The isolating side effect of civic participation

Abstract

Does widespread civic participation lead to more social capital? We argue that this is not necessarily the case. As more actors join voluntary associations, the pool of social resources available to non-active individuals is diluted, which can make it difficult for them to accrue social capital. We hypothesize that there is an insider/outsider effect, whereby widespread civic participation increases the risk that non-participants will be excluded from communal life. Using survey data from the German Freiwilligensurvey (German Volunteer Survey 1999–2014) and multi-level regressions, we examine the link between regional civic participation and individual social capital. Our findings reveal that civic participation rates on the macro level affect participants’ and non-participants’ friendship and support networks differently. While high macro-level participation is associated with more friendships and a higher expectation of receiving support for participants, it is also associated with fewer friendships and a lower expectation of receiving support for non-participants.

Keywords

civic participation Germany social capital social exclusion

Civic participation is a central aspect of modern life: we sing in a choir, play on a soccer team, play tennis at a club, meet on yoga courses, learn Italian in evening classes, tutor pupils after school, demonstrate against water pollution, volunteer in a soup kitchen, and contribute to the vitality of communal life in numerous other ways. Thus, the impact of civic participation on communal life is a topic that has attracted interest from policy-makers and researchers. Civic participation, inter alia, is said to create involved citizens and engender trust between them; to help generate social capital; and to help communities overcome dilemmas (Putnam, 1993, 2000; Salemink and Strijker, 2018; Uitermark, 2015). But scholars disagree about whether a high level of civic participation in voluntary associations generally unleashes socially integrating forces, or whether it mainly fosters relationships between socially similar individuals, while excluding dissimilar others (Glanville, 2004). Most political elites and some researchers of voluntary associations believe that civic activism is closely associated with a happy and socially integrated population (Roßteutscher, 2009; Roth, 2003; Stricker, 2006). Voluntary associations are perceived as tools reducing cleavages that create conflict, inequality, and distrust (Babchuk and Edwards, 1965; Olson, 1982; Putnam, 2000), and as such they are drivers of social cohesion (Gabriel, 2001; Roth, 2003; Wiertz, 2015). Based on these assumptions, many governments draw an idealized picture of civic participation, declaring civic self-organization a foundation of a democratic culture and social cohesion (Roßteutscher, 2002; Salemink and Strijker, 2018; Uitermark, 2015; van Deth, 2010). Yet there is also another perspective that describes voluntary associations as contexts in which people interact with socially similar others, thereby fostering the development of ties to similar others, rather than bridging social cleavages (Feld, 1982; Marsden, 1987; McPherson et al., 1992; Pollack, 2003; Popielarz and McPherson, 1995).

Empirically, however, the connection between widespread civic participation and social cohesion is anything but clear, and the evidence that such a connection exists is lacking (Lo Iacono, 2018). Research has shown that association membership increases with socioeconomic status (Warren, 2001: 73), and that migrants and foreigners are largely excluded from the world of leisure clubs and volunteer organizations (Roßteutscher, 2008: 398–401; Zimmer, 1996: 69). Thus, why should a strong culture of civic participation be tantamount to having a socially integrated community? Isn’t it also conceivable that civic activism leads to the clustering and delimitation of those who participate, and to the marginalization of those who do not?

The present study seeks to answer this question by investigating whether civic participation on the macro level increases the social capital of people in general, or just of those taking part in civic life. We argue that widespread civic participation does not necessarily lead to social cohesion,¹ and may instead lead to the exclusion of those who do not directly participate.

The article makes a three-fold contribution to this field of research. First, the perspective that civic participation does not necessarily produce positive outcomes for the community as a whole or promote social cohesion is strengthened by a theoretical argument, as well as by empirical findings. We begin not by distinguishing between different types of voluntary associations, but by making a more general argument: namely, that widespread participation systematically hinders the social inclusion of certain groups in society. And since there is a lack of empirical research on the validity of spill-over effects (Lo Iacono, 2018: 136), we strengthen the empirical foundation for this perspective. Second, we show that civic participation on the macro level is a feature of modern society that has to be taken into account. In the present case, the level of civic participation is a qualitative feature of the context in which individuals live and act. It has an impact on an individual’s chances of finding friends and of receiving support from others. Third, our ‘causal chain’ is not a macro-micro-macro link, but a micro-macro-micro link: that is, large-scale individual involvement transforms the social context in which the citizens live, thereby creating structural features that, in turn, affect the relationship between civic activism and social capital on the micro level.

The remainder of the article proceeds as follows. After introducing the concepts of civic participation and social capital, we argue that social capital − as indicated by social resources − is influenced by the contextual level of civic participation. We then describe the methodological design and the operationalization of the concepts. Two-level regressions are conducted to examine whether regional civic participation has an impact on the relationship between individual activity and individual social capital. The empirical results show that participants and non-participants in civic life are affected differently by contextual levels of civic participation. While high macro-level participation is associated with having more friendships and a higher expectation of receiving support for participants, low macro-level participation is associated with having fewer friendships and a lower expectation of receiving support for non-participants. Thus, it appears that while regional civic participation can have positive effects for participants, it can also lead non-participants to experience social exclusion – and, in a sense, social deprivation.

Civic participation and social capital

Civic participation implies not just membership in a group, but active involvement in public life. It is voluntary and encompasses all of the non-profit opportunities individuals have outside of their family or work: for example, singing in a choir, playing on a soccer team, or volunteering in a soup kitchen. Generally, civic participation reaches beyond the private sphere, is performed collectively and publicly, and includes both political and non-political activities.

In Germany, civic participation is a mass phenomenon. In 2019 − before the onset of the coronavirus pandemic − the German Football Association (Deutscher Fußball-Bund) numbered 7.1 million members and 149,735 soccer teams. The German Red Cross (Deutsches Rotes Kreuz) had nearly 3 million members. Nearly 70% of German citizens reported being a member of at least one club or society, and more than 37% reported that they volunteer regularly (Freiwilligensurvey [FWS; German Volunteer Survey] 1999–2014).

The value of civic participation is manifested as social capital. In the literature, the basic concept of social capital is defined fairly consistently, with the majority of scholars agreeing that social capital lies in social relationships,² and can be conceptualized as consisting of social resources and information benefits (Bourdieu, 1986; Burt, 2000; Coleman, 1988; Lin, 1999, 2001; Putnam, 1993, 2000). However, the exact interpretations and conceptualizations of social capital differ considerably. Robert D. Putnam is probably the best-known proponent of the social capital approach. In Making Democracy Work, Putnam (1993) argued that a vibrant civil society is of great importance for society as a whole. In horizontal networks – like choirs, leisure clubs, neighbourhood cooperation networks and church communication networks – the exchange of information and cooperation is encouraged. Through successful cooperation, joiners ‘become more tolerant, less cynical, and more empathetic to the misfortunes of others’ (Putnam, 2000: 288f). They develop mutual attachments, pro-social norms (solidarity, reciprocity) and mutual trust. Putnam refers to three characteristics of social life – social networks, pro-social norms and social trust – as social capital. Social capital is characterized as a social resource that allows individuals as well as collectives to overcome collective dilemmas and to pursue common goals. Thus, social capital is said to promote political performance, economic success, and social trust within the society. In Bowling Alone, Putnam (2000) deplored the decline of communal life, and, by extension, of social capital in US society. He argued that, in recent decades, people have been retreating from civic life, and that this trend has weakened social networks, pro-social norms and mutual trust, causing people to feel less attached to each other and to the broader community.

Another conceptualization of social capital was proposed by Pierre Bourdieu (1986), who conceptualizes it as a private good. Bourdieu interpreted social capital as the extent of potential resources held by the individual through the possession of stable relationships as a result of the individuals’ investment. Social capital expresses itself in the individual’s social relationships and the size of his/her social network. It enables him/her to exert power by mobilizing his/her network’s resources. From his perspective, social capital is attached to class and other forms of stratification, which are, in turn, associated with various forms of advantage. Thus, Bourdieu’s concept incorporates the social stratification and inequality perspective, which Putnam’s approach lacks.³

By interlinking civic participation, interpersonal ties and community well-being, the social capital approach assumes that there are spill-over effects for individuals who do not participate in civic activities themselves, but still benefit from the features described above (Putnam, 2000; Stolle, 2003; Van der Meer, 2003). In the present context, this would mean that a strong culture of civic participation in voluntary associations, leisure clubs, and initiatives is likely to be positively linked to social capital in general. But is this really plausible?

Indeed, studies have shown that individuals who participate in voluntary associations have more friends (primarily in these associations), are happier, and are more willing to trust others than non-participants (Davidson and Sanyal, 2017; Roßteutscher, 2008). Previous research has also found that social embeddedness is linked to informal work, especially in the form of money/barter exchange. Individuals who are engaged in informal work for money/barter most commonly do so with people they know well (Slack et al., 2017).

There is also evidence that − after the workplace and school − voluntary associations are the third-most important social foci for generating social relationships beyond kinship⁴ (Feld, 1982; Louch, 2000; McPherson et al., 2001). Many people join leisure clubs explicitly in order to meet new people (Fischer et al., 1977; McPherson and Smith-Lovin, 1987).

One study found that individuals living in communities with relatively dense social networks exhibit more social trust than those living in communities with less dense networks (Lo Iacono, 2018). Moreover, Benton (2016) was able to show that civic involvement moderates the relationship between social background and deficits in social capital. He argues that civic groups are often divided by the same status dimensions that predict social capital deficits. According to Benton, low-status actors typically lack access to certain civic groups and are therefore unable to build high levels of social capital. Thus, these low-status individuals suffer from profound social capital deficits, while joiners have more friends – and, in turn, more social capital (Benton, 2016). From our perspective, this argument is intuitively plausible.

However, while spill-over effects feature prominently in discussions of social capital, we know very little about them (Lo Iacono, 2018: 127, 136). These effects include the question of whether a participation culture affects participants and non-participants equally. We expect that voluntary associations act as containers within which social relationships and social networks are established and bonding social capital is accumulated. This, in turn, makes it more difficult for non-participants to build social capital, because they have less access to these social networks. In the next sections, we will outline the rationale behind this argument, and formulate our hypotheses.

Linking civic participation and social capital

Why would a strong culture of civic participation lead to social exclusion? Figure 1 displays the suggested micro-macro-micro link.

Figure 1.

Culture of participation and social cohesion

At first glance, individual civic participation seems to be associated only with individual social capital. However, the more individuals participate in leisure clubs and political initiatives, the more they bring the culture of participation into being. In addition, the more individuals meet and act in associational contexts, the more they build their social capital within these associations, including by making friendships and acquaintanceships, and accumulating potential social support through these friendships. These two processes transform the social context in which the individuals are acting into a situation in which stable social relationships are most likely to be established in associational contexts. Such a modified social context can, in turn, affect the relationship between participation and social capital on the individual level. Thus, it becomes more difficult for non-participants to establish new relationships and to find support from others, simply because their opportunities to do so have shrunk. The more voluntary associations that foster the development of internal networks there are, the more these associations function as containers that exclude non-participants.⁵ As a consequence, participants and non-participants may accrue different levels of social capital resulting in a form of social inequality that may affect societal cohesion (macro level).

We outline the two processes of structuring and moderating in more detail before deriving our hypothesis that the regional level of civic participation moderates the relationship between individual participation and individual social capital.

Figure 2 presents two different scenarios in which non-participants may or may not experience social exclusion. The graph on the left depicts a community in which social relationships are mainly organized outside of voluntary associations. Individuals can find and meet friends and acquaintances through all kinds of social mediators. Participants tend to maintain their relationships primarily by meeting in churches or leisure clubs, while non-participants tend to meet others in a wide range of contexts, including cafés, bars, restaurants, discotheques, shopping malls, cinemas, museums, libraries, parks, fitness clubs, and fun parks.

Figure 2.

Civic participation and individual embeddedness

A specific feature of voluntary associations is that they make experiencing positive social contacts more feasible, and thus facilitate the development of stable relationships among their members. The psychological attraction approach posits that liking someone in a new relationship tends to occur under conditions of spatial proximity and of perceived similarity (Festinger et al., 1950; Wrzus et al., 2012: 70). When people take part frequently in a voluntary association, like a leisure club or political initiative, both of these conditions are met. Individuals who regularly participate in a voluntary association have many opportunities to interact with each other. In addition, the members of these associations come together to act based on shared interests and needs. For example, the members may have a common interest in singing, learning a language, cooking, hiking, or fishing. Sharing such interests enhances the likelihood that the members will find that they have broader similarities and commonalities with each other, which may, in turn, lead to strong feelings of familiarity and connectedness (Kleiner, 2016). Thus, voluntary associations serve as containers that foster positive relationships.

Indeed, empirical studies have shown that voluntary associations, following the workplace and school, are the third-most important social foci for maintaining non-relational relationships (Feld, 1982; Louch, 2000, McPherson et al., 2001). Furthermore, research has discovered a positive link between frequency of church attendance and social network size (Bradley, 1995; Ellison and George, 1994; Hirschle, 2014).⁶

It is quite plausible to assume that such a link is not limited to religious organizations, but is inherent in all kinds of civic associations in which social relationships are produced and reproduced. However, this favourable situation for insiders can operateexclusionary for outsiders. As a consequence of this dynamic, internal cohesion within associations may not strengthen the social community as a whole.

As a lot of citizens start participating in voluntary associations, the quality of the social context begins to change. If individuals participate in voluntary associations and cultivate their contacts within these associations, their time and attention is taken up by these clubs and initiatives. But since the individuals’ capacities in terms of time and attention are limited, they cannot have an unlimited number of friendships. Research has shown that personal network size remains relatively consistent; that is, if new friendships are created, old ones are abandoned (Dunbar, 1992; Mollenhorst et al., 2008).

In Figure 2, the right-hand panel depicts a community with numerous social networks developed in the context of voluntary associations. Here, the social context is structured to a much greater extent by voluntary associations in which individuals regularly come together to share and engage in activities related to their interests and needs. The more participants are socially embedded within these associations, and the more time they spend within these associational contexts, the greater the extent to which their social networking is limited to those who actively participate in the associational contexts. This does not mean that the number of relationships in society changes; but rather that as growing numbers of people participate in leisure clubs, more of their relationships will exist within these associational contexts. As a consequence of this dynamic, non-participants will have fewer opportunities to establish new relationships or to cultivate existing friendships with participants, and may also have limited access to social resources, like material and immaterial support from others beyond their kin network. Thus, as the number of such relationship-promoting contexts increases, and the more they mature or expand, the more they are likely to produce a state of inward social closure in which those who do not join are excluded from social networking.

In such a context, individuals who live in a vibrant social environment, but do not participate themselves, run the risk of being left out. Their experience may be similar to that of free atoms that are not bound to the existing molecules and thus have a hard time finding another free atom for purposes of joint bonding.

Subnational regions as context categories

This mechanism of social exclusion is relevant not only on the national level, but on the subnational level as well. A region can be interpreted as a spatial unit with its own specific constellation of historical, social, political, cultural and intellectual elements (Hirschle and Kleiner, 2014; Kleiner, 2018). Due to their specific features – including their power relations, religious references, rural issues, economic and social factors, density of relationships, and regionally endogenous innovation potential – subnational regions develop their own regional characteristics, including their own values, mentality and lifestyle. This geographically limited lifestyle is generally cultivated and passed on from one generation to the next, and tends to be resistant to change (Charron and Lapuente, 2013; Fischer, 1975).⁷ Thus, the region in which individuals live, meet friends and go to work can be interpreted as their social context, which provides them certain opportunities while constraining them from pursuing others, and thus acts as a reference point for individual perception and behaviour.

In social science research, subnational regions have become increasingly important in recent years, and are now often used as categories for comparative analyses. Indeed, empirical studies have confirmed the existence of inter-regional cultural, social and political differences within countries (Charron and Lapuente, 2013; Kaasa et al., 2014; Kleiner, 2015; Minkov and Hofstede, 2012, 2014). Therefore, it is important to recognize subnational regions as units of analysis, and to use them as context categories to explain social behaviour.

Hypotheses

It has been argued that a strong culture of civic participation creates a situation in which social relationships are systematically forged and maintained within voluntary associations. Living as a non-participant in such an environment may mean systematic exclusion from opportunities to access social resources in the form of social relationships, and of material and immaterial support from others beyond the kin network. Following the theoretical considerations outlined above, the following assumptions are made:

H₁: The effect of individual participation on individual social capital is conditional on the regional level of civic participation.

H₂: Participants and non-participants in civic associations are differentially affected by the contextual level of civic participation. A higher level of regional participation is associated with a higher level of individual social capital for participants (H_2a), and with a lower level of individual social capital for non-participants (H_2b).

Data and operationalization

Unfortunately, we are not able to operationalize social capital from an objective perspective. We do not have the opportunity to measure a person’s actual social network or potential access to social resources. However, the German National Volunteer Survey (FWS), the only representative and comprehensive survey on civic participation and engagement in Germany, contains a subjective measurement of social capital: namely, the subjective size of a person’s social network, as well as the (im)material support the person can expect to receive from others. Thus, our analysis is based on Bourdieu’s idea of social capital. He proposes that the volume of social capital a person has depends on the size of the network of connections that s/he can mobilize and on the volume of capital – economic, cultural, and symbolic – that each person to whom s/he is connected possesses.

Design

Our data are drawn from the FWS, a representative survey conducted in 1999, 2004, 2009 and 2014 that asked respondents aged 14 and older living in Germany about their voluntary activities. For our analyses, we use the pooled cumulative data file.⁸ Multi-level analyses are performed in order to take regional differences into account. The question of whether regional civic participation affects an individual’s subjective social capital is addressed in two-level regressions with random intercept random slope coefficients, whereby German citizens (Level 1) are nested in subnational regions (Level 2). To model the context, we use German districts, called ‘Landkreise’.⁹ Only regions with at least 50 and 60 respondents with valid answers were included in our analyses, respectively.¹⁰ Since the dependent variables are both dichotomous, logistic regressions are performed.

Variables and controls

How much subjective social capital an individual has can be determined by his/her responses to two survey items. The first item is about the size of the individual’s social network outside of his/her own family (Wrzus et al., 2012). We assume that the more people an individual has relationships with, the greater his/her chances are of gaining access to the resources s/he needs (Borgatti et al., 1998). Thus, to measure the first dependent variable, we use answers to an FWS item that asks respondents whether their local circle of friends and acquaintances is large, medium, or limited in size (see Table 7 in the Supplemental Material for literal wording). However, this variable may be problematic because it is subjective in nature. First, each respondent is invited to estimate the size of his/her own social network depending on the local conditions. Second, we cannot be sure whether a person who reports having a limited number of friends has less social capital than another person who has a lot of friends. For example, having a large but very homogeneous social network may not provide an individual with more social resources than having a small network, because all relationships promise the same kinds of benefits. Unfortunately, we are not able to determine what exactly this item measures. We can, however, reduce the degree of differentiation by dichotomizing the outcome variable. Now, the variable no longer differentiates between the categories medium and large. Second, as a kind of validation test, we use a second variable to measure subjective social capital. If the results for both dependent variables point in the same direction, we can be more certain of this outcome variable. The FWS contains a filter variable asking whether there is anybody outside of the respondent’s household s/he can turn to for help if needed. A second dichotomous variable asks whether the respondent can turn to friends or acquaintances. This item serves as a second dependent variable.¹¹

In order to measure civic participation, we use an item in the FWS asking the respondent whether s/he has been active outside of work and family in one or more proposed social areas: for example, sports, culture, or nature protection (for the whole list, see Table 7 in the Supplemental Material). Merging the answers of this item battery, we construct a dummy variable indicating whether or not the respondent has been actively involved in a social group, a club, an initiative, or a project within the last 12 months. The percentage of active respondents is used to determine the level of civic participation¹² per region. Given that there are alternative explanations for differences in social capital, we control for several characteristics of individuals to exclude spurious effects: that is, age, gender, training status, education, financial situation, employment status, duration of residence, and whether the respondent lives in the western or eastern part of Germany (Currarini et al., 2009; Fischer, 1982; Slack et al., 2017; Wrzus et al., 2012, 2013).¹³ We also include a dummy for each point in time to control for time-varying features. The analyses incorporate all the valid answers from respondents holding German citizenship.

Analyses

Of the approximately 33,000 respondents processed, nearly 70% reported that they had been involved in a club, an initiative, a project, or a self-help group outside of work and family within the last 12 months. The histogram in Figure 3 shows an approximately normal distribution of regional participation. The bars depict the absolute frequency of the regions in which certain shares of people in this region are active. For example, the seventh bar from the left shows that in nearly 20 regions, about 60% of the respondents living in these regions have taken part in at least one form of civic activity.

Figure 3.

Distribution of regional participation

In the first step, we examine whether the regional degree of participation shows a conditional effect on the individual-level relationship between participation and having a lot of friends and acquaintances. Thus, we employ cross-level interactive terms in which active and non-active citizens are considered separately. As the question related to the dependent variable was not asked in 2014, and since we only use regions with at least 50 valid interviews, the number of regions dwindles to 114. Since the cross-level interaction term is of central interest, we grand mean-centred the macro variable and group mean-centred the Level-1 effect variable (see Bauer and Curran, 2005) in order to facilitate the interpretation of the results of the models. Since we are studying cross-level interactions and are interested in the variation in the lower-level relationships across contexts, we include the random slope (Heisig and Schaeffer, 2019). Table 1 shows the results of the first multi-level bivariate model (M01).

Table 1.

Circle of friends.

Note: Multi-level regression models (NUTS-3 level). Since the outcome variable is a binary indicator for having a medium/large circle of friends, logistics models were estimated. The bivariate model (M01) and full models (M02; M03). M01 and M02 contain the centred independent variables, and M03 contains the uncentred data. Significance: *p ⩽ 0.05; **p ⩽ 0.01; ***p ⩽ 0.001.

Source: Freiwilligensurvey (German Survey on Volunteering) 1999–2009. Author’s calculations.

We find a cross-level ‘participation-by-taking part’ effect that validates our first hypothesis (H₁). After controlling for individual participation as well as regional civic participation, the participation-by-taking part interaction term shows a positive and statistically significant effect; meaning that the gap between participants and non-participants in the likelihood of having a lot of friends is larger in regions with more widespread participation (M01). We also find that individuals have a better chance of having a large number of friends if they live in a region with a high level of regional participation. However, this effect vanishes as soon as we control for individual socioeconomic characteristics (M02). In contrast, individual participation enters the regression positively with a statistically significant coefficient (M01), which continues to hold when controlling for individual features (M02). The results also show that points in time make a difference. The respondents had significantly lower chances of having a lot of friends in 2004 and 2009 than in 1999.

Model 2 (M02) shows the results when individual characteristics are included in the regression model. The participation-by-taking part effect on the dependent variable is still positive and statistically highly significant, which means that the more vibrant the civic environment is, the better the active participants’ chances of having a large social network are relative to those of inactive citizens.

Since the effect size of coefficients from logistic regressions cannot be interpreted in an intuitively understandable way, we calculated the predicted probabilities, and plotted them across the full range of values of regional civic participation (Figure 4). This approach allows for an intuitive interpretation of the effect size, and provides an answer to the question of whether the level of regional participation makes a statistically substantial contribution to the relationship between an individual’s participation and the size of his/her social network. As a result of the centring of the central variables, we receive negative Level-2 values. In addition, the two groups in question – the active and the non-active citizens – cannot be plotted separately if we use the centred variables. Therefore, we used the regression model containing the uncentred variables to predict and plot the values (M03).

Figure 4.

Regional participation and circle of friends, and regional participation and expected help from friends or acquaintances.

The left panel in Figure 4 depicts the predicted probabilities for a large circle of friends. The crosses are the point estimates of the participants, and the hollow circles are the probabilities of the non-participants. The dashed lines above and below the point estimates reflect the 95% confidence intervals associated with the probability estimates. The figure clearly demonstrates that the active respondents are more likely than the inactive respondents to have a large social network. As the level of regional participation increases, the probability of having a large social network further diverges for these two groups. While the chances of having a lot of friends increase for active citizens, they decrease for non-active citizens. Although these trends are not statistically substantial, they are observable, and hint at the expected direction.

Table 2 shows the results for the item on the amount of help individuals can expect to receive from friends and acquaintances. As this question was asked in all four waves, we were able to use regions with at least 60 valid interviews, and still keep 130 regions to analyse. After controlling for the participation-by-taking part interaction term, as well as for individual participation, the level of regional participation is positively associated with the dependent variable (M04), but loses statistical significance when controlling for individual socioeconomic factors (M05). Again, compared to non-active individuals, active individuals have much greater subjective chances of receiving help from friends and acquaintances (M04), even when controlling for individual characteristics (M05). Likewise, after controlling for individual participation as well as regional civic participation, the participation-by-taking part interaction effect is positive and statistically highly significant. To provide an intuitive interpretation of these findings, we sketched the predicted probabilities based on the uncentred data by using the estimates of model M06. The right-hand panel in Figure 4 plots the substantive effects of regional civic participation on the relationship between individual participation and the expected level of help. As described earlier, the crosses are the point estimates for the active respondents, and the hollow circles are the predicted probabilities of receiving support from friends or acquaintances for the non-participants. The dashed lines reflect the 95% confidence intervals associated with the probability estimates.

Table 2.

Expected help from friends or acquaintances.

Note: Multi-level regression models (NUTS-3 level). Since the outcome variable is a binary indicator for expecting to receive help from friends, logistic models were estimated. Bivariate model (M04) and full models (M05; M06). M04 and M05 contain the centred independent variables, and M06 contains the uncentred data. Significance: *p ⩽ 0.05; **p ⩽ 0.01; ***p ⩽ 0.001.

Source: Freiwilligensurvey (German Survey on Volunteering) 1999–2009. Author’s calculations.

As long as the regional level of participation is relatively low (44%), participants have only a slightly higher probability than the non-participants of expecting to receive help. However, as the regional level of participation rises, the chances for the two groups develop differently: that is, while the probability of receiving support increases slightly for the participants, it decreases substantially for the non-participants. The probability of expecting support decreases from 79% [0.759; 0.816] to 73% [0.700; 0.754] for non-active respondents, as the regional level of participation increases from 44% to 84%.

We performed several robustness checks. First, we re-ran the analyses while varying the number of respondents per region; that is, by using 70, 80 or 90 respondents. Second, we re-ran the models by controlling the effects for various area-level features, like voter turnout, regional GDP, regional rate of unemployment, regional proportion of women, average age, and average duration of residence. The results are presented in Tables 5 and 6 in the Supplemental Material. Third, instead of using a dummy variable for individual participation and regional participation, it is possible to use an additive index for individual participation and the mean of this additive index to measure regional participation. From our perspective, using this measurement is not more valid than using a dummy and the share of participants per region. Nevertheless, we re-ran the models with the additive scale and the corresponding macro variable. The results are displayed in Tables 3 and 4, and the probabilities are plotted in Figure 5 (Tables 3 and 4 and Figure 5 are in the Supplemental Material). All of the findings are very similar to those we present in the main text.

Our results indicate that the non-participants’ networks and expectations of receiving help from friends – which are together interpreted as the subjective dimension of social capital − are negatively influenced by the rise in regional levels of civic participation. There are, of course, different places to find friends: in school, at university, at work, or at parents’ meetings. But we also know that clubs, initiatives and interest groups are the third-most important places to make friends (Feld, 1982; Louch, 2000; McPherson et al., 2001). Not everyone has the same opportunities to take part in civic life. Also, in periods characterized by individualization, not everybody has a big family and solid familial bonds. Thus, we can conclude that a culture of participation can make it more difficult for non-participants to accrue social capital, and can therefore lead to social exclusion.

Discussion

The present article has sought to answer the question of whether the contextual level of civic participation affects participants’ and non-participants’ social capital differently. We argued that large-scale increases in civic participation transform the quality of the social context in such a way that social relationships tend to occur within voluntary associations, leisure clubs, initiatives and interest groups. This dynamic can make it harder for inactive citizens to accrue social capital. Therefore, we expected to observe that participants in civic life are affected differently than non-participants by the contextual level of civic participation. Indeed, our findings illustrate that an increasing level of regional participation is associated with an increasing level of individual social capital for participants, and a decreasing level of individual social capital for non-participants.

Our study calls into question the assumption that there is a purely positive relationship between large-scale civic participation and social capital. We instead found that living in a vibrant environment, but not being interested in or not having the opportunity to take part in voluntary associations, can make an individual feel more isolated.

Correspondingly, having a high level of civic participation does not guarantee social cohesion; that is, if substantial parts of the population are systematically excluded from social life, both the personal sense of belonging (Fonseca et al., 2019: 247) and the ability to avoid societal inequalities will be affected. Studies have shown that members of specific social groups – such as migrants or individuals who are socially, economically or educationally disadvantaged − tend to participate less than the majority population in voluntary organizations (for example, Ishizawa, 2015). Likewise, many voluntary civic organizations – such as churches, neighbourhood organizations, and recreation clubs – are segregated by status dimensions like social class. Thus, these organizations tend to foster social network ties among in-group members while excluding outsiders, and thereby reinforce social capital deficits among low-status actors (Benton, 2016). It is also worth noting that while people who are well connected tend to be perceived as more trustworthy and trusting, people who are poorly connected tend to be more sceptical and suspicious of others (Lo Iacono, 2018: 136). When large parts of the population do not feel socially integrated, experience social deprivation or do not trust their fellow citizens, the social cohesion of the community is weakened. Our findings represent an important step towards understanding how civic participation, social capital and social cohesion are intertwined. Future research on social cohesion should take into account the possibility that civic participation may lead to social exclusion.

Another contribution of this study is that it strengthens the structural features of subnational regions in this research area. By making plausible the claim that individual behaviour first structures the context that forms a participation culture – and that this context, in turn, has retroactive effects on the individual level that influence the relationship between individual participation and an individual’s chances of accruing social capital – we show how important it is to consider the features of different action levels and to figure out how they interact with each other. In the end, for certain groups, accruing social capital becomes more difficult precisely because most people participate in civic life.

We also recognize the limits of the current study. First, we were not able to operationalize social capital in its objectified form, but instead had to rely on a subjective perspective. We are aware that the well-disposed reader may criticize this conceptualization of social capital. However, this is a problem that arises in all surveys and it cannot be solved here. In addition, we believe that the subjective approach reflects the objectified social capital a person possesses. What is most relevant, however, is that the results show that participants and non-participants are affected differently by contextual levels of civic participation in terms of their social networks and the levels of support they expect to receive from friends and acquaintances.

Another limitation relates to our use of cross-sectional data. The correlational nature of our study does not enable us to identify the precise causal direction that underlies our findings. Civic participation may lead to social exclusion and a lack of social capital, but the opposite is also possible: that is, the socially excluded may be less likely to join voluntary associations simply because they are not asked to do so. From political participation research, we know that in addition to having resources and being motivated, it is recruitment that leads to political activism (Verba et al., 1995: 513). This might also be the case for civic activism: that is, having a lot of friends and acquaintances may improve a person’s chances of being included in voluntary associations. However, a number of studies have shown that individuals also have intrinsic motivations to become engaged (e.g. Deci et al., 1999). While participation and social capital may be mutually reinforcing, we need panel data to solve this puzzle. Accordingly, our analysis offers another clear marker toward future research.

Supplemental Material

sj-docx-1-jos-10.1177_1440783320969906 – Supplemental material for The isolating side effect of civic participation

Supplemental material, sj-docx-1-jos-10.1177_1440783320969906 for The isolating side effect of civic participation by Tuuli-Marja Kleiner in Journal of Sociology

Supplemental Material

sj-docx-2-jos-10.1177_1440783320969906 – Supplemental material for The isolating side effect of civic participation

Supplemental material, sj-docx-2-jos-10.1177_1440783320969906 for The isolating side effect of civic participation by Tuuli-Marja Kleiner in Journal of Sociology

Supplemental Material

sj-xls-3-jos-10.1177_1440783320969906 – Supplemental material for The isolating side effect of civic participation

Supplemental material, sj-xls-3-jos-10.1177_1440783320969906 for The isolating side effect of civic participation by Tuuli-Marja Kleiner in Journal of Sociology

Footnotes

Acknowledgements

I am deeply grateful to the two anonymous reviewers and the JOS editors for taking the time to provide valuable comments and suggestions to improve the article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: I received financial support for editing from my employer, Johann Heinrich von Thünen Institute. Institute of Rural Studies, Bundesallee 64, 38116 Braunschweig, Germany.

ORCID iD

Tuuli-Marja Kleiner

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Tuuli-Marja Kleiner is a senior researcher at the public Institute of Rural Studies (Braunschweig, Germany). Previously, she worked as researcher and lecturer at different universities in Germany, studied for her doctorate at the University of Hamburg and habilitated on the role of culture in the relationship between citizens and the state at the Goethe University Frankfurt. Her main research areas are civic participation, political trust and public opinion polarization (published inter alia in Swiss Political Science Review, European Journal of Political Research, European Political Science). Her present research focuses on civic engagement, political participation, political culture and social cohesion. She is mainly interested in the question of how different aspects of citizenship, such as civic engagement or public opinion polarization, affect social cohesion and democracy.

References

Babchuk

Edwards

(1965) ‘Voluntary Associations and the Integration Hypothesis’, Sociological Inquiry 35: 149–62.

Bauer

D.J.

Curran

P.J.

(2005) ‘Probing Interactions in Fixed and Multilevel Regression: Inferential and Graphical Techniques’, Multivariate Behavioral Research 40: 373–400.

Benton

(2016) ‘Uniters or Dividers? Voluntary Organizations and Social Capital Acquisition’, Social Networks 44: 209–18, doi: 10.1016/j.socnet.2015.09.002.

Beugelsdijk

van Schaik

Arts

(2006) ‘Toward a Unified Europe? Explaining Regional Differences in Value Patterns by Economic Development, Cultural Heritage and Historical Shocks’, Regional Studies 40(3): 317–27.

Borgatti

S.P.

(1998) ‘SOCNET Discussion of the Origins of Social Capital’, Connections 21(2).

Bourdieu

(1986) ‘The Forms of Capital’, pp. 241–58 in J.G. Richardson (ed.) Handbook of Theory and Research for the Sociology of Education. New York: Greenwood Press.

Bradley

(1995) ‘Religious Involvement and Social Resources: Evidence from the Data Set “Americans’ Changing Lives”’, Journal for the Scientific Study of Religion 34: 259–67, doi: 10.2307/1386771.

Burt

R.S.

(2000) ‘Structural Holes versus Network Closure as Social Capital’, chapter 2 in R. Dubos, K. Cook and R.S. Burt (eds) Social Capital: Theory and Research. Taylor & Francis Group

Coleman

J.S.

(1988) ‘Social Capital in the Creation of Human Capital’, American Journal of Sociology 94: S95–S120.

10.

Charron

Lapuente

(2013) ‘Why Do Some Regions in Europe Have a Higher Quality of Government?’, Journal of Politics 75: 567–82.

11.

Council of Europe (2010) New Strategy and Council of Europe Action Plan for Social Cohesion. Approved by the Committee of Ministers of the Council of Europe, 7 July 2010. URL (consulted 28 October 2020): https://www.coe.int/t/dg3/socialpolicies/socialcohesiondev/source/2010Strategy_ActionPlan_SocialCohesion.pdf

12.

Currarini

Jackson

M.O.

(2009) ‘An Economic Model of Friendship: Homophily, Minorities, and Segregation’, Econometrica 77: 1003–45, doi: 10.3982/ECTA7528.

13.

Davidson

Sanyal

(2017) ‘Associational Participation and Network Expansion: Microcredit Self-help Groups and Poor Women’s Social Ties in Rural India’, Social Forces 95: 1695–24, doi: 10.1093/sf/sox021.

14.

Deci

E.L.

Koestner

Ryan

A.M.

(1999) ‘A Meta-analytic Review of Experiments Examining the Effects of Extrinsic Rewards on Intrinsic Motivation’, Psychological Bulletin 125: 627–68.

15.

de Tocqueville

Alexis

(2019 [1840]) De la démocratie en Amérique, Tome I et II of Oeuvres complètes. Paris: independently published.

16.

Dika

S.L.

Singh

(2002) ‘Applications of Social Capital in Educational Literature: A Critical Synthesis’, Review of Educational Research 72: 31–60.

17.

Dunbar

(1992) ‘Neocortex Size as a Constraint on Group Size in Primates’, Journal of Human Evolution 22: 469–93, doi: 10.1016/0047-2484(92)90081-J.

18.

Ellison

George

L.K.

(1994) ‘Religious Involvement, Social Ties, and Social Support in a Southeastern Community’, Journal for the Scientific Study of Religion 33: 46–61, doi: 10.2307/1386636.

19.

Feld

S.L.

(1982) ‘Social Structural Determinants of Similarity among Associates’, American Sociological Review 47: 797–801, doi: 10.2307/2095216.

20.

Festinger

(1950) ‘Informal Social Communication’, Psychological Review 57: 271–82, doi: 10.1037/h0056932.

21.

Fischer

C.S.

(1975) ‘Toward a Subcultural Theory of Urbanism’, American Journal of Sociology 80: 1319–1341.

22.

Fischer

C.S.

(1977) Networks and Places: Social Relations in the Urban Setting. New York: The Free Press.

23.

Fischer

C.S.

(1982) To Dwell among Friends: Personal Networks in Town and City. Chicago: University of Chicago Press.

24.

Fonseca

Lukosch

Brazier

(2019) ‘Social Cohesion Revisited: A New Definition and How to Characterize It’, Innovation: The European Journal of Social Science Research 32(2): 231–53, doi: 10.1080/13511610.2018.1497480

25.

Gabriel

(2001) Sozialkapital und Demokratie. Zivilgesellschaftliche Ressourcen im Vergleich. Wien: WUV.

26.

Glanville

J.L.

(2004) ‘Voluntary Associations and Social Network Structure: Why Organizational Location and Type Are Important’, Sociological Forum 19(3): 465–91.

27.

Heisig

J.P.

Schaeffer

(2019) ‘Why You Should Always Include a Random Slope for the Lower-level Variable Involved in a Cross-level Interaction’, European Sociological Review 35(2): 258–279.

28.

Hirschle

(2014) ‘Consumption as a Source of Social Change’, Social Forces 92: 1405–33, doi: 10.1093/sf/sou001.

29.

Hirschle

Kleiner

T.-M.

(2014) ‘Regional Cultures Attracting Interregional Migrants’, Urban Studies 51: 3348–64, doi: 10.1177/0042098013516687.

30.

Ishizawa

(2015) ‘Civic Participation through Volunteerism among Youth across Immigrant Generations’, Sociological Perspectives 58: 264–85, doi: 10.1177/0731121414556843.

31.

Kaasa

Vadi

Varblane

(2014) ‘Regional Cultural Differences within European Countries: Evidence from Multi-country Surveys’, Management International Review 54(6): 825–52.

32.

Kleiner

T.-M.

(2015) ‘Geographische Nähe, Landesgrenze und kulturelle Distanz. Eine Analyse regionaler Dyaden am Beispiel Österreichs’, Österreichische Zeitschrift für Soziologie 40: 281–97, doi: 10.1007/s11614-015-0173-7.

33.

Kleiner

T.-M.

(2016) ‘Why We Trust Other Nations: The Role of Cultural Values and Corruption in Generating Transnational Trust’, Comparative Sociology 15: 85–111, doi: 10.1163/15691330-12341375.

34.

Kleiner

T.-M.

(2018) ‘Public Opinion Polarization and Protest Behavior’, European Journal of Political Research 57: 941–62.

35.

Lin

(1999) ‘Building a Network Theory of Social Capital’, Connections 22(1): 28–51.

36.

Lin

(2001) Social Capital: A Theory of Social Structure and Action. Cambridge: Cambridge University Press.

37.

Lo Iacono

. (2018) ‘Does Community Social Embeddedness Promote Generalized Trust? An Experimental Test of the Spillover Effect’, Social Science Research 73: 126–45, doi: 10.1016/j.ssresearch.2018.03.001.

38.

Louch

(2000) ‘Personal Network Integration: Transitivity and Homophily in Strong-tie Relations’, Social Networks 22: 45–64, doi: 10.1016/S0378-8733(00)00015-0.

39.

Marsden

P.V.

(1987) ‘Core Discussion Networks of Americans’, American Sociological Review 52: 122–313.

40.

McPherson

J.M.

Smith-Lovin

(1987) ‘Homophily in Voluntary Organizations: Status Distance and the Composition of Face-to-face Groups’, American Sociological Review 52: 370–9, doi: 10.2307/2095356.

41.

McPherson

J.M.

Popielarz

Drobnic

(1992) ‘Social Networks and Organizational Dynamics’, American Sociological Review 57: 153–70.

42.

McPherson

Smith-Lovin

Cook

J.M.

(2001) ‘Birds of a Feather: Homophily in Social Networks’, Annual Review of Sociology 27: 415–44, doi: 10.1146/annurev.soc.27.1.415.

43.

Minkov

Hofstede

(2012) ‘Is National Culture a Meaningful Concept? Cultural Values Delineate Homogeneous National Clusters of In-country Regions’, Cross-Cultural Research 46: 133–59.

44.

Minkov

Hofstede

(2014) ‘Clustering of 316 European Regions on Measures of Values: Do Europe’s Countries Have National Cultures?’, Cross-Cultural Research 48(2): 144–76.

45.

Mollenhorst

Völker

Flap

(2008) ‘Social Contexts and Personal Relationships: The Effect of Meeting Opportunities on Similarity for Personal Relationships of Different Strength’, Social Networks 30: 60–8.

46.

Olson

(1982) The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities. New Haven, CT: Yale University Press.

47.

Pollack

(2003) Zivilgesellschaft und Staat in der Demokratie. Wiesbaden: VS Verlag für Sozialwissenschaften.

48.

Popielarz

McPherson

J.M.

(1995) ‘On the Edge or In Between: Niche Position, Niche Overlap, and the Duration of Voluntary Memberships’, American Journal of Sociology 101: 698–720.

49.

Putnam

R.D.

(1993) Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press.

50.

Putnam

R.D.

(2000) Bowling Alone: The Collapse and Revival of American Community. New York: Simon and Schuster.

51.

Roßteutscher

(2002) ‘Advocate or Reflection? Associations and Political Culture’, Political Studies 50: 514–28, doi: 10.1111/1467-9248.00382.

52.

Roßteutscher

(2008) ‘Undemokratische Assoziationen’, pp. 61–76 in A. Brodocz, M. Llanque and G.S. Schaal (eds) Bedrohungen der Demokratie. Wiesbaden: VS Verlag für Sozialwissenschaften.

53.

Roßteutscher

(2009) ‘Soziale Partizipation und Soziales Kapital’, pp. 163–80 in Kaina

Römmele

(eds) Politische Soziologie: Ein Studienbuch. Wiesbaden: VS Verlag für Sozialwissenschaften.

54.

Roth

(2003) ‘Die dunklen Seiten der Zivilgesellschaft’, Forschungsjournal Neue Soziale Bewegungen 16: 59–73.

55.

Salemink

Strijker

(2018) ‘The Participation Society and Its Inability to Correct the Failure of Market Players to Deliver Adequate Service Levels in Rural Areas’, Telecommunications Policy 42: 757–65.

56.

Schwadel

(2005) ‘Individual, Congregational, and Denominational Effects on Church Members’ Civic Participation’, Journal for the Scientific Study of Religion 44: 159–71.

57.

Simonson

Vogel

Tesch-Römer

(2017) Freiwilliges Engagement in Deutschland. Der Deutsche Freiwilligensurvey 2014. Wiesbaden: Springer VS.

58.

Slack

Cope

M.R.

Jensen

Tickamyer

A.R.

(2017) ‘Social Embeddedness, Formal Labor Supply, and Participation in Informal Work’, International Journal of Sociology and Social Policy 37: 248–64.

59.

Stolle

(2003) ‘The Sources of Social Capital’, in Hooghe

Stolle

(eds) Generating Social Capital: Civil Society and Institutions in Comparative Perspective. New York: Palgrave Macmillan, pp. 19–42.

60.

Stricker

(2006) Ehrenamt als soziales Kapital. Partizipation und Professionalität in der Bürgergesellschaft. Duisburg-Essen: Universität Duisburg-Essen.

61.

Uitermark

(2015) ‘Longing for Wikitopia: The Study and Politics of Self-organisation’, Urban Studies 52: 2301–12, doi: 10.1177/0042098015577334.

62.

Van der Meer

. (2003) ‘Rain or Fog? An Empirical Examination of Social Capital’s Rainmaker Effects’, in Hooghe

Stolle

(eds) Generating Social Capital: Civil Society and Institutions in Comparative Perspective. New York: Palgrave Macmillan, pp. 133–51.

63.

van Deth

. (2010) ‘Participation in Voluntary Associations: Dark Shades in a Sunny World?’, American Behavioral Scientist 53: 640–56.

64.

Verba

Schlozman

Brady

H.E.

(1995) Voice and Equality: Civic Voluntarism in American Politics. Cambridge, MA: Harvard University Press.

65.

Warren

M.E.

(2001) Democracy and Association. Princeton, NJ: Princeton University Press.

66.

Wiertz

(2015) A Bridge Too Far? Volunteering, Voluntary Associations, and Social Cohesion, PhD dissertation. University of Oxford.

67.

Wrzus

Hänel

Wagner

Neyer

F.J.

(2013) ‘Social Network Changes and Life Events across the Life Span: A Meta-analysis’, Psychological Bulletin 139: 53–80, doi: 10.1037/a0028601.

68.

Wrzus

Wagner

Neyer

F.J.

(2012) ‘The Interdependence of Horizontal Family Relationships and Friendships Relates to Higher Well-being’, Personal Relationships 19: 465–82.

69.

Zimmer

(1996) Vereine. Basiselement der Demokratie: Eine Analyse aus der Dritte-Sektor-Perspektive. Wiesbaden: Springer VS.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.86 MB

30.81 MB