The Impact of Social Capital on Migrants’ Social Integration: Evidence from China

Abstract

In recent decades, a large influx of migrants from rural areas to cities has brought numerous pressures to urban infrastructure, the environment, public services, and social stability. Therefore, promoting the social integration of migrants is crucial for the sustainable development of cities. Analyzing this situation from the perspective of social capital, as opposed to economic and human capital, can better reveal the underlying mechanisms behind social integration. Therefore, we sought to explore the impact mechanism of social integration among China’s migrants by establishing a comprehensive model that incorporates bonding social capital (BSC), bridging social capital (BRC), relative deprivation (RD), hometown attachment (HA), and social integration (SI) through research based on the 2017 China Migrant Dynamic Survey (CMDS). The results indicate that different types of social capital have varied impacts. Bridging social capital is positively correlated with social integration, with relative deprivation playing a mediating role between the two. In contrast, bonding social capital is negatively correlated with social integration, with hometown attachment moderating the relationship between them. This study enriches social capital theory in the field of migration research and provides recommendations for government efforts to promote social integration

1. Introduction

The issue of migrant social integration is particularly crucial in today’s globalized context. With the intensification of global population movements, difficulties in migrant integration can lead to social fragmentation, resource imbalances, and economic inequalities, thereby affecting societal stability and the goals of sustainable development [1].

In China, rapid industrialization and urbanization have prompted a large number of surplus rural laborers to migrate to cities. According to the Seventh National Population Census, as of 2020, China’s internal migrant population reached 376 million, accounting for 26.6% of the total population and representing an increase of nearly 70% compared to 2010 [2]. Although large-scale population mobility can increase the flexibility of the labor market and promote economic growth, it is challenging for migrants to truly integrate into society due to fragmentation caused by the urban–rural dualistic system in China, which leads to discrimination against them in many aspects, such as employment, housing, healthcare, and education [3].

With regard to addressing these issues, there is increasing recognition that social integration is not simply an economic adaptation process but a multidimensional composite process that includes economic integration, cultural acceptance, behavioral adaptation, psychological identification, and more. Economic capital provides migrants with basic living security and economic independence, helping them to become stable in their new environment [4]. Human capital enhances migrants’ employment opportunities and income levels through education, skills, and health [5]. With further research, it has gradually been discovered that the economic integration of the migrant population relies on social relationships; social capital, which focuses on cultural interaction and social networks, has become an important factor influencing social integration [6]. In the context of sustainable development, social capital is defined as the networks and relationships that encourage trust and reciprocity, which shape the quality and quantity of social interactions within a society [7]. Social capital can generally be categorized into “strong” and “weak” social capital or “original” and “new” social capital [8]. However, there is a lack of detailed research on the impacts of these types of social capital on the social integration of migrants.

Therefore, in this study, we primarily used social capital theory to examine the influences of different types of social capital on the social integration of migrants, as well as to explore its underlying mechanisms. By analyzing a sample of migrants in China, we sought to construct a comprehensive multidimensional framework for social integration, identifying significant mediating and moderating effects. This research holds significant implications for the self-development of migrants and the formulation of relevant policies, with the aims of facilitating better integration into society, stabilizing social order, and fostering a positive social atmosphere.

2. Literature Review

2.1. Social Capital and Social Integration

According to the development of social capital theory, social capital can be categorized into two levels: individual and collective. From an individual perspective, social capital is defined as resources embedded in a social structure that can be accessed or mobilized for purposeful actions, such as information, job opportunities, knowledge, social support, etc. [9]. Collective social capital is expressed as networks of residents, and the norms of reciprocity and trust that emerge from them, which help participants to foster cooperation and action, as well as improve social efficiency [10]. Despite various theoretical perspectives, the fact is that individual social capital is closely related to collective social capital, through a key element: the social network. The process of social integration involves building relationships between different groups, forming social networks, and aiding the integration of migrants into urban life. Essentially, it is also a process of cultivating new forms of social capital. Current research shows that promoting the accumulation of localized social capital among migrants is a crucial factor in facilitating their integration into urban life [11]. The changes in social capital before and after migration fit with Patnam’s social capital theory [12]; he categorizes social capital into bonding social capital and bridging social capital. The former is a form of homogeneous social capital, which emphasizes strong cohesion among internal members and exclusivity; meanwhile, the latter falls under heterogeneous social capital, emphasizing interactions across different groups and openness [13]. Therefore, it is more apt to elucidate the process and mechanisms of the integration of migrants from the theoretical context of Putnam.

In academia, scholars in the field of new economic sociology, have introduced social capital into the study of social integration [14]; one such scholar is Portes, who argues that resources within the social networks and relationships of new migrants will influence their willingness to migrate, their choice of destination, and their social adaptation after migration [15]. This influence is even more evident among foreign Chinese communities, such as Wenzhou City in Paris and Chinatown in New York [16]. People migrate to urban areas for a variety of reasons, mostly because of improved economic opportunities and access to services. Some research has shown that the original kinship, blood ties, and geographical connections of farmers can help them find non-agricultural jobs in urban areas [17]. There is also research indicating that, if migrants have frequent contact with familiar acquaintances in the local community, their frequency of participating in community activities is higher, a factor which can facilitate quicker integration into urban life [18]. Bonding social capital refers to the relationships and social networks that connect individuals who are similar in terms of their social identity, such as family members, close friends, or people from the same ethnic group [19]. It has been proven that, during the early stages of urban migration, the bonding social capital of migrants plays a positive role [20]. Beyond providing psychological support, it also enhances levels of adaptation in employment, economic aspects, and behavioral adjustments through social networks; however, social integration is a long-term and complex process. With increasing research depth, more and more scholars have found that, although bonding social capital facilitates strong cohesion and a high degree of solidarity among its members, it can also help to achieve intragroup support [21]. However, the excessive focus on “us” and “them” has led to the separation of migrant groups from local indigenous groups, thus hindering the construction of a new social network and affecting migrants’ identity and sense of belonging [22]. Some scholars refer to this internally formed social circle as a “subculture”, which is significantly different from the mainstream urban culture and hinders the urban integration of a migrant population [23].

Based on the abovementioned issues, some scholars believe that, when migrants face prolonged difficulties in integrating into urban life, and their existing social networks fail to meet their expectations for obtaining more valuable resources and improving their living standards, they will seek to construct new social networks with members possessing better resources through which to enhance their adaptability to urban life [24]. The social capital formed through these new cross-cultural and cross-group interactions is known as bridging social capital; it plays a crucial role in the subsequent deep integration of the migrant population [25]. Research reveals that bonding social capital aids migrants in securing low-skilled jobs with either non-technical or low-technical content; however, bridging social capital has the potential to reduce the likelihood of engaging in such low-skilled employment [26]. It has also been demonstrated that bridging social capital helps migrants to participate in collective bargaining and, thus, increase their bargaining power over wages, in addition to learning from each other through positive interactions with other groups, which can help to raise their incomes [27]. Additionally, scholars have found that groups with more bridging social capital tend to have higher levels of political participation and a stronger urban identity [28]. Based on these factors, one aim of this study is to determine the extent to which social capital is positive or negative in relation to social integration.

2.2. The Mediating Effect of Relative Deprivation

Social stratification is a significant societal phenomenon. Migrant groups often occupy lower social strata due to resource scarcity. When there is a discrepancy between perceived deserved status and actual attained status, a sense of relative deprivation arises [29]. The concept of relative deprivation was first introduced by Runciman, who suggested that individuals experience a sense of deprivation when the four following fundamental conditions are met in comparison to a reference group: (1) the individual lacks x; (2) the individual aspires to possess x; (3) the individual observes others within the reference group having x; and (4) the individual’s expectation of attaining x is considered reasonable and achievable [30]. Broadly, relative deprivation refers to a subjective psychological state experienced by those who feel a sense of unfairness and dissatisfaction after comparing themselves to others. Research has addressed the income–happiness paradox, in which individual income is positively correlated with subjective well-being. However, when an individual compares their income to those of others within their reference group, it tends to be negatively correlated with their own sense of happiness. This phenomenon is also known as the social comparison effect or relative deprivation effect [31]. Narrowly defined, relative deprivation refers to the objective economic state of individual relative deprivation, which introduces the concept of a reference group. This term is widely applied in the field of income distribution, reflecting the state of income inequality at the individual level [32].

Bridging social capital helps to alleviate income inequality for individuals by building positive social relationships that provide support and share resources. A study based on a sample of Malaysian and Bhutanese farming households demonstrated that social capital has a significant poverty-reducing effect and reduces income inequality among farming households [33]. There is also research based on an analysis of US Census data, which reveals that, for each 1% increase in residents’ social capital, the income inequality index measured with the Gini coefficient decreases by 0.2% [34]. Additionally, Chinese studies have drawn similar conclusions, indicating that the accumulation of social capital effectively alleviates income inequality caused by insufficient income mobility [35].

The deprivation of migrants’ income can lead to a series of psychological and social issues. For individuals, those with higher income deprivation are in a state of relative poverty, which hinders their access to basic medical services and health protection [36]. Moreover, as the sense of relative deprivation intensifies, it may induce more negative emotions. These stresses and anxieties can directly affect health, leading to conditions such as heart disease, hypertension, and suicide, and they may also indirectly lead to harmful behaviors like smoking, alcohol abuse, and drug use [37]. Additionally, relative deprivation is associated with the occurrence of crime and social conflict [38], which further affect social stability. For groups, the effects of deprivation vary among different populations; for example, migrant children who move with their parents tend to feel more alienation and insecurity compared to urban local children. This psychological state can lead to pessimism and negative emotions, thus affecting their sense of belonging to schools, communities, and the city [39]. Within the migrant worker population, there is also a hierarchy, in that those with lower economic income experience a stronger sense of relative deprivation. As they leave their rural ties behind and enter the city to reshape their social network, they endure greater psychological stress [40]. Therefore, we aim to uncover the extent to which relative deprivation mediates the association between bridging social capital and social integration.

2.3. The Moderating Effect of Hometown Attachment

Place attachment involves three aspects of interaction, among person, place, and process, and describes the positive emotional bonds individuals have with a place and their tendency to maintain close ties with it [41]. During the migration process, individuals not only retain deep affection for their original place of residence but also develop emotional attachments to new regions as they gradually integrate into new environments. However, migrants, due to their relatively disadvantaged position in terms of human and social capital, often require a longer time to form deep attachments to the places they move into [42].

In the context of urbanization in China, hometown attachment reflects the nostalgia of rural-to-urban migrants for their hometown’s way of life, customs, and moral beliefs. Although early studies indicated a negative correlation between hometown attachment and mobility [43], later research found that geographical mobility does not weaken migrants’ attachment to their homeland; they maintain deep connections with their place of origin [44]. In fact, migrants with more bonding social capital fit this characteristic, maintaining close ties with their home and through kinship and locality, reflecting their satisfaction and identification with their homeland. Moreover, connection and attachment to a place can increase the willingness to stay there, and may even lead to a tendency to return [45]. Moreover, during the migration process from rural to urban areas in China, due to the restrictions of the household registration system, many migrants are unable to access public services and welfare based on their household registration, which puts them in an unstable situation in the city and potentially increases the likelihood of their returning to their place of origin. In light of these issues, we pose the following research question: “Does hometown attachment positively moderate the negative impact of bonding social capital on social integration?”.

In summary, we constructed a comprehensive framework with which to explain the relationships between social capital, relative deprivation, hometown attachment, and social integration, as shown in Figure 1. Both bonding social capital and bridging social capital directly influence the degree of social integration for migrants. It is important to note that their mechanisms of action differ. Hometown attachment moderates the impact of bonding social capital, while relative deprivation mediates the influence of bridging social capital.

3. Methods and Variables

3.1. Methods

The data used in this study were collected from the 2017 China Migrants Dynamic Survey (CMDS). We chose the CMDS data because they met the needs of our research. Firstly, the CMDS was designed and implemented by the National Health Commission of China, using the PPS sampling method and covering 31 provinces, ensuring good representativeness. Secondly, the survey content includes background information, family status, employment situation, willingness to stay, and social integration, providing measurement indicators for our core variables. Thirdly, the survey targets migrants aged 15 and over who have resided in the local area for one month or more. After screening the total sample and excluding cases with missing values, we obtained an effective sample size of 169,894.

The empirical part of this study was divided into three steps. First, we conducted descriptive statistics on the data using SPSS 24.0. Second, considering that the variables related to social integration are continuous, while social capital is a binary variable, we conducted principal component analysis (PCA) and logistic PCA on these variables separately, using SPSS 24.0 and R 4.0. PCA is a statistical method used to reduce the dimensionality of data while preserving as much variance as possible [46]. While testing the reliability and validity of the scales, we generated observable social capital and social integration indices. Finally, we performed basic regression analysis and structural equation modeling (SEM) analysis in Stata to test the relationships between key variables [47], including the impacts of different types of social capital on various dimensions of social integration, as well as to examine the mediating effect of relative deprivation and the moderating effect of hometown attachment.

3.2. Variables and Measurement

3.2.1. Social Integration

Regarding the measurement of social integration for migrants, representative studies mainly divide it into economic integration, political integration, cultural integration, and social participation [48,49]. We constructed the social integration index for the migrant population considering four dimensions: economic integration (EI), cultural integration (CI), behavioral integration (BI), and psychological integration (PI). (1) In terms of economic integration, we measured the three following items related to income and expenditure: “Average monthly housing expenditure for your household in the local area over the past year”, “Average monthly total expenditure for your household in the local area over the past year”, and “Average monthly total income for your household over the past year”. Through understanding these factors, we can understand the stability and adaptability of migrants in the economic dimension, as higher income and reasonable expenditure levels usually indicate better economic integration. (2) Differences in cultural beliefs and living habits are important aspects of cultural integration. The higher the degree of cultural assimilation, the better the integration of migrants with the local culture. The measurement of cultural integration focuses on the differences in life views between migrants and locals. We asked respondents to rate their agreement with the following statements: “I feel that locals look down on migrants”. “It is important for me to adhere to the customs of my hometown when handling affairs.” and “There are significant differences in hygiene habits between me and local residents”. Responses ranged from completely disagree to completely agree, coded as 1 to 4, respectively. They were then reverse coded, so that higher values indicated higher levels of cultural assimilation. (3) Social participation is a concrete manifestation of behavioral integration. To measure behavioral integration, we used items related to the social participation of the migrant population. We asked respondents whether they had participated in specific activities since 2016, including “providing suggestions or supervising management at their workplace, community, or village”, “reporting situations or proposing policy suggestions to government departments”, and “engaging in donations, voluntary blood donation, and volunteer activities”. The level of participation in these activities was assessed by answering the question “from never to frequently”, with values ranging from 1 to 4. Higher values indicate higher frequencies of social participation and higher levels of behavioral integration. (4) Psychological integration reflects migrants’ subjective sense of identification with their current place of residence. In terms of psychological integration, we focused on the migrants’ subjective evaluation of their current locality, which included statements like “I like the city/place where I currently reside”, “I pay attention to the changes in the city/place where I live”, and “I am very willing to integrate into the local community and become one of them”. The responses range from “strongly disagree” to “strongly agree”, which were assigned values from 1 to 4, respectively, with higher numbers indicating a greater degree of psychological integration. Finally, the above items were subjected to principal component analysis (PCA) to generate indices for each of the four dimensions separately, as well as a composite index for overall social integration.

3.2.2. Social Capital

Bonding social capital refers to the connections within similar or closely linked groups, such as relationships between family, relatives, or close friends; it emphasizes strengthening internal ties, providing emotional support, and sharing resources. Bridging social capital refers to the connections and networks between different social groups, which can facilitate the transmission of information and a broader sense of identity; it emphasizes inclusiveness and connections that cross social boundaries [13]. Therefore, we measured these variables using the following five questions: (1) Have you participated in trade union activities locally since 2016? (2) Have you participated in volunteer association activities locally since 2016? (3) Have you participated in alumni association activities locally since 2016? (4) Have you participated in fellow villager association activities locally since 2016? (5) Have you participated in hometown chamber of commerce activities locally since 2016? Responses of “yes” were coded as 1, while responses of “no” were coded as 0. Finally, we used logistic principal component analysis to generate indices for bonding and bridging social capital.

3.2.3. Relative Deprivation

Some studies measure relative deprivation through subjective socioeconomic status assessments or self-reported evaluations, reflecting individuals’ subjective experiences when facing objective deprivation [50]. Other research employs objective measurement indicators, such as the Yitzhaki index, Kakwani index, Podder index, and Esposito index [51,52,53,54]. These indices primarily gauge relative deprivation based on income inequality among individuals. Considering potential biases in subjective evaluation information within survey data, we selected the widely recognized Kakwani index as an objective measurement method with which to measure relative deprivation. The Kakwani index, building upon the Yitzhaki index, further considers the average income of reference group members, fulfilling the requirement for scale invariance. The calculation formula is as follows:

$${Kakwani}_x=\frac{1}{Nx\mu z}{\sum }_{y=1}^{Nx}({income}_{zy}-{income}_x)$$

(1)

In this equation, income_x refers to the average monthly household income of a migrant individual, x (hereinafter referred to as individual x); income_zy represents the average monthly household income within the county (district) of z that is higher than that of individual x, belonging to another individual, y. Nx denotes the total number of other individuals whose household average monthly income is higher than that of individual x, and these Nx individuals constitute the reference group for individual x. By subtracting the average monthly household income of these Nx individuals in the county (district) of z from that of individual x, and then averaging the differences, we obtain the Yitzhaki index for individual x. μz is the average monthly household income of individual x’s reference group. The Kakwani index is confined to the range [0, 1], which avoids the large variations in the Yitzhaki index values that can occur with changes in individual income. Subsequently, we calculated the relative deprivation index for surveyed migrants based on the survey items, “average total monthly household income over the past year” and “current residential address county (district)”.

3.2.4. Hometown Attachment

Environmental perception refers to individuals’ cognition, emotions, subjective evaluations, and the special significance attributed to a particular place [55]. Hometown attachment typically refers to individuals’ positive emotional connections to their hometown. Therefore, by definition, environmental perception is considered the foundation of hometown attachment. Research has shown that the perception process is a psychological and sociocultural one. In this process, the more frequently individuals interact with the environment, the deeper their perception of it becomes [56]. For migrants, the frequency of interaction with their hometown often reflects the extent of their hometown attachment. Therefore, we can measure hometown attachment by considering the frequency with which respondents visit their hometown, with the question “How long has it been since you last visited your hometown?” Responses of one year or more are assigned a value of 0, while those less than a year are assigned a value of 1.

3.2.5. Covariates

According to the literature, we controlled for several demographic variables that affect migrant social integration [57,58]. These include: gender (female = 0, male = 1), age, educational level (no formal education = 1, primary school = 2, junior high school = 3, high school/technical school = 4, junior college = 5, undergraduate = 6, graduate = 7), and marital status (unmarried = 0, married = 1).

4. Results

4.1. Descriptive Statistics

The data in

Table 1. Descriptive statistics.

Variable	Specifications	Mean	Standard Deviation
Gender	Female = 0, male = 1	0.52	0.50
Age	≥15 years old	36.66	11.07
Marital status	Unmarried = 0, married = 1	0.81	0.39
Education level	1–7	3.44	1.16

indicate that the proportion of males and females in the sample is nearly equal. In terms of age distribution, the average age of the surveyed migrants is approximately 37 years old. Married migrants make up 80% of the sample, exceeding the proportion of unmarried migrants. The average educational level is 3.44, which corresponds to an educational attainment not higher than high school or technical school, suggesting that most migrants have a relatively low level of education.

4.2. Factor Analysis

Given the numerous items involved in social capital and social integration, it is necessary to perform dimensionality reduction on the data. Therefore, we used PCA to explore the potential structure of the data. The results in

Table 2. Factor analysis and validity test.

Variable	Items	Loadings	CR	AVE	$\sqrt{\mathbf{A}\mathbf{V}\mathbf{E}}$	Correlation Coefficient
EI	Housing expenditure	0.772	0.879	0.709	0.842	0.431–0.703
	Total expenditure	0.904
	Total income	0.845
CI	Social exclusion	0.708	0.754	0.506	0.711	0.209–0.371
	Customs and traditions	0.677
	Hygiene habits	0.747
BI	Community involvement	0.814	0.757	0.519	0.720	0.211–0.433
	Political participation	0.797
	Volunteer participation	0.510
PI	Favorability	0.861	0.883	0.716	0.846	0.579–0.635
	Attention	0.856
	Willingness to integrate	0.820
BSC	Trade union	0.704	0.753	0.606	0.778	0.332
	Volunteer association	0.846
BRC	Alumni association	0.500	0.749	0.523	0.723	0.141–0.315
	Fellow townsmen association	0.996
	Hometown chamber of commerce	0.572

show that the KMO values for the social integration and social capital scales are 0.691 and 0.62 (>0.6), respectively, and the p-values using Bartlett’s test of sphericity are <0.001, indicating suitability for factor analysis. Specifically, the factor loadings for each item on the scales are >0.5, with cumulative variance contributions of 62.7% for social integration and 61.1% for social capital (>60%), suggesting reasonable dimensionality and good representativeness of factors. All AVE values in the table are greater than 0.5, and all CR values are greater than 0.7, indicating good convergent validity of the scale. Overall, the square root of AVE is greater than the value of the correlation coefficient, which indicates a good discriminant validity between the variables [59]. In order to obtain a composite index of social capital and social integration, we calculated the composite factor scores for each dimension, using the factor score weighting sum formula, and then standardized these scores into a factor index between 1 and 100, which also makes the model construction more reasonable. The details are shown in Table 2.

4.3. Analysis of Main Effects

Table 3. Regression analysis.

Variable	Total	By Dimensions
	M1	M2	M3	M4	M5
	SI	EI	CI	BI	PI
Gender	−0.016	−0.040 **	−0.636 ***	0.819 ***	−0.339 ***
	(0.015)	(0.009)	(0.066)	(0.040)	(0.068)
Age	0.023 ***	−0.0003	0.011 **	0.034 ***	0.128 ***
	(0.001)	(0.0004)	(0.003)	(0.002)	(0.003)
Marital status	0.822 ***	0.455 ***	1.898 ***	0.624 ***	0.778 ***
	(0.007)	(0.004)	(0.031)	(0.019)	(0.032)
Education level	1.128 ***	0.965 ***	−0.379 ***	0.201 ***	1.532 ***
	(0.020)	(0.012)	(0.089)	(0.053)	(0.092)
BSC	−0.089 ***	−0.036 ***	0.031 ***	−0.223 ***	−0.163 ***
	(0.002)	(0.001)	(0.010)	(0.006)	(0.010)
BRC	0.147 ***	0.034 ***	0.048 ***	0.454 ***	0.293
	(0.002)	(0.001)	(0.011)	(0.007)	(0.011)
_cons	9.045 ***	2.725 ***	50.677 ***	9.373 ***	66.778 ***
	(0.051)	(0.030)	(0.231)	(0.139)	(0.237)
F	4899.471 ***	3394.660 ***	959.236 ***	2468.519 ***	615.419 ***
R2	0.147	0.107	0.032	0.080	0.021

Note: * p < 0.05; ** p < 0.01; *** p < 0.001. Standard errors are provided in parentheses.

shows the association between social capital and social integration. Model 1 finds that, after controlling for gender, age, educational level, and marital status, bonding social capital is significantly negatively correlated with social integration (β = −0.089, p < 0.001), while bridging social capital is significantly positively correlated with social integration (β = 0.147, p < 0.001). Looking at the dimensions separately, Models 2–5 find that bridging social capital is significantly positively correlated with economic integration (β = 0.034, p < 0.001), cultural integration (β = 0.048, p < 0.001), behavioral integration (β = 0.454, p < 0.001), and psychological integration (β = 0.293, p < 0.001). On the other hand, bonding social capital is significantly negatively correlated with economic integration (β= −0.036, p < 0.001), behavioral integration (β= −0.223, p < 0.001), and psychological integration (β= −0.163, p < 0.001). However, a different conclusion is presented in the dimension of cultural integration. Model 3 finds that bonding social capital is significantly positively correlated with cultural integration (β = 0.031, p < 0.01).

4.4. Mediation and Moderation Analysis

To further explore the mechanisms of social capital’s impact on social integration, we built upon previous research in this study, constructing a model based on the mediating effect of relative deprivation and the moderating effect of hometown attachment. To ensure the suitability of the structural equation model, we assessed the model’s fit and adjusted it according to the modification indices. The RMSEA = 0.026 meets the standard of less than 0.08, CFI = 0.993, TLI = 0.96 meets the standard of greater than 0.9, and SRMR = 0.002 meets the standard of less than 0.05, indicating a good model fit.

Table 4. Path coefficients of structural model and hypothesis testing.

Structural Path	Path Coefficient	S.E.	p-Values	Confidence Intervals (95%)
SI ← BSC	−0.224	0.006	<0.001	−0.236–0.212
SI ← BRC	0.361	0.006	<0.001	0.349–0.372
RD ← BRC	−0.018	0.002	<0.001	−0.023–0.013
SI ← RD	−0.210	0.002	<0.001	−0.215–0.206
SI ← BSC*HA	−0.011	0.002	<0.001	−0.015–0.005

Note: * indicates multiplication.

presents detailed results. We found that bridging social capital is significantly negatively associated with relative deprivation (β = −0.018, p < 0.001), indicating that bridging social capital can reduce the sense of relative deprivation among migrant populations. Meanwhile, relative deprivation has a significant negative impact on social integration (β = −0.210, p < 0.001). That is, the stronger the sense of relative deprivation among migrant populations, the lower their level of social integration. According to the stepwise method, the coefficients on the mediation path (bridging → deprivation → integration) are all significant, indicating the mediating effect of relative deprivation. Furthermore, we performed the bootstrap method to further investigate the mediating effects of relative deprivation within social integration. A total of 5000 bootstrap resampling iterations were conducted, establishing a 95% confidence interval for the analysis of mediating effects. The results indicate that the confidence interval does not include zero, indicating a significant mediating effect.

To test the moderating effect of hometown attachment, we constructed an interaction term between bonding social capital and hometown attachment. The results in Table 4 show that the path coefficient of the interaction term is significantly negative (β = −0.011, p < 0.001), indicating that the moderating effect of hometown attachment is established. To visually illustrate the moderating effect of hometown attachment, we plotted a simple slopes graph, as shown in Figure 2. From the graph, it can be observed that, as bonding social capital increases, social integration tends to decrease. Moreover, the slope under the condition of high hometown attachment is noticeably steeper than under the condition of low hometown attachment, suggesting that, as hometown attachment deepens, the negative impact of bonding social capital on social integration is strengthened.

5. Discussion and Conclusions

Based on the 2017 CMDS, we examined the relationship between social capital in the forms of bonding and bridging, as well as social integration among the migrant population. Additionally, we explored the mediating role of relative deprivation and the moderating effect of hometown attachment. The results indicate the following: Firstly, there is a significant direct correlation between social capital and social integration of migrants. Overall, bonding social capital has a generally negative impact on the social integration of migrants, while bridging social capital has a positive effect. On a dimensional level, bonding social capital hinders the economic, behavioral, and psychological integration of migrants, but promotes cultural integration; bridging social capital facilitates social integration across all four dimensions. Secondly, bridging social capital indirectly affects social integration through relative deprivation. Lastly, hometown attachment strengthens the negative impact of bonding social capital on social integration.

Our research indicates that different types of social capital have varying effects on the social integration of migrants. Bridging social capital facilitates social integration, while bonding social capital hinders it, a conclusion that is consistent with the viewpoint of social network theory. Social networks not only establish connections between actors, but also categorize them according to different relational networks [60]. Different types of relationships generate different resources, and Granovetter posited that weak ties between individuals might have a more significant and profound influence on shaping the attitudes and behaviors of social members compared to strong ties [61]. The diversity of social relationships is positively correlated with the quality of social capital [62]. Thus, bridging ties are often associated with positive social capital, facilitating social mobility and social integration [63]. The conclusions of many studies also agree with these views, suggesting that social relationships are most effective in connecting individuals with resources and knowledge [64,65,66]. However, the exclusivity of closed networks may hinder access to resources and valuable information [67]. Furthermore, closed bonding networks may be associated with negative social capital through racial segregation and ghettoization, thereby reducing social cohesion [68].

In terms of economic integration, social network relationships can provide migrants with a competitive advantage in the job market, enabling them to secure better employment opportunities. However, the value of social networks that migrants rely on in the labor market is contingent; for individuals with lower occupational status and limited personal resources, such as women and minority groups, this may restrict their access to opportunities and confine them to low-paying jobs. In other words, they may be trapped in inferior networks, thereby suffering from a deficit of social capital [69]. In the process of behavioral integration among migrants, organizations play a crucial role. They help migrants overcome political participation barriers stemming from socioeconomic resource inadequacies by providing resources [70]. Due to insufficient social capital, ethnic organizations’ involvement is often confined internally, with limited opportunities to enter broader governance networks. Bridging social capital, defined as connections between heterogeneous groups, is believed to be more conducive to democratic participation [71]. The impact of social capital on psychological integration also aligns with the aforementioned characteristics. Social support serves as a primary pathway connecting social capital and health, and it can be related to health through emotional, instrumental, or informational support, as well as through social companionship [72]. Studies have found that, in urban Chinese communities, relationships between individuals with different social identities become integral to residents’ daily lives [73]. Over-reliance on bonding social capital may hinder residents from accessing health-related resources from other groups, thereby jeopardizing their health [74]. In contrast, residents living in communities with abundant bridging social capital, whether in urban or rural areas, demonstrate better health outcomes [75].

It is worth noting the impact of social capital on cultural integration. Research indicates that bridging social capital plays a positive role in migrants’ cultural adaptation by facilitating connections with local residents [76]; however, some studies have found that establishing connections with locals is a difficult and lengthy process. During a certain stage of adaptation, already-integrated peers and fellow countrymen can help migrants understand various local cultural customs. While bonding social capital may have a lower degree of influence compared to bridging social capital, both, to some extent, foster attitudes conducive to integration, thereby promoting migrants’ cultural adaptation as a result of the interaction between the two types of capital among migrants [77].

Our research further elucidates the mechanisms through which social capital leads to social integration among migrants. Firstly, bridging social capital promotes social integration by reducing migrants’ relative deprivation. The theory of relative deprivation suggests that, while absolute deprivation threatens one’s survival, relative deprivation does not; instead, it limits one’s ability to fully participate in society [78]. For migrants, intergroup and intragroup relative deprivations are two sources that affect the willingness to integrate. According to social stratification theory, due to the disadvantaged economic, social status, and cultural levels of the migrant group, they face exclusion in education, healthcare, pension, and employment in the host society, forming a social stratification with urban residents and reducing their willingness to integrate [79]. Within the migrant group, social identity theory reveals the mechanism of the impact of relative deprivation. The stronger the sense of relative deprivation, the lower the migrants’ identification with their own group, leading to dissatisfaction with their group and a desire to dissociate from it [80]. Bridging social capital precisely promotes connections between migrants and different groups, especially urban residents, providing more information and resources and improving their quality of life, thereby reducing inequality and helping migrants integrate into society. Secondly, hometown attachment exacerbates the negative impact of aggregated social capital on migrants’ social integration. According to place identity theory, an individual’s emotional attachment to a specific place affects their behavior and attitudes [81]. When migrants form a strong hometown attachment in the host society, they are more likely to participate in community activities and establish social networks, thus promoting social integration. However, if the hometown attachment is too strong, it may hinder the migrants’ identification with and integration into the host society.

Objectively speaking, this study has some limitations that need to be addressed in future research. Firstly, our study relies on cross-sectional data, thus requiring further longitudinal research to deepen causal relationships. For instance, utilizing multi-wave panel data could explore variable interactions and developmental trends, further examining mediation and moderation effects. Secondly, this study does not account for the heterogeneous impacts across different age groups. Future research could conduct further categorization studies to better explain various impacts. Thirdly, due to data limitations, the measurement of relative deprivation focuses on economic deprivation, lacking subjective self-assessment indicators. Therefore, additional supplementary data are needed to support the conclusions.

Furthermore, there is a need to further explore the issue of discrimination faced by migrant populations. While bridging social capital can mitigate the impact of discrimination to some extent, in reality, discrimination may also drive migrants towards bonding social capital. This shift may occur because migrants, when confronted with discrimination, tend to seek support, identity, and resources within their own ethnic, cultural, or community ties. Therefore, in our future research, we need to comprehensively consider the interaction between bonding and bridging social capital to help reduce migrants’ experiences of discrimination, diminish feelings of relative deprivation, and, thereby, promote deeper social integration and sustainable societal development.

Although this study has certain limitations, it examines the differences in migrants’ social networks concerning resource acquisition and elucidates how these differences impact their social integration. This work enriches social capital theory in the field of migration research and provides recommendations for government efforts to promote social integration. The sustainable development of society relies on the active participation and contribution of all its members, and the social integration of migrants is crucial to achieving this goal. By promoting the social integration of migrants, we can not only enhance social cohesion but also drive economic and cultural diversity and innovation [82]. Firstly, migrants should be encouraged to actively interact and communicate with various social groups to enrich their social networks. Secondly, communities and social institutions should also organize cultural exchange and educational activities to help migrants understand and adapt to the culture of the new environment, as well as to allow local residents to learn about migrant cultures, thereby promoting mutual understanding and respect. Lastly, the government should provide migrants with more employment and educational opportunities and ensure equal access to public services [83], in order to promote their social integration and to construct a more inclusive, diverse, and sustainable social environment.