Introduction
Formal volunteering refers to activities performed through organizations to alleviate needs in society by individuals who receive little to no monetary compensation for their efforts (Wang et al. Reference Wang, Mook and Handy2017; Xu Reference Xu2014). Before the market-oriented reform in 1978, formal volunteering among the Chinese was repressed because of the omnipotent government. Since the 1980s, however, reform and opening policies have triggered a reconstruction of the relationship between the state and society, thus allowing the volunteer behaviors of individuals to gradually expand (Tan and zhou Reference Tan and Zhou2009). Furthermore, the Chinese government, which has encountered severe pressure from the rapid growth in social welfare needs, has changed its perspective toward volunteering and has begun to encourage citizens to engage in more volunteer service to improve the welfare of society (Tian Reference Tian2004). At the present time, the Chinese government is ascribing increasing importance to citizens’ volunteer behaviors. For example, on January 17, 2019, when General Secretary Xi Jinping visited the birthplace of Chinese community volunteer service organizations, he fully affirmed the important role played by Chinese volunteers in solving social problems and pointed out that volunteer service is an important force for enhancing the modernization of social governance in China (Du Reference Du2019). However, in contrast to Western countries, where grassroots volunteering dominates, the government of China initiates the emergence of and dominates the development of Chinese volunteer activities. Currently, the Chinese Young Volunteers Association (CYVA) and the China Volunteers Association (CVA) are the two largest volunteer service organizations (VSOs) in China. Led by the government, both the CYVA and the CVA use administrative powers to mobilize people to participate in volunteer activities. In 2010, for example, the CYVA and the Communist Youth League of China initiated the “Care for the Children of Migrant Workers Volunteer Activities” program and adopted a top-to-bottom mobilizing strategy. Volunteers were recruited through grassroots government units under the guidance of upper-level government units (Cheng and Zhang Reference Cheng and Zhang2015). In addition to government-organized VSOs, party and government organizations also play an active role in initiating and organizing volunteer activities. Volunteers are mainly recruited from among the employees of party and government organizations. However, there is a certain degree of mandate in the mobilization. For example, in a survey conducted in Jinan, Xu (Reference Xu2014) found that an officer in a subdistrict office (jiedao banshichu) in Jinan city was required to perform volunteer work. If he refused, deductions in his salary would result.
Over the past decades, there have been tremendous increases in government-organized volunteering in China. Meanwhile, grassroots VSOs have become a nonnegligible force in the field of volunteering, thus undermining the dominant role of the government in volunteer participation. For example, according to Tan (Reference Tan2008), during the 2008 Wenchuan Earthquake, millions of volunteers spontaneously participated in Wenchuan Earthquake-related volunteer activities, such as providing rescue assistance, cleaning up rubble and distributing rescue materials. This not only demonstrates to the world the volunteer spirit of the Chinese people but also suggests the great potential for volunteer services in China’s third sector (Shieh and Deng Reference Shieh and Deng2011). Thus, volunteering in China can be classified into two categories, namely government-organized volunteering and grassroots VSO-organized volunteering. While government-organized volunteering is mobilized based on a top-to-bottom mobilization strategy, grassroots VSO-organized volunteering is mobilized based on a bottom-up mobilization strategy (Smith Reference Smith2014; Xu Reference Xu2014).
Motivated by the scarcity of quantitative studies on volunteering in China, this study uses data from the 2013 Survey on Philanthropic Behaviors of Urban Citizens in China to explore who volunteers are, i.e., the predictors of volunteering in urban China. Although most existing studies on volunteering focus on Western societies, many researchers have begun to study this phenomenon on a global scale (Hustinx et al. Reference Hustinx, Handy and Cnaan2012). Issues related to who volunteers and why people in non-Western countries volunteer have aroused the intense interest of scholars. Considering that China is the most populous country in the world and that its sociopolitical structure differs vastly from those of Western structures, this paper expands our knowledge of volunteering in different cultural and political contexts.
Predictors of Volunteering
With research on volunteering in China in its infancy, two large-scale sample surveys on volunteering have been conducted in China. The first survey, which was conducted in 2001 and covered six provinces in China, revealed that the rate of participation in volunteering is as high as 85.2% (Ding et al. Reference Ding, Jiang and Tan2007). However, it is presumed that this is, to a great extent, an overestimation of the rate. The main reason for this is the survey’s broad definition of volunteering, as it included activities such as informal helping, e.g., helping a neighbor with errands. Furthermore, the results of this survey provided limited evidence on current volunteering given that many societal changes have occurred over the past two decades. The second survey was administered in 2010 and reported that 67.5% of people had never participated in any type of volunteer work. However, because the questionnaire was quite short, it prohibited a number of interesting multivariate analyses of volunteer behaviors from being conducted (Zhang Reference Zhang2011). In addition to these two sample surveys that were specifically designed to investigate volunteering, the 2012 Chinese General Social Survey (CGSS) contained several items regarding volunteering (Hu Reference Hu2017). Furthermore, some qualitative studies that have focused on community volunteers and youth volunteers as well as the relation between volunteering and social governance in China provide us with insights regarding volunteering in China (Lu Reference Lu2011; Luo and Ding Reference Luo and Ding2012; Tan Reference Tan2004; Xu Reference Xu2012, Reference Xu2014; Li Reference Li2019; Zhang Reference Zhang2019).
In this study, we employ social resource theory in accordance with Wilson and Musick (Reference Wilson and Musick1997, Reference Wilson and Musick1997) who argue that entry into the volunteer labor force requires a variety of resources, such as income and social network. Given that government-organized volunteering is common in China, political resources are perceived as having an important role in promoting volunteering in China. Hence, the following are suggested as influencing factors of volunteering in urban China: demographics (e.g., gender and age), socioeconomic status (e.g., education and income), social capital variables and political resources (Bekkers Reference Bekkers2004; Einolf Reference Einolf2011b; Gu et al.Reference Gu, Gu, Wei, Pan, Chen and Ge2013; Hackl et al.Reference Hackl, Halla and Pruckner2007; Smith Reference Smith2010, Reference Smith2014; Li Reference Li2019).
Demographic Factors
A survey conducted in the USA found that women volunteer at a higher rate than men across categories based on age and other major demographic characteristics (U.S. Bureau of Labor Statistics 2013). However, some multivariate studies have reported that the effect of gender is spurious because it disappears when we control for other variables, such as socioeconomic status variables (Einolf Reference Einolf2011a; Gallagher Reference Gallagher1994). Wilson and Musick (Reference Wilson and Musick1997) report that gender has a direct effect on informal helping but not on formal volunteering because formal volunteering includes a variety of activities that require different forms of capital. In contrast to the USA, no significant gender differences in volunteering have been found in Japan (Taniguchi Reference Taniguchi2010). In China, women are socialized into a caring role, which sensitizes them to other people’s needs (Huang and Ni Reference Huang and Ni2018). Additionally, the overall employment rate for women in 2010 was 60.8%, compared to 80.5% for men, a factor that may induce women to enter the volunteer workforce (Song Reference Song2011). Taken together, we expect that whether it is government-organized or grassroots VSO-organized volunteering, the probability of participation is higher among females than among males and that among those who volunteer, females contribute more time than do males.
In most Western societies, the relationship between age and volunteer participation is curvilinear; that is, it increases through middle age and tapers off later in life (Bekkers Reference Bekkers2004; Einolf Reference Einolf2009; Gallagher Reference Gallagher1994; Taniguchi Reference Taniguchi2010). However, among those who volunteer, older people often commit more hours to volunteering than younger people. For example, in the USA, the median time commitment for adults 65 years and older is 86 h/year, compared with 45 h/year for those between 35 and 44 years of age and 52 h/year for those between 55 and 64 years of age (U.S. Bureau of Labor Statistics 2013; Morrow-Howell Reference Morrow-Howell2010). Wilson and Musick (Reference Wilson and Musick1997) argue that there is an indirect impact of age on volunteering and that this relationship is mediated by physical health, cohort and life-cycle effects. Taniguchi (Reference Taniguchi2010), however, concludes that, in Japan, the relationship between age and volunteering is gender dependent. Specifically, men have the highest participation rate when in their 60s, whereas the rate for women is highest when they are in their 30s. With respect to the Chinese, according to the existing literature and given the fact that the elderly in China tend to care for their grandchildren after retirement (Chen and Lin Reference Chen and Lin2004; Yu Reference Yu2019), a curvilinear relationship between age and volunteering is expected for both the probability of participation in volunteering and time devoted to volunteering. This holds true for both government-organized volunteering and grassroots VSO-organized volunteering.
Studies have also determined that married individuals are more likely to volunteer than individuals who are separated, divorced or widowed (Rotolo and Wilson Reference Rotolo and Wilson2006). According to life course theories, marriage is related to adult roles such as parenting, steady socioeconomic status and increased social expectations, all of which are positively associated with volunteering (Oesterle et al.Reference Oesterle, Johnson and Mortimer2004; Wilson Reference Wilson2012). With respect to China, the Chinese people highly value their families. Therefore, marriage means more family responsibilities, not only for the newly formed family but also for the blood relatives of both the husband and the wife (Ding Reference Ding2001; Liu and Jia Reference Liu and Jia2020). Furthermore, in Chinese communities, spontaneous volunteering is not common (Zhang et al. Reference Zhang, Deng and Xin2016). Thus, while marrying and having children may increase informal mutual help among neighbors, it does not necessarily lead to an increase in formal volunteering. Based on these findings, we hypothesize that the association between marriage and grassroots VSO-organized volunteering is negative for both probability of participation and time devoted to volunteering and that marriage has no significant effect on government-organized volunteering.
A considerable number of empirical studies have documented a significant relationship between religion and volunteering. Religion affects volunteering positively through strengthening the sense of civic obligation and widening the social network of those who hold specific beliefs (Campbell and Yonish Reference Campbell, Yonish and Smidt2003; Lim and MacGregor Reference Lim and MacGregor2012; Omoto and Snyder Reference Omoto and Snyder1993; Saroglou Reference Saroglou2006; Son and Wilson Reference Son and Wilson2012). Son and Wilson (Reference Son and Wilson2012) argue that through the process of socialization, religious people acquire a sense of obligation that promotes prosocial behaviors such as volunteering. In a similar vein, Wilson and Musick (Reference Wilson and Musick1997) find that church attendance and the frequency of prayer have a positive effect on respondents’ volunteer behaviors. Furthermore, religion’s effect on volunteering can spread through personal networks to nonreligious people. For example, as concluded in a 2012 study, people who rarely or never attend religious services themselves are significantly more inclined to volunteer if they have religious friends (Lim and MacGregor Reference Lim and MacGregor2012). However, inconsistent with most existing studies, Saroglou (Reference Saroglou2006) suggests that religious people’s prosocial behaviors are only limited to certain close individuals. With regard to China, it must be considered that, in contrast to Western countries, China is a secular country with no strong, highly organized religions. Furthermore, all religious activities must be registered and approved by the government (Gilreath Reference Gilreath2008). Therefore, it can be concluded that religion plays a very limited role in promoting volunteering among the people of China.
Socioeconomic Factors
The effect of education on volunteering may be nonlinear, and the effect may vary based on certain variables, such as race (Huang et al. Reference Huang, Maassen van den Brink and Groot2009; Musick et al. Reference Musick, Wilson and Bynum2000). Education often facilitates volunteering through several means. First, education, including both formal schooling and informal learning, cultivates civic values and skills, heightens cognitive abilities and broadens horizons. Therefore, educated people are more acutely aware of social issues, e.g., chronic poverty and orphaned children, and are also more likely to be asked to volunteer. This awareness of needs and being asked to volunteer are two important mechanisms that promote volunteering (Bekkers and Wiepking Reference Bekkers and Wiepking2011; Brady et al. Reference Brady, Verba and Schlozman1995). Second, individuals with higher levels of education are more likely to achieve higher socioeconomic status and have a larger social network, both of which are positively associated with volunteering (Brand Reference Brand2010). Given certain circumstances, the relationship between education and volunteering is also curvilinear. Taniguchi (Reference Taniguchi2010) reports that in Japan, high school educated persons are more likely to volunteer than college-educated individuals and those with the lowest level of education. A similar curvilinear relation is also found among volunteer firefighters based on data from the USA (Thompson Reference Thompson1993). In China, formal schooling propagates traditional Chinese moral values, such as offering assistance and sympathy to the poor. Therefore, we expect that the effect of education on grassroots VSO-organized volunteering in urban China is significantly positive with respect to both participation probability and time devoted. In terms of government-organized volunteering, on the one hand, individuals with higher levels of education have more opportunities to work in party and government organizations and possess more social resources (Zhao and Li Reference Zhao and Li2006). On the other hand, the top-to-bottom mobilization strategies used to induce participation in government-organized volunteer activities are generally more inclined to motivate grassroots staff, many of whom generally have lower levels of education (Tan Reference Tan2004). Taken together, we predict that the effect of education on government-organized volunteering is minimal.
The empirical evidence on the relationship between income and volunteering is mixed (Feldman Reference Feldman2010; Lee and Brudney Reference Lee and Brudney2009; Pho Reference Pho2008). For example, Lee and Brudney (Reference Lee and Brudney2009), based on a rational choice approach, indicate that hourly wage has a negative effect on the probability of volunteering. When income is measured by household income, a curvilinear relation between income and the probability of volunteering is confirmed. Among those who volunteer, Freeman (Reference Freeman1997) argues that the value of time is inversely associated with volunteer hours, while Pho (Reference Pho2008) notes a positive association between wage and volunteer contributions with imputed wage values. Taniguchi (Reference Taniguchi2010), based on a resource approach, finds that in Japan, household income has a trivial effect on volunteer behavior. The absence of consensus among empirical studies may be because some study samples are solely composed of employed individuals, whereas others include all adults. As our study includes individuals with and without employment, household income is used to examine the association between income and volunteering. Since a top-to-bottom mobilization strategy is employed in government-organized volunteer activities, we predict that household income has an insignificant effect on volunteering organized by the government. Regarding grassroots VSO-organized volunteering, although spontaneous volunteering is not uncommon in China, it has not become a form of public participation recognized by the middle class. Hence, we predict that there is no significant association between household income and grassroots VSO-organized volunteering for either participation probability or time devoted.
Social Capital
Although there is agreement regarding the importance of social capital in promoting cooperation, there is no consensus regarding what constitutes social capital or how social capital should be measured (Coleman Reference Coleman1988; Fukuyama Reference Fukuyama1995; Putnam Reference Putnam1995). This paper adopts Putnam’s (Reference Putnam1995) definition of social capital, i.e., “features of social organization, such as trust, norms and networks that can improve the efficiency of society by facilitating coordinated actions.” In general, networks and trust are two dimensions of social capital used to explain the association between social capital and volunteering.
Participation in wider social networks facilitates formal volunteering because being recruited is an important driving mechanism behind volunteering (Bekkers and Wiepking Reference Bekkers and Wiepking2011; Freeman Reference Freeman1997). Musick et al. (Reference Musick, Wilson and Bynum2000) report that being recruited can increase the volunteer participation rate by 45%, whereas Paik and Navarre-Jackson (Reference Paik and Navarre-Jackson2011) argue that bonding social networks and bridging social networks have different effects on volunteer behaviors. Bonding social networks mobilize individuals to volunteer through social pressure, whereas bridging networks provide nonredundant sources of information related to volunteering. However, Brown and Ferris (Reference Brown and Ferris2007) suggest that social networks, which are measured by individuals’ associational networks, have no significant effect on volunteering. The mixed results on the relationship between social networks and volunteering may be due to the different measurements of social networks used in different studies. With regard to China, we expect that social networks have a significant positive effect on grassroots VSO-organized volunteering for either participation probability or time devoted. However, we hypothesize that social networks have no effect on government-organized volunteering, as it generally uses administrative powers to mobilize people.
The influence of trust on civic engagement has been the subject of previous studies (Brown and Ferris Reference Brown and Ferris2007; Delhey et al. Reference Delhey, Newton and Welzel2011; Uslaner Reference Uslaner2002; Wang and Graddy Reference Wang and Graddy2008; Wu et al.Reference Wu, Zhao, Zhang and Liu2018). For example, Fukuyama (Reference Fukuyama1995) argues that trust decreases the uncertainty inherent in different forms of exchanges and thus creates a foundation for cooperation. Brown and Ferris (Reference Brown and Ferris2007) suggest that a norm of trust, which is measured by trust in others and trust in the community, is positively related to participation in volunteering. Taniguchi (Reference Taniguchi2013) finds that in Japan, level of generalized trust is positively associated with irregular volunteering, whereas the association between generalized trust and regular volunteering may be weaker or even absent.
A Chinese person generally divides his/her social ties into three circles according to the distance between them. In other words, family members are the main constituents of the innermost circle. Familiar friends and other close persons belong to the middle ring, where particular trust can be developed through face-to-face social interactions. Most people in society, including strangers and unfamiliar friends, fit in the outermost ring, which is where generalized trust can be established (Chen Reference Chen1994; Fei Reference Fei1948; Luo Reference Luo2005). Given the findings of the existing research, we predict that generalized trust is significantly and positively associated with both types of volunteering for either participation probability or time devoted, whereas trust in friends and trust in relatives have no significant effect on either type of volunteering.
Wu et al. (Reference Wu, Zhao, Zhang and Liu2018) indicate that institutional trust plays an important role in facilitating philanthropic behavior in China. However, Wu et al. (Reference Wu, Zhao, Zhang and Liu2018) do not distinguish between trust in the central government and trust in the local government. In fact, Chinese people’s trust in the central government varies significantly from their trust in the local government (Liu and Du Reference Liu and Du2013). Therefore, this study uses the following three indicators to measure institutional trust: trust in the central government, trust in the local government and trust in neighborhood committees. Shi (Reference Shi2001) shows the Chinese people’s trust in the central government is based on a positive expectation because traditional Chinese culture defines the relation between individuals and the state as hierarchical rather than as “a reciprocal one in which the obligations of obedience and respect were contingent upon the model behavior of those with power.” In contrast, trust in the local government is based on what the local government has done. In short, the behaviors of the local government function as important signals to citizens regarding the moral standards of the society in which they live. Unfair, corrupt, inefficient, and biased practices in the administrative machinery of the local government force individuals to rely on themselves (Rothstein and Stolle Reference Rothstein and Stolle2008). Meanwhile, the citizens’ perception of the level of corruption within the local government is inversely related to their trust in government institutions (Chang and Chu Reference Chang and Chu2006). China’s economic reforms, which began in 1978, have resulted in rapid economic growth while also spawning corruption in many fields (Wang Reference Wang2005). Therefore, we predict that trust in the central government will significantly increase the probability of participating in and devoting time to both types of volunteering. Trust in the local government is negatively associated with grassroots VSO-organized volunteering, whether for the participation probability and time devoted. Furthermore, trust in the local government plays no role in influencing government-organized volunteering. For trust in neighborhood committees, since a large number of volunteer activities organized by the government require the assistance of the neighborhood committees, we expect a positive association between trust in neighborhood committees and the probability of participating in the volunteer activities organized by the government as well as the amount of time devoted to volunteer activities. The effect of trust in neighborhood committees on grassroots VSO-organized volunteering is trivial.
Finally, due to the popularity of the Internet in China, many grassroots VSO-organized volunteer activities are initiated through various Internet sites. Hence, we hypothesize that the effect of online trust on grassroots VSO-organized volunteering is significantly positive for both participation probability and time devoted but that it has no influence on government-organized volunteering.
Political Capital
In China, government-organized volunteering is common. People volunteer in response to the government or the CCP’s mobilization efforts (Xu Reference Xu2017;Yu Reference Yu2019;Hu Reference Hu2017). Consequently, members of the CCP and those who work for party and government organizations (dangzheng jiguan) or for state-owned enterprises have more opportunities to volunteer (Tan Reference Tan2014). Although a top-to-bottom mobilizing strategy is often adopted, volunteer activities organized by the government benefit their participants in several ways, including employment opportunities, postgraduate admissions and job promotions (She Reference She2013). Therefore, our study uses two variables, namely membership in the CCP and type of work unit, to measure political capital. Given the different motivational strategies adopted by government-organized and grassroots VSO-organized volunteering programs, we expect that members of the CCP or people working in party or government organizations or public institutions/state enterprises are more likely to participate in government-organized volunteering and commit more time to volunteer activities, whereas membership in the CCP and type of work unit have no significant effect on grassroots VSO-organized volunteering.
Method
Data
The data are derived from the 2013 Survey on Philanthropic Behaviors of Urban Citizens in China, which was conducted between August and December of 2013 by the School of Social Development and Public Policy of Beijing Normal University. Information collected includes demographic characteristics, occupation and income, social capital, participation in volunteer activities and donations. A stratified multistage sampling method was employed to select samples representative of Chinese urban residents. The first step was to randomly select 27 cities according to geographic location, population size and level of economic development. The second step was to randomly choose four residential communities in each city based on community type, namely commercial housing, Danwei housing,Footnote 1 public housing and old housing. Finally, in each community, 50 households were randomly selected, and one individual was then randomly selected from each household, i.e., the adult family member whose birthday was closest to July 1. Of the 5400 responses collected, 123 of the questionnaires were incomplete. Thus, 5277 samples were used in this study.
The survey was conducted using face-to-face interviews. One interviewer was selected for each of the communities, and each interviewer was to have a college degree and be familiar with the community. Before the survey was formally administered, the interviewers were sent to Beijing to receive rigorous training. Home interviews were then conducted with assistance from the director of the neighborhood committee, and the respondents were given a small gift in return for their participation.
Variables
Volunteer Behavior
There are four dependent variables considered in our study: (1) whether the respondent participated in any volunteer activities organized by the government in 2012, (2) the number of hours devoted to volunteer activities organized by the government in 2012, (3) whether the respondent participated in any volunteer activities organized by grassroots VSOs in 2012 and (4) the number of hours devoted to volunteer activities organized by grassroots VSOs in 2012. These four dependent variables are measured based on the following three questions: “Have you participated in volunteer activities launched by formal organizations in the past year?” Respondents who responded “Yes” were then asked how many hours they had devoted to volunteer activities in 2012 and through what types of channels they had volunteered. A value of zero was assigned to those respondents who stated that they had not volunteered at all in 2012. The volunteer channels were categorized into six types: (1) party and government organizations, (2) the Chinese Young Volunteers Association, (3) the China Community Volunteers, (4) grassroots VSOs, (5) religious organizations and (6) other NGOs. Government-organized volunteer participation refers to volunteer participations organized by the party and government organizations, public institutions, state enterprises or government-sponsored VSOs (such as the CYVA), while grassroots VSO-organized volunteer participation is defined as volunteer participations organized by grassroots VSOs, religious organizations and other grassroots NGOs.
Four types of independent variables are used in this study: (1) demographics (gender, age, marital status and religion), (2) socioeconomic status (education and household income), (3) social capital variables (social networks and trust) and (4) political capital variables (membership in the CCP and type of work unit).
Demographic Variables
This study codes gender such that a 0 is assigned for male and a 1 is assigned for female. Age is a continuous variable that measures experience. Given that existing evidence indicates that the relation between age and volunteering is curvilinear, we include age squared as a control variable (Bekkers Reference Bekkers2004; Einolf, Reference Einolf2009; Gallagher, Reference Gallagher1994). Marital status is measured using the following four categories: 1 = married (referent), 2 = divorced, 3 = widowed and 4 = never married. The religion variable is coded as 1 for those who claim to have a religion and 0 for those who claim to not have a religion.
Socioeconomic Variables
Education is measured based on the question, “What is the highest level of education you completed?” Three categories are considered: 1 = primary school or lower, 2 = junior or senior high school graduate (referent)Footnote 2 and 3 = at least some college. Household income is measured by quintiles of per capita disposable income according to the following five categoriesFootnote 3: 1 = low-income group, 2 = lower middle-income group, 3 = middle-income group, 4 = upper middle-income group and 5 = high-income group.
Social Capital Variables
Two dimensions of social capital are measured, trust and social network. Six types of trust are identified: (1) trust in the central government, (2) trust in the local government, (3) trust in neighborhood committees, (4) generalized trust, (5) particular trust and (6) online trust. Trust in the government is measured based on the respondents’ replies to three questions: (1) To what extent do you trust the central government? (2) To what extent do you trust the local government? (3) To what extent do you trust neighborhood committees? Generalized trust is determined based on the respondents’ replies to the question, “How much do you trust most people in society?” Particular trust is measured based on the degree of trust respondents have in their friends and relatives and online trust is measured based on the respondents’ degree of trust in blogs. Respondents rate their responses on a Likert scale ranging from “1 = do not trust at all” to “5 = trust completely.”
We use the Chinese New Year Greeting Network to measure network-based social capital; this measure was developed as a revised version of the position generator in the Chinese context by Bian and Zhang (Reference Bian and Zhang2001). Size of the network is measured by the number of people with whom the respondent interacted during the spring festival of 2012. The heterogeneity of the network is measured by the number of different occupations of all individuals with whom the respondent interacted during the Spring Festival of 2012.
Political Capital Variable
Two variables are used to measure the political capital variable, namely membership in the CCP and type of work unit. If the respondent is a member of the CCP, a value of 1 is assigned; if the respondent is not a member of the CCP, a value of 0 is assigned. Type of work unit is measured using the following five categories: 1 = party and government organizations (referent), 2 = public institutions/state enterprises, 3 = other work unit (including all full-time employment respondents whose work unit type does not belong to categories 1 and 2), 4 = unemployed or out of the labor force and 5 = flexible employment (including respondents with no fixed work units and part-time workers but not including college students who work part-time).
Models
Volunteer hours are nonnegative and have a cluster of observations at the value of zero, which is called a corner solution response. Wooldridge (Reference Wooldridge2010) discussed the relative merits of several approaches related to a corner solution response, such as the Tobit regression model, the Heckman sample selection model and the two-part model. For volunteer hours, Tobit regression is too restrictive because it assumes that a single mechanism governs the “participation decision” and the “amount decision.” Compared with Tobit regression, the Heckman sample selection model is less restrictive, as it estimates participation decision separate from the amount decision. Nonetheless, the Heckman sample selection model has often been used to deal with variables whose actual values are not observed for a large proportion of the cases. Therefore, it may not be appropriate for modeling volunteer hours because we are interested in the effects of covariates on actual volunteering as opposed to potential volunteering. In summary, the two-part model is used to study predictors of volunteering in urban China.Footnote 4 Specifically, logistic regression is employed to explore factors influencing the decision to participate in volunteering, and an ordinary least squares regression is then run to analyze predictors of volunteer hours among volunteers.Footnote 5 See Wooldridge (Reference Wooldridge2010) for further discussions on the two-part model. STATA 15.0 was used to carry out the two-part model in this study.
Results
Table 1 presents the summary statistics of the study variables grouped by volunteer status and the statistical significance of differences in the mean values between volunteers and nonvolunteers. Given the multistage sampling used in the 2013 Survey on Philanthropic Behaviors of Urban Citizens in China, standard errors are adjusted for within-cluster correlations. The data on volunteer hours are heavily skewed toward 0, with volunteers representing 40.2% of the total urban population. Among those who volunteer, the mean and median yearly hours volunteered are 20 and 6 h, respectively. Furthermore, there are significant differences in the means of many independent variables between volunteers and nonvolunteers. While volunteers tend to be younger than nonvolunteers, the difference is very small. Additionally, compared with nonvolunteers, the proportion of those who are married is higher among volunteers, whereas the proportion of those who are widowed is lower among volunteers. Because China is a secular country, there are few people who have an explicit religious belief. Accordingly, 9% of the volunteers stated that they have an explicit religion, which is three percentage points higher than nonvolunteers. The proportion of those who have some college education is higher among volunteers than among nonvolunteers, whereas among volunteers, the proportion having primary school education or less is lower. With regard to household income, while the mean of per capita disposable household income is slightly higher among volunteers than among nonvolunteers, the difference is not statistically significant.
Descriptive statistics of the study variables by volunteer status
Volunteers (n = 2121) |
Nonvolunteers (n = 3156) |
|||
|---|---|---|---|---|
M |
SD |
M |
SD |
|
Dependent variable |
||||
Hours volunteered [0, 1200]*** |
20.47 |
57.62 |
0 |
0 |
Demographic variables |
||||
Gender [0, 1]*** |
0.45 |
0.50 |
0.48 |
0.49 |
Age [18, 90]*** |
47.74 |
13.28 |
49.10 |
15.07 |
Marital status |
||||
Married [0, 1]** |
0.82 |
0.38 |
0.76 |
0.42 |
Divorced [0, 1] |
0.05 |
0.21 |
0.06 |
0.23 |
Widowed [0, 1]*** |
0.04 |
0.20 |
0.08 |
0.27 |
Never married [0, 1] |
0.09 |
0.28 |
0.10 |
0.30 |
Religion [0, 1]** |
0.09 |
0.29 |
0.06 |
0.24 |
Education |
||||
Primary school or less [0, 1]*** |
0.06 |
0.23 |
0.10 |
0.30 |
Junior or senior high school [0, 1]*** |
0.55 |
0.50 |
0.62 |
0.48 |
Some college [0, 1]*** |
0.39 |
0.49 |
0.27 |
0.45 |
Household income [1, 5]** |
2.96 |
1.44 |
2.82 |
1.43 |
Social capital |
||||
Generalized trust [1, 5]*** |
3.06 |
0.85 |
2.84 |
0.85 |
Trust in friends [1, 5]*** |
4.02 |
0.67 |
3.90 |
0.71 |
Trust in relatives [1, 5]*** |
4.57 |
0.65 |
4.51 |
0.66 |
Trust in the central government [1, 5]*** |
4.52 |
0.73 |
4.31 |
0.79 |
Trust in local government [1, 5]*** |
4.34 |
0.79 |
4.12 |
0.87 |
Trust in neighborhood committee [1, 5]*** |
4.40 |
0.70 |
4.10 |
0.81 |
Online trust [1, 5]*** |
2.39 |
0.90 |
2.29 |
0.87 |
Size of network [10, 98] |
37.01 |
20.81 |
36.18 |
18.36 |
Heterogeneity of network [2, 24]*** |
10.44 |
5.27 |
9.34 |
4.98 |
Party membership [0, 1]*** |
0.33 |
0.47 |
0.17 |
0.37 |
Work unit and employment status |
||||
Government/party agencies [0, 1]*** |
0.09 |
0.28 |
0.03 |
0.17 |
Public institutions/state enterprises [0, 1] |
0.16 |
0.36 |
0.11 |
0.32 |
Other work unit [0, 1] |
0.30 |
0.46 |
0.29 |
0.45 |
Unemployed or OLF [0, 1]*** |
0.36 |
0.48 |
0.42 |
0.49 |
Flexible employment [0, 1]*** |
0.09 |
0.29 |
0.15 |
0.35 |
OLF out of labor force
*P < 0.05; **P < 0.01; ***P < 0.001(two-tailed)
On average, all trust variables score significantly higher among volunteers than among nonvolunteers. Specifically, volunteers score highest on trust in relatives (4.57), followed by trust in the central government (4.52), trust in neighborhood committees (4.40), trust in the local government (4.34), trust in friends (4.02), generalized trust (3.06), and online trust (2.39). As Table 1 indicates, trust in relatives and trust in the central government are approximately the same and much higher than generalized trust. The same trust pattern is identified among nonvolunteers. The mean of the heterogeneity of the networks is slightly higher among volunteers than among nonvolunteers, and the difference is significant at the 0.01 level. However, there is no significant difference in size of network between volunteers and nonvolunteers.
Of the volunteers, 33% are members of the CCP, which is 16 percentage points higher than the result for nonvolunteers. The proportion of those who work for the party and government organizations is higher among volunteers than among nonvolunteers, whereas the proportion of those who are unemployed, out of the labor force or working as flexible employees is higher among nonvolunteers than among volunteers.
Table 2 presents the summary statistics for the study variables grouped by the type of volunteer participation. As indicated in Table 2, there are certain substantive distinctions between the two types of volunteering. First, participants who volunteered in activities organized by the grassroots VSOs spent an average of 26.88 h on volunteer activities in 2012, compared with 19.34 h spent by those who participated in volunteer activities organized by the government. Second, participants who volunteer for activities organized by grassroots VSOs are disproportionately female, younger, more religious and have a higher level of family income than those who volunteer for activities organized by the government. However, participants who volunteer for activities organized by the government are more likely to be members of the CCP and more likely to be employed in state-owned enterprises or public institutions. Third, participants in volunteer activities organized by grassroots VSOs exhibit greater levels of generalized trust and online trust than volunteers who participate in volunteer activities organized by the government. In contrast, volunteers who participate in activities organized by the government trust local governments and neighborhood committees more than participants who volunteer in activities organized by grassroots VSOs. It is important to note that both groups of volunteers have a high level of trust in the central government.
Summary statistics for the study variables by type of volunteer participation
Government-organized volunteer participation (n = 1760) |
Grassroots VSO-organized volunteer participation (n = 381) |
|||
|---|---|---|---|---|
M |
SD |
M |
SD |
|
Dependent variable |
||||
Hours volunteered [0, 1200]*** |
19.34 |
55.76 |
26.88 |
68.18 |
Demographic variables |
||||
Gender [0, 1] |
0.54 |
0.49 |
0.56 |
0.49 |
Age [18, 90]** |
47.98 |
12.97 |
46.56 |
14.48 |
Marital status |
||||
Married [0, 1]** |
0.83 |
0.38 |
0.79 |
0.40 |
Divorced [0, 1] |
0.05 |
0.22 |
0.04 |
0.19 |
Widowed [0, 1] |
0.04 |
0.20 |
0.05 |
0.21 |
Never married [0, 1]*** |
0.08 |
0.27 |
0.12 |
0.33 |
Religion [0, 1]** |
0.08 |
0.28 |
0.11 |
0.32 |
Education |
||||
Primary school or less [0, 1]*** |
0.06 |
0.24 |
0.03 |
0.18 |
Junior or senior high school [0, 1]*** |
0.54 |
0.49 |
0.60 |
0.49 |
Some college [0, 1]*** |
0.36 |
0.48 |
0.28 |
0.44 |
Household income [1, 5]** |
2.94 |
1.45 |
3.08 |
1.40 |
Social capital |
||||
Generalized trust [1, 5]** |
3.05 |
0.85 |
3.12 |
0.82 |
Trust in friends [1, 5] |
4.02 |
0.65 |
4.02 |
0.70 |
Trust in relatives [1, 5] |
4.57 |
0.66 |
4.56 |
0.62 |
Trust in the central government [1, 5] |
4.53 |
0.72 |
4.51 |
0.74 |
Trust in local government [1, 5]*** |
4.36 |
0.77 |
4.26 |
0.85 |
Trust in neighborhood committee [1, 5]** |
4.42 |
0.69 |
4.33 |
0.74 |
Online trust [1, 5]*** |
2.45 |
0.97 |
2.65 |
0.97 |
Size of network [10, 98]*** |
36.17 |
20.41 |
40.73 |
21.91 |
Heterogeneity of network[2, 24] |
10.42 |
5.22 |
10.59 |
5.46 |
Party membership [0, 1]*** |
0.34 |
0.47 |
0.29 |
0.46 |
Work unit and employment status |
||||
Government/party agencies [0, 1]*** |
0.09 |
0.28 |
0.06 |
0.24 |
Public institutions/state enterprises [0, 1]*** |
0.27 |
0.45 |
0.21 |
0.41 |
Other work unit [0, 1]** |
0.19 |
0.38 |
0.23 |
0.42 |
Unemployed or OLF [0, 1] |
0.36 |
0.48 |
0.38 |
0.48 |
Flexible employment [0, 1] |
0.09 |
0.28 |
0.11 |
0.31 |
OLF out of labor force
*P < 0.05;**P < 0.01; ***P < 0.001 (two-tailed)
The two-part model estimation results are presented in Tables 3 and 4. With respect to the logistic regression, the odds ratio (OR) is used to measure the strength of the association between the predictors and the probability of participating in volunteer activities. Regarding the OLS regression, the natural exponential of the coefficients is also presented because the dependent variable was log transformed.
Predictors of government-organized volunteering
Participation logistic |
Log-transformed volunteer hours OLS |
||||
|---|---|---|---|---|---|
OR |
SE |
Coeff. |
Exp (coeff.)a |
SE |
|
Demographic variable |
|||||
Gender |
1.39*** |
0.13 |
0.18** |
1.20** |
0.04 |
Age |
1.08*** |
0.02 |
0.03*** |
1.03*** |
0.01 |
Age squared |
0.99*** |
0.00 |
− 0.003*** |
0.99*** |
0.00 |
Marital status |
|||||
Married [ref.] |
|||||
Divorced |
0.93 |
0.14 |
− 0.02 |
0.98 |
0.08 |
Widowed |
0.61*** |
0.10 |
− 0.23*** |
0.79*** |
0.08 |
Never married |
0.92 |
0.18 |
0.02 |
1.02 |
0.07 |
Religion |
0.92 |
0.18 |
0.10 |
1.11 |
0.07 |
Education |
|||||
Primary school or less [ref.] |
|||||
Junior or senior high school |
1.15 |
0.17 |
0.05 |
1.05 |
0.07 |
Some college |
1.39 |
0.28 |
0.08 |
1.08 |
0.08 |
Household income |
0.99 |
0.05 |
− 0.03 |
0.97 |
0.21 |
Social capital |
|||||
Generalized trust |
1.26*** |
0.08 |
0.14*** |
1.15*** |
0.02 |
Trust in friends |
0.98 |
0.09 |
0.01 |
1.01 |
0.03 |
Trust in relatives |
0.94 |
0.10 |
0.01 |
1.01 |
0.03 |
Trust in the central government |
0.99 |
0.13 |
0.02 |
1.02 |
0.04 |
Trust in local government |
0.86 |
0.21 |
− 0.11** |
0.90** |
0.04 |
Trust in neighborhood committee |
2.01*** |
0.29 |
0.09*** |
1.09*** |
0.04 |
Online trust |
0.98 |
0.08 |
0.02 |
1.02 |
0.02 |
Size of network |
0.99 |
0.01 |
0.01 |
1.01 |
0.01 |
Heterogeneity of network |
1.03 |
0.02 |
0.03*** |
1.03*** |
0.01 |
Party membership |
2.11*** |
0.27 |
0.45*** |
1.57*** |
0.05 |
Work unit and employment status |
|||||
Government/party agencies [ref.] |
|||||
Public institutions/state enterprises |
0.80 |
0.16 |
− 0.28*** |
0.76*** |
0.08 |
Other work unit |
0.47*** |
0.09 |
− 0.49*** |
0.61*** |
0.08 |
Unemployed or OLF |
0.54*** |
0.11 |
− 0.45*** |
0.64*** |
0.09 |
Flexible employment |
0.39*** |
0.12 |
− 0.55*** |
0.58*** |
0.09 |
Pseudo-R-squared |
0.14 |
– |
|||
Log pseudolikelihood |
− 2481.82 |
– |
|||
Wald Chi-square |
11,876.03*** |
– |
|||
Adj R-squared |
– |
0.17 |
|||
OLF out of labor force
aNatural exponential of the coefficients
*P < 0.05; **P < 0.01; ***P < 0.001 (two-tailed)
Predictors of grassroots VSO-organized volunteering
Participation logistic |
Log-transformed volunteer hours OLS |
||||
|---|---|---|---|---|---|
OR |
SE |
Coeff. |
Exp (coeff.)a |
SE |
|
Demographic variable |
|||||
Gender |
1.40** |
0.20 |
0.09** |
1.09** |
0.03 |
Age |
1.01 |
0.04 |
0.01 |
1.01 |
0.01 |
Age squared |
0.99 |
0.00 |
0.001 |
0.99 |
0.00 |
Marital status |
|||||
Married [ref.] |
|||||
Divorced |
0.65 |
0.25 |
− 0.06 |
0.94 |
0.07 |
Widowed |
0.98 |
0.27 |
− 0.03 |
0.97 |
0.06 |
Never married |
1.09 |
0.24 |
0.05 |
1.05 |
0.05 |
Religion |
2.11** |
0.65 |
0.19*** |
1.21*** |
0.06 |
Education |
|||||
Primary school or less [ref.] |
|||||
Junior or senior high school |
2.25*** |
0.65 |
0.12* |
1.13* |
0.07 |
Some college |
2.17** |
0.85 |
0.17* |
1.19* |
0.07 |
Household income |
1.11 |
0.07 |
0.01 |
1.01 |
0.01 |
Social capital |
|||||
Generalized trust |
1.28** |
0.14 |
0.06** |
1.06** |
0.02 |
Trust in friends |
0.99 |
0.14 |
− 0.01 |
0.99 |
0.02 |
Trust in relatives |
0.92 |
0.16 |
− 0.01 |
0.99 |
0.03 |
Trust in the central government |
1.27** |
0.12 |
0.13*** |
1.14*** |
0.03 |
Trust in local government |
0.76** |
0.09 |
− 0.10 |
0.90** |
0.03 |
Trust in neighborhood committee |
1.52 |
0.34 |
0.11*** |
1.12*** |
0.03 |
Online trust |
1.21 |
0.13 |
0.04* |
1.04* |
0.02 |
Size of network |
1.12* |
0.01 |
0.02* |
1.02* |
0.01 |
Heterogeneity of network |
1.14* |
0.01 |
0.02* |
1.02* |
0.01 |
Party membership |
1.88*** |
0.35 |
0.21*** |
1.23*** |
0.04 |
Work unit and employment status |
|||||
Government/party agencies [ref.] |
|||||
Public institutions/state enterprises |
0.93 |
0.24 |
− 0.16 |
0.85 |
0.09 |
Other work unit |
0.88 |
0.26 |
− 0.16 |
0.85 |
0.10 |
Unemployed or OLF |
1.02 |
0.33 |
− 0.11 |
0.90 |
0.09 |
Flexible employment |
0.78 |
0.22 |
− 0.15 |
0.86 |
0.09 |
Pseudo-R-squared |
0.10 |
– |
|||
Log pseudolikelihood |
− 967.16 |
– |
|||
Wald Chi-square |
5050.91*** |
– |
|||
Adj R-squared |
– |
0.12 |
|||
OLF out of labor force
aNatural exponential of the coefficients
*P < 0.05, **P < 0.01, ***P < 0.001 (two-tailed)
Regarding demographic variables, as predicted, gender differences do exist. Specifically, females are more likely to participate and spend more time in both government and grassroots VSOs-organized volunteer activities than their male counterparts. Age has a nonlinear effect on government-organized volunteering, i.e., both the probability of participation and the time devoted increases with age until attaining their peaks, after which they decrease with age. Not consistent with our predictions, however, is that age has no significant effect on grassroots VSOs-organized volunteering. With respect to the effect of marital status, widowed persons are less likely to participate and spend less time in government-organized volunteer activities than are married persons. Inconsistent with our expectations, marital status plays no role in influencing grassroots VSO-organized volunteering. Whereas no significant relationship is found between religion and volunteering organized by the government, religion is significantly associated with volunteering organized by grassroots VSOs, a finding that is inconsistent with our expectations. Specifically, for people with explicit religious beliefs, the odds of participation is 2.11 times that of those without any explicit religious beliefs, and among those who volunteered, people with explicit religious beliefs contributed 21% more time than those without explicit religious beliefs.
When socioeconomic variables are considered, as predicted, household income plays no role in influencing either type of volunteering. For volunteering organized by grassroots VSOs, individuals with higher levels of education are more likely to participate in and commit more time to volunteering than are individuals with lower levels of education. However, consistent with our expectations, education has no significant effect on volunteering organized by the government.
Most interestingly, trust in local government has a significant negative impact on volunteering organized by grassroots VSOs. Specifically, with respect to volunteering organized by grassroots VSOs, for each one-unit increase in trust in local government, the odds of participation decrease by approximately 24% and the time devoted decreases by 10% among those who volunteer. In contrast to trust in local government, trust in the central government is positively associated with volunteering organized by grassroots VSOs. Specifically, with each one-unit increase in trust in the central government, the odds of participation increases by approximately 27%, and time devoted increases by 14% among those who volunteer. Inconsistent with our expectations, trust in the central government has no impact on government-organized volunteering, whereas trust in local government was negatively associated with the time devoted to volunteering organized by the government.
Generalized trust is positively associated with volunteering, whether organized by the government or by grassroots VSOs. With respect to volunteering organized by the government, with each one-unit increase in generalized trust, the odds of participation increase by approximately 26% and time devoted increases by 15% among those who volunteer. Regarding grassroots VSOs-organized voluntary activities, with each one-unit increase in generalized trust, the odds of participation increase by approximately 28% and volunteer hours increase by 6% among those who volunteer.
As predicted, there is a significant positive association between trust in neighborhood committees and volunteering organized by the government. However, inconsistent with our expectations, time devoted among those participating in grassroots VSO-organized volunteering increases by 12% for each one-unit increase in trust in the neighborhood committee. As hypothesized, although online trust has no role in influencing volunteering organized by the government, it is positively associated with time devoted among those participating in volunteering organized by grassroots VSOs.
As predicted, both the size of the social network and the heterogeneity of the network are positively associated with grassroots VSOs-organized volunteering for both participation probability and time devoted to volunteering. However, inconsistent with our expectations, there is a significant association between heterogeneity of the network and time devoted by participants in volunteer activities organized by the government.
As we hypothesized, membership in the CCP has a substantial positive effect on volunteering organized by the government. The odds of participating in volunteering organized by the government for members of the CCP is 2.11 times that of nonmembers of the CCP, and among those who volunteer, members of the CCP contribute 57% more time than nonmembers of the CCP. However, inconsistent with our hypothesis, but thought-provoking, membership in the CCP also has a significant positive impact on volunteering organized by grassroots VSOs. For members of the CCP, the odds of participating in volunteering organized by grassroots VSOs is 1.88 times that of nonmembers of the CCP, and among those who volunteer, on average, members of the CCP contribute 23% more time than nonmembers of the CCP.
Type of work unit is a factor when analyzing volunteering organized by the government. The odds of participating in volunteering organized by the government decrease by 47% for those who in work units outside of the party and government organizations, public institutions or state-owned enterprises compared to those who do work in the party and government organizations. Moreover, among those volunteer, on average, individuals working for other work units contribute 61% less time than those working for party and government organizations. As expected, type of work unit plays no role in influencing volunteering organized by grassroots VSOs.
Discussion
Volunteering has undergone rapid development in the course of political and economic reform in China. Values underlying the development of Chinese volunteer behavior include the benevolence valued in traditional Chinese culture and the spirit of Lei Feng,Footnote 6 which emerged during the planned economy. There is no doubt that the volunteer spirit from Western societies has had and is continuing to have a profound impact on volunteering in China. A salient feature of volunteering in China is that the Chinese government plays an important role in organizing and facilitating volunteer activities. Hence, in accordance with previous studies, our study distinguishes volunteering organized by the government from that organized by grassroots VSOs (Hustinx et al. Reference Hustinx, Handy and Cnaan2012; Smith, Reference Smith2014; Xu Reference Xu2014).
To mobilize volunteer work to efficiently meet the huge demand for social services and to promote the modernization of social governance in China, more studies on volunteer behaviors in China should be conducted. This study draws on a sample of 5277 people from 27 Chinese cities to determine who volunteers in urban China. Furthermore, our study indicates that volunteering organized by grassroots VSOs is very similar to volunteering in the USA and Japan. Demographic (e.g., gender), socioeconomic (e.g., education) and social capital variables (e.g., generalized trust) all help to explain volunteering organized by grassroots VSOs. Perhaps the most interesting finding of our study is that the association between trust in the central government and grassroots VSO-organized volunteering is significantly positive, whereas the association between trust in the local government and grassroots VSO-organized volunteering is significantly negative. While this finding is thought-provoking, it is not unexpected. China’s economic development, industrialization and postindustrialization periods have led to the rise of self-expression values that promote civic participation and reduce public trust in the government (Wang Reference Wang2005). Accordingly, the rise of self-expression values in China partially explains this finding, i.e., the relationship between volunteer behaviors and institutional trust. However, this relationship is complex, and the theories based on Western social experiences are limited in explaining social phenomena in China.
In China, volunteering organized by the government is still dominant, and the influencing factors of volunteering are significantly different from those observed in Western societies. First, volunteering has no significant association with education or socioeconomic status. Second, CCP membership and employment in party and government organizations/state-owned enterprises or public institutions are significant predictors of volunteering organized by the government (Jung and Kwon Reference Jung and Kwon2011; Taniguchi Reference Taniguchi2010).
The driving forces behind the two types of volunteer behaviors are very different. Specifically, the compulsory measures adopted by work units within the state system (tizhi) are the main driving mechanism of government-organized volunteer participation (Xu Reference Xu2014). The word “compulsory” is likely to cause confusion, particularly for Western scholars. For example, people may wonder why a compulsory behavior is recognized as a voluntary behavior. This study, however, defines such compulsory behavior as volunteering organized by the government because the Chinese government, academia and international institutions all recognize such behavior as volunteering behavior. In addition to compulsory measures, many volunteer activities organized by the government are, in practice, implemented by neighborhood committees, which, in China, have dual characteristics. On the one hand, according to the China’s constitution, neighborhood committees are mass organizations of self-management at the grassroots level. On the other hand, and in reality, neighborhood committees play the role of a “quasi-government agent.” As such, neighborhood committees connect the government and the residents of the country and serve as a buffer zone between them. In their daily work, neighborhood committees usually rely on community activists, and since there are no formal constraints between the neighborhood committees and community activists (Gui Reference Gui2007; Read Reference Read2003), the neighborhood committees rely on renqing (human sentiment) and guanxi (personal network established by a long history of personal interactions in daily life) to mobilize community activists. With respect to grassroots VSO-organized volunteering, its main resources are nearly the same as the resources utilized by volunteer organizations in Western countries, with the exception of trust in the central government and membership in the CCP.
The Fourth Plenary Session of the 19th Central Committee of the Communist Party of China puts forward clear requirements for the improvement in the social governance system and recognizes that volunteer service is an important way for people to participate in social governance, thus substantiating the premise that the Chinese government not only recognizes volunteers as important providers of social services but also envisages volunteers as an important force of mass autonomy (Guo,Reference Guo2019). The findings of this paper provide empirical evidence as the government strives to develop relevant management policies and facilitates NGOs in recruiting potential volunteers. With respect to the future of volunteering in China, it is expected that the two types of volunteering will coexist in China for a long time (Tan Reference Tan2015). Thus, a future study could explore the relationship between these two types of volunteering and the effects they may have on one another.
This study has several limitations. First, since the face-to-face interviews were conducted with the assistance of neighborhood committees, this study may overestimate the trust individuals have in their neighborhood committees. Second, while the data that support our study include information on physical health, physical health is poorly measured. Given that physical health is an important predictor of volunteering, its omission from our study may result in some estimation biases.
Funding
Research reported in this publication was supported by the National Social Science Foundation of China under Award Number 17ZDA095.
Compliance with Ethical Standards
Conflict of interest
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Open access
