Introduction

Voluntary action is growing all over the world, and volunteers have been recognized as a significant social force, contributing to social development (Korten 1990). Guided by their interest, passion, or moral obligation, rather than financial gain, volunteers provide assistance to target groups that are usually marginalized and excluded. Their purpose is to tackle social problems which are largely ignored or insufficiently addressed by official channels and to nurture the well-being of their communities. It has been acknowledged that volunteers have played an important role in economic development and social integration (Cheung 2006; Dekker and Halman 2003; Hodgkinson and Weitzman 1990; Malik et al. 2002; Prouteau and Wolff 2004; Salamon and Sokolowski 2001). A research team of United Nations Volunteers (1999) found that volunteers in some countries contribute between 8 and 14% of overall Gross Domestic Product in their nations. For the case of USA, Mowen and Sujan (2005) find that the contribution of volunteers in the USA reached 75 billion dollars (estimated) which is equivalent to the overall income of the whole entertainment and recreation industries. These data suggest that volunteering is a strong, dynamic, and unifying force for societies, because of both its economic and social roles (Cordingley and Bates 2004).

Saudi Arabia is always noted as a wealthy country, but economic disparity is widespread. Although there is no official data on Gini coefficient (World Bank 2016), many economic observations suggest economic disparity. Two to four million native Saudis are struggling with poverty (Aarts and Roelants 2016; Sullivan 2013). The overall unemployment rate reached 11.5% in 2015, and 34% among Saudi women (Ali 2014; Jiffry 2014). Sixty percent of Saudis do not own a house (Abuzaid, 2012). Given such an economic background, there is a social need for charity and volunteering. Saudi government strongly promotes voluntary actions to alleviate social tension, and the growth in the number and prominence of volunteers has been recognized as a significant social phenomenon in the Kingdom. Around 591 Saudi NGOs and NPOs have been active to help disadvantaged people, operating under the supervision of the General Directorate of Non-government Charitable Institutions and Associations (Bendania et al. 2012). Research on volunteers in Saudi Arabia focused on its political operation mainly. For instance, Montagu (2010) argued that NGOs, including volunteer organizations, in Saudi Arabia were not just social service providers, but also an agent for sociopolitical dialogue and social reform. There have also been political analyses in NGO studies in other Middle Eastern countries (Abdelrahman 2004; Gubser 2002; Stachowski 2005). It seems apparent that there has been an absence of social analysis on Saudi NGOs, especially the volunteer organizations, which are service-providing nonprofit organizations established by ordinary Saudis to deliver social benefits to locals with various ends and forms.

In this study, we are particularly interested in how social culture motivates Saudi volunteers’ continuous volunteering which is the most serious difficulty that all volunteer organizations face. As the first effort to empirically address volunteers’ continuous participation in Saudi Arabia, we take a very different approach by arguing that volunteer retention may not influenced by collectivistic culture directly but through mediation of individualistic cultural components. We first present prior studies on culture, particularly the relationship between collectivism, individualism, and volunteering, followed by the theoretical framework and hypotheses. Methodology is covered in the third section, including data and scales. The fourth section presents findings from the mediated mediation analysis and moderation analysis. The last sections present our conclusion and a discussion of the current findings with past findings in the literature.

Literature Review

Helping others without financial reward, volunteers are often asked what drives their actions. Likewise, this is a prominent theme in the volunteering literature. What motivates people to volunteer is a complex and vexing topic (Esmond 1997), but it is the basis of research on volunteers which cannot be ignored (McCurley et al. 1998; Vineyard 2001). In this study, we investigate motivations of volunteers from a perspective of collectivism and individualism. Most studies address volunteering motivation from an individualistic angle, ignoring collectivistic factors (Cusick 2007). Some studies tried to explore both collectivistic and individualistic motivations, but were limited to description and categorization, or argued their parallel influence on volunteering. For instance, Rehberg (2005) once found that both collectivistic and individualistic motivations play a role in volunteering and categorized them into “achieving something positive for others,” “quest for the new,” and “quest for oneself.” Finkelstein (2010, 2011) argued that individualists and collectivists differ, not in their willingness to volunteer, but in why they choose to volunteer; therefore, the individualism/collectivism construct is useful in clarifying why people help, but not in predicting who will volunteer. However, these researches were conducted within a single national culture, measuring individualism/collectivism as individual differences, so the findings are limited when making predictions between individualistic and collectivistic national cultures. The present study contributes to the field by exploring the mechanism by which collectivism and individualism influence volunteering with an analysis of mediated mediation.

Collectivistic Motivations

Collectivism, as a cultural orientation, emphasizes the significance of group identities, group goals, and group interests. Community, society, or nation is focus on the eyes of collectivists. Across various cultural investigations, Middle Eastern cultures were shown to be more collectivistic than Western cultures (Al-Saggaf and Weckert 2011; Buda et al. 1998). In various studies, Saudi Arabia scored much higher than the USA and the UK in Hofstede’s measurement of collectivism (Cassell and Blake 2012; ITIM 2016). Such collectivism influences every corner of Saudi life, such as purchases (Opoku 2012), success of business (Cassell and Blake 2012; Liu et al. 2010), formation of organizational culture (Al-Otaibi 2010), formation of personal network (Jiang et al. 2012), customer relationship management (Agourram 2014), formation of psychological contract (Aldossari and Robertson 2014), managerial coaching behavior (Noer et al. 2007), perception of performance appraisal (Harbi et al. 2016), decision-making (Alsuwaidi 2008), personalized e-learning (Alamri et al. 2014), and organizational commitment (Alqurashi 2009).

Collectivism is reflected in various aspects. One collectivistic component related to volunteering is prosocial personality, which includes other-oriented empathy (concerning prosocial thoughts and feelings, such as empathic concern, perspective taking, and a sense of social responsibility) and helpfulness (a stable and confident sense of identity as a helper) (Penner 2002). The two dimensions of prosocial personality are conceptually and empirically distinct, but significantly associated with each other (Penner 2002; Penner et al. 1995; Rioux and Penner 2001). It has been demonstrated that collectivistic cultural groups report higher levels of these personality qualities (Feygina and Henry 2015; Knafo-Noam et al. 2015). Studies further reveal that volunteers score higher on personality traits such as empathetic concern and agreeableness than non-volunteers (Paterson et al. 2009). These differences likely encourage voluntary behaviors, namely a greater tendency to consistently help others out of deep-rooted predispositions such as personal values, social responsibilities, and cognitive empathy across various situations (Batson 1991; Habashi et al. 2016; Mowen and Sujan 2005), although not all studies discovered statistically significant links between personality traits and volunteering (Carlo et al. 2005; Omoto and Snyder 1995). Particularly, empathy, the most salient source of the altruistic motivation to help, influences individuals’ volunteering intention by enhancing their sensitivity to others’ needs and emotions (Batson and Weeks 1996; Caprara et al. 2010; Sibicky et al. 1995). Volunteers with such self-determined motives are found to have higher levels of work engagement and well-being (Ramos et al. 2016).

Research has also identified community identity as another collectivistic value that motivates volunteering (Levy et al. 2012; Okun and Michel 2006; Omoto and Snyder 2002). Community identity can play an important role in facilitating volunteering in that people with strong community identity tend to link their own welfare to that of others (Janoski et al. 1998). The extent to which individuals’ community factors into their identities influences the extent to which individuals seek to aid the community. Thus, building on the previously elaborated theory that collectivism may promote volunteering through instilling stronger prosocial personality qualities, the role of community identity would suggest that the motivation to help should be especially strong when aiding the perceived in-group, as suggested by Barry et al. (2008). In collectivistic Saudi Arabia—and in Islamic societies, more broadly—a strong sense of community identity can be seen in many areas of society, such as prevalence of tribalism, ideological homogeneity, and the Saudi family structure (Helal and Coston 1991; Jiang et al. 2010/2011; Shelley 1981). Traditionally, Saudis live in extended families, which cultivates collectivism in kinship networks. Tribalism, as a survival strategy, also demands mutual help among its members, which further boosts collectivism. Even nowadays, Saudi government uses site exploitation, architecture of public buildings, and housing employing the principles of traditional Saudi settlements, and landscape architecture with local trees and shrubs to create community identity among residents (Alhathloul and Mughal 1999).

To fully understand the relationship between collectivistic culture and voluntary activities in Saudi Arabia, one must also consider the role of religion. Correlations have been established between religiosity and prosocial behavior, with recent research findings support for the theory that personal religiosity (across religions, Islam included) plays a causal role in prosocial behavior (Lin et al. 2016; Yeung 2017). In Saudi Arabia, religion, namely collectivistic Islam, impacts volunteering via community identity which is based on an Islamic concept called “Ummah.” The Ummah, referring to “Muslim community” literally, is both a product and a contributor of collectivism among Saudis (Afzal Dar 2013). Saudi Muslims are aware of their roles in their families and Ummah, as it is instilled through religious education. As Abadeer (2015) points out, a Muslim should maintain the ties of kinship by keeping in touch with their relatives from both parents’ sides and showing kindness to them. The Ummah is not just a community but also a system of mutual help and social assistance envisaged by Islam (Khalid and Arshad 2011). There are numerous instructions in the Quran, Hadith, and Sunnah to teach Muslims to donate to charity, and help the community without rewards. For instance, Quran (The Holy Quran 5:2) says, “Help ye one another in righteousness and piety, but help ye not one another in sin and rancour.” It further says, “They perform (their) vows, and they fear a Day whose evil flies far and wide. And they feed, for the love of God, the indigent, the orphan, and the captive, (saying), ‘We feed you for the sake of God alone: no reward do we desire from you, nor thanks.’”(Quran 76:7–9). Throughout their education, Saudi Muslims are instructed that individual interests are subordinate to collective interests, for in this way they can bind the Ummah together (Chatterjee 2011; Khomeini et al. 2002; Markus and Kitayama 1991). For Murad (1985), it means sacrifice of personal time, worldly possessions, money, familial love, friendships, views, emotions, ego or life for the community. As compared to collectivism more broadly, especially those forced forms of collectivism imposed by states (Khvorostianov and Remennick 2017), the Islamic collectivism of Saudi Arabia may be even more of a potent motivational force for volunteering. For Saudis, it is in line with religion that they engage in initiatives that improve quality of life for the community without asking for material rewards for it is a way to gain God’s pleasure and His acceptance to heaven. Such a well-established community-based identity among Muslims may lead to volunteering because of its meaningfulness (Okun and Michel 2006). As a result, community identity among residents will be stronger and community is created (Khan 2016).

Individualistic Motivations

Humans are not always altruistic. Drawing on underpinnings of classical economics, rational choice theory states that any individual is self-interested and behaves to maximize benefits by weighing means and ends, and costs and benefits (Cornish and Clarke 1987). Such an angle is particularly reflected in culture of individualism which emphasizes self-realization, autonomy, self-reliance, uniqueness, achievement orientation, competition, emotional independence, and social differentiation (Loose 2008). In a society where individualism prevails, individual’s interests, achievements, personal goals, personal uniqueness, and personal control are given tremendous importance, while communal aspects are secondary (Bellah et al. 2007; Hsu 1983; Kâğitçibaşi 1994; Kim 1994; Markus and Kitayama 1991; Sampson 1977; Triandis 1995). Volunteering is certainly a prosocial behavior, but this does not exclude utilitarian motives.

It has been discovered that volunteering is instrumental to Western volunteers in many aspects (Clary et al. 1996; Esmond 1997; Omoto and Snyder 1993; Tschirhart et al. 2001). The instrumental functions act as a strong motivator to prospective volunteers, especially the six functions which are based on various studies (Anderson and Moore 1978; Beale 1984; Day and Devlin 1998; Gidron 1978; King et al. 1970; Schwartz 1970) and proposed by Clary et al. (1998). They are enhancement function of volunteering (people may participate in volunteering in order to improve self-esteem or self-confidence), values function (volunteering can help people to express or act on important social values such as humanitarianism and helping those less fortunate), career function (people may seek a job prospect or try to build up his career through volunteering), learning function (people use volunteering as an opportunity to learn skills, technology, knowledge, and improve capabilities), social function (people may volunteer to meet their social needs, particularly friendship and social networking), and protective function (people may volunteer to release their guilty of being more fortunate than others). These individualistic functions have been found to be important in motivating volunteers in various studies. For instance, Finkelstein (2010), using similar motivational measures as the current study, showed that, on an individual difference level, individualism positively predicted more career-oriented utilitarian motives in volunteering.

Saudi society has faced a rise in individualism recently (De Jong and Moaddel 2013). Given such a cultural change, which places an emphasis on the self, volunteers’ utilitarian motives may be common in Saudi Arabia; their decision-making on volunteering may be more like a cost–benefit process. In reality, a volunteer may have two or more of these motives in their mind when participating in volunteering. For a specific volunteering event, one person’s particular motive may be stronger than other motives. In a big volunteering event, volunteers may choose specific tasks to meet their particular motives (Snyder et al. 2000). It is also very possible that volunteers involved in the same event activity have different goals or motives.

Making predictions about the role of these individualistic motives in Saudi volunteering is complicated by the fact that Saudi society is in a cultural transition. We tested the above six functions of volunteering (social, career, learning, enhance, protective, and values function) in a pilot study in Saudi Arabia and discovered that career function, protective function, and values function of volunteering are less important than other three functions compared to Western and East Asian societies. Our understanding is that, first of all, Saudis resist accepting what are perceived as lower-paying and lower-skilled jobs (Anderson et al. 2012). Unfortunately, volunteering work likely falls into these undesirable categories. Moreover, limited social capital of the people whom volunteers serve and limited scale of professionalism among Saudi volunteer organizations mean limited job opportunities to prospective volunteers who seek help on career through volunteering (Matic and AlFaisal 2012). Therefore, job and career are secured by other social institutions, such as wasta (personal connections), much more than volunteering in the Middle East (Barnetta et al. 2013; Smith et al. 2012; Tlaiss and Kauser 2013). Second, traditional Saudi society is hierarchical so that people feel normal to be advantaged over others, and such a tradition still prevails in contemporary Saudi society (Williams, 2009). Seeking psychological self-protection via volunteering has not been found in Saudi Arabia (Bendania et al. 2012), which is consistent with Erasmus and Morey (2016) finding at an Australian faith-based organization. Third, values of humanitarianism have been expressed in Islam to a great extent, so much so that they do not see the necessity of strengthening it through volunteering (Lauby and Stark 1988). Given this, these three functions were not integrated into the theoretical framework.

Theoretical Framework

As discussed above, it is typically assumed that people volunteer mainly for altruist motives (Burns et al. 2006; Carpenter and Myers 2010; Guerra et al. 2012; Unger 1991; Wells 2005), especially in collectivist cultural settings such as Saudi Arabia. It is also proved in various studies that instrumental considerations motivate people to volunteer (Barron and Rihova 2013; Bussell and Forbes 2002; Coghlan and Fennell 2009; Dean 2014).

Volunteering motives, either collectivistic or individualistic, are tackled by various theories. For instance, the self-determination theory (Deci and Ryan 2000) holds that the interplay of intrinsic motives, extrinsic motives and three human needs (autonomy, competence, and relatedness) explain most human behavior and foster engagement for specific activities, such as volunteering (Wu et al. 2016). The theory of planned behavior (Ajzen 1991) states that individual’s behavioral intentions can be explained with the concept of attitude toward behavior, subjective norms, and perceived behavioral control. Such a perspective has been shown to be effective in predicting volunteer’s behavior and understanding volunteer’s retention (Lee et al. 2014; Macgillivray and Lyndstevenson 2013). Both the self-determination theory and theory of planned behavior place equal emphases on internal factors and external factors when addressing humans’ behavioral intentions. The aforementioned collectivistic factor of prosocial personality works as an intrinsic motive in the self-determination theory and an attitude toward behavior in the theory of planned behavior. On the other hand, the collectivistic factor of community identity works as an extrinsic motive in the self-determination theory and as a subjective norm in the theory of planned behavior.

In addition, as the rational choice theory holds, individual’s decision-making on volunteering is calculative, rational, and instrumental; therefore, individualistic purposes may be the primary driver in volunteering intention. Perhaps due to the dominance of volunteering researching taking place in overwhelming individualistic cultures, there were few empirical attempts to integrate both collectivistic and individualistic motivations into a framework to explore the relationship between them. A few studies took into account both altruist and egoist motivations, but just compared and weighed them, or only focused on the role of individualistic and collectivistic factors on different types of volunteering without addressing the mechanisms between them (Clary et al. 1996; Dovidio et al. 1990; Hallmann and Harms 2013; Shye 2009; Smith et al. 2010). In this study, collectivistic and individualistic factors are not considered to work in parallel, but as a mechanism of mediated mediations.

A Mediated Mediation Model

To understand Saudi volunteering, we propose that there are both individualistic motives and collectivistic motives behind continuous volunteering, and the two groups of motives lead to intention of continuous volunteering through a mechanism of mediation rather than parallel contribution. Based on the assumption of rational choice theory, that humans are self-interested by nature, we argue that individualistic motives may best account for continuous voluntary actions, although may not in a fully direct way. Specifically, prosocial personality, which is collectivistic in the broadest sense, seems to motivate Saudis to volunteer continuously, but such a causal relationship may be mediated by a less broad collectivistic factor, community identity. In other words, beyond a direct causal relationship between the prosocial personality and volunteer retention, prosocial personality has an indirect impact on volunteer retention by influencing community identity, which in turn influences the intention of continuous volunteering. Community identity may largely account for the observed relationship between prosocial personality and the intention of continuous volunteering. This is predicted because Saudis feel strongly attached to their own communities, and as a result, want to contribute to it by engaging in continuous volunteering. In the same vein, collectivistic community identity seems to have a direct impact on Saudis’ intention of continuous voluntary action, but individualistic instrumental motives, such as socializing, learning skills, and self-enhancing, serve as mediating variables that drive the relationship between community identity and continuous voluntary actions. When the instrumental motives are taken into account, the prosocial personality and community identity may be found to have no effect on the intention of continuous volunteering. It is the underlying mechanism by which the prosocial personality and community identity influence future voluntary engagement through utilitarian purposes. Such mediated mediations can be called theory of progressive reduction and are depicted in Fig. 1.

Fig. 1
figure 1

Conceptual mediational path model

Full size image

Moderated Causal Relationships

We will test the moderating role of gender because in many settings, gender is anticipated to have an impact on various social behaviors, including volunteering (Giudice 2015). Men and women differ in volunteering, because of employment status (Taniguchi 2006). However, men and women may not have substantial differences in amount of time volunteering, because men score higher on measures of income, education, trust, and secular social networks but women have broader social networks through religious participation (Einolf 2011). Furthermore, there are no observed gender differences in the importance of instrumental motives for volunteering (Fletcher and Major 2004). When considering the moderating role of gender in Saudi Arabia, however, the prevalence of patriarchy should be considered. Sharia law and Saudi tribal culture determine women’s specific, less influential positions in society, and require gender segregation in different aspects of social life. For instance, the system of male guardianship limits females’ ability to travel independently and emphasizes domestic duties. Equal access to employment and economic opportunities for women is an ongoing struggle, as well. That is why it is argued that “Saudi women suffer from the exclusion from participating and playing vital roles in their society and the deprivation from getting their rights” (Alhareth et al. 2015, p. 123). Such a cultural component may suppress Saudi females’ intention of volunteering, in spite of the encouragement for volunteering determined in Islam.

Moderation effect of employment status on the relationship between prosocial personality and community identity and intention of continuous volunteering will be tested, too. Full-time students, full-time employees, and unemployed people have different work schedules and different amounts of spare time. Some volunteer work is carried out during the workweek, which pushes out full-time students and employees. Moreover, for the full-time students or employees, the time they can volunteer is weekend and holiday, but weekend and holiday are family time in Saudi cultural settings. Such a time conflict makes it difficult for them to participate in voluntary activities. Thus, status of employment may impact individuals’ intention of volunteering.

Age will also be tested as a moderator on the relationship between prosocial personality and community identity and intention of continuous volunteering, because a individuals’ values, expectations, and needs change during their lives (Van Willigen 2000). It was discovered that volunteers of different age groups varied in motivation, in terms of aforementioned six functions (Deery et al. 2011). People of different ages also respond to peer influences on volunteering differently (Van Goethem et al. 2014). The relationship between age and functions of volunteering may not be linear, either (Okun and Schultz 2003). Age could also moderate the relationship between job satisfaction and intention to continue volunteering (Bang 2015). However, due to the racial and religious homogeneity in Saudi Arabia, age may have no impact on the intention of continuous volunteering.

Family influence may also moderate both the relationship between prosocial personality and the intention of continuous volunteering and the relationship between community identity and intention of continuous volunteering. Influence from family members has been demonstrated to have a considerable effect on social behaviors (Hao and Matsueda 2006; Thomson et al. 2007; Wang et al. 1995) and may even cultivate social entrepreneurship (Kao and Huang 2015). A volunteering family member may act as a role model for other family members, and their positive experiences may motivate siblings to engage in volunteering (Friedman 2003; Reilly and Vesic 2002). Family also influences the intention of continuous volunteering positively through family support and family culture and negatively through family rewards and coercion (Shek 2009).

All these factors may not result in voluntary action directly, but may moderate the associations between prosocial personality and intention of volunteering, and the relationship between community identity and intention of volunteering (Fig. 2).

Fig. 2
figure 2

Conceptual moderational model

Full size image

Hypotheses

Hypothesis 1:

Prosocial personality, a collectivistic factor, motivates Saudis to volunteer continuously because a general prosocial personality induces altruist behaviors. This association between prosocial personality and intention of continuous volunteering is mediated by community identity, a less broad collectivistic factor.

Hypothesis 2:

Community identity, a collectivistic factor, motivates Saudis to participate in volunteering continuously, because they see themselves as a part of the community and, therefore, seek to contribute to the community via volunteering. This association is mediated by individualistic instrumental motives including socializing, learning skills, and psychological enhancing.

Hypothesis 3:

Gender, age, state of employment, and whether having a volunteer family member influence the direction and/or strength of the relation between prosocial personality and the intention of continuous volunteering. For instance, there is a significant difference between genders regarding the intention of continuous volunteering for the same level of prosocial personality.

Hypothesis 4:

Gender, age, state of employment, and whether having a volunteer family member impact the direction and/or strength of the relation between community identity and the intention of continuous volunteering.

Methods

Data

The data were collected from participants who were engaged in various volunteering events organized by Saudi volunteer organizations. They mainly served schools, hospitals, social service organizations, and religious institutions in the format of visiting the needy, helping students with homework, collecting charity, environmental activities, and providing recreational and educational activities. Some of them are long-term volunteers, and some were merely members of volunteer organizations. Two hundred ninety questionnaires were collected and processed with SPSS 21.

Of the 290 volunteers, most of them were aged between 19 and 25 (63.1%; Table 1). There were moderately more male respondents than female respondents (57.2 vs. 42.8%, respectively; Table 1). Saudi students were more represented in the sample than employed Saudis (67.9 vs 20.3%, respectively; Table 1), while some were neither students nor employed (11.4%, Table 1). Nearly half of subjects (49.6%, Table 1) had a family member that was engaged in volunteering.

Table 1 Descriptive statistics and reliability test result (N = 290)
Full size table

Questionnaire and Scales

To investigate the mechanism which underlies the relationship between collectivistic factors and intention of volunteering via individualistic factors, five scales were adopted or constructed. All continuous variables were assessed on 7-point Likert-type scales, with 1 representing “strongly disagree” and 7 representing “strongly agree.” Cronbach’s alpha was used to assess the internal consistency and reliability of the scales. The responses of every scale were consistent across respondents and met the internal reliability requirement (Bryman 2004).

Prosocial personality was assessed using 10 items derived from the Prosocial Personality Battery (Penner et al. 1995). The composite mean was 5.27, suggesting that Saudis, generally speaking, have a reasonably prosocial personality (Table 1).

Community identity was assessed using nine items constructed by the authors. Items included affiliation to the community, community concern, mutual help on each other, solidarity of community, and belief in collective action for community (α = 0.784). The composite mean is 5.24, indicating that Saudis have connections with local community and have relatively strong community identity (Table 1).

Learning functions of volunteering was assessed using five items developed by Clary et al. (1998). The reliability alpha is 0.758. The composite mean is 5.51, suggesting that Saudis believe they can learn substantially from volunteering (Table 1).

Social function of volunteering was assessed using five items developed by Clary et al. (1998); the Cronbach alpha of 0.746 suggests a reliability of the scale. The composite mean is 4.59, indicating that Saudis believe that volunteering may offer opportunities to be with his friends or to engage in an activity viewed favorably by important others (Table 1).

The enhancing function of volunteering was assessed using five items developed by Clary et al. (1998); α = 0.838. The composite mean is 5.30, suggesting that volunteering can help on the ego’s growth and involves positive strivings of the ego (Table 1).

Moderator variables are gender, age, state of employment, and whether having a volunteering family member. Among them, gender (male = 1, female = 0), state of employment (students and employed Saudis = 1, neither = 0), and whether having a volunteering family member (having one volunteering family ember = 1; not having a volunteering family member = 0) were dummy coded.

The outcome variable was operationalized as whether respondents like to participate in volunteering work continuously in the near future.

The composite means of three scales of function of volunteering are much higher than their mean showing Saudi volunteers placed a lot emphasis on the instrumental sides of volunteering intentionally or unintentionally. The prevalence of collectivism seems apparent among participants, because the means of prosocial personality and community identity are much higher than their mean.

Findings

Correlations

Before conducting tests of the hypotheses, correlations between the independent variable (prosocial personality), proposed mediator variables (community identity, social function of volunteering, learning function of volunteering, and enhancement function of volunteering), and dependent variable (like to volunteer continuously) were conducted using Pearson’s correlation. As shown in Table 2, prosocial personality is statistically significantly associated with intention to continuously volunteer (r = 0.176, p < 0.01). The correlation between prosocial personality (IV) and mediator variable (MV) community identity reached a statistically significant level (r = 0.526, p = 0.189). The correlation between the MV community identity and DV likelihood to volunteer continuously is statistically significant and positive (r = 0.219, p < 0.01). The MV social function of volunteering was statistically significantly correlated with the community identity (r = 0.228, p < 0.01), so is the MV learning function of volunteering (r = 0.146, p < 0.05), and enhancement function of volunteering (r = 0.204, p < 0.01). Also, the DV like to volunteer continuously was statistically significantly correlated with the MV social function of volunteering (r = 0.415, p < 0.01), the MV learning function of volunteering (r = 0.527, p < 0.01), and enhancement function of volunteering (r = 0.524, p < 0.01).

Table 2 Pearson correlations among variables in mediation models (N = 290)
Full size table

Mediation Analysis

Mediation analysis is a statistical method addressing how a causal agent (X) transmits its effect on (Y) via a mediator (M) (Wu and Zumbo 2008). In the transmission process, the mediator variable functions as a vessel through which casual effects occur (Hayes 2013). Hayes (2013) developed a SPSS PROCESS which uses an ordinary least squares path analytic framework to estimate direct and indirect effects of agent X on Y. This study used the PROCESS, which includes a bootstrapping procedure and Monte Carlo confidence intervals that are implemented for inference about indirect effects (Preacher and Hayes 2008).

The above correlations have fulfilled the precondition for mediating analysis. That is, the correlations between the predictor, mediators, and outcome variable (i.e., paths a, b, and c in Fig. 1) are statistically significant.

From a simple mediation analysis conducted using ordinary least squares path analysis, prosocial personality indirectly influenced intention of continuous volunteering through its effect on community identity. As seen in Table 3, having a stronger prosocial personality tends to have a stronger community identity (a = 0.390), and a stronger community identity predicted a stronger intention to volunteer continuously (b = 0.164). A biased-corrected bootstrap confidence interval for the indirect effect (c = 0.064) based on 1000 bootstrap samples was entirely above zero (0.0145–0.1152). When the factor of community identity was controlled for, the original magnitude of the effect of prosocial personality dropped to statistical non-significance (c′ = 0.059, p = 0.244). Therefore, hypothesis 1 was supported.

Table 3 Coefficients for mediational path models
Full size table

Collectivistic community identity seems to explain Saudi’s intention of volunteering most, but subsequent mediation analyses reveal that individualistic motives such as social function, learning function, and enhancement function of volunteering more strongly account for Saudis’ intention of continuous volunteering (Table 3). In the three subsequent mediation models, participants’ community identity significantly predicted the social function of volunteering (a = 0.2875, p < 0.001), the learning function of volunteering (a = 0.146, p < 0.001), and the enhancement function of volunteering (a = 0.204, p < 0.001). Participants motivated by the social function of volunteering are more likely to volunteer continuously (b = 0.30, p < 0.001). Similarly, participants motivated by the learning function of volunteering were more like to volunteer consistently (b = 0.477, p < 0.001), as were participants motivated by the enhancement function of volunteering (b = 0.470, p < 0.05). All the indirect effects were statistically different from zero, as revealed by a 95% BC bootstrap confidence interval (social function: c = 0.083, CI 0.0407–0.1417; learning function: c = 0.07, CI 0.0178–0.1313; enhancement function: c = 0.096, CI 0.0447–0.1643, Table 3), thus supporting hypothesis 2.

Moderation Analysis

Moderation explains when or under what situations X exerts an effect on Y (Hayes 2013). If the size, direction, or strength of effect of X on Y is influenced by M, M is regarded a moderator of X’s effect on Y (MacKinnon and Luecken 2008). The regression coefficient for the product of predictor variable X and moderator variable M quantifies how the effect of X on outcome variable Y changes as M changes by one unit. The SPSS PROCESS by Hayes (2013) uses an ordinary least squares or logistic regression-based path analytic framework to estimate two- and three-way interactions in moderation models along with simple slopes and regions of significance for probing interactions and was adopted in this study.

Results reveal that all regression coefficients for X*M (gender*prosocial personality, age*prosocial personality, state of employment*prosocial personality, family volunteer member*prosocial personality) were not statistically different from zero (Table 4), suggesting that the effect of prosocial personality on the intention of continuous volunteering does not depend on gender, age, state of employment, and whether having a family volunteer member, thus rejecting hypothesis 3. The effect of community identity on intention of continuous volunteering doesn’t depend on these factors either, which rejects hypothesis 4. Because gender, state of employment, and the presences of a family member volunteer are categorical variables, moderation tests were conducted with analysis of variance tests in this study, indicating that there were no statistically significant interactions across these variables.

Table 4 Results from regression analyses of moderation of the effect prosocial personality and community identity on intention of continuous volunteering by gender, age, employment, and whether having a family volunteer member
Full size table

Discussion

Rather than treating collectivistic and individualistic factors parallelly and comparing their weight in motivating volunteers as prior studies have done, we explored these cultural motivations in volunteering simultaneously. The mediated mediation analysis works like peeling an onion layer after layer, from the broadest level (prosocial personality), to less broad levels (community identity and egoistic considerations). Mediation analyses revealed that collectivistic factors have an influence on volunteering retention, but through individualistic factors.

It is not surprising that community identity has to influence Saudis through individualistic instrumental considerations in terms of motivating volunteers. For Saudis, there is a more significant collective identity, particularly tribe identity (Hui and Yee 1994). However, Saudi Arabia has been in the process of fast economic development and urbanization. Millions of Saudis have left their villages, received higher education, and settled down in cities (Saleh 2002). Together with the migration from rural areas to cities is the process of transformation from mechanical solidarity to organic solidarity (Durkheim 1893). A new community identity which is based on organic solidarity and spatial design emerged consequently. Compared to tribal identity, which is based on mechanical solidarity, new community identity is less emotional to Saudis and therefore may not motivate them to volunteer as strongly as individualistic factors. Although individualistic motives seemed to be stronger than collectivistic motives, it does not mean we can ignore collectivistic factors. In some situations, collectivistic motivations might mediate or moderate the effect individualistic motivations on volunteering.

Saudi Arabia is well known for its sex segregation (Doumato 1992). Saudi females face obstacles when interacting with the broader society. Despite these gender differences, the moderation analyses suggest that the above results are generalizable to both genders. Although our study revealed that there was no significant gender difference in terms of instrumental motives in an Islamic environment, as what had been discovered in Western studies (Burns et al. 2008; Fletcher and Major 2004), equal opportunities for volunteering does not exist across Saudis. Females in Saudi Arabia face more difficulties in participating in volunteering due to male guardianship, especially when their families are concerned about possible gender integration.

Also, contrary to our hypotheses, the effect of prosocial personality and community identity on the intention of continuous volunteering did not differ across people in different states of employment, different age groups, nor whether family members volunteer. These findings are inconsistent with research conducted in Western societies, where female and having a family volunteering member predict more continuous volunteering. This lack of interactions may be due to the strong collective culture, which suppresses individuality on marginal dimensions.

Many nonprofit organizations are experiencing difficulties in recruiting new volunteers. Our study has great policy implications. We found out individualistic factors were more attractive to prospective Saudi volunteers, which is opposite to what Hager (2014) discovered in an American study. Therefore, it is necessary to understand their needs and accept their instrumental purposes. Based on that, nonprofit organization managers should create organizational conditions that are inviting to prospective volunteers and meet their needs. After all, all public charities have malleable conditions, which Hager and Brudney (2011) call “nurture”; otherwise, they may lose prospective volunteers. Therefore, it is important to design a project which can satisfy both social needs and volunteers’ instrumental needs.

Conclusion

Our study reveals a different side of volunteering in Saudi Arabia, one that contrasts the mainstream literature. It is always assumed that individuals volunteer for interests and benefits of community, and it is these collectivistic factors that motivate them to devote time and energy to volunteering. Consistent with what research has revealed about volunteering motives in the West, Saudis also are motivated to volunteer by the collectivistic factor of prosocial personality (Berry-Bobovski 2007; Clary et al. 1998). However, our study shows that, for Saudis, community identity mediates the effect the prosocial personality on intention of continuous volunteering. This suggests that community identity mattered in the causal relationship between prosocial personality and intention of continuous volunteering. When individualistic considerations, such as purpose of socializing, learning skills, and self-enhancement, were taken into account, the role of community identity was shown to be statistically nonsignificant, because the individualistic motivations mediate its effect on intention of volunteering. Therefore, these individualistic instrumental motives that explain Saudis’ volunteering behaviors, which is consistent with Batson’s (1987) argument that prosocial motivation is egoistic. For certain, there is no absolute opposition between the collectivistic motives and individualistic motives. Any volunteer has both collectivistic and individualistic motives in participating volunteering events. The findings of this study only highlight that the individualistic motivations functioned as the conduit through which the causal effect between collectivistic motivation and intention of volunteering operates, rather than a simple rejection of role of collectivistic factors.

Although meaningful results have been obtained, two limitations are noted. The first is the data were collected from volunteering events, so generalization is limited. The second limitation is the validity of self-reported responses, due to cultural pressure in Saudi Arabia and possibility of self-presentation.