al. (2007) have shown that Belgian municipalities with less dense networks of voluntary associations constitute fertile ground for extreme right parties.

12To the best of our knowledge, this paper is the first attempting to combine the two components of Putnam’s argument. We examine the influence on extreme right voting of both (a) the direct participation of individuals to associations and social organizations, and (b) the presence of such associations in the context in which individuals live. To achieve this goal, we combine individual-level data from electoral surveys and contextual variables on the density of social associations.

13Furthermore, we link individual social capital with attitudes towards immigrants. Earlier works have widely underlined exclusionist and anti-immigrant attitudes as the best predictors of extreme right voting (Mayer, 2002; Billiet, 1998; Rydgren, 2008). Some have even argued that extreme right voting could be defined as issue voting. Yet, when it comes to political attitudes towards migrants, it has been shown that the link between them and social and economic factors is not always straightforward (Lubbers and Scheepers, 2000; Lubbers et al., 2000, 2002). Consequently, it would be interesting to study separately what causes both anti-immigrant attitudes and extreme right voting. We think that it would be an error to assume that they are totally similar.

14Building on these theoretical considerations, this article tests the impact of social capital on anti-immigrant attitudes and on extreme right voting. Our approach is synthesized in figure 1. In this paper, we focus on the bold arrows and we use an indicator of social trust as a proxy for the private social capital.

Figure 1: The influence of the different dimensions of social capital on attitudes toward immigrants and on extreme right voting

Figure 1: The influence of the different dimensions of social capital on attitudes toward immigrants and on extreme right voting

Extreme right and socialization processes in Belgium

15The choice of Belgium for this analysis could be explained by obvious pragmatic considerations. Unlike most other election studies, we have at our disposal data on voters’ direct and contextual social capital. However, there are also more substantive reasons to choose the Belgian case to analyse the influence of socialization processes on extreme right voting and on attitudes towards migrants.

  • 4 Lorwin, 1971; Marissal et al., 2007; Authors, 2011.

16First of all, Belgium offers an interesting case of deep change in its traditional socialization processes. Until the 1960s, social and political life was dominated by three traditional political families (Liberal, Catholic, and Socialist). Parties of these families were alternatively forming coalition governments and gained most parliamentary seats. Their dominance was historically associated with the strength of their respective pillar. Since the late nineteenth century, these traditional political families—though less strongly for the liberals—have been part of a dense network of associations and institutions that encapsulate individuals from the cradle to the grave. These networks are composed of trade unions, health insurance companies, leisure, community and socio-political associations, youth movements, schools, universities, hospitals, newspapers, and so on4. The strength of these pillars and their associated parties reached a peak in the first half of the twentieth century. Since the early 1950s, in parallel with long-term socio-economic evolutions, Belgian society started a process of depillarization (Billiet, 1984; Huyse, 1984). Although they remain influential (David, Van Hamme, 2011; Hooghe 1999), the decline of pillars and their parties has certainly made room for the emergence of new parties, including parties covering the extreme right of the political spectrum.

17In this context of social transformation, Belgium is definitely a favourable ground to investigate to what extent social capital remains significant for understanding extreme right voting in a context in which the traditional organizations of social encapsulation are thought to have lost much of their influence.

18The Belgian case is even more interesting since these transformations have apparently had a differentiated course in the two main regions of the country: Flanders and Wallonia. In both regions, traditional parties have been challenged, first by regionalist parties, and then by ecologists. However, the extreme right only became a serious political challenger in Flanders. Between 1981 and 2007, the Vlaams Belang (Flemish extreme right party) continuously increased its electoral appeal and progressively became one of the major parties. It won almost one-quarter of the valid votes in the 2004 regional elections. In contrast, the French-speaking extreme right has only had passing electoral success (Delwit et al., 1998; Aish et al., 1994), hardly gaining more than five percent of the votes except in the old industrialized province of Hainaut. In order to understand these differences, socialization processes have to be taken into consideration. Previous research has shown that the differentiated success could not be fully explained by differences in the characteristics of the voters in the two regions (Van Hamme, 2008), nor by the different economic situations in Flanders and Wallonia (Coffé, 2005). An alternative explanation could therefore be that socialization processes have been a key element for the electoral performances of the extreme right (Klandermans and Mayer, 1999).

Questions, Data and Method

19We analyze the impact of social capital on extreme right voting and on attitudes toward immigrants and immigration. Controlling for classical socio-demographic variables at the individual level (age, gender, social class, education, etc.), we use regression analyses to test the impact of being embedded in social networks both at the individual and the contextual levels.

20We focus on two questions regarding the attitudes toward immigration and immigrants and extreme right voting:

21To what extent does direct individual involvement in social networks (measured by association membership) reduce intolerance toward immigrants and extreme right voting?

22To what extent does the local presence of social networks (measured by the intensity of pillarization) reduce intolerance toward immigrants and extreme right voting? In contrast to the first question, we consider here only the associations, institutions or organizations, which are part of a pillar (i.e. related to liberalism, socialism or catholicism).

23At the individual level, we test the impact of individual embeddedness in social networks using the following variables:

24Association membership, distinguishing between associations that are linked to a particular pillar (labelled “political association”) and associations that are not politically oriented;

25Trade union membership, with a distinction according to the ideological nature of the trade union (Liberal, Socialist, or Catholic).

  • 5 Pôle interuniversitaire Opinion Publique et Politique on the French-speaking side and Institute of (...)

26The data at the individual level come from the Belgian post-electoral surveys led by ISPO and PIOP5. In order to increase the sample, we use both 1999 and 2003 surveys.

27Additionally, at the contextual level, we assess the effect of:

28the local density of social networks, taking into account trade unions, youth movements, and the presence of several types of associations (expressed as a share of the population being a member of one or more of these associations at the level of the community);

29the ideological nature of local pillarization; as shown by David and Van Hamme (2011), most associations can be related to one of the three pillars in Belgium (Socialist, Catholic or Liberal). Indicators of the ideological nature of the pillars are based on four variables: trade union membership, social election results, share of members of the health insurance company, and members of the youth movement associated to each pillar.

  • 6 The Canton groups one to several municipalities and is the lower level at which results can be obta (...)

30All variables are computed at the level of the canton6, which has been shown by David and Van Hamme (2011) to be an appropriate scale. These measures are similar to those used in David and Van Hamme (2011) but we use the absolute intensity instead of the relative intensity of the local presence of the pillars.

31In the regressions, we use the standard individual control variables (age, sex, socio-professional status, education and church attendance). At the contextual level, we control for degree of rurality (distinguishing between rural, urban and suburban), year of election, share of immigrants, unemployment rate, and three dummies for the region where the individual is located (Brussels, Wallonia, and Flanders) which largely captures the differences in terms of electoral supply. At the individual level, we control for age, gender, main language, employment status (unemployed, student, not active vs. employed), and socio-professional activity.

  • 7 The five questions used to measure individual social trust are: (1) I really don’t know anymore who (...)

32In some specifications, we include a control variable measuring respondents’ declared social trust. This variable is used as a proxy for the social isolation of individuals through channels other than participation in public life (mainly relation to family and friends). Hence, it is used as a variable to control the robustness of the analyses produced to answer our two key questions. The variable of social trust is obtained from a factor analysis based on five questions related to trust in others and on one’s ability to cope with social complexity7. We believe that introducing this variable in the model does not raise significant problems of endogeneity, since social trust is not directly part of the ideological discourse of the extreme right parties, as opposed to anti-immigrant attitudes or political trust.


The impact of social capital on extreme right voting

33The results presented in table 1 describe the influence of individual and contextual social capital on extreme right voting. The dependent variable (vote XD) equals one if the considered individual voted for an extreme right party (FN or VB in Belgium) in the 1999 or 2003 election.

Table 1 : Logit regression for extreme right voting (simplified table – full table in appendix)

Table 1 : Logit regression for extreme right voting (simplified table – full table in appendix)

Source des données : PIOP-ISPO, 1999, 2003

34As for the influence of individual embeddedness in social networks, we have used several model specifications. In models 1 and 2, no distinction is made between types of associations. In model 3, we differentiate between membership to trade unions and to other associations – because trade unions are traditionally more closely linked to political parties, and therefore membership could be endogenous to voting. In model 4, we differentiate between the ideological orientation of the trade union (catholic, socialist or liberal), and between political and non-political (leisure) associations.

35The first result to underline is that the influence of individual embeddedness in social networks has a much more limited effect on extreme right voting than argued in the theory and in most previous empirical analysis. When we do not distinguish between types of associations, the effect of individual membership is never significant. The second major finding is that, when distinguishing between types of associations, only two types display a significant relation with extreme right voting: being a member of a politicized association and being a member of the catholic trade union. The first type reduces the probability of voting for an extreme right party. Yet one could not argue of a real effect of social capital, since being an active member of a political association is very endogenous to voting behaviour. For the other significant factor - membership to the catholic trade union (ACV-CSC) – the effect is even more puzzling. It increases the probability to vote for the extreme right. So does, and even more strongly, being a regular churchgoer (see table A.1. in appendix). This result is in contradiction with previous analyses which consider religion as a deterrent for extreme right voting - at least in the recent period and in contrast to extreme right in the 1930s (Mayer, Perrineau, 1989 ; Marissal et al., 2007). Our interpretation is that, when controlled for the rural/urban context, this constraint does not hold.

36The introduction of social trust in the model does not affect the results, despite the very significant impact on extreme right voting. This suggests that forms of social life other than public social life are important for understanding extreme-right voting. However, we must be very careful when interpreting these results in the absence of a direct measure of, for example, family life and relations to friends.

37At the contextual level, we observe no significant impact of the presence in the canton of associations, trade unions, and youth movements, on extreme right voting. When we consider the ideological nature of the context (Table A.3.), we find that only the liberal context has a negative but weak impact on extreme right voting.

38Finally, we must note the strong negative impact of the rural context on extreme right voting (Table A.1.): the more rural the context, the weaker the probability to vote for the extreme right. Living in a suburban area (intermediate areas in table A.1.) also shows a significant negative effect on the probability to vote for the extreme right, although the effect is weaker. The difference between rural and urban areas has already been underlined, but in our model, the impact persists when controlling both for individual and contextual variables. While this cannot be interpreted as a demonstration of the impact of social networks on extreme right scores, it is true that the nature of social networks and social control are very different in rural and urban areas.

The impact of social capital on attitudes toward immigrants

39In the second table, the relation between social capital and attitudes towards immigrants is studied. To build our dependant variable, we conducted a factor analysis based on the following four questions, for which the answers range from 1 (“completely disagree”) to 5 (“completely agree”):

40“We cannot trust immigrants”;

41“Migrant workers take advantage of our national social security”;

42“Muslims are a threat to our culture”;

43“More conditions should be required to grant Belgian nationality to foreigners”.

44The eigenvalue associated with the first component is 1.8, and with the second 0.1. The loadings are very homogenous for the four original variables (between 0.61 and 0.72). This shows the coherence of the first component in capturing the attitudes of individuals toward immigrants.

Table 2: Linear regression for intolerance toward immigrants and immigration (simplified table – full table in appendix)

Table 2: Linear regression for intolerance toward immigrants and immigration (simplified table – full table in appendix)

Source des données : PIOP-ISPO, 1999, 2003

45As illustrated in Table 2, the picture is somewhat different when considering attitudes as the dependent variable, instead of behaviours (extreme right voting). The results are more straightforward. Individual membership to associations or trade unions significantly reduces negative attitudes towards migrants. The effect holds when decomposing between types of associations. Besides, being member of a trade union, a political association, or a non-political association has the same negative and significant effect on attitudes. The only discrepancy is when one distinguishes between trade unions. Only membership to the two largest trade unions – catholic and socialist – significantly reduces negative attitudes towards migrants.

46Conversely, the intensity of social networks in the individual’s context does not have a significant effect on attitudes towards migrants. When decomposing the contextual effect and looking at the three pillars, we find a weak positive impact of the presence of liberal and catholic pillar in the context, depending on the model (table A.4.). Like for extreme right voting, these results hold even when we introduce the social trust variable. This factor is significantly and negatively correlated with the dependent variable, but its introduction in the model does not affect the effect of the other independent variables capturing the role of social capital.

47Finally, in contrast with the previous analyses, rural context and being Flemish have no impact on attitudes toward immigrants (Table A.2.). Although, the extreme right obtains much better electoral results in Flanders and urban areas, attitudes toward immigrants are not significantly different in Flanders and Wallonia, or in urban compared to rural areas. This finding suggests that in Flanders and in urban areas, negative attitudes toward migrants result in extreme right voting more often than in Wallonia and in rural areas. While the difference between Flanders and Wallonia is well known (Coffé, 2005) and has been discussed above, it is more difficult to interpret the contrast between rural and urban areas. One hypothesis could be that urban anomy plays a stronger role than rural anomy in the translation of anti-immigrant attitudes into a vote for the extreme right. But this hypothesis would need further testing to be convincingly confirmed.

Robustness checks

48One could discuss the potential exogeneity of variables such as “being member of a politicized association” or of variables concerning trade union membership in an analysis of extreme right voting. Therefore we ran two sets of regressions without these variables (columns 1 and 2 of tables A.3 and A.4 in appendix). Removing these variables does not affect the coefficient and the significance of the other effects.

49Another robustness check consists in the decomposition of the contextual effect between the three ideological families.

50Finally, we decided to run a regression with every single variable taken into consideration in the paper. It is worth to note that all the variables seem to have separate effects, as nor the coefficients, nor their significance obtained when the variables are introduced “one by one” seem to be significantly affected.


51Individualization and the weakening of social ties have long been associated with the rise of extreme right voting (Arendt, 1951; Bendix, 1952). This paper provides a test of this hypothesis by considering the two facets of Putnam’s social capital argument: direct membership to associations, and the intensity of public life in the individual’s environment. This is to our knowledge the first study that combines data on individual social capital with information on the strength of social networks and institutions in the voters’ environment.

52We find that this hypothesized effect of associational membership is far from being confirmed. In Belgium, it appears that the decision to vote for the extreme right is not influenced much by direct membership of voluntary associations nor, indirectly, by the strength of voluntary association in the voter’s environment. Only being a member of a political association reduces the probability to vote for the extreme right, raising the issue of endogeneity. In contrast, being a member of an association or a trade union, either of a political or non-political nature, significantly reduces the attitudes toward immigrants and immigration, although it is not the case for the individual’s context. We show that contextual variables rarely impact extreme right voting or attitudes toward immigrants.

53Our results are in sharp contrast with previous works (Coffé et al., 2007; Coffé, 2002). Therefore, the first question to address is how can we account for these divergent results.

54The most logical answer is the methodological innovation of this contribution. We offer the first analysis that attempts to deconstruct the social capital hypothesis. We combine indicators of individual and contextual social capital, and we distinguish between types of associations (non-political, political or pillarized, and trade unions). It appears that this subtler operationalization, which was not possible in previous works due to lack of data availability, makes a significant difference in the results obtained. First, it shows that the contextual effect of social capital disappears when controlled for individual membership. These results directly challenge Coffé et al’s observations (2007), who had no data available at the individual level. Second, it shows that membership in political and non-political associations should be differentiated. The former has an impact on both voting behavior and attitudes toward migrants, although there might be an issue of endogeneity with voting behavior. The latter only influences attitudes, reducing anti-immigrant sentiments.

55These elements confirm that any future study linking social capital and associational membership to extreme right voting should apply a more subtle operationalization of all components of this theoretical approach. One could even argue that additional components of social capital should be added. In particular, the database used in this article does not include information on contacts with family and friends, which are an important component of social anomy (Durkheim, 1997 [1951]) and social trust. In our analyses of extreme right voting and negative attitudes toward migrants, social trust has a constant significant negative effect. We may hypothesize that our index of social trust partly captures the intensity of these other forms of social life, particularly related to family and friends.

56Besides challenging previous research, other elements in our analysis raise interesting theoretical questions. First, as recently demonstrated by Poznyak et al. (2011), factors explaining extreme right voting vary over time. One factor could be decisive for the emergence of extreme right parties, but be much less decisive when the party is consolidated. In particular, processes of social anomy and demassification might play a decisive role in explaining the emergence and success of extreme right parties. It would be interesting to test the evolution of the impact of these processes. Poznyak and colleagues only had data on social capital for one election and were not able to fully test the evolution of its effect over time. The fact that our results contradict those of earlier electoral studies in Belgium may indicate that the role of social capital evolves over time. Social anomy could have been vital in the emergence of Vlaams Belang in Flanders, but became less important after the party consolidated. Further tests would be required to address this question.

57Finally, one of the most striking results lies in the effect of rural vs. urban context on extreme right voting and attitudes toward migrants. While the rural context has no impact on attitudes toward migrants, meaning that rural residents are not more tolerant than urban inhabitants or vice-versa, to reside in a rural area significantly reduces the probability to vote for the extreme right, even when controlling for the unemployment rate and the share of migrants in the area. It means that the weaker probability to vote for the extreme right in rural areas is not explained by different economic conditions (Givens, 2002; Betz, 1994), or less frequent contacts with migrants (Golder, 2003). Therefore, other explanations must be provided to understand this specific voting behaviour in rural areas. These findings confirm that the context has a real impact on the translation of attitudes into voting behaviours. Yet, existing theoretical models do not account for this. On the one hand, it is worth investigating more in-depth the process of socialization and the potential differences between urban and rural areas. On the other hand, it would also be worth looking at supply-side factors, in particular the local strength of the extreme right parties. Indeed, attitudes are translated into votes because of the capacity of a party to mobilise on these attitudes. The organizational strength of a party, not only nationally but also locally, is crucial in that respect. These new elements should be investigated further in future research.


The data/tabulations utilised in this publication were made available by th ISPO and PIOP – Interuniversity Centres for Political Opinion Research, sponsored by the Federal Services for Technical, Cultural and Scientific Affairs. The data were originally collected by Jaak Billiet, Marc Swyngedouw (ISPO) for the Flemish voters and André-Paul Frognier, Pierre Baudewyns (PIOP) for the French-speaking. Neither the original collectors of the data nor the Centre bears any responsability for the analysis or interpretations presented here”.

Haut de page


Aish-Van Vaerenbergh, A-M., Swyngedouw, M., 1994, Stabilité et instabilité du vote en Wallonie: désalignement et réalignement des électeurs, in Frognier, A-P., Aish-Van Vaerenbergh (eds.),. Elections. La fêlure. Brussels, De Boeck, p. 61-76.

Arendt, H., 1951, The origins of totalitarianism, Orlando, Harcourt Brace & Company.

Arzheimer, K., 2009 “Contextual Factors and the Extreme Right Vote in Western Europe, 1980–2002”, American Journal of Political Science, 53, 2, p. 259-75.

Bendix, R., 1952, ‘Social stratification and political power’, American Political Science Review, 46, 2, p. 357-75.

Betz, H-G., 1994, Radical Right-Wing Populism in Western Europe, Basingstoke, Macmillan.

Betz, H-G., Immerfall, S., (eds.), 1998, The new politics of the right: Neo-populist parties and movements in established democracies, Basingstoke, Macmillan.

Billiet, J., 1984,  ‘On Belgian Pillarisation : Changing Perspectives’, Acta Politica, 19, 1, p. 117-28.

Billiet, J., 1998, Qui vote pour le Vlaams Blok ? in Delwit, P., De Waele, J-M., Rea, A. (eds), L’extrême droite en France et en Belgique, Bruxelles, Editions complexe.

Billiet J., De Witte, H., 1995, ‘Attitudinal dispositions to vote for a ‘new’ extreme rightwing party: The case of the ‘Vlaams Blok’, European Journal of Political Research, 27, 2, p. 181–202.

Billiet, J., De Witte, H., 2001, ‘Wie stemde in juni 1999 voor het Vlaams Blok en waarom?’, Tijdschrift voor Sociologie, 22, 1, p. 5-35.

Chapin, W.D., 1997, ‘Explaining the electoral success of the new right: the German case’, West European Politics, 20, 2, p. 53-73.

Coffé, H., 2002, ‘De invloed van de levensbeschouwelijke en maatschappelijke betrokkenheid op een positieve beoordeling van het Vlaams Blok’, Tijdschrift voor Sociologie, 23, 2,p. 161-92.

Coffé, H., 2005, Extreem-Rechts in Vlaanderen en Wallonië. Het verschi,. Roeselaere, Roularta Books.

Coffé, H., Heyndels, B., Vermeir, J. (2007) ‘Fertile grounds for extreme right-wing parties: Explaining the Vlaams Blok’s electoral success’, Electoral Studies, 26, 1, p. 144-55.

Dalton, R.J., 1984, ‘Cognitive mobilization and partisan dealignment in advanced industrial democracies’, The Journal of Politics, 46, 2, p. 264-84.

David Q., Van Hamme G., 2011, “Pillars and electoral behavior in Belgium: the neighborhood effect revisited”, Political geography, 30, 5, p. 250-262

Durkheim, E. (1997) [1951] Suicide : a study in sociology. London: The Free Press.

Eatwell, R., 2003, ‘Ten Theories of the Extreme Right’, in Merkl, P., Weinberg, L. (eds.), Right-Wing Extremism in the Twenty-First Century, London, Frank Cass, p. 47-73.

Eatwell, R., Mudde, C., (eds.), 2004, Western Democracies and the New Extreme Right Challenge, London, Routledge.

Fennema, M., 1997, ‘Some conceptual issues and problems in the comparison of anti-immigration parties in Western Europe’, Party Politics, 3, 4, p. 473-92.

Flanagan, S., 1987, ‘Value change in industrial societies’, The American Political Science Review, 81, 4, p. 1303-19

Franklin, M., Mackie, T., Valen, H. (eds), 1992, Electoral change: response to evolving social and attitudinal structure in western countries, Cambridge, Cambridge University Press.

Givens, T.E., 2002, ‘The role of socioeconomic variables in the success of radical right parties’, in Schain, M., Zolberg, A.R., Hossay, P. (eds.), Shadows over Europe: the Development and Impact of the Extreme Right in Western Europe, New York: Palgrave, MacMillan, p. 137-158.

Golder, M., 2003, ‘Explaining variation in the success of extreme right wing parties in Western Europe’, Comparative Political Studies, 36, 4, p. 432-66.

Hainsworth, P., 2008, The Extreme Right in Western Europe,London, Routledge.

Hooghe, M., 1999, ‘De persistentie van verzuiling op microniveau in Vlaanderen. Een analyse van surveydata over lidmaatschap, zuilintegratie, stemgedrag en maatschappelijke houdingen’, Res Publica, 4, p. 391-420.

Hooghe, M., 2008, ‘Voluntary Associations and Socialization’, in Castiglione, D., van Deth, J.M., Wolleb, G. (eds.), The Handbook of Social Capital, Oxford, Oxford University Press, p. 568-93.

Huyse, L., 1984, ‘Pillarization Reconsidered’, Acta Politica, 19, 1, p. 145-58.

Ignazi, P., 1999, ‘Les partis d’extrême droite : les fruits inachevés de la société postindustrielle’, in Perrineau, P. (ed.), Les croisés de la société fermée, Paris, Cevipof.

Jackman, R.W and Volpert, K. (1996) ‘Conditions favouring parties of the extreme right in Western Europe’, British Journal of Political Science, 26 (4): 501-21.

Kitschelt, H., Mc Gann, A.J., 1997, The radical right in Western Europe, a comparative analysis, Ann Arbor, University of Michigan Press.

Klandermans, B., Mayer, N., 1999,  « Militer à l’extrême droite », in Perrineau, P. (ed.), Les croisés de la société fermée, Paris, Cevipof.

Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., Frey, T., 2008, West European Politics in the Age of Globalization, Cambridge, Cambridge University Press.

Lorwin, V.R., 1971, ‘Segmented Pluralism. Ideological Cleavages and Political Cohesion in the Smaller European Democracies’, Comparative Politics, 3, 2, p. 141-75.

Lubbers, M., Gijsberts, M., Scheepers, P., 2002, ‘Extreme Right-Wing Voting in Western-Europe’, European Journal of Political Research, 41, 3, p. 345-78.

Lubbers, M., Scheepers, P., 2000, ‘Individual and contextual characteristics of the German extreme right-wing vote in the 1990’s: A test of complementary theories’, European Journal of Political Research, 38, 1, p. 63-94.

Lubbers, M., Scheepers, P., Billiet ,J., 2000, ‘Multilevel modeling of Vlaams blok voting: Individual and contextual characteristics of the Vlaams blok vote’, Acta Politica, 35, 4, p. 363-98.

Marissal, P., Medina Lockhart, P., Vandermotten, C., Van Hamme, G., 2007, Atlas de Belgique, Géographie politique, Bruxelles, SPFC.

Mayer, N., 1987,  « De Passy à Barbès. Deux visages du vote Le Pen à Paris », Revue française de science politique, 37, 6, p. 891-906.

Mayer, N., Perrineau, P., 1992, ‘Why do they vote for Le Pen?’, European Journal of Political Research, 22, 1, p. 55-81.

Mayer, N., 2002, Ces Français qui votent Le Pen. Paris, Flammarion.

Mudde, C., 2007, Populist Radical Right Parties in Europe. Cambridge, Cambridge University Press.

Mutz, D., 2002, ‘Cross-Cutting Social Networks: Testing Democratic Theory in Practice’. American Political Science Review, 96, 1, p. 111-126.

Newton, K., Delhey, J., 2003, ‘Who Trusts? The Origins of Social Trust in Seven Nations’. European Societies, 5, 2, p. 1-45.

Newton, K., 2007, ‘Social and Political Trust’. in Dalton, R.D., Klingemann, H.D. (eds.), The Oxford Handbooks of Political Sciences, Oxford, Oxford University Press, p. 241-272.

Norris, P., 2005, Radical Right: Voters and Parties in the Regulated Market, Cambridge, Cambridge University Press.

Pakulski, J., 2005, ‘Foundations of a post-class analysis’, in Wright, E.O. (ed), Approaches to class analysis, Cambridge, Cambridge University Press, p. 152-79.

ISPO/PIOP. 1991 – 1995 – 1999 – 2003 General Election Study – Belgium. Codebook and Questionnaire. K.U.Leuven/U.C. Louvain”.

Poznyak, D., Abst, K., Swyngedouw, M., 2011, ‘The Dynamics of the Extreme Right Support: a Growth Curve Model of the Populist Vote in Flanders-Belgium in 1987-2007’, Electoral Studies, 30, 4, p. 672-88.

Putnam, R., 1993, Making democracy work. Civic traditions in modern Italy, Princeton, Princeton University Press.

Putnam, R., 2000, Bowling Alone. The Collapse and Revival of American Community, New York, Simon & Schuster.

Putnam, R., Pharr, S.J., Dalton, R.J. (2000), ‘Introduction: What’s Troubling the Trilateral Democracies?’, in Pharr, S.J., Putnam, R. (eds.), Disaffected Democracies: What’s Troubling the Trilateral Democracies?, Princeton, Princeton University Press, p. 3-27.

Rink, N., Phalet, K., Swyngedouw, M. (2009), ‘The Effects of Immigrant Population Size, unemployment, and Individual Characteristics on Voting for the Vlaams Blok in Flanders 1991–1999’, European Sociological Review, 25, 4, p. 411-24.

Rossteutscher, S., 2008, ‘Social Capital and Civic Engagement: a Comparative Perspective’, in Castiglione, D. van Deth; J.M., Wolleb, G. (eds.), The Handbook of Social Capital, Oxford, Oxford University Press, p. 208-240.

Rydgren, J., 2008, ‘Immigration sceptics, xenophobes or racists? Radical right-wing voting in six West European countries’, European Journal of Political Research, 47, 6, p. 737-765.

Rydgren, J., 2009, ‘Social Isolation? Social Capital and Radical Right-wing Voting in Western Europe’, Journal of Civil Society, 5, 2, p. 129–150.

Schain, M., Zolberg, A., Hossay, P. (eds.), 2002, Shadows Over Europe: The Development and Impact of the Extreme Right in Western Europe, New York, Palgrave Macmillan.

Stolle, D., 1998, ‘Bowling Alone, Bowling Together: Group Characteristics, Membership and Social Capital’, Political Psychology, 19, 3, p. 497-526.

Swyngedouw, M., 1998, ’The extreme right in Belgium - of a non-existing Front National and an omnipresent Vlaams Blok’, in Betz, H., Immerfall, S. (eds.), New party politics of the right. Neo-populist parties and movements in established democracies. USA, St. Martins Press, p. 59-75.

van der Brug, W., Fennema, M., Tillie, J., 2005, ’Why some anti-immigrant parties succeed and others fail? A Two-step model of aggregate electoral support’, Comparative Political Studies, 38, 5, p. 537-573.

van der Brug, W., Fennema, M., 2007, ’Causes of voting for the radical right’, International Journal of Public Opinion Research, 19, 4, p. 474-87.

van Deth, J.M., 2008, ‘Measuring Social Capital’, in D; Castiglione, van Deth, J.M., Wolleb, G. (eds.), The Handbook of Social Capital, Oxford, Oxford University Press, p. 151-176.

Van Hamme, G., 2008, « La géographie de l’extrême-droite peut-elle s’expliquer par l’ « ethnocentrisme » des territoires ? Une analyse multiscalaire en Europe occidentale », Espace, Populations, Sociétés, 2008, 3, p. 441-452.

Verba, S., Nie, N.H, et al., 1978, Participation and political equality: a seven-nation comparison. Cambridge, Cambridge University Press.

Haut de page


1 Mayer, 1987; Betz and Immerfall, 1998; Kitschelt and Mc Gann, 1997; Chapin, 1997; Eatwell, 2003; Eatwell and Mudde, 2004; Schain et al., 2002; Norris, 2005.

2 Arzheimer, 2009; Betz, 1994; Givens, 2002; Golder 2003; Rink et al., 2009, Jackman and Volpert, 1996.

3 Ignazi, 1999; Lubbers and Scheepers, 2000; Billiet and De Witte 1995; Lubbers et al., 2002.

4 Lorwin, 1971; Marissal et al., 2007; Authors, 2011.

5 Pôle interuniversitaire Opinion Publique et Politique on the French-speaking side and Institute of Social and Political Opinion Research on the Flemish side.

6 The Canton groups one to several municipalities and is the lower level at which results can be obtained for general elections.

7 The five questions used to measure individual social trust are: (1) I really don’t know anymore who or what to trust. (2) These days, things are so complicated that I don’t know what to do. (3) Most people disappoint you once you get to know them better. (4) Nowadays, I can’t understand what is going on. (5) Nowadays, one can still trust most people. For each item respondents can answer on a scale from 1 (completely agree) to 5 (completely disagree). The factor analysis on these five dimensions gave an eigenvalue above 2, for the first factor, and around 0.15, for the second one. Combined with homogenous loadings associated to the first factor, we decided to keep this factor only.

Haut de page

Table des illustrations

Titre Figure 1: The influence of the different dimensions of social capital on attitudes toward immigrants and on extreme right voting
Fichier image/png, 113k
Titre Table 1 : Logit regression for extreme right voting (simplified table – full table in appendix)
Crédits Source des données : PIOP-ISPO, 1999, 2003
Fichier image/png, 117k
Titre Table 2: Linear regression for intolerance toward immigrants and immigration (simplified table – full table in appendix)
Crédits Source des données : PIOP-ISPO, 1999, 2003
Fichier image/png, 120k
Haut de page

Pour citer cet article

Référence électronique

Quentin David, Jean-Benoit Pilet et Gilles Van Hamme, « Does Associational Membership Affect Extreme Right Voting and Anti-Immigrant Attitudes? An Individual and Contextual Analysis in Belgium »L’Espace Politique [En ligne], 23 | 2014-2, mis en ligne le 03 juillet 2014, consulté le 27 septembre 2023. URL : ; DOI :

Haut de page


Quentin David

Maître de conférence
EQUIPPE, Université de Lille 1

Jean-Benoit Pilet

CEVIPOL, Université Libre de Bruxelles (ULB)

Gilles Van Hamme

IGEAT, Université Libre de Bruxelles (ULB)

Articles du même auteur

Haut de page

Droits d’auteur


Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

Haut de page
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search