Why do people vote by party line

Elections in Germany: principles, procedures, analyzes

The four most important explanatory approaches show the different approaches to explaining voting behavior. However, the difficulties of electoral research in appropriately depicting the complex process of voting decisions in theoretical models also become clear.

In principle, even extensive and complex data sets, considered on their own, are of no value in explaining voting behavior. They only acquire their meaning, their meaning and their explanatory power within theoretical explanatory models (Ulrich Eith / Gerd Mielke, 2016). These establish a relationship between voting behavior and corresponding upstream influencing factors, which must be both theoretically plausible and empirically verifiable. The four most important explanatory models for voting behavior are presented in more detail below and compared with one another. The first three are based on studies of American voting behavior in the 1940s and 1950s, based on famous research traditions. Their fundamental informative value has been proven time and again to this day. The fourth developed in the 1980s.

Sociological explanatory approach

a) Microsociological perspective

The service union ver.di on a warning strike. Belonging to social groups ... (& copy version-foto.de)

... and about churches or religious communities can help determine the voting decision. Closing service of the Evangelical Church Congress in June 2015 in Stuttgart. (& copy picture-alliance / dpa)


A first classic approach emphasizes the behavior-relevant importance of the social environment: voting behavior is group behavior. The study of the opinion-forming process during the American presidential election in 1940 in Erie County (Ohio) by Paul F. Lazarsfeld and his colleagues at Columbia University (Harald Schoen, 2014) set the tone for this explanatory model, which is referred to in the scientific literature as the microsociological, social-structural or group-theoretical approach ).

The financial crisis in the headlines. Economic issues also influence voting behavior. Photo: Daniel Biskup / laif


Individual voting decisions were determined less by the originally assumed influence of the mass media or election propaganda than by belonging to different social groups with fixed political norms of behavior. The interaction of the various group affiliations could be demonstrated to a high degree with the help of the characteristics of socio-economic status, religious affiliation and size of the place of residence. The more uniform the voting norms of the groups to which the individual eligible voter belonged, the lower the probability of an individually deviating voting decision. However, if the electorate was superimposed with contradicting demands for loyalty (cross-pressures), they generally reacted by reducing their political interest and delaying the decision to vote. In this case, a decision had to be made first about which group membership should now be viewed as the most important - and therefore as behavior-relevant.

Up to the most recent election analyzes, more recent studies have repeatedly shown the great influence of the social environment on the individual political opinion-forming process. Membership in a trade union, strong ties to the Catholic Church or roots in the Protestant self-employed or craftsman milieu still have a high predictive value for voting decisions - especially when several factors are coupled. The explanatory model sees the individual ideally at the center of concentric, mutually reinforcing circles of social influence and, from this perspective, illustrates in a special way a stable voting behavior that is constant over a longer period of time. However, short-term changes in voting decisions can only be explained inadequately in this way.

b) Macrosociological perspective

The basic ideas of the microsociological explanatory approach can also be transferred to the analysis of the formation and development of party systems (Harald Schoen, 2014). This macro-sociological perspective and expansion emphasizes long-term stable alliances between certain population groups and political parties, as can be observed in Western Europe since the 19th century. Despite all the electoral and institutional differences in the various countries, Christian Democratic, Socialist or Social Democratic, liberal and, more recently, green parties, each with their own electorate, can be found in almost all party systems. Seymour M. Lipset and Stein Rokkan developed a two-stage model in their study of over a dozen countries in the 1960s, with the help of which the characteristics and changes in party systems can be explained in their basic features.

Accordingly, the development of the Western European party systems is closely related to the democratization process in the 19th and 20th centuries. The different countries went through a comparable development. Viewed ideally, there were four fundamental problems to be overcome in each case: first, the relationship between center and periphery in the course of the establishment of the national state, second, the conflict between church and secular power over political and cultural supremacy in the new state, third, the interests that diverged with the beginning of industrialization of rural-agrarian and urban-craft areas and fourthly, the clashes between capital and labor. In all of these cases, political elites formed a coalition with the affected, politicized population groups in order to be able to vigorously represent their interests. These initially looser connections developed into stable party organizations and so anchored the so-called lines of conflict (cleavages) permanently in the national Western European party systems. Liberal, Christian Democratic and also socialist parties have their roots in these far-reaching disputes.



The four classic lines of conflict mentioned have modernized over time. Today, a contrast between management and workforce or market liberals and supporters of the welfare state (old: capital and labor), conservative and modern (old: church and state, center and periphery, rural-agricultural and urban-craft) as well as materialistic and post-materialistic can start out (Heiko Geiling / Michael Vester, 2007).

Lipset and Rokkan also systematized the conditions under which new groups - such as the ecological parties that emerged in the 1980s - can assert themselves in existing party systems. Your success basically depends on four factors. Firstly, there is the question of the legitimacy of the new protest and, secondly, that of the political rights and possibilities of those who support it. Both hurdles are no longer any particular obstacles for new parties, at least in Western democracies. Then, thirdly, it remains to investigate whether existing parties might be able to take up the new protest on their part, and fourthly, it is important to take into account the persistence of the established structures. With this approach, Lipset and Rokkan have developed a model that attaches great importance to the development and change of party systems: the actions of political elites - their ability to formulate political projects and to secure permanent support from the electorate for them - as well as the willingness of the population for active political participation and debate.



Election analyzes with a sociological approach emphasize the importance of politicized milieu or group structures to which the individual voters feel connected. Depending on the type of data, they mainly use the common economic and cultural categories of social structure such as occupation, income, education, denomination, age and size of residence as appropriate indicators, as well as characteristics of the regional industrial structure, individual union membership or frequency of church attendance. The empirical verification of sociological explanatory hypotheses on individual voting behavior can, however, only be carried out with individual data. Structural shifts in the relationship between the party system and social structure, on the other hand, can be uncovered in a special way with the help of aggregate data analyzes.

Individual psychological explanatory approach



The second classic explanatory approach undertakes a clear change in perspective: voting behavior is an expression of an individual psychological relationship with a party (Ulrich Eith / Gerd Mielke, 2016). In their studies of the American presidential elections in the 1950s, Angus Campbell and his colleagues at the University of Michigan at Ann Arbor developed a new, more dynamic, and far more suitable way of explaining electoral behavior in dealing with the more static sociological model. The core of this approach, known as the individual psychological "Ann Arbor" or "Michigan model", is individual party identification. This means that voters have a longer-term emotional bond with their party (Harald Schoen / Cornelia Weins, 2014).

It is acquired during political socialization through parental home, friends or membership in political groups and - once pronounced - influences the perception and evaluation of political events to a high degree. In addition to this long-term party identification, there are two short-term influencing factors: the evaluation of the candidates and attitudes to current political issues (the so-called issue orientation). The individual voting decision now results from the specific interaction of these three factors (party identification, candidate orientation, political issues), which the authors compared with a decision-maker. As a rule, the voting decision is made according to the long-term stable party identification. However, there may well be brief dissonances between the three variables in individual elections. Personnel or factual issues that are subjectively perceived as decisive can cause the selective voting decision to fail, even contrary to the long-term effective party identification (cf. graphic Determining factors in the voting decision, in the book, p. 114).

On the one hand, the individual psychological explanatory model also emphasizes the social anchoring of the political opinion-forming process, shown here as an individual party identification developed through family and social institutions. On the other hand, in comparison to the sociological model, situational moments of the voting decision are also taken into account, namely the candidate evaluation and attitudes to current political issues. By combining these different variables, the individual psychological model has the necessary prerequisites to be able to reflect the complex processes of voting on a high theoretical level. In addition, the model statements can be empirically verified. Comparable series of surveys have been available for several decades. In addition to party identification, data on the popularity of top politicians, the ranking of the most important political problems including the corresponding attribution of competencies, satisfaction with the political system and the economic situation as well as the so-called Sunday question about the election decision ("How would you decide, if next Sunday would be the federal election? "). Various theoretically and empirically demanding studies on the basis of this model were able to repeatedly demonstrate its prognostic capability and its knowledge output for understanding the interplay of short-term and long-term influences on voting behavior.

Model of rational voting behavior

Another way of looking at the voting process characterizes the third approach. In his economic theory of democracy, developed in the 1950s, Anthony Downs concentrates entirely on the analysis of individual decision-making calculations.

The personal voting decision is determined by the maximum political benefit that can be achieved (Ulrich Eith / Gerd Mielke, 2016). A "rational voter" therefore chooses the party whose policy he expects to gain the greatest advantage from. The economic concept of rationality used in Downs' study, however, never refers to the goals of the acting person, but always only to the use of the available means, i.e. to the economically effective (rational) pursuit of a self-chosen goal that is subject to one's own value judgment. A rational person therefore first arranges his alternative courses of action with regard to his given goals. He then selects the most effective alternative and always comes to the same result under the same framework conditions. A concept of rationality understood in this way is of course very different from the colloquial notions of rationality, be it connected with normative demands of an obligation of political action for the common good or decision-making by means of purely logical, intersubjectively verifiable thought processes.

In electoral research, rational voting behavior is generally equated with the orientation of voters to current political issues and issues (issue voting) (Kai Arzheimer / Annette Schmitt, 2014). The "rational voter" determines his voting decision by setting up a so-called utility differential. To this end, he compares the work of the government in the previous legislative period with the presumed outcome of the opposition had it been in power. He then decides in favor of the party that he believes is most likely to achieve his individual goals. The decision to vote is therefore ultimately dependent on current political problems and the appearance of the government and the opposition, and to a large extent also on economic indicators such as inflation rates, unemployment figures or growth rates. From this perspective, social loyalties or long-term emotional party affiliations only play a subordinate role.

Meeting point for artists in Berlin's trendy Prenzlauer Berg district. Those who meet in the art scene are looking for spaces to identify with a certain lifestyle, which can also be expressed in the decision for a party. (& copy picture-alliance / dpa)


In everyday research, the analyzes of the rationality of voting behavior fall back on the same polls and data sets that are also used in individual psychological studies. In the rational model, the attitudes asked about current political problems are of central importance for declaring the election; in the individual psychological approach, however, they are viewed as short-term influencing factors that have already been filtered through party identification. This interest in knowledge resulting from both theoretical approaches has led to a large number of detailed studies on the effectiveness of short-term, above all economic, factual questions in voting. However, the rational explanatory model also has its limits. Within his narrow model assumptions, the question of why someone even takes part in an election cannot be conclusively answered. The effect of one's own voice, i.e. the benefits of participation, is negligible compared to the costs and efforts involved. The explanation of the voting decision in favor of small parties that have no chance of participating in government also remains unsatisfactory.

Model of social milieus

i

Change in values

Change in values ​​is generally understood in the social sciences as a change in fundamental social value orientation; thus it covers an important part of the cultural change. In general, we are interested in the connection between cultural and social change [...]. Of the new theories of changing values, the theory of Ronald Inglehart has achieved the greatest importance in the field of election research.

Inglehart assumes that the value priorities of a person are essentially shaped by the socio-economic situation during the socialization phase, whereby in the sense of the scarcity thesis, those values ​​that are relatively scarce always have the highest priority. With the transition to a post-industrial society, physiological and security needs are satisfied at a relatively high level, so that so-called post-materialistic values ​​are gaining in importance. From the socialization thesis it can be deduced that the increase in post-materialists is primarily a cohort phenomenon because economic prosperity only affects value priorities during the phase of primary socialization. The increase in post-materialistic value orientation is therefore relatively quiet in the way of generation succession. The scarcity thesis establishes a connection between economic prosperity and that originally proposed by Abraham H.Maslow postulated a hierarchy of needs, according to which basic human needs (food, sexuality, physical integrity) must first be satisfied before humans can turn to so-called higher needs.

Alternative explanations of the change in values ​​are on the one hand general theories of modernization, which postulate a secularization of society as a necessary consequence of modernization, and on the other hand theories that emphasize more specific institutional socialization conditions, for example in the educational system or in the family. Helmut Klages assumes a change from the values ​​of duty and acceptance (obedience, performance, order, etc.) to values ​​of self-development, both in terms of hedonism and idealistic social criticism.

Theories of cyclical change in values, according to which progressive and conservative phases alternate with a certain periodicity, must be distinguished from linear modernization (progress) theories.

From: Pappi, Franz Urban: Wertewandel, in: Dieter Nohlen / Florian Grotz (eds.): Kleines Lexikon der Politik, Bonn 2008, p. 637 f.


In the 1980s, a new approach to explaining voting behavior was developed: the division of voters into socio-moral milieus. In 1984 the Sinus Institute presented a study that claimed to describe and predict changes in behavior and attitudes in the German population against the background of a change in values ​​that was taking place (Ulrich Eith / Gerd Mielke, 2016). The study, which originally looked at consumer behavior, was updated in 1992 on behalf of the SPD by the Sinus Group and the Polis Institute and transferred to voting behavior.

The term "social milieus" is distinguished from conventional definitions in two ways: Social milieus do not necessarily correspond to economically defined social classes. They are also not milieus in the traditional sense that are characterized by joint action or joint communication, such as the working class milieu.

Rather, society is divided into social milieus through the identification of fundamental value orientations that determine the prevailing lifestyles and strategies. Attitudes to work, family or consumer behavior are also included as well as wishes, fears or future expectations.



The Sinus Group distinguishes ten social milieus, which are characterized by common basic values ​​and ways of life. The developers see the advantage of the milieu concept in the fact that it enables political parties to act in a more target-group-oriented manner and in this way to tap new potential voters.

The Sinus researchers recommended the SPD to try harder to create the "new employee milieu". According to the result of the research group, it is a young milieu, both demographically and in terms of research history. The average age is well below 50 years. Its representatives are characterized by intermediate educational qualifications and service-oriented or technological occupational fields.

You strive for a self-determined and materially secure life. In this milieu, politics is not perceived as a question of right or wrong ideology, but rather as pragmatic problem-solving management. Among the members of this milieu, the willingness to vote in a social democratic manner is particularly evident due to "factual loyalties".

Basically, the SPD has the problem that its supporters are represented in numerous and heterogeneous milieus, which makes a programmatic target group orientation difficult. The respective electoral focus of the CDU / CSU, FDP and the Greens, however, concentrated in a few milieus.

The CDU, for example, recruits a significant proportion of its electorate from the "petty-bourgeois" and "upwardly mobile" milieu. This circumstance is an "excellent starting position in the electoral market" from the point of view of maximizing votes.

The division of the (elected) population according to socio-moral values ​​and lifestyles has also been carried out by other scientists (Michael Vester, 2006). The category schemes vary, as do the forecasts for future voter potential of the various parties. The utility of social and moral milieu categories for election campaign strategists is not without controversy, even if the SPD's "New Middle" campaign in 1998 could be interpreted as a confirmation of the concept.


Possibilities and limits of the explanatory models

According to the study "Living in Europe" by the European Union, the risk of poverty in Germany has increased significantly over the past two decades. According to this, in 2015 16.7 percent of people lived on the verge of poverty or were poor - around half more than in 1998. However, economically defined classes do not necessarily correspond to a specific social milieu. (& copy picture-alliance / dpa)


The previous presentation of the most important explanatory approaches has shown the different approaches to explaining voting behavior. However, the difficulties of electoral research in appropriately depicting the complex process of voting decisions in theoretical models also became clear. Individual action calculations, group-specific interests as well as long-term and short-term influences can hardly be taken into account in a single model and can also be empirically distinguished (Ulrich Eith / Gerd Mielke, 2016). The explanations are limited to different facets. They cannot therefore be compared directly with one another. Their respective premises, procedures and questions are too different.

The sociological approach primarily examines the politically relevant group interests that exist in a pluralistic society. By analyzing the historically evolved conflict pattern, it is possible to identify those political, economic and cultural factors that have essentially contributed to the development and stabilization of these group interests. Various regional studies have repeatedly demonstrated that the cultural aspects are of great importance.

From the investigation of income, wishes, future expectations and the identification with value orientations, electoral research gains important insights into voting behavior. Photo: Kirchgessner / laif


The political elites have a high priority in this explanation. In the long term, they formulate and represent the worldviews of their respective followers. In the short term, they update these in the form of daily political (group) demands in the elections. The sociological approach illustrates the political impact of basic social structures and milieus, which are only subject to a process of change that is progressing step by step. For political parties, the initial potential and its medium-term change can therefore be estimated. In contrast, short-term mood swings largely elude the analytical instruments of the sociological approach.

The model of rational voting behavior takes an opposite starting point. It does not ask about the interests of the electorate, which are taken for granted. The question is about the external conditions and restrictions under which the voters have to make their decision. Short-term mood swings can thus be interpreted as a reaction to changed conditions - such as the inflation rate, the unemployment rate, economic growth rates or a changed supply of personnel. In this model, too, the behavior of the political elites plays a key role. In the context of the model, the political parties are first and foremost oriented towards a strategy of maximizing votes, towards winning the next elections. To do this, they work on the topics that are most favorable for their respective purposes in order to influence public political discussions. However, this competitive model sharpens the awareness of the electorate's share of responsibility for the state of the political system. Suppliers are always geared towards the wishes of their potential customers.



On the other hand, this explanatory model does not adequately explain longer-term shifts in the political balance of power and regional differences in voting behavior. Both are ultimately based on changes or differences that develop in the interests of the electorate.

The individual psychological approach occupies a certain middle position. The interests of the electorate are reflected in the form and distribution of individual party identifications. However, this approach asks less about the socio-structural causes and the longer-term changes. Rather, the focus of the analytical interest is on the interplay of long-term basic orientation and short-term problem assessment in current election decisions.



The model of social milieus mixes the sociological model with the individual psychological model in a certain sense. In this way, groups are formed that are not viewed as a historically evolved unit or that have common characteristics due to their origin, education or the like. Rather, individual ideas about life and value orientation form common characteristics of different, different milieus. On the basis of these admittedly fuzzy definitions, attitudes and thus voting intentions towards and for parties can be examined.

Election research is one of the most developed disciplines in empirical social research. The secured knowledge of voting behavior has grown rapidly over the past decades. However, a comprehensive theory of voting behavior is not in sight - possibly not even desirable (Ulrich Eith / Gerd Mielke, 2016). It is precisely the existence of several powerful explanatory approaches with different perspectives and focal points that enables electoral research to examine the complex process of voting decisions in a differentiated manner in all of its various facets. However, the accuracy of the predictions will not improve significantly.