Keywords

Introduction

Motivation to volunteer (MTV) is one of the most frequently researched topics in the field of volunteering research (Handy and Hustinx 2009). Understanding why people volunteer can provide important cues to organizations in their recruitment and retention of volunteers . The literature on why people choose and continue to volunteer is rich but mostly limited to a single country, industry, or organization. Moreover, the dominant approach is a functional one, treating MTV as an expression of preexisting needs and dispositions that precede and drive the act of volunteering. For example, the “Volunteer Functions Inventory” (VFI) developed by Clary and colleagues (Clary et al. 1998; Clary and Snyder 1999), one of the most frequently used instruments for measuring multiple motivational dimensions, assumes that MTV originates from a basic set of universal human needs that can only be met through volunteer activities. Clary and Snyder (1999) pointed out that although different volunteers pursue different goals and that a single volunteer may have multiple important motivations, all reasons for volunteering can be traced back to the universal psychological functions volunteering generally serves.

This prevailing understanding of MTV as originating from inner human drivers explains why few studies have examined how volunteer motivations are shaped by contextual characteristics. However, other social-constructionist perspectives on MTV do exist. Such perspectives consider motivational accounts as a reflection of a larger set of cultural understandings, that is, the prevailing values and beliefs in a society (Dekker and Halman 2003; Wuthnow 1991). For instance some motivations give expression to a culture of volunteering that emphasizes selfless and compassionate acts and disapproves of self-oriented or egoistic orientations. In this perspective, motives do not precede action, but help to frame and justify our actions by referring to the broader set of cultural understandings. Motives, specifically “motive talk” (Wuthnow 1991) , are “constitutive of action, part of a discourse giving meaning to and helping to shape behavior” (Wilson 2000, p. 218).

An essential assumption therefore is that the context influences the use of motives and hence that important differences depending on the societal context may occur. In his classic book Acts of Compassion , Wuthnow (1991) very extensively describes how the unique context of “American individualism” makes volunteers struggle to find a balance between altruistic and utilitarian accounts of their caring activities. As Wuthnow notes, “an adequate language of motivation is thus one of the critical junctures at which the individual and the society intersect: being able to explain why is as important to our identity as a culture as it is to our sense of selfhood as individuals” (Wuthnow 1991, p. 50– emphasis added).

A more contextual understanding of MTV further explains changes in the prevalence of certain motivations. In Western European societies, there has been a growing conviction that, due to modernization and secularization , “traditional,” religiously inspired and other-oriented volunteering is gradually being replaced by “new,” more individualized and self-interested types of involvement . As a result, volunteers, especially from younger generations, are less inclined to provide altruistic reasons for volunteering (Hustinx and Lammertyn 2003).

In this study, we aim to extend our contextual understanding of MTV by examining cross-national differences in the motivations of volunteers . We assess if and how specific societal characteristics are associated with self-reported motivations to volunteer. In particular, we will focus on the role of religion (individual religiosity and religious context), positing a major link with the importance of altruistic MTV. Some of the alternative and competing hypotheses that will be explored within the context of more secular societies are the broader cultural framework, focusing on the dominant value pattern as well as the extent of institutional variations in welfare state regimes and the characteristics of the nonprofit sector .

To examine our hypotheses, we use the second wave of the World Values Surveys (WVS 1990) , which includes a series of questions on participation in voluntary work and the main reasons for doing so. For our analysis, we selected 18 countries based on the availability of contextual data for the year 1990. To our knowledge, the 1990 WVS survey is the only cross-national survey that included a question on volunteer motivation. Given that the data was collected more than two decades ago, our study has a major limitation: Our findings do not reflect the motivational accounts of contemporary volunteers or current contextual factors, but rather provide a test for a number of theoretical assumptions. As a result, the main contribution of this study will be to improve our understanding of contextual factors influencing MTV, an approach that is underdeveloped in the current literature.

Literature Review

MTV is a well-researched topic (Wilson 2000). Much of the research has been conducted either at the national level, using representative samples, or at the organizational or sector level, using volunteers both in specific activities and those involved with particular organizations (Musick and Wilson 2008). Regardless of this diversity in the study of MTV , scholars have consistently found MTV to be a complex interplay that includes both altruistic and self-interested accounts (Cnaan and Goldberg-Glen 1991; Wuthnow 1991) .

However, existing research indicates that the importance attached to both of these motivational dimensions differs across individuals and groups. For example, it is well established in the literature that the MTV of youth differ from other age- groups (Handy et al. 2010). Gillespie and King (1985) found that a greater ­proportion of older volunteers reported giving time for altruistic reasons such as to “help others” and “contribute to the community.” By contrast, younger volunteers more often expressed MTV in order to acquire training and skills. In a national survey of Canadians, volunteer rates were highest among youth, who also put stronger emphasis on self-interested motivations than other age cohorts. For example, 65 % of 15- to 19-year-olds versus 13 % of those 25 and older reported volunteering to improve their job opportunities (Hall et al. 2006). Among the student population, Winniford et al. (1995) found that American college students said that they volunteered primarily because of altruistic concern for others, although they also stated that they sought to satisfy self-fulfillment and development needs (e.g., affiliation, sense of satisfaction and development of career skills). In addition, Dickinson (1999) reported that in the UK, students who volunteer interpreted it as a conscious attempt to enhance their chances of success in finding postgraduate employment.

In explaining variations in motivational accounts, altruistic reasons for volunteering are primarily connected to religion and religious belief . Altruism is a key value taught by many religions. A sense of selflessness and duty towards the poor is central to all major religions. In essence, it urges religious people to engage in social activities such as volunteering on behalf of others in need (Batson et al. 1993; Cnaan et al. 1993; Ellison 1992; Graham 1990; Wuthnow 1990, 1991; Wymer 1997) .

Thus, religious involvement may change the nature or priority of people’s motives (Weiss Ozorak 2003; Wilson and Janoski 1995). Its role is educational, sensitizing people to social concerns on which they might not otherwise focus (Weiss Ozorak 2003). There has been some tendency to relate the spirit of altruism to particular religious traditions, most commonly the Judeo–Christian tradition rooted in the Old Testament commandment to “treat your neighbor as yourself” (Leviticus, Chap. 18; in Salamon and Sokolowski 2009). Wuthnow (1991) found a strong relationship between familiarity with the story of the Good Samaritan (Luke 10:30–36) and doing good oneself.Footnote 1

The opportunity to express religious beliefs and values is thus an important function of volunteering (Wood and Hougland 1990; Wymer 1997), and it also predicts whether volunteers complete their expected period of service (Clary and Miller 1986; Clary and Orenstein 1991). In the USA, expressing religious beliefs or responding to a moral obligation based on religious beliefs is among the top three motives for giving and volunteering (Wymer 1997).

Research on the relationship between religion and volunteering has, however, revealed that it is not religious conviction but rather religious practice that constitutes a key determinant for volunteering. In other words, religious convictions are fostered through active participation in a religious community (Lam 2002; Lim and MacGregor 2012; Ruiter and De Graaf 2006; see also the chapter by Bennett in this book). It is through religious practice that social networks among fellow members of the religious community are built and that information is shared. Active members of religious communities are thus more likely to learn about volunteering ­opportunities and to be asked to volunteer. Consequently, in research on religion and volunteering, religious attendance is used as a key predictor (Lim and MacGregor 2012).

In this chapter, we do not focus on volunteer behavior but on motivations to volunteer . Nevertheless, we also expect to see some influence of religious attendance: Through social interaction and interpersonal influence among individuals within a “moral community” (Stark and Bainbridge 1996), shared norms and values are strengthened and motivations and discourses are regulated.

Given that the endorsement of altruism is universal among all the world’s major religious traditions, we hypothesize:

  • H1a: Religious people express a stronger support for altruistic MTV and a weaker support for self-oriented MTV . We expect a positive association with both personal beliefs and service attendance.

  • H1b: There is no difference between the various religions with respect to their effect on altruistic MTV.

The (increasing) importance of self-interested MTV, on the other hand, could be explained on the basis of theories of modernization and value change. It has been argued that as a result of processes of advanced modernization, secularization, and individualization, present-day volunteers put increasing emphasis on self-oriented reasons for their involvement, at the expense of altruistic reasons (Hustinx and Lammertyn 2003). From this perspective of social change, “traditional” volunteering was embedded in a religious tradition of benevolence and altruism . Dedication to the common good was a highly esteemed asset to which deviating individual motivations were easily subordinated. By contrast, in a more individualized context, traditional loyalties weaken and the interaction between an individualized biography and volunteer experience intensifies . As volunteering becomes increasingly embedded in self-authored individualized narratives, it becomes a tool for self-actualization or “life(-style) politics” (Bennett 1998). The volunteering field is seen as a “market of possibilities” (Evers 1999) for self-realization and the setting of personal goals.

This shift in motivational accounts can be linked to theories of value change in general, most notably Inglehart’s theory on postmaterialist value change (Inglehart 1971, 1997; Inglehart and Welzel 2005) . The basic assertion of Inglehart is that among Western populations, a gradual change in values, from materialist to postmaterialist, has been occurring through generational replacement. These generational differences can be traced back to different socialization experiences during the formative years. Whereas older cohorts experienced the economic deprivation of wartime as well as the Great Depression and the mutual efforts to rebuild society, younger cohorts were raised in times of economic prosperity and a growing emphasis on individual autonomy and self-expressive values (Inglehart 1997; Inglehart and Welzel 2005).

Not only has the economic well-being of the average citizen increased objectively, but so has their sense of existential security. As a consequence, citizens develop new value priorities (Delhey 2009). Value change is observed as occurring along two axes: from traditional authority to secular rational values, and from survival values to self-expressive values. The younger age cohorts no longer stress values such as economic growth, the fight against rising prices or crime rates, obedience and trust in (religious) hierarchies; rather, they prioritize more secular and self-expressive values such as tolerance, freedom of speech, environmental protection and individual fulfillment. Support for freedom of expression, in addition to tolerance of ethnic or sexual minorities, is found to be stronger and more widespread among the younger age cohorts (Stolle and Hooghe 2005) .

While self-expression values are associated with higher levels of individualism , Welzel (2010) notes that disagreement exists about whether these values are of a civic nature or not. Scholars have argued in both directions. On the one hand, individualism is easily equated with more self-oriented attitudes and behavior, hence with egoism . On the other hand, Welzel argues that since self-expression values imply a basic sense of human equality, it enables universal feelings of solidarity. Therefore, Welzel argues for self-expression values as a civic or socially responsible form of modern individualism. Other authors have also argued in favor of a “solidary individualism” (Berking 1996) or “altruistic individualism” (Beck 1997) that can constitute a seemingly contradictory motivational basis for present-day volunteering (Hustinx 2001; Hustinx and Lammertyn 2003) .

Using data from the World Values Study, Welzel (2010) found that the association between self-expression values and altruism is mixed. Higher country levels of self-expression were strongly associated with higher levels of altruism . At the individual level, however, the association with egoism/altruism was U-shaped: In the lower range of the self-expression values scale (i.e., respondents that scored weaker on this scale), increasing support for self-expression values was associated with stronger egoism, whereas in the upper range of the scale (i.e., respondents that scored stronger on this scale), increasing support for self-expression values was associated with stronger altruism. While this pattern confirms neither the civic nor the “uncivic” interpretation of self-expression values, Welzel argues that it more clearly supports the civic interpretation because stronger self-expression values are associated with stronger altruism, not egoism, especially at high levels of these values (Welzel 2010, pp. 13–15).

In sum, the emergence of a self-expressive value pattern could be linked to altruistic as well as self-interested reasons for volunteering. According to the “civic” interpretation, both types of motives could be easily combined, while the “uncivic” interpretation considers them as mutually exclusive. We therefore formulate two competing hypotheses:

  • H2a: Volunteers with a self-expressive value pattern will put less emphasis on altruistic motivations and more emphasis on self-interested reasons for volunteering.

  • H2b: Volunteers with a self-expressive value pattern will put a stronger emphasis on both altruistic motivations and self-interested reasons for volunteering.

Beyond these individual factors, important contextual influences may be at play as well. Ruiter and De Graaf (2006, pp. 193–194) note that the relation between the national religious context and volunteer work is somewhat neglected in the literature. Based on Kelley and De Graaf (1997), they develop arguments to predict a positive impact for the degree of devoutness of a society on volunteer participation. Kelley and De Graaf (1997) found that people who were raised by secular parents in relatively devout countries were more religious than people who grew up with similar parents in more secular countries. The authors explained this through people’s exposure to religious culture and their pools of potential friends, teachers, colleagues, and marriage partners who would be predominantly devout. Ruiter and De Graaf (2006) expected this “spillover” effect on nonreligious people for volunteering as well. Moreover, a religious context exerts a socialization effect on secular people as the likelihood of encountering religious people in one’s personal social network would be greater while the impact of individual religiosity would be weaker in more devout societies . We further argue that having more religious people in one’s network also increases the exposure to religious beliefs and values such as altruism. Hence, we expect a higher likelihood of altruistic motivations to volunteer in a more religious national context.

While Ruiter and De Graaf (2006) found support for the “spillover hypothesis” based on data from the World Values Study, their findings were not reproduced using the Gallup World Poll data that includes a larger number of countries, pointing to higher rates of volunteering in both secular and highly devout societies (Lim and MacGregor 2012, Bennett in this volume). While this shows that the results are sensitive to the countries included in the analysis (Van Der Meer et al. 2010), Lim and MacGregor (2012) further argue that although the average service attendance in a country is commonly used to test the network spillover hypothesis, such a national average is a poor proxy for the influence of religiosity in the personal networks of individuals. They formulate several reasons: Religion is not evenly distributed geographically; homophily among the nonreligious may be higher in religious environments, hence levels of segregation might be higher in devout areas compared to secular ones; recruitment efforts of religious organizations are more likely to be targeted at religious people; and finally, interpersonal influence is based on a shared identity and thus might be more effective when individuals share a religious faith. Furthermore, Lim and MacGregor (2012) indicate that the average service attendance of a country could relate to individual volunteering through mechanisms other than network spillover. In more devout countries, where a national religious culture may influence people’s likelihood of volunteering through public discourse and the media, a higher organizational density may exist with more volunteer opportunities. Thus, even while contextual effects are present, other mechanisms than network spillover could exist and it is very difficult to disentangle these different mechanisms. In their own study, using data from the Gallup World Poll, Lim and MacGregor (2012) found evidence for the existence of a national religious culture , rather than a spillover effect.

Taking into account different contextual mechanisms, we can safely assume that secular volunteers in a devout society will express more support for altruistic values compared to their secular counterparts in secular societies . Altruism is a more central part of the prevailing religious culture and could be fostered through the higher likelihood of religious persons being present in the personal networks of individuals (Bellah et al. 1985 ; Lim and MacGregor 2012; Ruiter and De Graaf 2006; Wuthnow 1991):

  • H3: A religious national context will be associated with a stronger emphasis on altruistic MTV

Besides the religious context, we also expect the dominant value pattern to influence MTV. Following Inglehart (Inglehart 1997; Inglehart and Welzel 2005) , we expect self-expressive values to prevail in more secular societies. Parallel to the hypotheses of network spillover and national culture in the case of a religious national context, similar mechanisms could apply with respect to the dominant value pattern in a country. As argued above, however, existing perspectives predict an association with altruism or egoism. We therefore formulate two competing hypotheses:

  • H4a: A national context in which the dominant value pattern is postmaterialist will be associated with a stronger emphasis on self-interested MTV.

  • H4b: A national context in which the dominant value pattern is postmaterialist will be associated with a stronger emphasis on both altruistic and self-interested MTV.

Besides these variables involving cultural context, institutional explanations for motivational differences could be formulated. For this argument, we relied on three cross-national examinations of MTV that looked at differences in the particular welfare regime of a country (Hustinx et al. 2010; Hwang et al. 2005; Ziemek 2006). First, Hwang et al. (2005) compared MTV between Canada and the USA and found Americans more likely to mention altruistic MTV, while Canadians were more likely to emphasize self-interested reasons. To explain these differences, the authors argued that while both countries are liberal democracies, Canada’s government provides more extensive social welfare programs (such as universal health care and aid to vulnerable groups) than the US government. Thus, volunteers in the USA see helping the poor and disadvantaged as part of their role as citizens and are more likely to report altruistic MTV than Canadian volunteers who see this role fulfilled by their government (the authors controlled for individual religiosity, for it should be noted that the USA is a far more religious country than Canada). A second study by Ziemek (2006) examined MTV across countries with different levels of economic development , namely, Bangladesh, Ghana, Poland and South Korea. Clustering MTV into three categories, “altruism,” “egoism,” and “investment in human capital,” she tested the differences in MTV through the volunteer’s perceived level of public spending. Perceptions of high public spending were found to negatively influence altruistic MTV and positively influence investment motivation .

A more recent study on student volunteers across six countries suggests that MTV is also influenced by regimes, albeit partially (Hustinx et al. 2010). The latter study applied the social origins theory , advanced by Salamon and Anheier (1998) and predicated on Esping-Andersen’s (1990) “worlds of welfare capitalism.” This theory explains the size and development of the nonprofit sector as an outcome of broadly defined power relations among social classes and social institutions. In brief, social origins theory identifies four different regimes: liberal, social ­democratic, corporatist, and statist, with corresponding levels of government social welfare spending and nonprofit sector size ranging from high to low. In addition, the social origins theory examines the role nonprofit organizations serve in a society (Salamon and Sokolowski 2003). Depending on the regime, the nonprofits are more likely to provide some of the services that have an instrumental value to society or expressive services that are the actualization of values or preferences, such as the pursuit of artistic expression, preservation of cultural heritage or the natural environment.

At one end, in the liberal model, low government spending on social welfare services is associated with a relatively large nonprofit sector mainly focused on service provision . At the opposite end is the social-democratic model in which high government spending on social welfare results in a limited role for service provision by nonprofits, but a larger role for the expression of political, social, or recreational interests. In addition, corporatist and statist models are both characterized by strong states, in which the state and nonprofits are partners in the corporatist model while the state maintains the upper hand in many social policies in the statist model. In both models, the service role is dominant.

Across these four types of regime, the relationship with volunteering is not linear (Salamon and Sokolowski 2003). There are two regime types in which the amount of volunteering is high. The social-democratic regime has a distinct pattern of high levels of volunteering, but mostly in expressive rather than service roles. In the liberal regime, participation in volunteering is also very high yet mainly located in serviced-oriented sectors that are underserved by public workers. The corporatist regime also produces a much more service-oriented pattern of volunteering , yet with moderate levels of volunteering given the substantial amount of paid staff. Finally, in the statist regime, volunteering is largely underdeveloped.

In addition to the varying rates of volunteering, we suggest that MTV will also differ in different regimes, and that a systematic link can be found between the regimes and the primary MTV. Following Hwang et al. (2005), we hypothesize that volunteers are most likely to report altruistic MTV when they provide services that are underserved by government, that is, when nonprofits fulfill a primary role in the welfare production of a country. Based on social origins theory , this will most likely be the case if a nonprofit regime is characterized by (1) a revenue structure with low government spending; (2) a large nonprofit sector with a small paid workforce and a large unpaid workforce; and (3) service provision as the dominant volunteering type. These characteristics correspond to the liberal regime. In clear contrast are the social-democratic and corporatist regimes, which both heavily rely on government support for the sector. In the former, volunteering is largely expressive in form, while in the latter, a majority of volunteers is involved in service provision but their role is moderate and auxiliary. Hence, we expect altruistic MTV to be the weakest in the social-democratic regime and moderate in the corporatist regime. Finally, the statist regime is characterized by limited growth in both government social spending and nonprofit activity; moreover, nonprofit organizations lack the type of autonomy and resources typical of Western democracies. Nevertheless, existing volunteers mainly provide services that are underserved by government; thus, we predict moderate support for altruistic MTV. Given that the social-democratic regime is the only regime in which the expressive role of volunteering is dominant, we predict that self-interested reasons for volunteering will be the most prevalent in this regime .

In sum, we hypothesize that:

  • H5a: Support for altruistic MTV is the weakest in the social-democratic regime, and the strongest in the Liberal regime. Corporatist and statist regimes express moderate support for altruistic MTV.

  • H5b: Support for self-interested reasons for volunteering is the strongest in the social-democratic regime.

Data and Methods

Sample

We use data from the 1990 wave of the WVS . To our knowledge, this is the only cross-national survey which includes a question on motivations to volunteer. As already mentioned in the introduction, the data mainly allows for testing theoretical assumptions about contextual influences on self-reported MTV (i.e., a test of cultural vs. institutional explanations), and does not provide an up-to-date empirical picture.

The 1990 wave includes data from 40 countries worldwide. We selected only those countries for which valid measures on all dependent and independent variables were available. More specifically, the countries selected for this study are those countries (1) in which questions on volunteer participation and motivations to volunteer were asked; and (2) that were included in the early wave of the Johns Hopkins Comparative Nonprofit Sector Project . This made data available for the size of the nonprofit sector, the composition of the workforce, dominant functions, and sources of revenue within the national nonprofit regimes, all as close to the year 1990 as possible. As a result, 18 countries were included in the analysis (West and East Germany still counted separately), with a total of 7,186 respondents who indicated a willingness to volunteer at the time of the survey (27.1 % of the total sample). Appendix A6.1 provides an overview of the number and proportion of volunteers per country.

Variables

Our key dependent variable is the motivation to volunteer. In the WVS survey, the first question determined whether respondents were currently doing unpaid work for any organization taken from a list of 16 types of organizations. In a subsequent step, for those respondents who indicated that they were doing unpaid work in any of these organizations, their reasons for doing voluntary work was asked using a 5-point Likert-type question format (1 = unimportant, 5 = very important). The 14 reasons for doing voluntary work included statements ranging from “a sense of solidarity with the poor and disadvantaged” to “purely for personal satisfaction,” thereby reflecting both “altruistic” and “self-interested” dimensions of the motivation to volunteer . A principal component analysis confirmed that the different MTV clustered into these two dimensions. We treat both measures as summated scales.

Altruistic reasons for volunteering (Cronbach’s alpha.81) include seven items: (1) a sense of solidarity with the poor and disadvantaged; (2) compassion for those in need; (3) an opportunity to repay something, give something back; (4) a sense of duty , moral obligation; (5) identifying with people who were suffering; and (6) to help give disadvantaged people hope and dignity. We dropped an item about “religious beliefs” as altruistic MTV as this might have artificially inflated the effects of the religion variables on altruistic MTV.

Self-interested reasons for volunteering (Cronbach’s alpha.66) include four items: (1) time on my hands, wanted something worthwhile to do; (2) purely for personal satisfaction; (3) for social reasons, to meet people; and (4) to gain new skills and useful experience.

The independent variables are situated both at the individual and the country level.

Background Characteristics

First of all, we account for relevant socio-demographic variables: gender (ref = male), age (continuous), education level (age at which education completed, divided into ten categories ranging from 1 = 12 years of age or earlier; 10 = 21 years of age or older), marital status (ref = married/cohabiting vs. single, divorced/separated/widowed), employment status (ref = working vs. unemployed, student, housewife/husband, retired, other), and political orientation (10-point scale, 1 = left, 10 = right).

Religiosity

Next, we measure individual religiosity by means of three measures. Firstly, we look at religious membership , which is measured by questioning whether people belong to a religious denomination and, if so, which one. The questionnaire included the following options: Roman Catholic (41,8 %), Mainline Protestant (25,2 %), Fundamentalist Protestant (3,6 %), Jew (0,4 %), Muslim (0,1 %), Hindu (0,4 %), Buddhist (0,8 %), other (5,8 %), never (21,6 %), and no answer (0,4 %). We assign respondents to the following five categories: Catholic , other Christian, other (non-Christians), none (nonreligious), and missing. While we acknowledge that these broad categories do great injustice to the existing religious diversity , most religious denominations have too low a number of observations to be considered separately in the analysis.

Secondly, we assess the influence of religious service attendance by asking respondents how often they attended religious services, apart from weddings, funerals and christenings (ranging from more than once a week to never or practically never).

Finally, we consider professed closeness to God as a measure of individual religiosity, by including a question that asked how important God is in the respondent’s life, which was assessed on a scale from 1 (not important at all) to 10 (very important).

At the contextual level, we include the mean religious attendance and the mean level of closeness to God in a particular country.

Value Patterns

Besides religiosity, we look at individual and collective value patterns. As argued above, while we hypothesize that religiosity is correlated with altruistic MTV , support for postmaterialist values will be associated with self-oriented reasons for volunteering. We use the Inglehart measure of postmaterialism , which was included in the 1990 wave of the WVS. This scale is based on a series of three questions about what the respondent thinks that the aims for his/her country should be for the next 10 years. In each of the questions, the respondents are presented with two choices that represent a materialist value pattern (e.g., “maintaining order in the nation”) and with two that represent a postmaterialist one (e.g., “protecting freedom of speech”). The final score on the postmaterialism scale is the count of the number of postmaterialist items over these three questions that were mentioned as first or second choice (“high” priority) from the given group of four goals.

At the country level, we integrate the mean postmaterialism score in the analysis.

Nonprofit Sector Regime

A final contextual variable assesses cross-national variation in MTV as a function of institutional variations in the national nonprofit sector regime. As indicated above, Salamon and Anheier’s social origins theory is predicated on Esping-Andersen’s (1990) “worlds of welfare capitalism”; hence, the different nonprofit regimes are embedded in the broader welfare state regimes . Therefore, we first include a measure of the welfare state regime types based on Esping-Andersen’s seminal work (1990). He identified three models: liberal (Anglo-Saxon), social democratic (Scandinavian), and conservative corporatist (continental Europe). We add an Eastern European type to include these countries in our analysis .

In addition, to assess the influence of the country-specific nonprofit sector regime more exactly, and based on the discussion above, we use four indicators that are available from the Johns Hopkins Comparative Nonprofit Sector Project (CNP; Salamon et al. 1999, 2004) and which were gathered in the early period of the project, so as to match the WVS data of the 1990 wave as closely as possible. The indicators included in our analysis are:

  • Percentage of Civil Society Organization (CSO) workforce as share of economically active population

  • Percentage of volunteer share of CSO workforce (expressed in full-time equivalents, FTEs)

  • Percentage of source of CSO revenue: government (vs. fees and philanthropy)

  • Dominant function of the nonprofit sector: service or expressive

Because the CNP estimates the amount of volunteer labor in an aggregated way, generating a count of the total amount of volunteer effort in terms of FTEs, we ­additionally look at the mean percentage of volunteers in the population based on the WVS survey.

Results

In a first step, we look at the distribution of MTV across countries. Figure 6.1 shows that the motive mix differs depending upon the national context. If we consider a mean score above 3.50 on a scale from 1 to 5 as a measure of the importance of one of the two motivational dimensions, it is only in Austria that both MTVs are (very) important. Altruistic motivations are also important in the USA, East Germany, Great Britain and Northern Ireland, and self-interested reasons for volunteering play an important role in Finland as well. There is also one country, Romania, in which none of the motivations are important (mean score below 2.75 for both motivational dimensions).

Fig. 6.1
figure 1

Scatterplot of motivation to volunteer (MTV) by country

Full size image

When one considers the mean scores for both dimensions we can clearly discern a cluster of Anglo-Saxon countries (USA, Great Britain, and Ireland) that score average to high on altruism and low to average on self-interested benefits. Most other countries present the reversed pattern, with a low to average score on altruism and an average to high score for self-interested reasons for volunteering.

In a second step, we aim to explain variations in MTV by means of a causal analysis. In Table 6.1, we present the results of a set of multilevel linear regression models, with altruistic and self-interested reasons for volunteering as dependent variables. Model 1 includes individual-level predictors only, model 2 contains both individual-level and country-level variables, and model 3 shows the most parsimonious model with only significant individual and country-level variables.

Table 6.1 Multilevel linear regression model for altruistic motivations and self-interested motives for volunteering
Full size table

At the individual level, looking at the indicators of individual religiosity , there is no influence of denominational differences. Religious service attendance and the importance of God in one’s life, on the other hand, are significantly and positively associated with altruistic MTV, while a more frequent service attendance also weakens the importance attached to self-interested MTV. These findings confirm hypotheses H1a and H1b.

Next, a postmaterialist value pattern does not relate to altruistic MTV and is weakly and negatively associated with self-interested motivations. This disconfirms H2a, but as there is no pronounced relationship between self-expressive values and altruism, it does not support H2b either. In our analysis, we also included political orientation as a measure of broader value orientations. The results show that volunteers who identify themselves as more left wing report more altruistic MTV .

We furthermore observe a number of significant relationships with the socio-economic background characteristics of volunteers. Female and older volunteers put significantly stronger emphasis on altruistic motivations than male and younger ones, and younger volunteers indicate significantly more frequently that they are motivated by self-interested reasons than older volunteers. Higher educated volunteers express significantly weaker support for both altruistic and self-interested reasons for volunteering in comparison with lower educated ones.

Employment status is mainly associated with self-interested reasons for volunteering. In comparison with volunteers who are employed, unemployed volunteers, students, housewives, and retired volunteers are significantly more strongly motivated by self-interested reasons for volunteering. Housewives also value altruistic MTV higher. A higher income scale is negatively associated with both altruistic and self-interested reasons for volunteering. There is no relationship between marital status and MTV.

At the country level, we have included the mean importance of God and the mean religious service attendance as measures of the level of devoutness in a particular national context. The mean importance of God has a negative influence on both altruistic and self-interested reasons for volunteering; the mean religious service attendance does not influence self-reported MTV. Hypothesis H3 could thus not be confirmed. Our analysis further reveals that a national context in which the dominant value pattern is postmaterialist has a positive influence on self-interested MTV, confirming hypothesis H4a and disconfirming H4b.

While these variables measure the influence of the cultural context, our analysis also includes measures of national differences in institutional welfare regimes. As hypothesized in H5a, support for altruistic MTV is the weakest in the social-democratic regimes; however, no differentiation exists among the other regimes. Contrary to H5b, welfare state regimes do not differ in their population’s support for self-interested MTV. More specific characteristics of the particular nonprofit regime partially relate to MTV. The revenue structure of the nonprofit sector influences MTV: As predicted, the more government support for the nonprofit sector, the less volunteers express altruistic MTV. The size of the CSO workforce negatively influences self-interested MTV yet is not associated with altruistic MTV. The relative share of volunteers in the CSO workforce (measured in terms of FTEs) relates to altruistic MTV but in the opposite direction of what we hypothesized: the larger the share of volunteers in the CSO workforce, the weaker volunteers’ support for altruistic MTV. The proportion of volunteers in the population, on the other hand, has a positive influence on both altruistic and self-interested reasons for volunteering. The dominant function of the nonprofit sector is not related to reported MTV.

Discussion and Conclusion

In this study, we examined whether and how motivations to volunteer are shaped by contextual characteristics. In contrast to prevailing understandings of MTV in terms of inner human drivers, we approached MTV in a social-constructionist way, considering motivational accounts as a reflection of prevailing values and beliefs in society. A more contextual understanding of MTV further allows examination of how changes in the emphasis put on certain motivations are linked to broader social transformations. In the volunteering literature, a shift from altruistic to more self-interested or instrumental MTV has been described, which can in turn be linked to broader societal processes of secularization and changes in values.

As a result, in this chapter, we examined the influence of national context on motivations to volunteer using data from the second wave of the WVS 1990 , including 18 countries in the analysis and with a total of 7,186 respondents who volunteered. Up to the present, only few studies have endeavored to carry out such a cross-national examination of MTV. On the one hand, we focused on the broader cultural framework, understood in terms of the national religious context and the dominant value pattern. On the other hand, we assessed how institutional variations in terms of welfare state regimes and characteristics of the nonprofit sector affect motivations to volunteer.

Across all countries studied, people who volunteered expressed both altruistic and self-oriented motivations, a finding that is consistent with previous studies. Nevertheless, we observed important variations in the emphasis that was put on both motivational dimensions depending on individual background characteristics and variations in the national context. In Table 6.2, the key hypotheses and corresponding empirical findings are summarized.

Table 6.2 Summary of the findings
Full size table

First, at the individual level, we assessed the role of religion and religiosity in explaining MTV. As discussed above, altruism is a value that is central to all religions; hence, religious involvement may influence or change people’s chief reasons for volunteering. As expected, we did not find any differences between belonging to different religious traditions (however, we noted limited variation in the religious membership in the countries selected for our analysis in which Christian affiliations dominate), but personal closeness to God and religious service attendance increased the importance of altruistic motivations to volunteer. A more frequent service attendance reduced the emphasis put on self-interested MTV. Thus, religious people seem to have internalized other-oriented values, and they acquire these values in places of religious worship and through their stronger integration in religious networks . At the individual level, motivational accounts, and more specifically the emphasis put on altruistic versus self-interested reasons for volunteering, can therefore be related to both religious conviction and practice .

At the country level, we examined the influence of a national religious context, assuming that in a devout society both religious and secular volunteers will express more support for altruistic MTV. This hypothesis was based on both the religious culture and network arguments : In a religious country, altruism will be more central to the general value pattern and it will be more likely that religious people are part of the social network of secular volunteers. Contrary to our expectations, our analysis showed, first, that the mean religious service attendance in a country did not affect reported MTV, and second, that a more religious national context in terms of a larger segment of the population emphasizing the importance of God in their life had a negative influence on both altruistic and self-interested MTV. There are two possible explanations for the negative effect on altruistic MTV: First, when you know that there are a lot of religious people in your environment, you may assume that helping the poor and disadvantaged will be part of their role as citizens; hence, you feel less inclined or obliged yourself to volunteer for altruistic reasons. Another explanation could be that individual religiosity is contained in the private sphere, as something personal; hence, there is little religiously inspired “motive talk.” On the other hand, the negative influence on self-interested reasons for volunteering could be explained by the fact that, in a religious context, to “gain” something from volunteering is met with disapproval. In other words, in a more devout country, self-oriented MTV is considered inappropriate and volunteers are less likely to report such motivations. This hampering effect of a high level of religiosity in a country on the support for self-interested MTV could be interpreted as a contextual effect of national religious culture . Surprisingly, however, there is no association between a high level of religiosity in a country and support of volunteers for altruistic reasons. Combined with the lack of influence of mean church attendance, the strong connection that exists between individual religiosity and altruistic MTV is not strengthened at the contextual level. That average church attendance is not associated with support for altruistic MTV suggests that more extended religious networks do not necessarily lead to a greater exposure to religious culture or greater interpersonal influence. This seems to confirm Lim and MacGregor’s (2012) argument that such a national average is a poor proxy for the influence of religiosity in the personal networks of individuals.

While, on the one hand, we expected a strong link between national religious context and altruistic MTV, on the other hand, a postmaterialist value pattern, commonly associated with higher levels of individualism , was linked to self-interested motivations. However, arguments could be made for a positive association between self-expressive values and altruistic MTV as well. In our analysis, we observed effects both at the individual and country level, but these effects go in opposite directions. Individuals with a stronger postmaterialist value pattern put less emphasis on self-interested reasons for volunteering, while countries with a stronger postmaterialist value pattern are more likely to express self-interested motivations. Similar to Welzel (2010), we thus found no conclusive evidence regarding the “civic” or “uncivic” nature of postmaterialist values. Furthermore, Welzel found mixed associations. While Welzel observed that higher country levels of self-expressive values were associated with higher levels of altruism, we found an opposite pattern. Based on our analysis, we may in any case conclude that both at the individual and country level, postmaterialist values do not seem to be at odds with altruistic MTV; they neither stimulate nor hamper support for such other-oriented reasons for volunteering.

Finally, we also looked at variables determined by institutional context as an alternative explanation for cross-national variations in MTV . Here we find partial evidence. As suggested in earlier exploratory studies, we find that when government social spending is high, as in social-democratic welfare regimes, altruistic MTV receives less support. This finding could also be linked to the fact that in a social-democratic regime, the nonprofit sector performs a more expressive role, given that most services are provided by government (Salamon et al. 2004). Our findings also show that the higher a government’s share in the revenue structure of the nonprofit sector in a particular country is, the less likely volunteers are to express altruistic motivations. The total size of the nonprofit sector, in terms of the CSO workforce as a share of the economically active population, has a negative influence on ­self-interested MTV. While we had argued that a larger nonprofit sector would increase the support for altruistic MTV, this is not the case; the presence of a large organizational universe that represents social goals and values, embodied by (un-)paid workers, seems to moderate self-interested accounts of volunteer service. Contrary to our expectations, the dominant function of the nonprofit sector did not have an effect on MTV. Regarding the relative share of volunteers within the CSO workforce , the higher the total amount of volunteer labor , in terms of full-time equivalents, the less emphasis that is put on both altruistic and self-interested reasons for volunteering—a finding for which we do not have an explanation; possibly, this may be too abstract a measure that is not connected to individual volunteers’ perceptions of the characteristics of the nonprofit sector. As already noted, the relative share of volunteer labor in the total CSO workforce is an aggregate measure, estimating the total number of hours given by volunteers. This measure creates an abstraction of the actual number of volunteers. Indeed, the effort of several volunteers is necessary to arrive at one FTE. Therefore, we also looked at the mean percentage of volunteers in the population. The higher the percentage of volunteers in the population, the more inclined volunteers are to emphasize altruistic motivations to volunteer. This may point to the embedded nature of volunteering to help others: a more general culture of voluntarism, as we hypothesized .

Although not the focus of our study, our results also revealed the influence of various individual background characteristics on MTV. We observed that female volunteers are more altruistically motivated than their male colleagues, a finding that resonates with earlier research that found women to be more disposed to care and to express stronger altruistic concerns and empathy than men. This is invariably explained in terms of either biology or socialization (Gerstel 2000; Musick and Wilson 2008). Younger volunteers are significantly more motivated by self-interested reasons than older volunteers, which is also consistent with earlier research. For example, as discussed above, young people are more focused on career-related reasons for volunteering, a typical life course effect. Surprisingly, higher educated volunteers express less support for both altruistic and self-interested reasons for volunteering. A possible methodological explanation may be that the self-interested reasons for volunteering measured in the survey do not match the self-oriented MTV of higher educated volunteers. Finally, we could discern a more instrumental use of volunteering by people who are not in a full-time employment position: Unemployed volunteers, students, housewives, and retired volunteers more frequently expressed self-interested reasons for volunteering; thus, for these categories, volunteering may perform functions otherwise served by a paid job, such as to gain new skills or useful experiences, or be undertaken for social reasons.

To conclude, some general insights can be drawn from our findings. Firstly, based on our cross-national analysis , we found evidence for a contextual understanding of the motivation to volunteer. Religious context, national value patterns, and welfare/nonprofit regimes influence the support for altruistic and self-interested reasons for volunteering. It thus makes sense to situate motivational accounts at the intersection between individual and society (cf. Wuthnow 1991) , and not just treat them as a matter of inner psychological needs—as dominant theories of MTV claim. The emphasis put on certain types of motivations is clearly influenced by broader cultural and structural patterns. As a result, further cross-cultural examination of volunteer motivations is a fruitful option.

Secondly, the “traditional” beneficial relationship between religion and altruistic motivations holds at the individual level. While studies predicting participation in volunteering have mainly pointed to the importance of active religious networks, the reported reasons for volunteering are influenced by both religious beliefs and practices. Altruistic orientations are fostered through religious teachings and through active participation in a religious community. While we found some evidence for the influence of a religious national context, the evidence was partial and in an unexpected direction: On the one hand, no relationship was found between extensive religious networks and support for altruistic motivations; on the other, strong religious beliefs among the general population were negatively associated with both altruistic and self-interested MTV. In a more devout country, self-interested reasons for volunteering are therefore less culturally accepted, yet surprisingly the expression of altruistic motivations is also hampered. In other words, when there are more religious people in one’s environment, there is less religiously inspired “motive talk.” Just as traditional volunteering was a “good deed” that lost its sincerity when being “shown off” too much (Beck 1997), it seems that to expose one’s “good intentions” too much undermines the credibility of these intentions in a more devout context.

Our multilevel findings on the relation between religion and MTV consequently imply that the process of secularization cannot be linked in a straightforward way to the weakening of altruistic MTV. Although a decline in individual church practice and individual beliefs would decrease the support for altruistic motivations, and increase the approval of self-interested MTV, in a more secular national context the support for altruistic MTV would, on the contrary, be stronger. Furthermore, we found other variables that had a positive effect on altruistic MTV. At the individual level, gender (females), age (older people) and political orientation (left wing) were positively correlated with altruistic MTV. At the country level, the prevalence of a postmaterial value pattern, which stands in opposition to traditional religious beliefs, did not represent a threat to feelings of altruism, and produced mixed findings concerning self-interested MTV. Furthermore, we may carefully conclude that welfare states with lower social spending, a large nonprofit sector with little revenue from government and an active citizenry, in terms of a high rate of volunteering, all stimulate the expression of altruistic motivations. On the other hand, there are also factors that hinder altruistic motivational accounts and stimulate the expression of self-interested MTV. Higher educated people are less likely to support altruistic motivations (yet surprisingly also less frequently mention self-interested MTV). The employment status of volunteers also plays a role: The nonemployed approach their volunteer involvement in a more instrumental way, as a means to acquire skills and experience, and to do something worthwhile. At the contextual level, a “crowding out” effect seems to occur: When government social spending is high, and nonprofit organizations to a large extent depend on government subsidies, volunteers are less inclined to express support for altruistic MTV.

In sum, the assumed transition from altruistic to self-interested motivations, which is claimed to result from processes of secularization and value change cannot therefore be confirmed unambiguously. Religion is not the only and unmistakable source of altruistic inspiration. In a secular context, there are also individual and contextual factors that are positively associated with altruistic MTV. While higher levels of individual religiosity will continue to foster altruism, more secular contexts will also continue to express a mix of altruistic and self-interested motivational accounts.