The Effect of Folk Belief on Social Trust: The Mediating Role of Social Support and the Sense of Identity

Abstract

Although the influential factors of social trust have aroused heated discussion, the research on the influence of religious belief, especially Chinese folk belief, on social trust and the mechanism between the two is relatively insufficient. This study aims to explore the influence of folk beliefs on Chinese residents’ social trust and the mediating role of social support and the sense of identity. The empirical analysis of 23,823 Chinese residents shows that there is a significant positive correlation between folk belief and social trust. Social support and a sense of identity play a mediating role between folk beliefs and residents’ social trust. People with folk beliefs can significantly promote their social trust by improving their degree of social support and their sense of identity. In addition, the influence of folk belief on the social trust of residents in eastern, northern China and rural areas is more significant, showing regional and registered residence heterogeneity. Therefore, we should correctly interpret and popularize the core spirit and social and cultural significance of folk belief to form a bond of social trust. At the same time, folk beliefs can be returned to the people, which will help people find a sense of belonging and sense of identity and enhance their social trust.

1. Introduction

Over the past decade, social scientists have put a lot of effort into the study of social trust (Cook 2001; Fukuyama 1995; Sztompka 1999). Social trust is defined as “a social expectation that arises from a community of regular, honest, and cooperative behavior based on common norms” (Francis 1996). It is the foundation of social life and an important promoter of social capital and support. Social trust is the fundamental basis of social interaction within a society, and it is also an informal rule that supplements formal institutions such as laws and regulations to maintain social order (Zak and Knack 2001). There is a lot of evidence to show that trust between people reduces transaction costs and promotes cooperation between people, so it is very important for economic and social development (Beugelsdijk et al. 2004; Knack and Keefer 1997). The existence of social trust in China is different from that in Western countries, which is manifested by a high degree of trust in “in-group” members except for family members, while a low degree of social trust in “out-of-group” members. Long-term honesty and reliability are the way to gain trust, so “in-group” members have a close and specific relationship (Feng et al. 2016). Social trust can be enhanced by the consensus formed by communication and sharing through social media (Kim 2015). Social trust is also affected by a variety of factors, including political trust (Tao et al. 2014), socioeconomic status (Brandt et al. 2015), social networks (Sherchan et al. 2013), and ethnic diversity (Dinesen and Sonderskov 2015).

Among various influencing factors, cultural consensus brought by NGOs can also enhance social trust. Among them, religious belief is an important resource of social cohesion, and its relationship with social trust has been studied a lot. An examination of the connection between these two variables can be established by going back to Durkheim’s concept of mechanical solidarity. Mechanical solidarity is to emphasize the collective consciousness or the similarity within the group (Durkheim 1893). From the perspective of folk belief, among the groups with the same belief or belief, they have the similarity of “belief.” This similarity enables them to form a collective consciousness under a certain belief, which is the basis of most religious beliefs. Most studies have also found, as noted, that religious affiliation is associated with greater social trust, and if one belongs to a religion that is mainstream and widely seen as legitimate or if one belongs to a religion that is practiced by almost all members of the community, this will favor increased social trust. If you belong to a radical or extreme religion, belonging can put you on the margins of society and make you an outcast, which can lead to a decrease in social trust. It can be said that the influence of religious belief on social trust is complex. Existing studies have discussed the relationship between religious belief, including folk belief, and social support (Keating 2013; Moxey et al. 2011). In fact, compared with the influencing factors of social support, the influencing factors of social trust are more worthy of study and will highlight their advantages. First, conceptually, social trust is an important driving force in achieving social support. Social support usually refers to the individual’s ability to consult and guide in connection with the social environment, which is achieved through extensive social interaction and communication (Vaux and Harrison 1985); social trust arises in an environment based on common norms and is the foundation of social life (Francis 1996). It promotes the realization of social support between people and between people and the environment. Second, the effect of religious belief on social support will ultimately affect the establishment of social trust. The ability of a member of a society to obtain social support, whether individual or institution, is based on trust and performs its basic social function. The prosocial value of religious belief can enhance social trust by increasing trust among people who share the same religious identity (Yilmaz 2021), but the increase in religious diversity will aggravate the negative effect of religious belief on social trust (Berggren and Bjornskov 2011). This can be explained by religious identity theory. In a pluralistic society, religion tends to make use of distinction, participation, conflict and threat in relation to others by defining that individuals involved are not in relation to others around them, thus developing a stronger sense of group boundaries in a pluralistic background. Under the influence of this sense of group boundaries, the level of social trust will not improve. Existing studies have shown that religious belief has an impact on people’s social trust. Adam Smith also noted in “The Wealth of Nations” that religious belief can be important in increasing trust and credibility in a group of people. Religious belief is an underlying determinant of social trust, and its teachings teach believers ethics, morals, and social behavior. In situations of social conflict and the emergence of social insecurity, religious belief is increasingly valued as a constructive force for building social trust at different levels of interaction (Ruben 2011). However, most existing studies focus on Western countries and do not examine the relative influence of different religious traditions on the formation of social trust in non-Western countries. In this study, we attempt to extend this line of research by assessing the influence of folk beliefs on the social trust of residents in China. This question is important because the effects of different types of religion on social trust among members of a society can vary. China is a good case study. As the basis of China’s traditional religious landscape, China’s folk belief is a highly integrated system, including elements of Buddhism, Taoism and other traditional religious beliefs. China’s folk beliefs originated from traditional religions. With the development of politics, economy and culture, diversified religions have adjusted the original objects of belief and worship ceremonies by adapting different local social production modes (Zhu and Li 2021). Belief objects and worship rituals have become part of local folk belief traditions and customs. Therefore, compared with Western countries, Chinese folk belief is more interdependent and collectivist. Even in foreign countries, Chinese people can gather together to establish social organizations because of folk beliefs (Xu and Hamamura 2014; Yang and Hu 2012). In addition, unlike the monotheism of other religious countries, China tends to be more polytheistic. China stresses harmony but diversity. In front of folk beliefs, people can hold the idea of “believing is something, not believing is nothing.” Without formal rules, different people in the same community may have different religious beliefs (Yang and Hu 2012), and even worship and praying for good luck are not consistent. Research indicated that folk belief is indeed the mainstream of China’s religious market. About 70% of China’s population is a follower of a folk belief, while only five percent are members of an institutional religion, and 25% are not religious (Zhang et al. 2021). According to the World Values Survey (WVS), the proportion of Chinese residents who are religious has risen from 3% in 1990 to 15% in 2012. Therefore, the number of religious believers in contemporary China is growing rapidly (Yang 2010). However, the influence of religion in contemporary China has not been extensively studied, and given the increasing popularity of religion at present, China urgently needs more religion-based studies.

In conclusion, we take sample data from China as an example to study the relationship between folk belief and social trust, as well as the influencing mechanism of the relationship between the two, in order to fill the gap in existing research. First, we used data from 23,823 residents in China to verify the influence of folk beliefs on social trust through robustness tests to ensure that the conclusions obtained were representative. Then, the econometric model was used to investigate the mediating mechanism of the social support variable and the sense of identity variable on the relationship between folk belief and social trust. Finally, we examined whether the influence of folk belief on social trust was heterogeneous with one’s locality and registered residence.

2. Literature Review and Hypotheses

In the process of trust resource construction, the influencing factors of social trust have become a topic frequently discussed by scholars in recent years. As the product of combining regional characteristics and history, folk belief is not only the ideological and cultural sustenance independently chosen by individuals but also the way of building social relations. There are few research results on folk belief in related fields, but the influence of religious belief on social trust is mostly discussed. Domestic and foreign scholars have studied the relationship between religious belief and social trust through data from various countries. Although some researches indicate that religion has a negative impact on social trust (Berggren and Bjornskov 2011; Bjornskov 2007; La Porta et al. 1997), others believe that religion plays a vital role as a source of social networks and has a positive impact on social trust (Chuah et al. 2016; Delhey and Newton 2005). Religious beliefs and practices encourage individuals to develop values including morality, kindness and charity, which contribute to social trust (Hopkins 2011). For example, using the data from the third phase of China’s rural-urban migration project, some selected 13,732 eligible sample sizes. Through the ordered Probit regression, it was found that among Chinese migrant workers, religious believers had a stronger degree of social trust than non-believers, and religious participation could significantly improve the social trust of Chinese migrant workers. Some study estimated the impact of religious belief on the trust level of Turkish people by using the third-phase WVS survey data and found that religious belief and the frequency of participation in religious activities both had a positive and significant impact on social trust (Yilmaz 2021). Others conducted field laboratory experiments in Bangladesh and West Bengal, India, and randomly selected villagers to investigate the relationship between religion and relative status on social trust and credibility, and found that highly religious ethnic groups in the two places showed stronger intra-group social trust than those with a low religious degree. That is, the high-religious group trusts its members more (Gupta et al. 2018).

In addition, from the perspective of religious heterogeneity, some research compared the total sample and the group of highly religious believers and found that the degree of trust of Catholics and evangelical Protestants was significantly lower than that of mainline Protestants (Mencken et al. 2009). Isaac Addai et al. (2013) explored the relationship between Ghanaians’ religious affiliation, religious importance and social trust based on the personal data in the 2008 African Democracy Dynamics Survey. He adopted stratification and multiple regression technology and found that the overall influence of religion on social trust was weak, and compared with non-belief and traditional religious belief, There is a positive correlation between Catholic, Protestant, Muslim, Pentecostal/evangelical faith and social trust (Addai et al. 2013). R. Traunmuller (2011) used the survey data of the GSOEP wave in 2003 to conduct a multi-level analysis of 97 small-scale regions in Germany to explore the influence of different religious beliefs on social trust and found that Protestants were not only more likely to trust others but also that the social environment dominated by Protestants would increase social trust (Traunmuller 2011).

Research indicated that keeping religious commitments can be a means of building cooperation. Religious commitments create cooperation more effectively than other commitments based on secular networks. Rich religious ritual activities lead to the formation of various interreligious groups, which can facilitate cooperation and social trust formation on a larger scale (Irons 2004). At the same time, religious systems usually consist of eight core elements, namely authority, meaning, moral obligation, myth, ritual, sacred, supernatural, and taboo. While all elements may not interact directly, they all interact with the ritual. Based on the importance of ritual in the religious system, ritual performance can be understood as a barometer of social cooperation and trust (Sosis 2020). Studies have also shown that participation in religious services helps to build social support networks with fellow religious people and to receive more help from fellow religious people (Shaver et al. 2020). The religious beliefs and rituals of Christianity and other supernatural forces can unite people in a wider network and push people to constantly expand the scope of social interaction, which further emphasizes the interaction between religious beliefs and the content of cultural forms such as social-ecological environment (McNamara and Henrich 2018).

On the basis of the above conclusions, we propose the following hypotheses:

Hypothesis 1 (H1).

Folk belief has a significant positive effect on residents’ social trust. In addition, folk belief has a more significant positive effect on the social trust of people who have folk beliefs and are members of religious groups than people who have folk beliefs but are not members of religious groups.

Social support refers to the perception of the availability of help or support from others in their social network. In a high-trust environment, social trust helps to expand the range of social support (Sendroiu and Upenieks 2020). According to the affective theory of social exchange (Lawler 2001), more social support generated by more frequent social interaction will trigger people’s positive emotions, and these emotions, as reinforcement stimuli, will lead to primary and sustained interaction, thus enhancing people’s perception of network cohesion (Emerson 1976), and thus generating stronger “in-group” social trust (Lawler and Yoon 1996). Thus, the longer one attends church, the more opportunities one has to interact with other church members, accumulate more social support, and in turn, generate positive feelings about the group. This positive feeling is reflected in the fact that we tend to trust people we know more than people we don’t know well (Macy and Skvoretz 1998). As Coleman (Coleman 1988) theorizes, a closed network characterized by highly dense and overlapping sets of social relationships is an important factor in creating “in-group” trust in relationships.

Meanwhile, according to the social trust theory, individuals develop trusting attitudes when they interact with others in society (Daniels and von der Ruhr 2010). Religious activities can provide a platform, interactive network and communication resources between people, enhance their degree of social support, and thus enhance “in-group” social trust (Wuthnow 2002). For example, using survey data from the Portrait of American Life Study, research indicated that the positive relationship between religiosity and trust is driven not by religious belief or practice but by specific characteristics of the microscopic processes that take place in the church (e.g., closeness to religious leaders, the density of congregational relationships, and giving and receiving help from other parishioners) (Seymour et al. 2014).

In contrast to in-group trust, studies have also shown that religious believers increase their trust in strangers through social support and participation in religious ritual activities, namely “out-of-group” social trust. Using data from about 5000 rural-urban migrant families in 15 Chinese cities, the research concluded that religious teachings and opportunities for social support contributed to increased “out-of-group” social trust (Niu and Zhao 2018). In addition, research indicated that religious participation also helps rural society break the restrictions of kinship networks, provide new networks and social support, and thus enhance their “out-of-group” social trust (Zhang et al. 2019). And social support is in place precisely to contend with concerns regarding the free rider.

On the basis of the above conclusions, we propose the following hypotheses:

Hypothesis 2 (H2).

The association between folk belief and strangers’ social trust is mediated by social support.

Religion influences trust, and the way in which trustworthiness is achieved depends on people’s identity (Chuah et al. 2016). According to social identity theory, identity competition plays a central role in maintaining relationships between groups and is considered to be a byproduct of individuals’ efforts to satisfy basic human needs, including various psychological needs. In the process of achieving identity, religion is often more helpful than other factors in building and maintaining the cultural significance of individual and group identity and more effectively meeting these psychological needs (Seul 1999). As a system to understand primitive philosophical thoughts, religion can influence the construction of group identity and life meaning and ensure the function of social cohesion and legitimacy (Diaz 2008). Of the total number of Christian participants in Tan and Vogel’s study, the 44 people known to be highly religious received greater trust from others, especially from those who shared the same identity (Tan and Vogel 2008). On the other hand, some collected three samples of African American and Hispanic respondents from the United States, with a total sample size of 570, and conducted two experimental studies and a questionnaire study. Research has found that the sense of belonging to a minority group’s social identity itself leads, in turn, to an increase in social trust and social relationships (Valcke et al. 2020). The research examined the effect of the complexity of the trustor’s social identity on the degree of trust of others, out-group trust and in-group trust. The results showed that the complexity of the trustor’s social identity was positively correlated with its interpersonal trust and out-group trust. Activating the high complexity of the trustor’s social identity would produce a high level of interpersonal trust (Xin et al. 2016).

On the basis of the above conclusions, we propose the following hypotheses:

Hypothesis 3 (H3).

The association between folk belief and residents’ social trust is mediated by the sense of identity.

3. Methodologies

3.1. Data and Study Sample

The data used in this study are from the China Family Panel Studies (CFPS). The CFPS was carried out by China Center for Social Research at Peking University. The samples in the CFPS database are nationally representative. The survey team will constantly update the survey methods every year to ensure the authenticity and reliability of the survey results. Finally, the following databases are formed: family members, family economy, individual self-response, children’s parents’ surrogate response and individual surrogate response. The data covered 25 provinces (municipalities and autonomous regions) in China, reflecting the sociodemographic characteristics, way of life, and social opinions of Chinese citizens. In this study, data from CFPS in 2018 were used. Combined with the variables required by the research in this study, individual sample code is taken as the matching index to complete the horizontal data merger. The data contents include variables such as family size, family income at the family level and variables such as age, gender, marriage and education at the individual level. The object of this study is the entire population. After deleting the missing values of the variables required for the study, a total of 23,823 effective samples were obtained.

3.2. Measures

3.2.1. Dependent Variable

The dependent variable in this study is social trust, which refers to out-of-group trust or generalized trust. The question to measure the dependent variable was “How much do you trust strangers?”. The answers to the question were the continuous variable, ranging from 0 to 10. The higher the number, the stronger the sense of social trust of residents.

3.2.2. Independent Variable

The independent variable in this study is folk belief. The questions to measure the independent variable included: “Do you believe in gods?”, “Do you believe in ancestors?”, and “Do you believe in Feng Shui?”. In this study, the above questions were summed up. People who believed in at least one of them were regarded as having folk beliefs, which made folk beliefs become categorical variables, namely “0 = no belief” and “1 = belief”.

3.2.3. Mediating Variables

The mediating variable in this study is social support and the sense of identity. Social support use was measured with the question, “How well connected do you think you are?”, The answers to the question were the continuous variable, ranging from 0 to 10. The higher the value, the more social support the residents receive. The sense of identity was measured by using the question, “How would you rate your local social class status?.” The answers to the question were the continuous variable, ranging from 0 to 10. The larger the value, the stronger the sense of social class identity of residents.

3.2.4. Control Variables

Based on the characteristics of residents gathered through CFPS surveys, we chose control variables. Gender is the categorical variable: female is 0, male is 1. Age is a continuous variable. Marriage is the categorical variable, unmarried is 0, and married is 1. The highest level of education is the ordinal variable, “0 = below primary school”, “1 = primary school”, “2 = middle school”, “3 = university and above”. Personal income is a continuous variable. The registered residence is the categorical variable, the rural account is 0, and the urban account is 1. Health level is the ordinal variable, coded as “1 = unhealthy”, “2 = average”, and “3 = healthy”. Family size is the continuous variable, measured by the number of family members. Working status is the categorical variable, not working is 0, and working is 1. The level of medical care is the ordinal variable, and the question for measuring it is, “How do you think the level of medical care there?”; the “very bad” and “bad” options in the answer are combined and coded as “1 = bad,” “average” options as “2 = average;” the “good” and “very good” options are combined and coded as “3 = good”. Medical insurance is the categorical variable; 0 is not participating and 1 is participating.

3.3. Statistical Analysis

In this study, Stata16.0 software was used to analyze the influence of folk beliefs on residents’ social trust. First of all, this study made a descriptive analysis of dependent variables, independent variables, mediating variables and control variables. Secondly, the OLS model was used to analyze the influence of folk beliefs on residents’ social trust, and the ordered Logistic model and PSM method were replaced to test the robustness of the baseline regression results. Finally, the step-up method and KHB method were used to verify the mediating role of social support and subjective class identity between folk belief and residents’ social trust.

4. Results

4.1. Descriptive Analysis

Table 1. Descriptive statistics of ordinal/categorical variables.

Variable	Variable Type	Sort Percent
Folk belief	Categorical variable	0/unbelief 28.05%; 1/belief 71.95%
Gender	Categorical variable	0/female 47.66%; 1/male 52.34%
Marriage	Categorical variable	0/unmarried 15.69%; 1/married 84.31%
Highest level of Education	Ordinal variable	0/below primary school 21.32%; 1/primary school 20.48%; 2/middle school 44.26%; 3/university and above 13.94%
Registered residence	Categorical variable	0/rural 77.33%; 1/urban22.67%
Health status	Ordinal variable	1/unhealthy 15.06%; 2/general 12.98%; 3/healthy 71.96%
Working status	Categorical variable	0/no 18.80%; 1/yes 81.20%
The level of medical care at the point of the visit	Ordinal variable	1/bad 12.78%; 2/general 33.90%; 3/good 53.32%
Medical insurance	Categorical variable	0/not participating 7.55%; 1/participating 92.45%

and

Table 2. Descriptive statistics of Continuous variables.

Continuous Variable	Mean Value	Standard Deviation	Min	Max
Social trust	2.287	2.225	0	10
The sense of identity	3.098	1.079	1	5
Social support	7.115	1.942	0	10
Age	46.079	14.655	16	96
Personal income	4.822	5.146	0	13.641
Household size	4.245	2.053	1	21

show the descriptive statistics of each variable. The mean value of social trust is 2.287, which means that the social trust level of the whole sample is low. Those who did not believe in folk beliefs account for 28.05%, and those who believed in folk beliefs account for 71.97%. The mean value of the sense of identity is 3.098, which means that the sense of identity of the whole sample is at a medium level. The mean value of social support is 7.115, which means that the social support of the whole sample is at an above-average level. Females account for 47.66%, and males 52.34%. The average age of the sample is 46. Among the marital status of the sample, 15.69% are unmarried, and 84.31% are married. In terms of the highest level of education, those who were below primary school account for 21.32%, 20.48% in primary school, 44.26% in middle school, and 13.94% in university or above. The rural samples account for 22.67%, and the urban samples account for 77.33%. In terms of health level, the proportion of unhealthy samples is 15.06%, the proportion of generally healthy samples is 12.98%, and the proportion of healthy samples is 71.96%. The mean household size is 4.245. The sample proportion of not working is 18.80%, and the sample proportion of working is 81.20%. 92.45% of the residents in the survey sample are covered by health insurance.

4.2. Impact of Folk Belief on Social Trust

The dependent variable in this study is social trust, which ranges from 0 to 10. Therefore, the OLS model is used to analyze the influence of folk beliefs on residents’ social trust. The specific regression results are shown in

Table 3. OLS model of the influence of folk beliefs on social trust.

	Model 1	Model 2	Model 3
	Social Trust	Social Trust	Social Trust
Folk belief	0.047 **	0.058 **	0.068 **
	(0.03224)	(0.03)	(0.032)
Gender (The control group was female)
Male	0.496 ***	0.497 ***	0.508 ***
	(0.029)	(0.029)	(0.030)
Age	−0.000 ***	−0.000 ***	−0.000 ***
	(0.000)	(0.000)	(0.000)
Marriage (The control group was unmarried)
Married	−0.093 **	−0.089 **	−0.087 **
	(0.040)	(0.040)	(0.041)
Highest level of Education (The control group was below primary school)
Primary school	−0.298 ***	−0.299 ***	−0.281 ***
	(0.045)	(0.045)	(0.045)
Middle school	−0.076 *	−0.078 *	−0.060
	(0.043)	(0.043)	(0.043)
University and above	0.819 ***	0.816 ***	0.802 ***
	(0.062)	(0.062)	(0.062)
Personal income	0.012 ***	0.012 ***	0.012 ***
	(0.003)	(0.003)	(0.003)
Registered residence (The control group was urban)
Rural	−0.097 ***	−0.096 **	−0.110 ***
	(0.037)	(0.037)	(0.037)
Health status (The control group was unhealthy)
General	0.027	0.028	0.061
	(0.054)	(0.054)	(0.054)
Very healthy	0.237 ***	0.238 ***	0.217 ***
	(0.042)	(0.042)	(0.042)
Household size		−0.004	−0.004
		(0.007)	(0.007)
Working status (The control group was no)
Yes			0.062 *
			(0.037)
The level of medical care at the point of the visit (The control group was bad)
General			0.032
			(0.046)
Good			0.286 ***
			(0.044)
Medical insurance (The control group was not participating)
Participating			0.087
			(0.054)
Constant	1.994 ***	2.008 ***	1.734 ***
	(0.085)	(0.090)	(0.107)
N	23,823	23,823	23,823
R2	0.052	0.052	0.056

Note: 1. * p < 0.05, ** p < 0.01, *** p < 0.001; 2. The value in brackets is the standard error.

.

Table 3 reports the results estimated using the OLS model. Model 1 is the regression result after only adding control variables at the level of individual characteristics; Model 2 is the regression result after both adding control variables at the level of individual characteristics and family characteristics; Model 3 is the regression result after both adding control variables at the level of individual characteristics, family characteristics and social characteristics. Through the regression results of Model 3, it can be found that folk belief positively affects residents’ social trust at the significance level of 1%, indicating that the group with folk belief has a higher level of social trust than the group without folk belief. Among the control variables, gender positively affects residents’ social trust at the significance level of 0.1%, and the level of social trust of males is higher than that of females. Age negatively affects social trust at a significance level of 0.1%, meaning that people’s level of social trust decreases as they age. Marriage negatively affects social trust at the 1% significance level, and unmarried people have higher levels of social trust than married people. In terms of the highest level of education, it is found that the group with a higher education level has a higher level of social trust. In personal income, the significance level of personal income at 0.1% positively affects residents’ social trust. The higher the level of personal income, the higher the level of social trust. The significance level of registered residence at 0.1% negatively affects residents’ social trust, and urban residents have higher social trust than rural residents. In terms of health level, the healthier the body, the higher the level of social trust. In terms of work status, work status positively affects social trust at the significance level of 5%, and the level of social trust is higher among those with jobs than those without jobs. In terms of medical treatment level, the higher the level of medical treatment, the higher the level of social trust of residents.

It can be seen from

Table 4. Test results for members of a religious community.

Variables	Social Trust
	Member of a Religious Community	Not Member of a Religious Community
Folk belief	0.134 **	0.174
	(0.067)	(0.262)
Control variables	All control variables	All control variables
N	463	16,677
R2	0.034	0.021

Note: 1. ** p < 0.01; 2. The value in brackets is the standard error.

that among members of religious communities, folk beliefs have a significant positive effect on their social trust. In the non-religious community, folk belief has no significant effect on their social trust. Whether he is a member of a religious community determines, to a certain extent, how often he participates in religious ceremonies and activities.

As can be seen from

Table 5. Test results by specific folk belief.

	Social Trust
Believe in gods	0.125
	(0.332)
Believe in ancestors	0.506 **
	(0.253)
Believe in fengshui	0.987 ***
	(0.372)
Control variables	All control variables
N	23,823
LR Statistic	1512.36
Pseudo R2	0.027

Note: 1. *** p < 0.001, ** p < 0.01; 2. The value in brackets is the standard error.

, in specific types of folk beliefs, ancestor belief and feng shui have positive responses to their social trust at the significance level of 5% and 0.1%, respectively. Belief in God, on the other hand, had no significant effect on social trust. In addition, among those who believe in God, 27.3% are registered members of religious activities, while among those who believe in ancestry, 56.9% are registered members of religious activities, and among those who believe in Feng Shui, 42% are registered members of religious activities.

4.3. Robustness Test

4.3.1. Replace the Model

The dependent variable in this study is social trust, which ranges from 0 to 10 and can be regarded as a discrete and ordered variable. Therefore, an ordered Logistic model is constructed to test the robustness of the model results in

Table 6. Robustness test.

	Social Trust
Folk belief	1.082 ***
	(0.028)
Control variables	All control variables
N	23,823
LR Statistic	1534.36
Pseudo R2	0.017

Note: 1. *** p < 0.001; 2. The value in brackets is the standard error.

to verify the credibility of the regression results in Table 3. The robustness test results are shown in Table 6. The regression results in Table 6 are consistent with those in Table 3, showing a positive correlation between folk trust and the social trust level of the samples at the significance level of 0.1%. This illustrates the robustness of the findings.

4.3.2. The Results of Propensity Score Matching

Due to confounding factors influencing the connection between folk belief and social trust, our results were affected by a sample selection bias. Consequently, propensity score matching was employed to rectify the results of the regression. First, given that the independent variable used for propensity score matching was a dummy variable, the sample was separated into two groups: those with folk belief and those without folk belief. Second, we matched the samples using the techniques outlined below: (1) we transformed the control variables into an indicator and determined the probability that each sample belonged to the experimental group. (2) To match the experimental and control groups, we used five matching methods: 1–1 nearest neighbor, 1–4 nearest neighbor, radius matching, kernel matching, and local linear regression matching, and divided individuals with similar values into the same group to ensure that group members had similar characteristics. (3) We determined the average treatment effect (ATT) of the participants, which measures the difference in happiness between the experimental and control groups.

Table 7. Propensity score matching results (ATT).

Dependent Variable: Social Trust
		ATT	SE	t
Nearest Neighbor Matching	1–1 Matching	4.79 ***	0.13	4.28
	1–4 Matching	4.56 ***	0.14	4.35
	Radius Matching	4.27 ***	0.15	4.28
The whole Matching	Kernel Matching	4.43 ***	0.13	4.33
	Local Linear Regression Matching	4.28 ***	0.14	4.32

Note: *** p < 0.001.

displays the ATT calculated using various matching strategies. In all models, the ATT value is more than zero, which is statistically significant, demonstrating that folk belief has a statistically significant beneficial effect on residents’ social trust.

4.4. The Mechanism Analysis of the Influence of Folk Belief on Social Trust

In this study, the step-up method proposed by Baron and Kenny (Baron and Kenny 1986) and the KHB method were used to test the mediating effect of social support and the sense of identity on the relationship between folk belief and social trust. The results are shown in the table below.

4.4.1. Testing the Moderating Effects of Social Support

In Model 1 of

Table 8. Folk belief and residents’ social trust: The mediating role of social support.

	Model 1	Model 2	Model 3
	Social Trust	Social Support	Social Trust
Folk belief	0.068 **	0.139 ***	0.062 **
	(0.032)	(0.028)	(0.032)
Social support			0.047 ***
			(0.007)
Gender (The control group was female)
Male	0.508 ***	−0.072 ***	0.511 ***
	(0.030)	(0.026)	(0.030)
Age	−0.000 ***	0.000 ***	−0.000 ***
	(0.000)	(0.000)	(0.000)
Marriage (The control group was Unmarried)
Marriage	−0.087 **	0.046	−0.089 **
	(0.041)	(0.036)	(0.041)
Highest level of Education (The control group was below primary school)
Primary school	−0.281 ***	−0.007	−0.280 ***
	(0.045)	(0.040)	(0.045)
Middle school	−0.060	0.002	−0.060
	(0.043)	(0.039)	(0.043)
University and above	0.802 ***	0.247 ***	0.790 ***
	(0.062)	(0.055)	(0.062)
Personal income	0.012 ***	0.002	0.011 ***
	(0.003)	(0.003)	(0.003)
Registered residence (The control group was urban)
Rural	−0.110 ***	0.005	−0.110 ***
	(0.037)	(0.033)	(0.037)
Health status (The control group was unhealthy)
General	0.061	0.006	0.061
	(0.054)	(0.048)	(0.054)
Very healthy	0.217 ***	0.364 ***	0.200 ***
	(0.042)	(0.037)	(0.042)
Household size	−0.004	0.019 ***	−0.004
	(0.007)	(0.006)	(0.007)
Working status (The control group was no)
Yes	0.062 *	0.169 ***	0.054
	(0.037)	(0.033)	(0.037)
The level of medical care at the point of the visit (The control group was bad)
General	0.032	0.023	0.031
	(0.046)	(0.041)	(0.046)
Good	0.286 ***	0.247 ***	0.274 ***
	(0.044)	(0.039)	(0.044)
Medical insurance (The control group was not participating)
Participating	0.087	0.246 ***	0.075
	(0.054)	(0.048)	(0.054)
Constant	1.734 ***	5.774 ***	1.461 ***
	(0.107)	(0.095)	(0.115)
N	23,823	23,823	23,823
R2	0.056	0.022	0.058

Note: 1. * p < 0.05, ** p < 0.01, *** p < 0.001; 2. The value in brackets is the standard error.

, folk belief positively affects residents’ social trust at the significance level of 1%, and holding folk beliefs can improve residents’ social trust level. In Model 2, folk belief positively affects residents’ social support at the significance level of 0.1%, indicating that folk belief can improve residents’ social support. In Model 3, folk belief positively affects residents’ social trust at the significance level of 1%. Social support has a significant positive effect on residents’ social trust at the level of 0.1%. The influence of folk belief on residents’ social trust decreased from 6.8% in Model 1 to 6.2% in Model 3, indicating that social support is one of the mechanisms through which folk belief affects residents’ social trust. Having folk beliefs can enhance residents’ social trust by improving the level of social support.

In order to test whether social support plays a mediating role in the influence of folk belief on residents’ social trust, this study also uses the KHB method to test. In

Table 9. Test results of the mediating effect of social support.

Mediating Variables	Effect of Type	Coefficient	Standard Error	Z	p
Social support	Total effect	0.068	0.032	2.17	0.030
	Direct effect	0.062	0.032	1.96	0.050
	Indirect effect	0.007	0.002	3.94	0.000

, the total effect of folk belief on residents’ social trust is 0.068, passing the significance test of 5%. The direct effect was 0.062, which passed the significance test of 5%. The indirect effect of folk belief on residents’ social trust through social support is 0.007, which passes the significance test of 1%, indicating the existence of the mediating effect of social support. In summary, both the stepwise method and the KHB test found that social support plays an intermediary role in the relationship between folk belief and residents’ social trust, and social support is the mechanism influencing the relationship between folk belief and residents’ social trust.

4.4.2. Testing the Moderating Effects of the Sense of Identity

In Model 1 of

Table 10. Folk belief and residents’ social trust: The mediating role of the sense of identity.

	Model 1	Model 2	Model 3
	Social Trust	The Sense of Identity	Social Trust
Folk belief	0.068 **	0.032 **	0.064 **
	(0.032)	(0.015)	(0.032)
The sense of identity			0.144 ***
			(0.014)
Gender (The control group was female)
Male	0.508 ***	−0.007	0.509 ***
	(0.030)	(0.014)	(0.030)
Age	−0.000 ***	0.000 ***	−0.000 ***
	(0.000)	(0.000)	(0.000)
Marriage (The control group was unmarried)
Married	−0.087 **	0.115 ***	−0.103 **
	(0.041)	(0.019)	(0.041)
Highest level of Education (The control group was below primary school)
Primary school	−0.281 ***	−0.119 ***	−0.264 ***
	(0.045)	(0.022)	(0.045)
Middle school	−0.060	−0.176 ***	−0.035
	(0.043)	(0.021)	(0.043)
University and above	0.802 ***	−0.040	0.808 ***
	(0.062)	(0.030)	(0.062)
Personal income	0.012 ***	−0.010 ***	0.013 ***
	(0.003)	(0.002)	(0.003)
Registered residence (The control group was urban)
Rural	−0.110 ***	0.130 ***	−0.129 ***
	(0.037)	(0.018)	(0.037)
Health status (The control group was unhealthy)
General	0.061	0.136 ***	0.042
	(0.054)	(0.026)	(0.054)
Very healthy	0.217 ***	0.273 ***	0.177 ***
	(0.042)	(0.020)	(0.042)
Household size	−0.004	0.010 ***	−0.005
	(0.007)	(0.003)	(0.007)
Working status (The control group was no)
Yes	0.062 *	0.072 ***	0.052
	(0.037)	(0.018)	(0.037)
The level of medical care at the point of the visit (The control group was bad)
General	0.032	0.130 ***	0.014
	(0.046)	(0.022)	(0.046)
Good	0.286 ***	0.295 ***	0.243 ***
	(0.044)	(0.021)	(0.044)
Medical insurance (The control group was not participating)
Participating	0.087	0.100 ***	0.073
	(0.054)	(0.026)	(0.054)
Constant	1.734 ***	2.066 ***	1.436 ***
	(0.107)	(0.051)	(0.110)
N	23,823	23,823	23,823
R2	0.056	0.081	0.060

Note: 1. * p < 0.05, ** p < 0.01, *** p < 0.001; 2. The value in brackets is the standard error.

, folk belief positively affects residents’ social trust at the significance level of 1%, and holding folk beliefs can improve residents’ social trust level. In Model 2, folk belief positively affects residents’ sense of identity at the significance level of 1%, indicating that folk belief can enhance residents’ sense of identity. In Model 3, folk belief positively affects residents’ social trust at the significance level of 1%. At the level of 0.1%, the sense of identity has a significant positive effect on residents’ social trust. The influence of folk belief on residents’ social trust decreased from 6.8% in Model 1 to 6.4% in Model 3, indicating that the sense of identity is one of the mechanisms through which folk belief affects residents’ social trust. Having folk beliefs can improve residents’ social trust by improving their sense of identity.

In order to test whether class identity plays a mediating role in the influence of folk belief on residents’ social trust, this study uses the KHB method to test. In

Table 11. Test results of the mediating effect of the sense of identity.

Mediating Variables	Effect of Type	Coefficient	Standard Error	Z	p
class identity	Total effect	0.068	0.032	2.17	0.030
	Direct effect	0.064	0.032	2.03	0.042
	Indirect effect	0.005	0.002	2.06	0.039

, the total effect of folk belief on residents’ social trust is 0.068, passing the significance test of 5%. The direct effect was 0.064, which passed the significance test of 5%. The indirect effect of folk belief on residents’ social trust through the sense of identity is 0.005, passing the significance test of 5%, indicating the existence of the mediating effect of the sense of identity. In summary, both the stepwise method and the KHB test found that the sense of identity played an intermediary role between folk belief and residents’ social trust, and the sense of identity was the mechanism influencing the relationship between folk belief and residents’ social trust.

4.5. Heterogeneity Analysis

Different empirical phenomena show that the relationship between residents’ folk beliefs and their social trust is affected by their locality and registered residence. Therefore, this study attempts to analyze the heterogeneous effects of regional factors and residence factors on the relationship between residents’ folk beliefs and social trust.

First, this study divides the samples into three parts based on the eastern, central and western regions, respectively, to investigate the influence of folk beliefs on residents’ social trust in different regions. The sub-sample test results in

Table 12. Test results by region of East, Middle and West.

Variables	Social Trust
	East	Middle	West
Folk belief	0.126 ***	0.085	−0.098
	(0.045)	(0.061)	(0.064)
Control variables	All control variables	All control variables	All control variables
N	10,743	6172	6908
R2	0.070	0.060	0.039

Note: 1. *** p < 0.001; 2. The value in brackets is the standard error.

show that folk belief has a positive and significant impact on the population in the eastern region, and the eastern residents with folk belief have a higher level of social trust than those without folk belief. Folk belief had no significant effect on the social trust of middle and western residents. The difference in folk beliefs in social trust between East, middle, and West may be due to the influence of population size and economic development. Chinese folk belief includes the belief in ancestors and feng shui, which are the sustenance of people’s smooth life or good wishes. Meanwhile, there are many kinds of folk beliefs in China, and people can choose more spiritual sustenance. The eastern region has a large population scale and high population density. Meanwhile, the economic development in the eastern region is fast. People have a higher pursuit of career and financial resources, so they have a stronger belief in folk religions. In many circumstances, religious societies provide a source of comfort for those who are in uncertain situations. People who attend religious communities experience a greater sense of spiritual comfort. In contrast, the middle and western region has a small population, sparse residence and a small scale of human interaction. At the same time, affected by economic development, the belief in folk religions is less than that of the eastern region, which lacks the desire for an economy of work and life. Therefore, the belief level may be affected, resulting in a lower level of social trust.

In addition, we divide the samples into two parts based on the north and south, respectively, to investigate the influence of folk beliefs on residents’ social trust in different registered residences. The sub-sample test results in

Table 13. Test results by region of North and South.

Variables	Social Trust
	North	South
Folk belief	0.127 ***	−0.091
	(0.04)	(0.153)
Control variables	All control variables	All control variables
N	13,390	10,433
R2	0.054	0.060

Note: 1. *** p < 0.001; 2. The value in brackets is the standard error.

show that folk belief has a positive and significant impact on the social trust of the population in the north, and the north residents with folk belief have a higher level of social trust than those without folk belief.

Then, we divide the samples into two parts based on the rural and urban, respectively, to investigate the influence of folk beliefs on residents’ social trust in different registered residences. The sub-sample test results in

Table 14. Test results by registered residence.

Variables	Social Trust
	Rural	Urban
Folk belief	0.058 **	0.104
	(0.028)	(0.162)
Control variables	All control variables	All control variables
N	18,422	5401
R2	0.075	0.068

Note: 1. ** p < 0.01; 2. The value in brackets is the standard error.

show that folk belief has a positive and significant impact on the population in rural, and the rural residents with folk belief have a higher level of social trust than those without folk belief. In rural areas, the level of social trust of folk believers is 0.058 units higher than that of non-believers. Folk belief had no significant effect on the social trust of urban. Chinese religious believers in rural areas lack more opportunities for social interaction than those in urban areas. Religion-related activities can enhance their social trust by providing them with opportunities to exchange and share experiences in an atmosphere of equality. Furthermore, urban communities have robust institutions in place to police the free riders. This kind of perfect institutional system is conducive to the construction of a trusting environment in urban communities so as to mobilize community residents to really participate in community practice and finally achieve the goal of mutual supervision and mutual building of trust relationships.

5. Discussion

Folk belief, as the basis of Chinese traditional religion, is a highly integrated religious system. Therefore, China’s religious characteristics are different from those of other religious countries in the world. Therefore, the study of Chinese folk beliefs is of great significance for understanding the diversity of world religions. However, in China, there are very few empirical studies on the influence of folk beliefs on residents’ social trust. In addition, in China, there are few differences in the impact of folk beliefs on social trust based on specific beliefs of folk beliefs, social support and identity, as well as specific geographical and social backgrounds (such as different regions and registered residence, etc.). On this basis, it discusses the influence of folk belief on Chinese residents’ social trust and the intermediary mechanism of the relationship between the two. In fact, in the process of folk belief’s influence on social trust, the power of religious participation cannot be ignored. Belief is not as important as action. People tend to trust those who are willing to bear irrational short-term costs, such as religious sacrifices and rituals, for the long-term collective good. Irrational actions are more persuasive than irrational beliefs. Sam Harris argues that we cannot afford to pay lip service to a modest commitment to religious reconciliation, which only blinds us to the real dangers of fundamentalism, and that we should adopt secular humanistic approaches to the world’s problems (Harris 2005). Therefore, when we study the influence of folk beliefs on social trust, irrational behavior and participation also need to be paid attention to and explained.

5.1. Folk Belief Has a Significantly Positive Affect on Social Trust in China

In this study, we used data from 23,823 Chinese samples to examine the relationship between folk beliefs and residents’ social trust. We tested the hypothesis with the OLS model and concluded that folk belief has a significant positive impact on Chinese residents’ social trust. This conclusion is completely consistent with a large number of Western empirical studies on the relationship between religious belief and social trust (McCullough et al. 2016; Upenieks and Orfanidis 2022). In addition, among members of religious communities, folk belief has a significant positive effect on their social trust. In the non-religious community, folk belief has no significant effect on their social trust.

5.2. Social Support Plays a Mediating Role in the Influence of Folk Belief on Residents’ Social Trust

We found that social support plays a mediating role in the influence of folk beliefs on residents’ social trust; that is to say, folk beliefs have a significant positive impact on residents’ social trust through the improvement of residents’ social support degree. Folk belief has the cohesive advantage of strengthening social support. It comes from social members, serves social members, provides them with a place for communication, exchange and support, and promotes the generation of benign interpersonal bonds. Folk believers hold some activities and rituals of folk belief, such as weddings and funerals, offering sacrifices to ancestors, paying tribute to gods, building beams, farming and praying for rain, etc., so that various voluntary groups and networks are formed among each other to meet different needs. Through these groups or networks, social solidarity and social support are strengthened, and social trust between people is strengthened. Therefore, we should also emphasize that religious participation plays a significant role in promoting social support. In a study of 115 community seniors aged 65–90, research indicated that religious participation was more important in alleviating the distress of those with less social support (Dulin 2005).

5.3. The Sense of Identity Plays a Mediating Role in the Influence of Folk Belief on Residents’ Social Trust

We found that the sense of identity plays a mediating role in the influence of folk belief on residents’ social trust; that is to say, folk belief has a significant positive impact on residents’ social trust through the improvement of the sense of identity of residents. The sense of identity refers to the recognition and emotional attachment of social members to a certain group, which is based on cultural identity. Hogg proposed a subjective identity theory based on his previous work on subjective uncertainty and group identification (Hogg 2001). He believes that subjective uncertainty makes people’s identity become significant in highly substantive groups with ideological belief systems. The more ideological a belief system is, based on group membership, the better it will be at reducing uncertainty (Hogg 2005). Common ancestor worship, totem worship and festival ceremonies based on folk beliefs are all important cultural factors for the generation of the sense of identity. Through its unique forms of external expression, it makes its cultural ideas deeply rooted in the hearts of the people and forms deep-rooted cultural psychology and habits. By participating in ritual activities, social members with folk beliefs enhance their sense of identity, generate team centripetal force and cohesiveness, and enhance social trust for social members within the group in the process of contact with different cultures. Therefore, we should also emphasize that religious participation plays a significant role in the formation of social identity. The research analyzed the effect of religious participation on the social identity of 97 Istanbul immigrants. The results showed that religious participation improved the level of social identity. People with high attendance scored higher on social identity (Cetin 2019).

5.4. The Positive Effects of Folk Belief on Social Trust of Social Members in Rural and Eastern China Are More Significant

The influence of folk beliefs on Chinese residents’ social trust is heterogeneous. Among them, the influence of folk belief on the social trust of rural residents is more significant than that of urban residents. This may be because China’s social foundation is a rural society. The folk beliefs originating in rural society itself have a strong rural nature. It is based on the initial religious worship in rural society. It constantly absorbs the elements of other belief forms and accumulates and evolves to form a set of god worship concepts, behavior habits and corresponding ritual system and organization mode. People living in rural social fields are more likely to be affected by folk beliefs than those living in cities, and the degree of trust in social members is more different due to the influence of folk beliefs. In addition, compared with the central and western regions of China, folk belief has a more significant positive impact on the social trust of social members in eastern China.

5.5. Limitations

There are some limitations to this study. First of all, this paper focuses on the relationship between folk belief and social trust in the context of China. First of all, folk belief actually covers many types and dimensions and is not the only standard type summarized in this article. Second, there are different measures of the concept of social trust. However, due to limited data and questionnaire questions, this study did not study the impact of folk belief types on social trust in more detail. In addition, richer measures of social trust will be further improved in the future. Secondly, due to the limitation of public survey data, although this paper reveals the correlation between folk belief and residents’ social trust at the individual level, it is unable to explore the psychological process behind their interaction at the more micro-cognitive level. In future studies, the relationship between the two can be studied from different dimensions through secondary data research supplemented by quantitative methods, such as original data surveys and experiments.

6. Conclusions

The results show that there is a significant positive correlation between folk belief and Chinese residents’ social trust, and social support and a sense of identity play an intermediary role in the relationship between folk belief and social trust. Based on this, we propose the following suggestions: First, correctly interpret and popularize the core spirit and social and cultural significance of folk beliefs so that contemporary people can better understand and respect these traditional cultural elements, respect and fear the wisdom and character of ancestors, and form a bond of social trust; Second, we need to bring back folk beliefs. Some traditional practices based on folk beliefs have now become an important link between history and reality, and one of the key points is that it enables traditional folk culture to be passed on from generation to generation. The return of folk belief to the people will help people find a sense of belonging and identity and enhance their social trust.