Social Structure and Voting Choice
Summary and Keywords
The linkage between voters and political parties is to some degree based on stable social cleavages. Such cleavages express important and lasting societal divisions, allow parties and voters to establish long-term ties, and provide incumbents with clear representative and policy-making tasks against which they can be evaluated. Most research on cleavages has been based on the classic cleavages that were outlined in the Lipset-Rokkan model for social cleavages in industrial societies. These are:
(1) the center–periphery cleavage, which is anchored in geographical regions and related to different ethnic and linguistic groups as well as religious minorities; (2) the religious conflict between the Church and the State, which pitted the secular state against the historical privileges of the churches; this cleavage has more recently polarized the religious section against the secular section of the population; (3) the class conflict in the labor market, which involved owners and employers versus tenants, laborers, and workers; and (4) the conflict in the commodity market between buyers and sellers of agricultural products, or more generally, between the urban and the rural population.
Other social cleavages, such as gender, educational differences, and new divisions within the large new middle class, have been focused upon during the last decades. The new divisions within the new middle class are “horizontal” conflicts and can be conceptualized as a basic conflict between public and private employees, and as an alternative way of conceptualization, between those who work within technical, organizational, or interpersonal service environments.
Some of the cleavages have declined in importance over time, while others have increased. Some cleavages have changed character such as the class cleavage where part of the new middle class has voted for the New Left and part of the working class has voted for the New Right in the last decades. Changes in the impact and character of different cleavages have resulted in strategic reconsideration of important policies and changing location of the parties in the political space.
Comparative research on the relationship between social structure and party choice is discussed, with a focus on the structural variables that are included in the famous Lipset-Rokkan model for structural cleavages: class, religion, and urban-rural residence. In addition, gender and different versions of divisions within the service class or the new middle class are considered. These new conflict variables are the structural variables that are most frequently examined in the literature.
The cleavage concept has been much discussed since Bartolini and Mair’s (1990) important work. Bartolini and Mair argue that cleavages are more deep-seated than just having a structural anchorage. They favor a cleavage concept that has three distinct elements: social structure, values and beliefs, and political organizations. A traditional notion of social cleavage that is often implicit in much of the literature is that a cleavage basically reflects broadly based and long-standing social and economic divisions within society.
A social structure is a group of people that can be demarcated. It must furthermore have a certain permanency, or act with a certain regularity. Social structure comes near the start of what is termed the funnel of causality that is used to understand the causal mechanisms that contribute to vote choice. This contrasts with other variables that come later in the causal process, such as political issues, economic conditions (Hellwig & Marinova ORE), candidate images (McAllister in ORE), political scandals, and party identification (Dalton in ORE). Structural cleavages then express lasting societal divisions that allow parties and voters to establish long-term ties, and many other variables can be considered as intermediate variables between social structure and party choice.
In their seminal essay on the conflict structure in Western democracies, Lipset and Rokkan (1967, pp. 15–23) focused on the historical origins of the major conflicts between the political parties. They identified four central cleavages anchored in the social structure:
1. The center–periphery cleavage was anchored in geographical regions and related to different ethnic and linguistic groups as well as religious minorities. Although ethnicity is not explicitly addressed in the Lipset-Rokkan model, it can be subsumed under the center-periphery cleavage. The same applies to language conflicts.
2. The conflict between the Church and the State pitted the secular state against the historical privileges of the churches and over control of the important educational institutions. This cleavage has more specifically polarized the religious sectors of society against the secular sector. Conflicts between various religions and between various religious denominations within a country are also a central part of the structural religious conflict.
3. The conflict in the labor market involved owners and employers versus tenants, laborers, and workers. This is also labeled the class cleavage.
4. Finally, the conflict in the commodity market was between buyers and sellers of agricultural products, or more generally, between the urban and the rural population.
Research on each of these four cleavages is examined, along with gender and possible new divisions within the middle class. Included are two comparative longitudinal studies based on a set of West European countries from the early 1970s to the late 1990s. A recent cross-national survey is analyzed to compare the influence of various social cleavages in 2002–2006. The evidence includes Western and post-Communist countries. Differences between the Nordic, central western, southern, and east European countries and the islands (Ireland and United Kingdom) are examined.
Most works within this field require survey data, and such data are widely available from Western Europe and Northern America. These regions and Latin America are focused on as a possible example for less developed, democratizing party systems. There is continuing debate about whether the cleavages that Lipset and Rokkan developed are relevant beyond Western Europe, but no alternative structural cleavage model has been formulated.
To compare the location of voters for equivalent political parties, parties are grouped into party families (Mair & Mudde, 1998). Differentiations are made between the established or old party families such as Liberal, Conservative, Religious, Social Democratic, Communist, Agrarian, and Ethnic-Regional parties. The new party families are frequently associated with New Politics and are the Green, Left Socialist and Radical Rightist parties. The two former party families are frequently referred to as the New Left, while the Radical Right is often referred to as the New Right.
One important perspective in Lipset and Rokkan’s work is the persistent impact of social structure on party choice, which they called the freezing of party alignments: With few but significant exceptions, the party systems of the 1960s reflected the cleavage structure of the 1920s (Lipset & Rokkan, 1967, pp. 50–54). The freezing hypothesis reflected the strong relationship between the socio-structural variables and party choice that Lipset and Rokkan emphasized. This contributes to stability in the party system and fairly stable support for the various parties over time. Lipset and Rokkan’s freezing perspective is a perspective of stable alignment.
From around 1970s there has been considerable change in the electoral behavior of voters in Western democracies. Instead of stable alignment, researchers began using the notions dealignment and realignment (Dalton, Flanagan, & Beck, 1984).
Dealignment means that the impact of the structural cleavages has become smaller. The increased instability in the party system is caused by the fact that voters do not vote according to their location in the social structure to the same degree as previously.
Realignment implies the eclipse of old cleavages and the rise of new ones. There is first a dealignment from the old cleavages and then a new alignment related to the new cleavage structure. While Lipset and Rokkan focused on the national and the industrial revolutions, Dalton, Flanagan, and Beck (1984, pp. 455–456) indicated, for example, a third post-industrial revolution, which might create a new basis of social cleavages.1
Two kinds of new conflicts are focused on: new structural cleavages and value based conflicts, which to a larger degree than in industrial society have become important. Gender and new structural divisions within the new middle class are examples of such new structural divisions. Another type of realignment follows directly from the changes in social structure. Demographic or ecological realignment implies that changes in party support follow directly from the changes in social structure. Demographic realignment contributes also to changing political agendas and party strategies. Parties try to appeal to some of the new expanding social groups.
Region and Ethnicity
In their seminal article, Lipset and Rokkan identified the regional division as a center-periphery cleavage, associated with the National Revolution. It was a “conflict between the central nation-building culture and the increasing resistance of the ethnically, linguistically and religious subject population in the provinces and peripheries” (Lipset & Rokkan, 1967, p. 14, emphasis in the original). It was a conflict along a territorial axis during the nation-building process that found “local oppositions to encroachments of the aspiring or the dominant national elites and their bureaucracies: the typical reaction of the peripheral regions, linguistic minorities and cultural threatening populations to the pressure of the centralising, standardising and ‘rationalising’ machinery of the nation-state” (Lipset & Rokkan, 1967, p. 10).
The origin of regional conflicts is then associated with the nation-building process, and the regional conflict is essentially equivalent to the degree to which linguistic, religious, and ethnic groups are opposed along territorial lines.
In the most extensive study of the regional cleavage in Western Europe, Caramani (2004) examined the impact of the territorial cleavage from the first democratic elections to 1999 in 17 countries by using data on electoral results in general elections from territorial units. The main finding in Caramani’s work on territorial variations in electoral support is the long-term weakening of the territorial cleavage, a change consistent with hypotheses about the nationalization of party politics. The large decline in the impact of the territorial cleavage took place in the period up to World War I, and the period since the 1920s was characterized as a stable territorial configuration. The period since World War II has—in a long-term perspective—been a period of fundamental stability of territorial configurations. Caramani’s findings therefore support the Lipset and Rokkan hypothesis of “freezing of party alignments” from around the 1920s.
Caramani’s basic research interest is a long-term perspective, but in an important section, he analyzes average scores for all postwar elections (1945–1999) and for only the 1990s (Caramani, 2004, pp. 84–90). The measures based on data from the whole postwar period show that Switzerland and Belgium have the most heterogeneous party systems according to regions, followed by Finland, Spain, and Germany. The most homogeneous party system in this respect is found in Sweden, and then in Austria, Denmark, and Greece. The data based on only the elections in the 1990s show basically the same comparative patterns.
Caramani (2004, pp. 90, 92–93) also examines trends within the postwar period and finds that there is an average decline in the strength of the regional cleavage from the late 1940s to the 1960s, but then there is an increase in the 1980s and 1990s in some countries. Of the countries included in this study, this applies to Belgium, Great Britain, and Italy, and partly also to Spain. The regional realignment in these four countries has attracted considerable attention in the political science literature. The trend for the postwar period is, however, first and foremost stable alignment, but there is also evidence of further nationalization or dealignment in some countries, but also regional realignment.
The regional differences in support for various party families based on average figures from after World War II was clearly largest for the Ethnic-Regional parties, and then Religious parties, followed by Liberal, Communist, and Agrarian parties, and smallest for Conservative and Social Democratic parties.2
Race and Ethnicity
In the United States race and ethnicity has been a major cleavage that has increased over time. The black population has been much more inclined to support the Democrats than the Republicans. The same applies to the Hispanic population, although their inclination to support the Democrats has not been as strong as for African Americans. For the four presidential elections 2000–2012 44% of the white population on average supported the presidential candidates of the Democratic Party, 66% of the Hispanic population, and 94% of the black population.3
Latin America is a region with considerable ethnic diversity. Descendants of the original indigenous inhabitants coexist with people of European descent whose ancestors migrated in several waves during the last five centuries. Black and mulatto populations descending from Africans make up sizable minorities in many countries, and in some countries Asians form significant minority groups. Most available data suggest that around 10% of the population in the region is indigenous, while another 10–15% are blacks or mulattos of African descent, but these figures varies significantly from source to source (Morales, 2016, pp. 122–123).
In general, ethnic identity is not a strong determinant of party choice in Latin America. Indigenous identity seems, however, to be more relevant for voting than black and mulatto identification. There is also evidence indicating that the presence of candidates with a particular ethnic origin increases the level of ethnic voting. The role of ethnicity varies substantially across the region, and the main explanation seems to be the “politicization of the ethnic cleavage,” meaning that ethnicity has become an important issue in the political system at a particular moment (Morales, 2016).
Ethnicity plays a central role in voting behavior in Africa. In fact, the impact of other structural variables on party choice has been difficult to find in available comparative literature.4 The strength of the ethnic cleavage in sub-Saharan Africa seems to vary with the present and prior politicization of ethnicity and with one party dominance. Low degree of politicization of ethnicity and one-party dominance decreases ethnic voting (Basedau, Erdmann, Lay, & Stroh, 2011).
Another study focuses on various dimensions of ethnicity for explaining support for the incumbents versus the opposition. The most important dimensions are belonging to the same ethnic group as the incumbent president, feeling of ethnic discrimination from the state authorities, and the salience of ethnicity versus national identity. Salience of ethnicity reduces the support for the incumbents (Bratton, Bhavnani, & Chen, 2012).
The religious cleavage has two aspects. Religious denomination identifies the various religious communities of which people are members (or not a member of any religious community). Religiosity taps how religious people are—independent of their religious community; this is normally measured by frequency of attending religious services.
Many researchers have noted that there is a somewhat paradoxical situation related to the importance of the religious cleavage at least in advanced democracies. Only a small number of political issues clearly follow the religious-secular conflict line. By the same token, there are very few issues that are completely divorced from it. Despite the paucity of explicitly religious issues and the lack of religious themes in most campaigns, religious beliefs have proved to be a strong predictor of party choice in many West European democracies. The religious cleavage has therefore been characterized as a passive rather than an active force in shaping political behavior.
Perhaps the most important reason why religion continues to play an influential role for voter choice is that religious conflicts helped determine the structure of the modern party system and therefore still affect the electoral choices open to the voter. The religious cleavage5 is also important because it reflects deeply held human values, which have a great potential for influencing behavior. Although religious issues are not very prominent on the political agenda, religious values are related to a wide range of social and political beliefs: work ethics, achievement aspirations, lifestyle norms, parent-child relations, morality, social relations, attitudes toward authority, and acceptance of the state. Religion signifies a Weltanshauung that extends into the political area (Dalton, 1990, p. 86). Religious faith is strongly connected not only to party choice. The connection encompasses political ideology, issue outlook, and attitudes toward a wide range of political objects (Norris & Inglehart, 2004).
Lipset and Rokkan (1967) were impressively detailed about the development of the religious cleavage in Western Europe. The religious cleavage was first shaped by the Protestant Reformation, which created divisions between Catholics and Protestants. These divisions had political consequences because the control of the nation-building process often became intermixed with the religious cleavage. Protestants frequently found themselves allied with nationalist forces in the struggle for national autonomy. In Anglican England and the Calvinist Netherlands, the Protestant church supported national independence and became a central element of the emerging national political identity. In other nations, religious conflicts also ran deep, but these differences sidetracked the nation-building process (Dalton, 1990, p. 66).
Gradually the political systems of Europe accommodated themselves to the changes wrought by the Reformation. The French Revolution renewed religious conflicts in the 19th century. Religious forces—both Catholic and Protestant—mobilized to defend church interests against the Liberal, secular movement spawned by the events in France. Conflicts over church-state control, the legislation of mandatory state education, and disestablishment of state religion occurred across the face of Europe. In reaction to these Liberal attacks, new religious political parties formed in many West European countries. The party alignments developed at the start of the 20th century institutionalized the religious cleavage in politics, and many basic features of these party systems have endured to the present.
The religious structure in European countries has three divisions regarding the religious denominations. Protestantism is dominant in Britain and the Nordic countries, Roman Catholicism is dominant in southern Europe and several countries in central Europe, including many of the post-communist countries, while there are three countries with a predominant religious mixed population (Germany, the Netherlands, and Switzerland). Membership in denominations varies considerably and is largest in southern Europe (84% on average) and is 51–57% in the other regions. The degree of religiosity, indicated by frequency of church attendance, also varies considerably. Religiosity is highest in southern Europe and lowest in the Nordic countries and central Western countries, placing the East European countries in the middle.
Empirical research on mass behavior has underscored the continuing importance of the religious cleavage. Rose and Urwin (1969) conducted one of the first comparative analyses of the topic, examining the social basis of party support in 16 Western democracies. Their finding was that, contrary to conventional wisdom, “religious divisions, not class, are the main social basis of parties in the Western world today” (Rose & Urwin, 1969, p. 12). In a comparative study that included most West European countries, Rose (1974, pp. 16–18) compared the impact of religion, social class, and region on left-right voting on the basis of data from mainly the 1960s, and found that religion was much more important in all the Catholic and religiously mixed countries. Only in Britain and the Scandinavian countries was social class the most important predictor for left-right party choice.
In Western Europe the main changes over time in the strength of the religious cleavage are somewhat mixed. Although there has been a considerable change in the distribution on the religious variables in the direction of a more secular mass public, the correlation with party choice has shown a surprising persistence at a high level. For example, Dalton (2008, p. 159) compares the impact of religion on voting with the impact of social class in a comparative longitudinal study and concludes that “the trends for religious voting do not show the sharp drop-off found for class voting.”
Another longitudinal study shows, however, considerable decline in the impact of religion on party choice in the countries where the religious cleavage was most pronounced in the 1970s: Belgium, France, Italy, and the Netherlands.6 These declines produced a convergence in the impact of the religious variables on party choice across Europe. There are, however, also signs of a considerable persistence in the impact of religion in some other countries (Knutsen, 2004, chap. 2, 3, pp. 234–236).
According to the 24-nation study cited throughout, the religious cleavages are strongest in central western Europe and the Nordic countries and then in southern and eastern Europe and smallest on the islands.7
Contrary to some researchers who frequently assume that religion is not sufficiently widespread as a cleavage, comparative research shows that this cleavage is present in all Western democracies, but of different strength.
Table 1: Party choice and religious variables.
A. Religious denomination
B. Church attendance
Data Source: Cumulative file of European Social Surveys 1, 2, and 3.
The Christian parties are clearly closest to the religious pole on the various aspects of the religious cleavage, and then the Conservative, Ethnic-Regional, and Agrarian parties, while all the leftist party families (including the Greens) are closer to the secular pole. The Radical Rightist parties are close to the center on both of the aspects of the religious cleavage.
In the United States the Protestant population has been dominant, and there have been a large Catholic minority and a small Jewish population. Support for the Democrats has been largest from the Jewish population and from those who are not affiliated with any denomination. Catholics have leaned more toward the Democrats than have Protestants. The impact of religious denomination on party choice has fluctuated significantly over time; it increased from the 1980s to the 1990s and has declined considerably from 2000 to 2012.
Church attendance also has a moderate and fluctuating impact on voting in the United States. The Republican Party gains its strongest support among regular churchgoers, while the Democrats garner more support among those who never attend church. The impact of church attendance on party choice was low from the 1950s to the 1980s but increased significantly in the 1990s and declined somewhat in the 2000s (2000–2012), although being at a higher level than before the 1990s.
Catholicism has been dominant in Latin America, but the region cannot be identified as Catholic by default. Catholicism is losing ground to other Christian denominations and to secularism. AmericasBarometer surveys from 2012 indicated that 70% still considered themselves as Catholics, 19% identified with non-Catholic denominations, and 9% clamed no religious affiliation. Across Latin America, 8% of citizens are classified as mainstream or historical Protestants and 10% as evangelical or Pentecostal.
The two aspects of the religious cleavage are manifested in the following way in Latin America: Religious voters vote for rightist parties while secular voters vote for leftist parties. The differences in voting choice between Catholics and Protestants manifest themselves first and foremost in “identity politics.” Each group prefers candidates of its own faith (Boas & Smith, 2016).
Social class represents the classic structural cleavage in industrial society. Lipset and Rokkan linked the class cleavage first and foremost to divisions in the labor market between owners and employers on the one side and tenants, laborers, and workers on the other. It sprang out of the Industrial Revolution and proved much more uniformly divisive than other major cleavages (Lipset & Rokkan, 1967, pp. 14, 21, 35). The rising masses of workers resented their working conditions and the insecurity of their contracts. The result was the formation of a variety of labor unions and the development of nationwide socialist parties. The fact that the labor market cleavage was so uniformly divisive in a comparative setting implied that it tended to bring the party systems closer to each other in their basic structure. While conflicts and compromises along the other cleavages, especially the center-periphery and the state-church cleavage lines, tended to generate national developments of the party systems in divergent directions, the owner-worker cleavages moved the party system in the opposite direction. “the owner-worker cleavage tended to bring the party system closer to each other in their basic structure” (Lipset & Rokkan, 1967, p. 35). The Russian Revolution in 1917, however, brought about a more divisive structure among parties that articulated the interests of the workers. In some countries significant new Communist parties created a split among the socialist parties, while the Communists became an insignificant force in other countries (Bartolini, 2000, pp. 86–120, chap. 9; Lipset & Rokkan, 1967, pp. 46–50).
Different Types of Class Voting Studies
Recent research on class voting differentiates between three types of class voting:
1. “Traditional (Left-Right) class voting” examines the Left-Right division of parties and only two social classes (the manual-nonmanual division).
2. “Overall Left-Right class voting” examines the Left-Right voting across all social classes.
3. “Total class voting” considers class differences (based on a detailed class schema) in voting between all the parties without an implied Left-Right structure.
The party choice variable has (nearly) always been dichotomized into Left-Right in class voting research. This division can be questioned in advanced industrial societies. There is some evidence that social cleavages, and the class cleavage in particular, cut across the Left-Right division of parties. The New Left parties gain stronger support from the higher educated strata and the new middle class, while the New Right parties gain strongest support from the less educated and the workers. Therefore, newer research on class voting should consider all parties as separate categories. This applies increasingly to all studies of the relationship between social structure and party choice (see Knutsen, 2004, 2006).
Jansen (2011, pp. 24–28) has systemized research of recent trends in the study of class voting. He emphasizes that recent research focuses upon explaining the decline on class voting by socio-structural and political factors (such as changes in the positions of the political parties on important policy dimensions), and that multinomial and conditional logistic regressions increasingly are used as statistical methods. There is considerable scholarly debate about which statistical methods should be used to examine class voting and also other social cleavages.8
The first studies of class voting (traditional Left-Right class voting mentioned previously) used a traditional two-class schema between manual workers and all other classes (Nieuwbeerta, 1995). More recent class voting studies use more detailed class schemas. Prominent among these schemas are the so-called Erikson-Goldthorpe (hereafter EG) class schema originally developed in connection with social mobility studies (Goldthorpe, 1980; Erikson & Goldthorpe, 1992). The third generation of class voting studies typically used this class schema.
A main division exists between the predominantly salaried professional—higher technical, administrative, and managerial—positions, and the predominantly wage-earning manual occupations. The former are positions with a service relationship, and thus constitute the basis of the “service class” or the “salariat” of modern industrial society. The latter, where the labor contract usually prevails, constitute the basis of the working class. The service class comprises administrators and managers, employed professionals, higher-grade technicians, and supervisors of nonmanual workers. It is divided into a higher and a lower level according to administrative responsibility and educational training.
Routine nonmanual employees in the EG schema do not belong to the new middle class or the service class. This includes routine nonmanual positions, usually involving clerical, sales, or personal-service tasks, which exist on the fringes of professional, administrative, and managerial bureaucracies (Goldthorpe, 1980, p. 40).
The working-class group is grouped into skilled and unskilled and are manual wage-earners in all branches of industry. Supervisors of manual workers (foremen) and lower-grade technicians are grouped as skilled workers.
In addition to these classes among the employees, there are separate classes for the petty bourgeoisie (self-employed and professionals) and for the farmers and other self-employed in the primary sector.
Comparative Levels of Class Voting and Trends in Class Voting
Nieuwbeerta’s (1995) pioneering work is the most extensive cross-national analysis of class voting. He studied class voting in 20 countries over time and found a clear decline in class voting based on the two first types of class voting outlined previously. Class voting was largest in the Nordic countries, then in the continental and southwest European countries. Class voting varies a lot among the Anglo-American countries: Class voting was high in Britain and Australia and very low in Canada, Ireland, and the United States. In fact class voting was comparatively lowest of all 20 countries in the latter three countries. Nieuwbeerta also found that various statistical measures produced very similar results.
Knutsen’s 24-nation study showed that all types of class voting are largest in the Nordic countries while the comparative strength of the various types of class voting varies between the other regions. Class voting is generally smallest in eastern Europe and fairly similar in the other regions.
Table 2: Different types of class voting.
A. Traditional class voting
B. Total Left-Right class voting
C. Total class voting
The Alford index (PDI)
Notes: PDI – Percentage differences.
Kappa index is based on the standard deviations of log-odds ratios.
Data Source: Cumulative file of European Social Surveys 1, 2, and 3.
Similarly, Knutsen found a decline in class voting across eight west European democracies (2006). The average decline was 47% of the original strength in the late 1970s based on the traditional contrast of manual-nonmanual employment influencing Left-Right party support, and 36% for Left-Right class voting with several social classes. The decline in class voting was largest in Denmark and the Netherlands, then Britain, France, and Italy, and smallest in Belgium, Germany, and Ireland. There are also strong correlations between the various measures of class voting.
Both Nieuwbeerta’s and Knutsen’s longitudinal study found the decline in class voting is greatest in the countries where class voting has been largest, in particular the Nordic countries. Thus, class voting is converging to a fairly low level across Western democracies.
Total left-right class voting according to the EG-classes shows that on average the two working class categories are most likely to support the leftist parties, followed by the routine nonmanuals and then the lower and higher level of the service class. Farmers and petty bourgeoisie are the classes decisively least likely to support the leftist parties.
The Old Left (Social Democrats and Communist parties) gains strongest support from the working class, while the New Left (Left Socialist and Greens) gains strongest support from the service class and routine nonmanuals. The Liberal parties gain strongest support from the higher-level service class and the petty bourgeoisie, while the Conservative parties gain strongest support from the farmers and the petty bourgeoisie. The Christian Democrats have a very even class base but nevertheless gain strongest support from the farmers. The same pattern characterizes the Agrarian party family, but the difference in support is nevertheless much larger because it gains very limited support from classes other than the farmers. The class differences in support for the Radical Right are not large, but these parties gain strongest support from the workers and the petty bourgeoisie.
Class voting in the United States is very low in a comparative perspective. It has, however, been stable over time when the EG classes are used, and follows the same pattern as indicated for Europe. The farmers and the petty bourgeoisie are relatively more likely to support the Democrats compared to support for leftist parties in Europe. The petty bourgeoisie has, however, turned toward the Republican Party in more recent elections.
Class voting has traditionally been much weaker in nearly all of the Latin American countries than in Western Europe. However, this has changed since the mid-1990s. Class voting has increased in several Latin American countries, and there is huge variation in class voting in the region. Mainwaring, Torcal, and Somma (2016) find that the main reason for why class voting has increased in some countries, but not in others, is the mobilization of viable left-of center presidential candidates and parties and the leftist policies implemented by left-of center governments. In countries with leftist candidates and implementation of leftist policies class voting has increased, but not in countries without such candidates and policies. It is overall Left-Right class voting that has increased in several countries; however, it is not the skilled workers but the self-employed and the unskilled workers who have contributed to this change.
According to Lipset and Rokkan, another cleavage is the conflict in the commodity market between peasants and others employed in the primary sector and those who want to buy the products from the primary sector, particularly the urban population. The peasants want to sell their wares at the best possible prices and to buy what they need from the industrial and urban producers at low costs, while the urban population often had opposite economic interests (Lipset & Rokkan, 1967, pp. 20–21).
This is then essentially an urban-rural conflict. Such conflicts did not invariably prove party-forming. They could be dealt with within broad party fronts or could be channeled through interest organizations into narrower arenas of functional representation and bargaining. In many countries the religious interests of the rural population were more influential than the strictly economic ones, and the economic interest articulation took place within the Christian parties. Distinct Agrarian parties emerged only in some countries where strong cultural opposition had deepened and embittered the strictly economic conflicts (Lipset & Rokkan, 1967, pp. 44–46).
There are generally two aspects of the commodity market conflict: (a) how the class of farmers (and other self-employed in the primary sector) vote compared to other social classes and (b) the differences in voting behavior between the urban and the rural population. Both of these factors can be a continuing source of division in contemporary party systems.
The population in rural areas is generally more conservative and religious than the urban population, and vote for Christian and also Conservative parties. There is often a difference in character as well as size between rural areas and large cities. In rural areas small communities have centuries of preindustrial history and have been least affected by great population changes consequent to industrialization. “Traditional” values have a greater chance of survival in the countryside, even though some people may work in a modern environment (Knutsen, 2004, pp. 132–133).
As to the urban-rural contrasts in voting behavior, the conclusions of Knutsen’s eight-nation study were that the urban-rural cleavage is still of considerable importance in west European countries, but it is declining somewhat. The 24-nation study showed that the correlation between party choice and urban-rural residence is particularly large in Finland and then in Norway and Poland. The strength of the correlation is on average largest in the Nordic countries, weakest in southern Europe, and fairly similar in the other regions.
The patterns for the party families are fairly consistent across countries. All the leftist party families and also the Liberals and Conservatives get strongest support from the urban population. The Left Socialist and the Greens have a very urban electorate, while the Agrarian, and then the Christian, Ethnic-Regional, and Radical Rightist party families have the most rural electorate. The Radical Right has somewhat greater support among the rural than among the urban electorate. As to the support from the farmer class, the Christian parties gain their strongest support from farmers (50–70%) and in the Nordic countries the (Agrarian) Center parties are clear class parties. In the east European countries there is a strong and consistent tendency for the Liberal parties to gain stronger support from the urban population.
In the United States the Democrats have their strongholds in the larger cities while the Republicans have strongest support in the suburban areas and in the smaller towns and countryside. These differences are fairly stable over time, although there is considerable trendless fluctuation.
Until the end of the 1960s, women tended to have more conservative and traditional political orientations than men. Thus women were more inclined to vote for Religious and Conservative parties and less inclined to vote for socialist parties. According to the traditional gender gap, women were expected to be more conservative or center-right than men (Rose, 1974).
Traditional women’s values emphasizing “private” orientations associated with religion and family responsibilities were identified as the basis for these differences. Moreover, women have been less integrated in trade unions and working-class culture, and have thus been less solidaristic and collectively oriented than men. The most important explanation was their higher degree of religiosity, because the major differences were found with regard to support for the Christian parties (Rose, 1974).
Beginning in the 1980s or 1990s, however, women in many Western countries have changed from being more conservative than men to being more radical. The term modern gender gap has been used to characterize these new gender-based value differences and differences in voting patterns between women and men in many Western democracies (Norris, 1999, p. 150; Kittilson in ORE).
Researchers have advanced various explanations for how and why gender differences occur and what they imply. The two main explanations are those emphasizing structural accounts or economic interests, and those emphasizing cultural and value differences between women and men.
Explanations that emphasize structural factors and interests see changes in the gender-based division of work associated with the labor market and the family as the most important reasons for changes in women’s and men’s interests. The transition from an economy based on one breadwinner to one based on the two-income family has meant than women have increasingly become independent economic actors. Paid employment directly exposes women to gender inequalities that they are less likely to experience as homemakers, while also providing them with a means of economic independence that may shape their political behavior. Women are also more dependent on the public sector and the welfare state for employment than men, and they tend to depend more on social welfare to support and subsidize their families.
Cultural or value-based explanations have also been applied to observed differences in political attitudes. This approach emphasizes that there are extensive and deep-rooted value differences, or differences with regard to central political issues, between women and men. This applies both to traditional economic Left-Right issues and moral and political questions like peace, welfare, the environment, and social care.
Another explanation focuses on generational changes among women. The traditional gender gap is expected to be found among the older cohort, while the postwar cohorts, who gained their formative experiences in the 1960s and 1970s, have been more strongly influenced by the transformation of sex roles, the women’s movement, and changes in political attitudes and values (Norris, 1999, p. 154).
Inglehart and Norris (2000, 2003, chap. 4) formulated a developmental theory of the gender gap or of gender realignment. According to this theory there will be (a) systematic differences in the gender gap between societies based on their level of political and economic development, (b) within societies there will differences between generations, and (c) the explanations of the gender differences will be found in structural and cultural factors.
Comparative survey data support the various elements of this theory. The traditional gender gap in voting diminishes as societies reach an advanced industrial phase, and in post-industrial societies women are more likely to vote for the Left. In advanced industrial societies, cultural factors seem to explain the gender gap better than structural factors (Inglehart & Norris 2000, pp. 453–457).
According Knutsen’s my longitudinal study of eight Western democracies, in the early 1970s men were more likely to support the leftist parties in all of the eight countries although the differences were small in some of the countries. The traditional gender gap was particularly strong in Italy, Germany, and Ireland. In the late 1990s there is a modern gender gap in five of the countries and in 2008 the modern gender gap is found in all countries apart from Ireland. The changes from a traditional to a modern gender gap have been particularly large in Italy and Germany (Knutsen, 2013).
Knutsen’s 24-nation study finds the modern gender gap is largest in the Nordic countries and then in the central western countries, while there are on average no significant gender gap in the other regions. The gender differences fitting a modern gender gap are modest: the average percentage points for the Nordic and central western Europe are 7 and 5 percentage points with women are more likely to support the leftist parties. The left–right gender gap in eastern and southern Europe is about zero. The average percentage for all 24 countries is 4 percentage points. A major finding then is that the modern gender gap is found in countries with Protestant influence, while a more traditional gender gap is found in countries with a large portion of Roman Catholics (Knutsen, 2012).
Women are inclined to support the Green and the Social Democratic parties to a larger extent than men, while men tend to support the Agrarian, Communist, and, in particular, the Radical Rightist parties. Somewhat surprisingly, there are no significant gender differences between the Christian and Conservative parties. Gender differences are largest for support for the Greens and the Radical Right.
The United States is a case for a transformation from a traditional to a modern gender gap. In the 1950s men supported the Democratic Party to a larger extent than women (4 percentage points on average). There is a gradual change to a modern gender gap, which implies that women are more inclined to support the Democrats. This gender gap was on average 4, 7, and 10 percentage points in the 1970–1990s, and 7% in 2000–2012.
Women in Latin America have traditionally voted for Conservative and confessional parties to a higher degree than men. More recent studies show that this is still the case, but the gender gaps are moderate. A comparative study of 18 Latin American countries found that motherhood and lack of workforce participation contributed to a larger traditional gender gap, as did a polarized party system. A large share of professional workers at the societal (country) level decreased the traditional gender gap. Women also tend to vote for female candidates across the ideological spectrum (Morgan, 2016).
Sector Employment and Work Logics
Another potential new cleavage is the contrast between the public versus the private sector. This functional cleavage can be explained on the basis of the different economic interests of employees in the two sectors, as well as the cognitive conditions connected with different educational backgrounds and occupational experiences.
The extent to which one’s economic interests are directly linked to political decisions is perhaps the most noticeable difference between working in the public and working in the private sector. The public employee has clear self-interests connected with roomy public budgets, a well-developed welfare state, and market restrictions. A large public sector means more jobs and greater possibilities for a better career and higher economic rewards. By contrast, the interests of many private-sector employees are connected with the market and with capturing parts of it for the organizations or firms where they work (Knutsen, 2005, p. 594).
Many public sector employees are confronted with social problems in their work, something that can be expected to create attitudes that favor social reforms and more public initiatives. These aspects of work can also be expected to create an awareness of the weaknesses of market-oriented solutions. Furthermore, the ethics of welfare occupations, coupled with professional norms about what is best for the clients, often lead to a focus on the lack of resources and to demands for more vigorous public initiatives, resulting in a stronger public sector ideology (Knutsen, 2005, pp. 594–595).
Large numbers of middle-class public employees are recruited from educational institutions and educational backgrounds characterized by the values of public sector ideology (education for the professions, social science education) (Knutsen, 2005, p. 595; Tepe, 2012).
A macro-approach to political conflicts in advanced industrial democracies related to the size and structure of welfare states and differences in employment structures is to use the model of “trilemma of the service economy,” put forward by Iversen and Wren (1998). The neoliberal model emphasizes budgetary restraint and high private sector employment, accepting large wage inequality wherein the wages are low in the (expanding) private service sector (Iversen & Wren, 1998, pp. 514–515).
The second model also emphasizes budgetary restraints but places greater weight on equality. The model has its origins in corporatist and Christian democratic thought. High employment levels are a relatively low priority in this model because “women are viewed as the guardian of the traditional family and hence are encouraged to stay at home and care for children and spouse” (Iversen & Wren, 1998, p. 515).
The third model gives priority to earning equality and high employment performance at the expense of budgetary restraint. This is the social democratic model, which combines a strong egalitarian ethos with a work ethic that emphasizes employment as the root of collective identity and pride.
Each model involves its own costs and political conflicts, according to Iversen and Wren. Suffice it to mention here that the liberal model creates additional income inequality and perpetuates class division; the Christian democratic model creates high unemployment and breeds labor market exclusion and resentment among outsider classes; and the social democratic model generates its own political conflicts about the size of the public budgets, and structural conflicts between public and private employees (Iversen & Wren, 1998, pp. 517–518, 539–540, 544–545).
1. Public employees vote for the parties of the Left, and primarily Left Socialist and Green parties.
2. The impact of sector employment is largest within the service class. It is within the service class that the arguments about the awareness of social problems are most clearly coupled with work in the public sector, and the arguments about professional ethics and educational background are also most relevant for more highly educated personnel, that is, the service class.
3. The impact of sector is largest in social democratic welfare regimes (the Nordic countries) due to the higher level of political conflict coupled with the welfare services in these countries.
4. Those who work in the health, social, and educational sectors vote to the largest degree for leftist parties compared to those who work within other parts of the public sector. This difference is greatest in the Nordic countries. It is, then, not those working in public administration, which is the most left-wing.
A different approach emphasizes horizontal divisions based on the distinction between technical, organizational-administrative, and interpersonal “work logics.” The main idea is that each of these groups has different values, identifications, and interests based on educational background, work experience, and roles in the workplace (Oesch, 2006a, 2006b). The technical work logic is based on technical work and expertise. The organizational work logic is based on administration and bureaucratic structures that emphasize the leadership and coordination of the organization for those at the top. The interpersonal work logic is based on face-to-face interaction with clients in service organizations. The work takes place largely outside the clear command lines, and orientations to clients (as, for example, patients, children, and students) are central.
The three horizontal logics connects to EG-class schema’s hierarchical levels (workers, routine service workers, and the service class) so that one gets a comprehensive class schema with both horizontal and vertical-hierarchical components. For example, the service class in the EG-class schema is divided into three classes, namely technical professions, management-bureaucratic professions, and sociocultural professions.
An analysis from four West European countries based on this class schema finds that the basic conflict between the New Left (Green and Left Socialist parties) and Radical Right parties are based in the sociocultural professions and workers who work within the technical work logic (Oesch, 2013).
A more comprehensive analysis based on 18 Western European countries comes to similar conclusions (Knutsen & Langsæther, 2015). The Green parties are based in the sociocultural professions, with the lowest support among production worker. It is quite the opposite for the support for the Radical Right parties. For the Social Democratic and Liberal parties, the hierarchical divisions are dominant (that is, workers versus service class, regardless of work logic). In comparison, the Conservatives and the Left Socialist parties seem to be rooted in both the vertical and horizontal divisions in a rather complicated way. The Left Socialist parties in the Nordic countries, however, have a very similar class location to the Green parties with great support among the sociocultural professions and significantly less support among the production workers.
Changes in the relationships between socio-structural variables and party choice after a long period of stability that Lipset and Rokkan characterized as freezing of party alignments have been examined. Changes in the cleavage structure in relation to the concepts of stable conflict lines (stable alignment), dealignment and realignment were discussed.
Class voting in Western democracies is perhaps the clearest example of structural dealignment. Traditional and overall Left-Right class voting differences have declined over time in advanced industrial democracies. There are indications that differences in religious voting and urban-rural voting have also narrowed.
Realignment in the form of some variables that have gained greater importance for a party preference over time is most clearly expressed for gender and employment sector, and probably also for work logics. Even though sector employment is a stronger predictor for voting choice than gender, the change from a traditional to a modern gender gap is important. Gender is, however, a weak predictor of party choice even in the countries where the gender gap is largest. The increase in class voting in many Latin American countries is another evidence of realignment.
At the same time that social class has diminished in importance to voting on the major Conservative, Liberal, and Social Democratic parties, social class has become more important for support for the Green, Left Socialist, and Right Radical parties. The Green and the Left Socialist parties receive increasingly greater support from the service class, while the Right Radical parties have workers and the petit bourgeoisie as their primary class basis. These party families get clearer class basis over time, and this increase is an important explanation of why overall Left-Right class voting is declining.
Class variables show examples of realignment because an increasing share of the service class is voting for leftist parties. Analysis based on the effects of the sector and the various work logics is a fruitful way to identify the structural basis to explain especially the differences in voting within the large and diverse service class.
Demographic or ecological realignment is associated with changes in the social structure. Some parties get increased support because the social groups where they have the greatest support increase as a share of the population. Parties with the greatest support from secular citizens, the service class, urban residents, the public sector, and the sociocultural specialists receive increased support as a result of changes in social structure.
Religious cleavages seem to be more resistant to decline compared with the class conflict line. This can be explained by the fact that religious voting reflects deep-seated values. Thus, religious voting is to a large degree associated with stable alignment.
Caramani’s study shows that in a long-term perspective, the linkages between the regions and party choice have been quite stable in the postwar period (stable alignment). The highly visible examples of increased regional differences in voting are not representative of the main patterns of Western European countries.
The political parties have adapted to both the structural and value changes in affluent democracies. The structural changes in these societies are important explanations of the programmatic changes that the political parties have made.9 The competition for the votes of growing social groups, such as the population in the cities, the secular, the highly educated, and the service class, requires political parties that are adaptable. The parties must have the ability to make strategic, programmatic changes to capture the central interests, attitudes, and values in these large and often disparate social groups.
Abramson, P. R., Aldrich, J. H., Gomez, B. T., & Rohde, D. W. (2015). Change and Continuity in the 2012 and 2014 Elections. Thousand Oaks, CA: CQ PressFind this resource:
Bartolini, S. (2000). The political mobilization of the European left, 1860–1980: The class cleavage. Cambridge, U.K.: Cambridge University Press.Find this resource:
Bartolini, S., & Mair, P. (1990). Identity, competition, and electoral availability: The stabilisation of European electorates 1885–1985. Cambridge, U.K.: Cambridge University Press.Find this resource:
Basedau, M., Erdmann, G., Lay, J., & Stroh, A. (2011). Ethnicity and party preference in sub-Saharan Africa. Democratization, 18(2), 462–489.Find this resource:
Boas, T., & Smith, A. E. (2016). Religion and the Latin American voter. In R. E. Carlin, M. M. Singer, & E. J. Zechmeister (Eds.), The Latin American voter: Pursuing representation and accountability in challenging contexts (pp. 99–121). Ann Arbor, MI: University of Michigan Press.Find this resource:
Bratton, M., Bhavnani, R., & Chen, T. (2012). Voting intentions in Africa: Ethnic, economic or partisan? Commonwealth & Comparative Politics, 50(1), 27–52.Find this resource:
Caramani, D. (2004). The nationalization of politics: The formation of national electorates and party systems in Western Europe. Cambridge, U.K.: Cambridge University Press.Find this resource:
Dalton, R. (1990). Religion and party alignment. In R. Sänkiaho et al. (Eds.), People and their polities (pp. 66–88). Jyväskylä, Finland: The Finnish Political Science Association.Find this resource:
Dalton, R. (2008). Citizen politics: Public opinion and political parties in advanced Western democracies (5th ed.). Washington, DC: CQ Press.Find this resource:
Dalton, R., Flanagan, S. C., & Beck, P. A. (Eds.). (1984). Electoral change in advanced industrial democracies. Princeton, NJ: Princeton University Press.Find this resource:
Erikson, R., & Goldthorpe, J. H. (1992). The constant flux: A study of class mobility in industrial societies. Oxford: Clarendon.Find this resource:
Evans, G., & de Graaf, N. D. (Eds.). (2013). Political choice matters: Explaining the strength of class and religious cleavages in cross-national perspective. Oxford: Oxford University Press.Find this resource:
Goldthorpe, J. H. (1980). Social mobility and class structure in modern Britain. Oxford: Clarendon.Find this resource:
Gunther, R., Lobo, M. C., Belluci, P., & Lisi, M. (2016). The changing determinants of the vote. In Voting in old and new democracies (pp. 150–192). New York: Routledge.Find this resource:
Inglehart, R., & Norris, P. (2000). The developmental theory of the gender gap: Women’s and men’s voting behavior in a global perspective. International Political Science Review, 21(4), 441–463.Find this resource:
Inglehart, R., & Norris, P. (2003). Rising tide. Gender equality and cultural change around the world. Cambridge, U.K.: Cambridge University Press.Find this resource:
Iversen, T., & Wren, A. (1998). Equality, employment, and budgetary restraint. The trilemma of the service economy. World Politics, 50, 507–546.Find this resource:
Jansen, G. (2011). Social cleavages and political choices: Large-scale comparisons of social class, religion and voting behaviour in Western democracies. Nijmegen, The Netherlands: Ipskamp Drukkers BV.Find this resource:
Kitschelt, H., & Rehm, P. (2015). Party alignments: Change and continuity. In P. Beramendi, S. Häusermann, H. Kitschelt, & H. Kriesi (Eds.), The politics of advanced capitalism (pp. 179–201). Cambridge, U.K.: Cambridge University Press.Find this resource:
Knutsen, O. (2004). Social structure and party choice in Western Europe. A comparative longitudinal study. Houndsmills, U.K.: Palgrave Macmillan.Find this resource:
Knutsen, O. (2005). The impact of sector employment on party choice: A comparative study of eight West European countries. European Journal of Political Research, 44, 593–621.Find this resource:
Knutsen, O. (2006). Class voting in Western Europe: A comparative longitudinal study. Lanham, MD: Lexington.Find this resource:
Knutsen, O. (2007). The decline of social class? In R. J. Dalton & H.-D. Klingemann (Eds.), The Oxford handbook of political behavior (pp. 457–480). Oxford: Oxford University Press.Find this resource:
Knutsen, O. (2012). Party choice. In O. W. Gabriel & S. I. Keil (Eds.), Society and democracy in Europe (pp. 244–269). London: Routledge.Find this resource:
Knutsen, O. (2013). Structural determinants of party choice: The changing impact of socio-structure variables on party choice in comparative perspective. In W. C. Müller & H. M. Narud (Eds.), Party governance and party democracy (pp. 175–203). New York: Springer.Find this resource:
Knutsen, O., & Langsæther, P. E. (2015). Class voting in Western Europe—Do various class schemas make a difference? Paper presented at the 6th Conference of the European Survey Research Association (ESRA), July 13–17, 2015, in Reykjavik, Iceland.Find this resource:
Lipset, S. M., & Rokkan, S. (1967). Cleavage structure, party systems, and voter alignments: An introduction. In S. M. Lipset & S. Rokkan (Eds.), Party systems and voter alignments (pp. 1–64). New York: The Free Press.Find this resource:
Mainwaring, S., Torcal, M., & Somma, N. (2016). The left and the mobilisation of class voting in Latin America. In R. E. Carlin, M. M. Singer, & E. J. Zechmeister (Eds.), The Latin American voter: Pursuing representation and accountability in challenging contexts (pp. 69–98). Ann Arbor, MI: University of Michigan Press.Find this resource:
Mair, P., & Mudde, C. (1998). The party family and its study. Annual Review of Political Science, 1, 211–229.Find this resource:
Manza, J., & Brooks, C. (1999). Social cleavages and political change: Voting alignment and U.S. party coalitions. Oxford: Oxford University Press.Find this resource:
Morales, D. E. M. (2016). Ethnicity and electoral preferences in Latin America. In R. E. Carlin, M. M. Singer, & E. J. Zechmeister (Eds.), The Latin American voter: Pursuing representation and accountability in challenging contexts (pp. 123–142). Ann Arbor, MI: University of Michigan Press.Find this resource:
Morgan, J. (2016). Gender and the Latin American voter. In R. E. Carlin, M. M. Singer, & E. J. Zechmeister (Eds.), The Latin American voter: Pursuing representation and accountability in challenging contexts (pp. 143–167). Ann Arbor, MI: University of Michigan Press.Find this resource:
Nieuwbeerta, P. (1995). The democratic class struggle in twenty countries 1945–1990. Amsterdam: Thesis Publishers.Find this resource:
Norris, P. (1999). Gender: A gender-generation gap? In G. Evans & P. Norris (Eds.), Critical elections: British parties and voters in long-term perspective. London: SAGE.Find this resource:
Norris, P., & Inglehart, R. (2004). Sacred and secular: Religion and politics worldwide. Cambridge, U.K.: Cambridge University Press.Find this resource:
Oesch, D. (2006a). Redrawing the class map. Stratification and institutions in Britain, Germany, Sweden and Switzerland. Basingstoke, U.K.: Palgrave Macmillan.Find this resource:
Oesch, D. (2006b). Coming to grips with a changing class structure: An analysis of employment stratification in Britain, Germany, Sweden and Switzerland. International Sociology, 21, 263–288.Find this resource:
Oesch, D. (2013). The class basis of the cleavage between the New Left and the Radical Right: An analysis for Austria, Denmark, Norway and Switzerland. In J. Rydgren (Ed.), Class politics and the Radical Right (pp. 31–51). London: Routledge.Find this resource:
Rose, R. (Ed.). (1974). Electoral behavior: A comparative handbook. New York: The Free Press.Find this resource:
Rose, R., & Urwin, D. (1969). Social cohesion, political parties and strains in regimes. Comparative Political Studies, 2, 7–67.Find this resource:
Tepe, M. (2012). The public/private sector cleavage revisited: The impact of government employment on political attitudes and behaviour in West European countries. Public Administration, 90, 230–261.Find this resource:
(1.) More recent discussion of dealignment and realignment is found in Kitschelt and Rehm (2015). Kitschelt and Rehm (2015, p. 183) associate the post-industrial dealignment perspective with an occupational diversification that makes the organization of collective interests increasingly difficult, accelerated social mobility, and breakdowns of stable social networks, neighborhoods, and social “milieus.” As to perspectives related to voting, dealignment is coupled with the increased importance of voting on the basis of perceived competence of parties and politicians and issue ownership, and not so much with voting on the basis of position issues. For Kitschelt and Rehm (2015, pp. 180–182) the postindustrial realignment perspective implies that parties act strategically and realign with the evolving preference distribution in the population induced by changing occupational and sociodemographic group sizes. Given the high level sophistication of many postindustrial voters who can discriminate between the programmatic positions of the parties and the multidimensionality of the space on salient competitive issue dimensions, party systems tend to fragment through programmatic diversification. Postindustrial party systems are therefore highly fragmented if the electoral system allows this, and voters gravitate to parties with positions and appeals that are closest to the voter’s preferences in a multidimensional space. The postindustrial realignment perspective may then generate cross-nationally quite distinct party system configuration and multidimensional space.
(2.) The New Politics party families were not included in this part of Caramani’s analysis.
(3.) The analyses of American voting behavior are based on a cumulative file of American National Election Studies (ANES) 1948–2012, Manza and Brooks (1999) and Abramson, Aldrich, Gomez, and Rohde (2015).
(4.) In Gunther, Lobo, Belluci, and Lisi (2016), which report results from the Comparative National Election Project (CNEP), which covers some countries from the various continents in the world, two African countries are included, South Africa and Mozambique. The comparative analyses comprise 15 countries, and the explanatory power of sociodemographic variables without the inclusion of ethnicity on Left-Right party choice is largest in South Africa and considerably smaller in Mozambique, but larger than in 7 of the countries included in the study. The single structural variables that were significantly correlated with party choice in the two African countries differed substantially.
The reason why Africa is not included in the review for the other structural variables is because available data from the literature are from only two countries, and these data also point in different directions regarding how they influence party choice.
(5.) When the notion religious cleavage is used, it refers to both aspects mentioned previously.
(6.) These eight-nation longitudinal comparisons are based on West European countries in the cumulative file of Eurobarometer data from the early 1970s to the late 1990s. The Eurobarometers were collected semi-annually by the Commission of the European Communities. The file includes Belgium, Britain, Denmark, France, West Germany, Ireland, Italy, and the Netherlands.
(7.) The 24-nation results are based on a cumulative file of the 2002–2006 rounds of the European Social Survey. The surveys can compare the influence of the various social cleavages across European party systems, including both Western and post-Communist countries. These analyses are reported in greater details in Knutsen (2012, 2013). This European Social Survey allows comparisons across European regions. The Nordic countries are Denmark, Finland, Norway and Sweden. The western central European countries are: Belgium, France, Luxembourg, the Netherlands, West Germany, Switzerland, and Austria. “The islands” is Ireland and the United Kingdom. The southern European countries are Greece, Italy, Portugal, and Spain. The eastern European countries are Estonia, Poland, Slovenia, Slovakia, the Czech Republic, Hungary, and East Germany.
(9.) It is evident that the policy positions and strategies of the political parties are important explanations for the relationship between social structure and voting choice, but this topic is not discussed in this chapter. See Evans and de Graaf (2013) for thorough theoretical discussions and empirical analyses of party positions and class voting in particular in Western democracies.