Demographics and the social bases of voter turnout

Authored by: Eric Plutzer

The Routledge Handbook of Elections, Voting Behaviorand Public Opinion

Print publication date:  September  2017
Online publication date:  September  2017

Print ISBN: 9781138890404
eBook ISBN: 9781315712390
Adobe ISBN:




The term “demographics” conjures up the idea of personal characteristics whose categories are typically reported in national censuses and government reports. In every nation, sex, age, marital status, educational attainment and some measure of economic status comprise the core concepts typically reported. In addition, the “essential core” might include language spoken at home in Canada and Switzerland; race, religion and residential mobility in the United States; or ethnicity in Uganda and Nigeria.

 Add to shortlist  Cite

Demographics and the social bases of voter turnout

Demographics as the essential core of political behavior research

The term “demographics” conjures up the idea of personal characteristics whose categories are typically reported in national censuses and government reports. In every nation, sex, age, marital status, educational attainment and some measure of economic status comprise the core concepts typically reported. In addition, the “essential core” might include language spoken at home in Canada and Switzerland; race, religion and residential mobility in the United States; or ethnicity in Uganda and Nigeria.

For scholars of political behavior generally, demographic characteristics represent highly accessible data that are often correlated with behaviors such as turnout, participation, issue preferences and party affiliation. Although scholars may consider different approaches to coding and measurement, the operationalization of demographic concepts has been relatively uncontroversial in comparison to the definition and measurement of concepts such as efficacy, alienation and other psychological traits believed to be causes of political behavior. The relative stability of many demographic characteristics also makes them attractive because reverse causality is (often) less plausible. For example, voting in a recent election is unlikely to fundamentally alter a citizen’s reported age, sex, education or income.

For these reasons – ease of access, uncontested conceptualization and exogeneity – demographic characteristics have formed the essential core of any analysis that seeks to explain why citizens differ in their political behaviors.

The roots of this can be seen in all the classic studies of political behavior. Lazarsfeld, Berelson and Gaudet (1944), for example, examined age, education, occupation, age, place of residence and sex as predictors of political participation and engagement. Demographic predictors were used at least to some extent by Centers (1949) in The Psychology of Social Classes, Key’s (1961) Public Opinion and American Democracy (1961), The American Voter (Campbell et al. 1960) and most notably in the essays Lipset published in the volume Political Man (1960).

The declining usefulness of simple demographics

Demographic variables’ ease of use, however, can easily lull even experienced scholars into superficial – dare I say lazy – social research. Indeed, there are five common and recurring threats to valid inference.

First, lurking beneath the apparent consensus on measurement are multiple and conflicting meanings – for example, what exactly do we mean when we say one citizen is more highly educated than another? Might someone who attended Oxford for two years without attaining a degree have “higher” education than a graduate of my own Penn State University? Is a degree earned in 1980 comparable to a degree from the same school in 2015?

Second, observed differences such as gender gaps, ethnic cleavages and socio-economic gradients are the result of complex social processes that ideally should be understood and modeled directly.

Third, differences in measurement quality can sometimes tip the scales in favor of the better measured concept, leading researchers to incorrectly conclude that one variable is “more important” than another (for a modern review and some novel solutions, see Blackwell, Honaker and King 2015).

Fourth, some demographics are less stable than others, which can mislead scholars in interpreting their apparent effects. Instability can be of two types. First, there is instability in the actual construct – when a citizen is classified as married, divorced or single this might reflect a condition that has influenced current behavior for a few weeks, a few years or a few decades. In short, two individuals with the same demographic classification could have very different exposures to the actual causal factors that the demographic proxies are designed to capture.

Instability can also be a function of survey response – as in answering questions about sexual, racial or religious identification differently from time to time. Someone may identify today as African American and tomorrow as mixed race and this type of instability poses a different challenge to inference – with implications for questionnaire design as well as interpretation.

Fifth, and most fundamentally, society has changed dramatically since the pioneers of behavioral research began using demographics to predict political behavior. Social life was more rigid in 1950s and 1960s Europe and North America than it is today. Four examples illustrate this well.

First, men’s and women’s ideal roles were highly proscribed, and this gave the binary variable “sex” a clear referent that was reflected in observable differences in political engagement (Andersen 1975). Second, a sizable blue collar work force was not only a “class in itself” but, because of strong unions and labor-oriented political parties, blue collar workers comprised a class “for itself.” For these reasons, the blue–white collar distinction was highly predictive of many political behaviors, such as political party affiliation, and authoritarian attitudes (Lipset 1960). Third, in Europe especially, university education was restricted to a small group and connoted privileged status. But the democratization of higher education and the growing diversity of higher education options create significant heterogeneity within traditional groupings of educational attainment. Finally, race and ethnicity tended to also put individuals “in their place” with relatively rigid expectations; in contrast, a growing number of citizens now claim multiple racial and ethnic identities – identities that might differ in salience from one day to another. This, along with rapidly changing competition and coalitions among cultural minorities, complicates any attempt to measure and assess the political consequences of racial and ethnic affiliation.

In short, simple demographic classifications were never perfect, but they represented justifiable ways of operationalizing the social bases of politics in the 1950s and 1960s. They may still be useful, but we have graduated beyond their uncritical use.

If “demographic effects” are not as simple as they seem, they remain essential to any effort to understand political behavior. Humans are social animals and the political choices we make are arguably circumscribed by the social worlds we inhabit. For example, our workplaces structure political information we receive (Mutz and Mondak 2006); our neighborhoods and our income determine the extent to which political parties seek to mobilize us (Huckfeldt and Sprague 1995) or seek contributions (Verba, Schlozman and Brady 1995). Thus my goal in writing this chapter is to provide a guide to the social bases of political behavior that will help scholars – both novice and established – to engage in rigorous research that does justice to the complexity of social life and the group affiliations that demographic variables connote. Scholars armed with an appreciation for the complexity underlying these deceptively simple variables will be better equipped to conduct research that is creative, that more closely models the underlying social processes, and is more useful in guiding reforms in public policy.

Ascription, achievement and the fluidity of identity

Anthropologists and sociologists traditionally (see, for example, Foner 1979) distinguish between ascribed traits that do not change (e.g., year of birth, sex) from achieved traits that are the result of individual and social processes (such as education and occupation). That distinction may have been useful as a simplification, but now obscures the importance of identity in translating social location into politics. The binary distinction between male and female has given way to sexual and gender identities that challenge traditional roles and challenge traditional notions of biological determinism (Waylen 2013). While European Jews and African Americans in the 1930s had little power over the categories that others placed them in, racial and religious identities are now more fluid (Huddy 2001; Junn and Masuoka 2008; Putnam and Campbell 2012) and more personal (Bellah et al. 2007).

While binary distinctions may be useful as simplifying assumptions for empirical research, we can no longer assume uncritically that these distinctions do a good job of capturing the politically relevant aspects of personal characteristics. Yet this is exactly what a good deal of contemporary research does. Whether as “control variables” or as potential causes of political behavior, many contemporary scholars continue to use traditional demographic variables and interpret them along outdated, culturally conventional lines.

In that light, my goal in this chapter is to provide readers with a review that empowers them to engage in research that does justice to the underlying social dynamics that give many demographics their explanatory power. Demographics matter – but determining how and why is hard work that can no longer rely on social conventions about “people’s place” in the social order.

To do so, I will focus on four of the core clusters of variables that animate politics today. (1) I will begin with demographics that help us to understand political development over the life course: variables such as age, family formation, retirement. (2) I then discuss the cluster of demographics that comprise socio-economic status (SES): especially social status, education and income. (3) We will then move to variables that often interact in domestic life: sex, gender, family structure and labor force participation. (4) Finally, I turn to variables that represent national and geographic heritage – race, ethnicity and national origin. In each case, issues of conceptualization, operationalization and measurement are key considerations.

This is not, of course, an exhaustive list. This chapter will not discuss religion, residential mobility or place of residence (e.g., the urban–rural nexus). However, these variables are strongly illustrative of the key demographic variables that influence political behavior and the general principles that have led to major advances in the other clusters can be applied to other domains.

Scholars armed with an appreciation for the complexity underlying these deceptively simple variables will be better equipped to conduct research that is creative, that more closely models the underlying social processes and that will be more useful in guiding public policy.

Political participation over the life course

Of all demographic factors, none is more prevalent than age. Writing more than 50 years ago, Milbrath (1965: 134) cited eight papers that showed the same empirical pattern: very low turnout characterized the youngest eligible voters, followed by a steep rise in young adulthood followed by gradual increases until voters reach their sixties. Similar patterns are seen in all established democracies (Franklin 2004; Melo and Stockemer 2014) and emerging democracies (Potgieter 2013). Recently, fine-grained analyses suggest a slight fall in turnout when young voters leave the homes of their parents rather than a uniform monotonic rise (Bhatti and Hansen 2012a), but the broad pattern remains. This broad pattern, however, probably over-estimates the causal impact of age because age-specific mortality is higher for low-SES, low-turnout citizens, so each cohort becomes slightly more economically advantaged as it ages (Rodriguez 2012).

Whether and to what extent turnout falls off during old age is less often examined than the rise in young adulthood. This is partly due to the fact that many electoral surveys limit their sampling frame to the non-institutionalized population, thereby excluding citizens who live in nursing facilities. The result of this exclusion is that the small number of surveyed respondents who are among the old and oldest-old are disproportionately healthy, mobile and fit. For this reason, analyses of American National Election Studies, for example, fail to find a falling off in old age (Miller and Shanks 1996). However, when scholars have access to government electoral data that can be linked to census and health statistics, there is clear evidence of a steep decline in old age as well (Bhatti and Hansen 2012b).

Three broad explanations have been advanced to account for age effects. The first lies in the gradual clarification of interests and preferences. Younger citizens may not have a clear sense of their own current and future political interests (Achen 2006) – at the age of 18–20 they may be in university, in military service or in an apprentice position and lacking a clear sense of whether they are on track to benefit or not from high tax rates, tough workplace regulations or most government policies. These clarify over time, and seem to do so in ways consistent with Bayesian updating (Achen 2006). Likewise, new citizens may not fully appreciate how political parties differ (Plutzer 2002) so, even with clear values, it is difficult to map those values on to particular candidates or parties – diminishing the motivation to vote.

The second broad explanation is the gradual internalization of societal norms – adults are expected to start a family, keep up with the news, settle into a permanent residence of their own and participate in the civic life of the community. Markers of maturation and the adoption of adult roles, therefore, may be better predictors of turnout than age, per se (Highton and Wolfinger, 2001). The challenge of examining key “life cycle” events, however, is that many are not simply markers of maturation, but also of competing demands for time and attention (such as the raising of young children – Stoker and Jennings 1995; Plutzer 2002).

The third explanation focuses on how aging is associated with changes in social networks and peer groups. Young adults may first become eligible to vote while living with parents and while in schools that encourage civic engagement. But they soon find themselves living among other young people who lack electoral experiences – they go from a high turnout context to a low turnout context, with a corresponding reduction in social cues and expectations of civic participation (Bhatti and Hansen 2012a). As citizens near the ends of their lives, this process reverses as frailty and mortality both reduce the size of networks and impair the political involvement of close ties (Bhatti and Hansen 2012b).

No doubt, these three processes are all in play, sometimes in ways that are mutually reinforcing. The demographic variable of age, therefore, provides a window into the operation of key social and cognitive determinants of voting. There are important opportunities to enhance our understanding of participation by focusing theories on these underlying dynamics and seeking to model them directly, rather than by inferring their impact via the variable of age.

Of course, while “age” has a clear theoretical connotation, the variable of age in a crosssectional data set is confounded with generational effects. For example, we often refer to “baby boomers,” children of the welfare state, or Thatcher’s or Reagan’s children. Coming of age in particular political epochs has been explored extensively with respect to values (see, for example, Inglehart 1990), and vote choice (see, for example, van der Brug 2010), but less so for turnout (though see Miller and Shanks 1996).

The S in SES: social status

Socio-economic status, or SES, is not a variable but, rather, a catch-all term that encompasses a wide range of potential characteristics. In the 1960s and 1970s, when data and computational capacity exploded to create the modern discipline of political behavior, most researchers assumed that the various aspects of SES were so highly correlated that it made sense to think of them as comprising a single, unidimensional concept (Duncan 1961). For example, the occupational gradient of unskilled, skilled blue collar, clerical, management/professional was highly correlated with education and income (and, in the US, race as well). Scholars found it attractive to subsume these characteristics into a single concept because this would simplify both theory and analysis. Indeed, many data sets shipped with pre-calculated composite measures of SES – such as “Duncan’s SEI.” Today, few scholars find the unidimensional concept useful and our understanding of political behavior has been enhanced by close examination of the various components.

Social status and social exclusion

Max Weber famously wrote that collective political power can derive from class (the control of capital), status (one’s group-based social standing), and party (the power of numbers) (Weber 1978 [1922]). For the purposes of political behavior research, however, these are amorphous concepts. The “S” in SES has given rise to many different concepts and operational variables. It is useful to recognize that, for Weber, “status” was a categorical variable. He used the word, “standt,” which is best translated as “estate,” and Weber was likely thinking of the political power that was wielded by the first estate (monarchs ruling by the social convention that they have a “divine right”), the second estate (clergy, to whom people defer even when they lack coercive power of the state or the power of money) and the third estate (landowners, who derive certain rights, most notably suffrage, as a consequence of their status).

The concept was later Americanized. America, many believed, had no social classes and no state church, but a continuum of statuses based on occupation – white collar professionals derived prestige and influence that was not necessarily linked to their income. A school teacher held more prestige than a unionized factory worker who might actually have higher earnings.

The key takeaway is that power and privilege can derive from the formal and informal prestige that a society grants to groups because of esteem, respect or tradition; and that this power and privilege is analytically distinct from power deriving from the group’s economic resources.

Recently, this idea has found its expression most clearly in studies of social exclusion – when societal norms discourage the active participation of groups deemed to be outsiders or insufficiently integrated into the dominant culture. For example, Heath and his colleagues (2013, Chapter 8) show that ethnic minorities in the United Kingdom abstain from voting at especially high rates even after accounting for differences due to lower education, class and age; Aasland and Fløtten (2001) show similar impacts of Slavic minorities in Estonia and Lativa. The notion of social exclusion also has motivated former felons (Pettit and Sykes 2015; Bevelander and Pendakur 2011) and alienated youth (Bentley and Oakley 1999).


Other than age, no demographic variable has received as much attention as education (see Burden 2009 for an excellent review). It is almost always the first or second strongest predictor of voter turnout and political participation, and education’s presumed causal influence was unchallenged for four decades in spite of the obvious paradox that turnout in many Western nations declined even as educational attainment soared (Brody 1978; Franklin 2004). As late as 1992, Miller and Shanks (1996: 84) could write of the pre-New Deal generation that “education completed a half century earlier apparently subsumes all other social differentials as an influence promoting turnout.” Miller and Shanks (1996: 85–92) similarly report education to be the strongest influence on turnout for New Dealers and the largest effects were found among the post-New Deal generation.

Early attempts to discern the causal influence of education by instrumental variable methods produced numerous contradictory findings. Milligan, Moretti and Oreopolus (2004) and Dee (2004) showed statistically significant net effects of education, while Tenn’s (2007) analysis shows education’s impact to be spurious – at least during the period of completing one’s education.

Some analyses show that the impact of education is not absolute but apparently relative. Persson’s (2013) cross-sectional analysis of 37 countries shows relative education is more predictive than absolute. At the individual level, Nie, Junn and Stehlik-Barry (1996) show that education effects are relative to one’s generational cohort. The notion that education effects are relative can be interpreted in both causal and non-causal terms.

Hence it is relevant that a series of empirical analyses in the last decade have seriously undermined the idea that education has a causal effect on turnout. Persson’s (2013) panel analysis of Swedish voters shows education effects present before education is complete. Highton (2009), looking at political sophistication using US panel data, finds exactly the same thing. Berinsky and Lenz’s (2011) clever use of regression discontinuity designs shows no causal impact of education among young Americans in the Vietnam generation. While more research is yet to come, it seems that the stronger the research design in terms of causal inference, the less likely it is to find any impact of education whatsoever.

This then raises the question of what accounts for the strong correlation between education and turnout. It does not seem to be the result of products of education, such as cognitive ability or future earnings – for if it were, unmediated causal analysis would show an impact of education. Rather, it points us to precursors of education located in the family of origin, in personality and early socialization. In short, at this time, it appears that education effects likely reflect selection mechanisms and social forces that sort people into tracks of relative privilege.

The E in SES: class analysis and income effects

The economic components of SES have received considerably less attention than education in recent decades, with studies usefully located in two broad traditions. The first is more commonly employed in European political science and in North American political sociology, and relies heavily on the traditional notion of social class. The second is focused on consumption and therefore typically focuses on family or household income.

Class analysis

The Weberian and Marxist tradition of class analysis views politics in historical context and social classes as collective actors that make political history. Most of the research on social class derives from the presumed conflicts of interest that are created by an individual’s location in the economic order. Anticipating Marx, Madison (1787) famously wrote in Federalist #10 that “The most common and durable source of factions has been the various and unequal distribution of property. Those who hold and those who are without property have ever formed distinct interests in society.”

During the 1960s, 1970s and 1980s, class was most typically operationalized based on whether the head of a household could be classified as a white collar or blue collar worker (see, for example, Burnham 1987). Starting in the late 1980s, the operationalization of social class became more nuanced. The two leading approaches augmented occupational titles with information on where a worker fit in the class structure, such as whether a worker owned a business, or whether she/he supervised lower level workers (see Manza, Hout and Brooks 1995 for a review). By and large, class effects on turnout tend to be modest in size in the United States and minimal in Europe. Drawing on three decades of evidence, Gallego (2010: 239) concluded “inequality in election turnout is not universal.” According to records, in many countries, particularly European democracies, education and income are not associated with voter turnout. The small effect of class is in part due to the effective mobilization by trade unions and union-based political parties. Class effects are more evident for broader measures of civic participation (e.g., Li, Pickles and Savage 2005).

A second aspect of class analysis is class identification – do individuals think of themselves as being members of a particular class? The classical notion that class identity creates a class “for itself” implies that when lower income individuals self-identify with a working class movement, the group identification can be empowering (Mann 1973). Yet, the limited empirical evidence does not support this idea. For example, Walsh, Jennings and Stoker (2004) examined a multi-generational panel study and found that political participation was negatively related to working class identification (though they did not directly test the implied class × class identification interaction).

Income effects

Individual-level income effects are small or non-existent in Europe (Franklin 2002). When such effects are identified, they are typically modest or conditional on other factors – for example, Martikainen, Martikainen and Wass (2005) show small income effects on turnout among older Finnish citizens.

In contrast, income matters quite a bit for political participation in the United States. Leighley and Nagler (2013) demonstrate persistent effects of income on turnout in the United States over a 40-year span. In the US, the role of economic status has seen something of a renaissance in recent research. All scholars working in this area recognize that money is not a resource that can be directly utilized to increase participation, except in the narrow case of campaign contributions (Verba, Schlozman and Brady 1995). Rather, income is a proxy for more difficult-to-measure factors that directly or indirectly impact on participation in general and the turnout decision in particular.

Most of this work has focused on the lower end of the income distribution. Pacheco and Plutzer (2007, 2008) argue that economic hardship in young adulthood is a precursor to brushes with the law, crime victimization, interrupted education and early parenthood, all of which – in turn – disrupt the development of habits of civic engagement. Recent work by Holbein (2017) suggests that these experiences likely presage low turnout through their correlation with social skills such as inter-personal problem solving, emotional and impulse control, and overall life management – a finding consistent with Plutzer and Wiefek’s (2006) speculation that income mobility effects on turnout might reflect non-cognitive traits such as personal organization and perseverance, and with Ojeda’s (2015) demonstration that mental illness – in this case, depression – can impact voter turnout.

In short, recent advances have sought to unpack the complex social and psychological processes that are the proximal causes of observed income gradients in political participation. As Holbein shows, doing so can uncover actionable findings – methods to address the underlying causes may be far more practical than seeking to enhance participation by changing national distributions of income.

Measuring income

Even as income’s effect is presumed to be indirect, the presence of substantial total effects of income is important, because it has important implications for the functioning of democracy and social justice (Bartels 2009). Ojeda (2017) posited that if income effects reflect an individual’s or family’s purchasing power, then political science research might benefit from work in economics that seeks to better measure this accurately. In particular, Ojeda shows that operationalizing income in terms of the ability to meet basic needs reveals larger income effects than would be the case using traditional ordinal or log-scaled measures. His approach of adjusting income by creating a ratio of current income to the relevant poverty threshold for an individual’s family represents a promising approach that can be implemented in most electoral data sets.

Sex and gender, work and family

As recently as 1975, Kristi Anderson (1975: 439) could observe that “politics almost everywhere is a male dominated enterprise. … Sex differences in political participation are enormous.” Yet even a half century ago, sex differences in most forms of participation were small. Indeed, Campbell et al. (1960: 484) show only a 10 percent difference in turnout in the 1956 presidential election. Ten years ago, Paxton, Kunovich and Hughes (2007: 264) could review the extant literature and conclude that “Women’s fight for formal political representation is mostly won … women are more likely to vote than men in the United States … and across most countries women vote at rates fairly similar to men.”

If turnout differences are small, there are some additional gender gaps in other dimensions of participation. For example, women are less expressive in open-ended questions and more likely to give “don’t know” answers in political surveys (Atkeson and Rapoport 2003), they are less active in deliberative forums (Karpowitz, Mendelberg and Shaker 2012) and they score lower on scales measuring political knowledge, political interest and political efficacy (Verba, Burns and Schlozman 1997).

Family structure is another topic that has been investigated, with consistent findings that married citizens have higher turnout than unmarried citizens (Stoker and Jennings 1995; Denver 2008), with especially low levels for single parents (Wolfinger and Wolfinger 2008; Plutzer and Wiefek 2006). In addition, parental divorce during childhood can have lasting impacts that depress turnout in young adulthood (Sandell and Plutzer 2005).

By and large, these factors – gender, family structure and work life – have been studied in isolation, even though it might be expected that marriage, divorce, parenthood and career orientation are experienced quite differently for men and women. It is surprising then, that few have examined within-gender differences through the lens of a gendered society that differentially channels politically relevant resources to men and women. The major exception is the groundbreaking book by Burns, Schlozman and Verba (2001). They found that women whose work life was more career oriented had higher levels of participation than those with jobs that were less demanding or who had no employment at all. Though frequently cited, however, few have sought to build on this approach to develop a mature and cumulative research program on how the intersection of these factors may impact upon political participation.

Race and ethnicity

The enslavement of African Americans, their systematic disenfranchisement under Jim Crow laws and the slow implementation of the Voting Rights Act of 1965 creates a particular historical context for the study of race and politics in the United States. Not surprisingly, the electoral participation of African Americans lagged behind that of whites at the dawn of the era of behavioral political science. The authors of The American Voter, for example, estimated that black turnout was less than half the rate of whites even in the northern United States (Campbell et al. 1960: Table 11–6). Based on data from the American National Election Studies, Rosenstone and Hansen (2003) estimate an average 15-point racial gap in turnout between 1952 and 1988. That gap, however, has closed with African American turnout exceeding that of whites in the 2008 presidential election.

Critically, however, early studies examining the black–white gap in participation found that the effect disappeared or reversed after accounting for the much lower SES of African American voters (see, for example, Guterbock and London 1983). This then represented a puzzle: why did African Americans “outperform” the levels of participation that might be expected for white Americans of similar age, education and income? The research that sought to solve this puzzle is broadly generalizable. Though empirical results have been mixed, they represent approaches that have been applied to other minority groups in the United States, and have the potential for application to minority politics throughout the world.

Linked fate

The concept of linked fate was initially proposed in Dawson’s seminal book, Behind the Mule. Gay, Hochschild and White (2014) describe it this way:

due to historical and contemporary experiences of group disadvantage and discrimination, one’s own life chances depend heavily on the status and fortunes of Black Americans as a whole. That perception has led, in turn, to the rational substitution of group utility for individual utility in political decision-making, and often to a strong moral and emotional commitment to the group.

The similarity of linked fate to the earlier ideas of class consciousness – how a class transforms from the class in itself to a class for itself – is clear, and the concept can in principle be applied to any group in any country that has endured periods of disadvantage and discrimination. Indeed, it has been applied to many groups in the United States by adapting Dawson’s original question, “Do you think what happens generally to Black people in this country will have something to do with what happens in your life?,” by substituting other groups in place of the word “Black.” The empirical results, however, have been mixed. Effects for African Americans (Philpot, Shaw and McGowen 2009; Gay, Hochschild and White 2014), Latinos (Stokes 2003; Stokes-Brown 2006; Sanchez 2006; Gay, Hochschild and White 2014) and Asians (Kim 2015; Junn and Masuoka 2008; Gay, Hochschild and White 2014) have typically been small or not statistically significant in the expected direction.

Political empowerment

A long-standing argument is that members of racial and ethnic minorities may be mobilized by political empowerment – that is, descriptive representation can lead to feelings of pride and efficacy, and a reduction in cynicism, thereby spurring higher turnout (Bobo and Gilliam 1990). This argument is more complicated than it seems at first glance, however, because of feedback to majority member citizens. The political empowerment model is silent on (for example) the impact of an African American office holder on the mobilization of white voters (Washington 2006). If the political empowerment thesis is tested by examining black–white gaps, a significant effect could be due to larger black mobilization, white demobilization, or both. Likewise, a conclusion of “no effect” could be the result of black political empowerment and white reaction of similar magnitudes.

Most recently, Logan, Darrah and Oh (2012) confirmed the empowerment thesis for African Americans, showing a small but significant effect for Latinos, but a negative impact for Asian Americans.

Religious mobilization

A third explanation with broader applicability focuses on religious mobilization. When racial and ethnic groups attend religious institutions that cater specifically to members of their own members, the mobilization effect of religious attendance may be amplified – a conclusion reached by Segura and Bowler (2012) based on their review of the empirical literature (see Harris 1994 and Jones-Correa and Leal 2001).

In sum, every racial and ethnic minority group has a unique history – a history that may begin with enslavement, escape from war or civic violence, or in economic migration. The social and political exclusion of the first generations of the group are also critical in establishing patterns of political participation – and of course the institutional framework of the national and subnational governments play outsized roles. Amidst such diversity, can we build a broader theoretical framework that can guide our research and interpretation of racial and ethnic demographic variables? The three approaches reviewed here – linked fate, political empowerment and religious mobilization – represent promising ideas on which to build.

Summary and conclusion

Demographic variables represent an essential core of quantitative research in political behavior. Despite the rapid pace of social change, we continue to observe turnout gaps related to age, ethnicity and education. As a consequence, any serious analysis must control for these key variables, which structure opportunities, mobilization and information acquisition in modern democracies. Yet in every instance, a close look at the best empirical research suggests that no demographic variable embodies a simple causal story. In some cases, such as education, new research is overturning decades of political wisdom. In other cases, such as the linked fate and class consciousness paradigm, the empirical effects are relatively small in comparison to qualitative accounts of these processes. In every instance, we are reminded that society never holds “all else equal” so it becomes difficult to isolate the effects of age from income, income from race, race from class, class from work status, or work status from gender.

Because it is not possible to experimentally manipulate demographic variables, advances will necessarily come from unpacking the mechanisms that link social group membership to political engagement – mechanisms that will vary within demographic groups and which, in some cases, might be good candidates for experimental research. Better use of longitudinal research may also be promising, and clever exploitation of quasi-experiments (e.g., changes in voter context due to redistricting) may provide the basis to better specify theories such as those related to political empowerment. Finally, attention to measurement – especially for income, education and group identity – has substantial promise to enhance our understanding of how and why demographic variables consistently predict differences in voter turnout.


Aasland, A. and Fløtten, T. (2001) “Ethnicity and Social Exclusion in Estonia and Latvia,” Europe-Asia Studies, vol. 53, no. 7, November, 1023–1049.
Achen, C. H. (2006) “Expressive Bayesian Voters, Their Turnout Decisions, and Double Probit: Empirical Implications of a Theoretical Model,” Unpublished Manuscript, Princeton University.
Andersen, K. (1975) “Working Women and Political Participation, 1952–1972,” American Journal of Political Science, vol. 19, no. 3, August, 439–453.
Atkeson, L. R. and Rapoport, R. B. (2003) “The More Things Change the More They Stay the Same: Examining Gender Differences in Political Attitude Expression, 1952–2000,” Public Opinion Quarterly, vol. 67, no. 4, Winter, 495–521.
Bartels, L. M. (2009) Unequal Democracy: The Political Economy of the New Gilded Age, Princeton: Princeton University Press.
Bellah, E. R. N., Tipton, S. M. , Sullivan, W. M. , Madsen, R. and Swidler, A. (2007) Habits of the Heart: Individualism and Commitment in American Life, Berkeley: University of California Press.
Bentley, T. and Oakley, K. (1999) The Real Deal: What Young People Really Think About Government, Politics and Social Exclusion, London: Demos.
Berinsky, A. J. and Lenz, G. S. (2011) “Education and Political Participation: Exploring the Causal Link,” Political Behavior, vol. 33, no. 3, September, 357–373.
Bevelander, P. and Pendakur, R. (2011) “Voting and Social Inclusion in Sweden,” International Migration, 49, no. 4, August, 67–92.
Bhatti, Y. and Hansen, K. M. (2012a) “Leaving the Nest and the Social Act of Voting: Turnout Among First-Time Voters,” Journal of Elections, Public Opinion and Parties, vol. 22, no. 4, November, 380–406.
Bhatti, Y. and Hansen, K. M. (2012b) “Retiring from Voting: Turnout Among Senior Voters,” Journal of Elections, Public Opinion and Parties, vol. 22, no. 4, November, 479–500.
Blackwell, M. , Honaker, J. and King, G. (2015) “A Unified Approach to Measurement Error and Missing Data Overview and Applications,” Sociological Methods and Research, available online:
Bobo, L. and Gilliam, F. D. (1990) “Race, Sociopolitical Participation, and Black Empowerment,” American Political Science Review, vol. 84, no. 2, June, 377–393.
Brody, R. A. (1978) “The Puzzle of Political Participation in America,” in Beer, S. H. (ed.), The New American Political System, Washington, DC: American Enterprise Institute: 287–324.
Brug, W. van der (2010) “Structural and Ideological Voting in Age Cohorts,” West European Politics, vol. 33, no. 3, May, 586–607.
Burden, B. C. (2009) “The Dynamic Effects of Education on Voter Turnout,” Electoral Studies, vol. 28, no. 4, December, 540–549.
Burnham, W. D. (1987) “The Turnout Problem,” in Reichley (ed.) Elections American Style, Washington, DC: Brookings Institution: 97–133.
Burns, N. , Schlozman, K. L. and Verba, S. (2001) The Private Roots of Public Action, Cambridge, MA: Harvard University Press.
Campbell, A. , Converse, E. , Miller, W. and Stokes, D. (1960) The American Voter, Chicago: University of Chicago Press.
Centers, R. (1949) The Psychology of Social Classes, Princeton: Princeton University Press.
Dawson, M. C. (1994) Behind the Mule: Race and Class in African American Politics, Princeton: Princeton University Press.
Dee, T. S. (2004) “Are There Civic Returns to Education?” Journal of Public Economics, vol. 88, no. 9, August, 1697–1720.
Denver, D. (2008) “Another Reason to Support Marriage? Turnout and the Decline of Marriage in Britain,” British Journal of Politics and International Relations, vol. 10, no. 4, November, 666–680.
Duncan, O. D. (1961) “A Socioeconomic Index for All Occupations,” in Reiss, A. J. (ed.) Occupations and Social Status, New York: Free Press: 109–139.
Foner, A. (1979) “Ascribed and Achieved Bases of Stratification,” Annual Review of Sociology, vol. 5, August, 219–242.
Franklin, M. N. (2002) “The Dynamics of Electoral Participation,” in Ludec, L. , Niemi, R. and Norris, P. (eds.) Comparing Democracies 2: Elections and Voting in Global Perspective, Thousand Oaks: Sage: 148–168.
Franklin, M. N. (2004) Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945, Cambridge: Cambridge University Press.
Gallego, A. (2010) “Understanding Unequal Turnout: Education and Voting in Comparative Perspective,” Electoral Studies, vol. 29, no. 2, June, 239–248.
Gay, C. , Hochschild, J. L. and White, A. (2014) “Americans’ Belief in Linked Fate: A Wide Reach but Limited Impact,” Unpublished Manuscript, Harvard University.
Guterbock, T. M. and London, B. (1983) “Race, Political Orientation, and Participation: An Empirical Test of Four Competing Theories,” American Sociological Review, vol. 48, no. 4, August, 439–453.
Harris, F. C. (1994) “Something Within: Religion as a Mobilizer of African-American Political Activism,” Journal of Politics, vol. 56, no. 1, February, 42–68.
Heath, A. F. , Fisher, S. D. , Rosenblatt, G. , Sanders, D. and Sobolewska, M. (2013) The Political Integration of Ethnic Minorities in Britain, Oxford: Oxford University Press.
Highton, B. (2009) “Revisiting the Relationship Between Educational Attainment and Political Sophistication,” Journal of Politics, vol. 71, no. 4, October, 1564–1576.
Highton, B. and Wolfinger, R. E. (2001) “The First Seven Years of the Political Life Cycle,” American Journal of Political Science, vol. 45, no. 1, January, 202–209.
Holbein, J. (2017) “Childhood Skill Development and Adult Political Participation.” American Political Science Review, 1–12. doi:10.1017/S0003055417000119
Huckfeldt, R. R. and Sprague, J. (1995) Citizens, Politics and Social Communication: Information and Influence in an Election Campaign, Cambridge: Cambridge University Press.
Huddy, L. (2001) “From Social to Political Identity: A Critical Examination of Social Identity Theory,” Political Psychology, vol. 22, no. 1, March, 127–156.
Inglehart, R. (1990) Culture Shift in Advanced Industrial Society, Princeton: Princeton University Press.
Jones-Correa, M. A. and Leal, D. L. (2001) “Political Participation: Does Religion Matter?” Political Research Quarterly, vol. 54, no. 4, December, 751–770.
Junn, J. and Masuoka, N. (2008) “Asian American Identity: Shared Racial Status and Political Context,” Perspectives on Politics, vol. 6, no. 4, December, 729–740.
Karpowitz, C. F. , Mendelberg, T. and Shaker, L. (2012) “Gender Inequality in Deliberative Participation,” American Political Science Review, vol. 106, no. 3, August, 533–547.
Key, V. O. (1961) Public Opinion and American Democracy, New York: Knopf.
Kim, D. (2015) “The Effect of Party Mobilization, Group Identity, and Racial Context on Asian Americans’ Turnout,” Politics, Groups, and Identities, vol. 3, no. 4, November, 592–614.
Lazarsfeld, P. F. , Berelson, B. and Gaudet, H. (1944) The People’s Choice: How the Voter Makes Up His Mind in a Presidential Campaign, New York: Duell, Sloan and Pearce.
Leighley, J. E. and Nagler, J. (2013) Who Votes Now? Demographics, Issues, Inequality, and Turnout in the United States, Princeton: Princeton University Press.
Li, Y. , Pickles, A. and Savage, M. (2005) “Social Capital and Social Trust in Britain,” European Sociological Review, vol. 21, no. 2, April, 109–123.
Lipset, S. M. (1960) Political Man: The Social Bases of Politics, Garden City: Doubleday.
Logan, J. R. , Darrah, J. and Oh, S. (2012) “The Impact of Race and Ethnicity, Immigration and Political Context on Participation in American Electoral Politics,” Social Forces, vol. 90, no. 3, March, 993–1022.
Madison, J. (1787) “The Federalist No. 10,” The New York Packet, November 22, available online:
Mann, M. (1973) Consciousness and Action Among the Western Working Class, London: Macmillan.
Manza, J. , Hout, M. and Brooks, C. (1995) “Class Voting in Capitalist Democracies Since World War II: Dealignment, Realignment, or Trendless Fluctuation?” Annual Review of Sociology, vol. 21, August, 137–162.
Martikainen, P. , Martikainen, T. and Wass, H. (2005) “The Effect of Socioeconomic Factors on Voter Turnout in Finland: A Register-Based Study of 2.9 Million Voters,” European Journal of Political Research, vol. 44, no. 5, August, 645–669.
Melo, D. F. and Stockemer, D. (2014) “Age and Political Participation in Germany, France and the UK: A Comparative Analysis,” Comparative European Politics, vol. 12, no. 1, January, 33–53.
Milbrath, L. W. (1965) Political Participation: How and Why Do People Get Involved in Politics?, Chicago: Rand McNally.
Miller, W. E. and Shanks, J. M. (1996) The New American Voter, Cambridge, MA: Harvard University Press.
Milligan, K. , Moretti, E. and Oreopoulos, P. (2004) “Does Education Improve Citizenship? Evidence from the United States and the United Kingdom,” Journal of Public Economics, vol. 88, no. 9, August, 1667–1695.
Mutz, D. C. and Mondak, J. J. (2006) “The Workplace as a Context for Cross-Cutting Political Discourse,” Journal of Politics, vol. 68, no. 1, February, 140–155.
Nie, N. H. , Junn, J. and Stehlik-Barry, K. (1996) Education and Democratic Citizenship in America, Chicago: University of Chicago Press.
Ojeda, C. (2015) “Depression and Political Participation,” Social Science Quarterly, vol. 96, no. 5, November, 1226–1243.
Ojeda, C. (2017) “The Measurement of Income in the Study of Political Behavior.” Working paper, Knoxville: University of Tennessee.
Pacheco, J. S. and Plutzer, E. (2007) “Stay in School, Don’t Become a Parent: Teen Life Transitions and Cumulative Disadvantages for Voter Turnout,” American Politics Research, vol. 35, no. 1, January, 32–56.
Pacheco, J. S. and Plutzer, E. (2008) “Political Participation and Cumulative Disadvantage: The Impact of Economic and Social Hardship on Young Citizens,” Journal of Social Issues, vol. 64, no. 3, September, 571–593.
Paxton, P. , Kunovich, S. and Hughes, M. M. (2007) “Gender in Politics,” Annual Review of Sociology, vol. 33, August, 263–284.
Persson, M. (2013) “Is the Effect of Education on Voter Turnout Absolute or Relative? A Multi-Level Analysis of 37 Countries,” Journal of Elections, Public Opinion and Parties, vol. 23, no. 2, March, 111–133.
Pettit, B. and Sykes, B. L. (2015) “Civil Rights Legislation and Legalized Exclusion: Mass Incarceration and the Masking of Inequality,” Sociological Forum, vol. 30, no. S1, June, 589–611.
Philpot, T. S. , Shaw, D. R. and McGowen, E. B. (2009) “Winning the Race: Black Voter Turnout in the 2008 Presidential Election,” Public Opinion Quarterly, vol. 73, no. 5, 995–1022.
Plutzer, E. (2002) “Becoming a Habitual Voter: Inertia, Resources, and Growth in Young Adulthood,” American Political Science Review, vol. 96, no. 1, March, 41–56.
Plutzer, E. and Wiefek, N. (2006) “Family Transitions, Economic Status, and Voter Turnout Among African-American Inner-City Women,” Social Science Quarterly, vol. 87, no. 3, September, 658–678.
Potgieter, E. (2013) Predictors of Political Participation in New Democracies: A Comparative Study, MA Thesis, Stellenbosch University.
Putnam, R. D. and Campbell, D. E. (2012) American Grace: How Religion Divides and Unites Us, New York: Simon and Schuster.
Rodriguez, J. M. (2012) The Effects of Mortality and Health Inequalities Over Disparities in Political Behavior, Doctoral Dissertation, University of California.
Rosenstone, S. J. and Hansen, J. M. (2003) Mobilization, Participation, and Democracy in America, New York: Longman.
Sanchez, G. R. (2006) “The Role of Group Consciousness in Political Participation Among Latinos in the United States,” American Politics Research, vol. 34, no. 4, July, 427–450.
Sandell, J. and Plutzer, E. (2005) “Families, Divorce and Voter Turnout in the US,” Political Behavior, vol. 27, no. 2, June, 133–162.
Segura, G. and Bowler, S. (2012) The Future is Ours: Minority Politics, Political Behavior, and the Multiracial Era of American Politics, Thousand Oaks: Sage.
Stoker, L. and Jennings, M. K. (1995) “Life-Cycle Transitions and Political Participation: The Case of Marriage,” American Political Science Review, vol. 89, no. 2, June, 421–433.
Stokes, A. K. (2003) “Latino Group Consciousness and Political Participation,” American Politics Research, vol. 31, no. 4, July, 361–378.
Stokes-Brown, A. K. (2006) “Racial Identity and Latino Vote Choice,” American Politics Research, vol. 34, no. 5, September, 627–652.
Tenn, S. (2007) “The Effect of Education on Voter Turnout,” Political Analysis, vol. 15, no. 4, Autumn, 446–464.
Verba, S. , Burns, N. and Schlozman, K. L. (1997) “Knowing and Caring About Politics: Gender and Political Engagement,” Journal of Politics, vol. 59, no. 4, November, 1051–1072.
Verba, S. , Schlozman, K. L. and Brady, H. E. (1995) Voice and Equality: Civic Voluntarism in American Politics, Volume 4, Cambridge, MA: Harvard University Press.
Walsh, K. C. , Jennings, M. K. and Stoker, L. (2004) “The Effects of Social Class Identification on Participatory Orientations Towards Government,” British Journal of Political Science, vol. 34, no. 3, July, 469–495.
Washington, E. (2006) “How Black Candidates Affect Voter Turnout,” The Quarterly Journal of Economics, vol. 121, no. 3, August, 973–998.
Waylen, G. (2013) The Oxford Handbook of Gender and Politics, Oxford: Oxford University Press.
Weber, M. (1978) [1922] “The Distribution of Power Within the Political Community: Class, Status, Party, ” in Roth, G. and Wittich, C. (eds.) Economy and Society: An Outline of Interpretive Sociology, Berkeley: University of California Press: 926–940.
Wolfinger, N. H. and Wolfinger, R. E. (2008) “Family Structure and Voter Turnout,” Social Forces, vol. 86, no. 4, June, 1513–1528.
Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.