1. Introduction

Since the reform and opening up, population mobility, as one of the largest geographical processes in China [1], has provided sustained momentum for the socio-economic prosperity of urban areas. Previous studies have found that over the past thirty years, the annual average contribution rate of migrant population to China’s economic growth reached as high as 19.99% [2]. However, hindered by social isolation, institutional barriers, and skewed social welfare, migrant populations have not fully integrated into the mainstream society of their destination areas. To address these issues, the “National New Urbanization Plan (2014-2020)” proposed actively promoting a “people-oriented” new urbanization, encouraging the integration of migrant populations into the social life of their destination areas, and enhancing the quality of life of migrant populations [3]. The improvement of urbanization level and the enhancement of urban environmental quality are conducive to the improvement of residents’ quality of life and health levels [4]. China’s distinctive new urbanization embodies the connotations of people-oriented, synergy, inclusiveness, and sustainability [5]. Promoting the social integration of migrant populations and improving their quality of life in destination areas not only contributes to the development of new urbanization and the construction of a harmonious society but also is an important practical issue concerning the comprehensive construction of a moderately prosperous society in China. Subjective well-being is an important indicator reflecting individuals’ quality of life and health status. Exploring the influencing factors of well-being will provide policy “leverage” for improving residents’ quality of life. Therefore, studying the influencing mechanism of migrant populations’ well-being can explore effective ways to enhance their well-being and provide a scientific basis for promoting the implementation of new urbanization.

In recent years, some scholars have conducted research on the subjective well-being of migrant populations in China [6-7]. The research indicates that with the increase of economic income, the subjective well-being of migrant populations shows a trend of first rising and then declining [8], and their level of well-being is lower than that of local residents in the destination areas [6]. Among them, social comparison is an important factor influencing the well-being of migrant populations. As migrant populations gradually interact with local residents, they tend to choose local residents in the destination areas as the main reference group for comparison in terms of socio-economic status, living conditions, and other aspects. However, the significant gap with local residents in the destination areas may likely reduce the well-being of migrant populations. Regarding research on the well-being of migrant populations, existing studies have seldom explored the mediating effect of social comparison on the well-being of migrant populations, and mainly focused on small sample case studies. Therefore, based on the data from the 2014 National Dynamic Monitoring Survey of Migrant Population’s Health and Family Planning, this study uses multilevel linear regression models, mediating analysis, and moderating analysis techniques to reveal the association between the socio-economic status of urban migrant populations in China and their subjective well-being, with particular attention to the mediating effect of social comparison and the moderating effect of urban integration.

2. Literature Review

According to research by Diener et al., subjective well-being refers to an individual’s overall evaluation of their quality of life, including cognitive components and affective components [9]. Specifically, the cognitive component, namely life satisfaction, is a comprehensive assessment of individuals’ own life conditions, while the affective component represents individuals’ momentary emotional experiences, which can be further divided into positive and negative emotions [10]. Previous studies have explored the impact of residential environment (such as community-built environment, community social capital) and personal characteristics (such as socio-economic attributes, demographic characteristics, life events) on subjective well-being. Research results indicate that: (1) Regarding the residential environment: Residents living in affluent and well-equipped communities tend to have higher levels of happiness [11-12]. Good community services and accepting environments can effectively promote the integration of migrant populations into urban areas [13-14]. However, neighborhood poverty and population density have negative correlations with residents’ life satisfaction and positive emotions [15]. Lin Jie et al. believe that it is necessary to actively promote communication and trust among residents in the community and to build friendly and harmonious communities [16]. (2) Concerning personal characteristics: Individual socio-economic characteristics (such as income, occupation, household registration, etc.) are significantly positively correlated with subjective well-being [17]. For example, Yang Dongliang et al. found that an increase in job income can significantly enhance the happiness of the floating population [18], while Dang Yunxiao et al. argue that income can mitigate the negative impact of urban scale on residents’ happiness [19]. In addition to the attribute of high income, socio-economic attributes such as being elderly and having household registration also significantly affect individuals’ life satisfaction [20]. In terms of intergenerational differences, Cheng et al. found that the second generation of the floating population generally has higher expectations for the future than their own capabilities, resulting in their subjective well-being being significantly lower than that of the first generation of the floating population [21]. (3) Concerning social comparison: Knight et al.[22] and Gelatt [23] both point out that changes in reference groups can affect the subjective well-being of migrant populations. Migrant populations who have lived in their destination areas for a long time tend to compare themselves with local residents in terms of income, occupation, etc., resulting in lower levels of subjective well-being. From existing research, it can be inferred that social comparison plays a mediating and moderating role in the relationship between individuals’ socio-economic status and subjective well-being.

This introduces the concept of Social Comparison Theory (SCT): Social Comparison Theory, initially proposed by psychologist Leon Festinger, suggests that individuals have a tendency to compare themselves with others [24]. When the social status of the reference group is higher than that of oneself, despite possible weakening of self-esteem and self-confidence, individuals may benefit to some extent from the positive motivational effect of the reference group [25]. Anderson et al. introduced the concept of Sociometric status (SMS), which refers to an individual’s respect, envy, and influence, and found through the MacArthur ladder method that sociometric status significantly predicts individuals’ subjective well-being [26]. Similarly, after studying the relationship between social comparison and well-being, Diener et al. also found that social comparison can lead to changes in individuals’ levels of subjective well-being [27]. On the other hand, domestic scholars have compared the two types of social status and found age differences in the degree of attention to these statuses: older people place more emphasis on sociometric status, while younger people are more likely to be influenced by economic status comparisons [17].

In research on the social integration of migrant populations, Wang Mingfeng et al. use factors such as social relationship integration and economic integration to explain its connotation [28]. Zhu Yu et al. argue that social protection based on urban integration neglects the unique housing needs of migrant populations [29]. Compared with residents who own housing property, renters tend to have lower levels of happiness [30]. Liu et al. found a positive correlation between community integration and the subjective well-being of migrant populations, but it cannot effectively mitigate the negative effects of neighborhood deprivation on migrant populations [31]. Domestic scholars have mainly focused on institutional barriers, market exclusion, relative deprivation, and other aspects in studying the relationship between the social integration of migrant populations and subjective well-being, identifying factors influencing the subjective well-being of migrant populations [6-7].

In summary, existing research has the following deficiencies: Firstly, when exploring the relationship between objective socio-economic status and subjective well-being, many studies use single objective measurement indicators to characterize objective socio-economic status (such as income, occupation, etc.), which cannot accurately measure the relationship between the two. Secondly, there is less research that considers subjective socio-economic status as a mediating variable to explore the mediating effect of social comparison on subjective well-being. Thirdly, there is scarce research on the impact of migrant populations’ social integration in their destination areas on their subjective well-being. Given these deficiencies, it is necessary to further explore the influencing mechanism of subjective well-being of migrant populations based on existing research. This study is based on the data from the 2014 National Dynamic Monitoring Survey of Migrant Population’s Health and Family Planning. It uses multilevel linear regression models, mediating analysis, and moderating analysis techniques to reveal the association between the socio-economic status of urban migrant populations and their subjective well-being, with particular focus on the mediating effect of social comparison and the moderating effect of urban integration. The study aims to address the following questions: 1) Can the subjective evaluation of their own socio-economic status by migrant populations play a mediating role in the relationship between objective socio-economic status and subjective well-being? 2) As the degree of social integration and local acceptance of migrant populations in their destination areas increases, is the relationship between their subjective evaluation of their own socio-economic status and their subjective well-being stronger?

3. Research Methodology

3.1. Data Source and Sampling Method

The data used in this study are from the social integration and mental health questionnaire section of the 2014 National Dynamic Monitoring Survey of Migrant Population’s Health and Family Planning (CMDS 2014). The sampled communities for the survey are distributed across eight districts or prefecture-level cities nationwide (Figure 1), including Chaoyang District, Beijing Municipality; Qingdao City, Shandong Province; Xiamen City, Fujian Province; Jiaxing City, Zhejiang Province; Shenzhen City, Guangdong Province; Zhongshan City, Guangdong Province; Zhengzhou City, Henan Province; and Chengdu City, Sichuan Province. The survey team used a multi-stage sampling method to select the survey sample: first, townships and streets were selected using stratified probability proportionate to size sampling (PPS); second, neighborhood committees were selected within the sampled townships and streets according to PPS; finally, individuals were selected as sampling units within the neighborhood committees, with 20 individuals sampled per committee. The survey targeted male and female migrant populations who have resided in urban communities within the survey scope for more than one month and do not have local household registration (i.e., excluding migrant populations in rural communities). A total of 11,839 valid samples were obtained from the survey. After excluding those with incomplete information, the sample size for empirical analysis was 10,771, distributed across 591 communities in the eight districts or prefecture-level cities. The questionnaire collected information on demographic characteristics, socio-economic status, living conditions, physical health status, objective and subjective socio-economic status, degree of social integration, and subjective well-being of the respondents.

3.2. Research Methods and Model Setting

Given the hierarchical structure of the data, with the need to analyze changes in subjective well-being at both community and individual levels, this study intends to use a multilevel linear regression model to reveal the relationship between the socio-economic status of migrant populations and their subjective well-being. The model is set as follows:

Yijc=αXijc+βZijc+γc+μj+εij (1)

Where Yijc represents the subjective well-being score of individual i residing in community j in city c; Xijc represents objective socio-economic status; Zijc represents a series of control variables; γc represents city fixed effects; μj and εij respectively represent residuals at the community and individual levels.

Mediating analysis and moderating analysis will be employed to respectively reveal the mediating role of migrant populations’ subjective evaluation of their own socio-economic status in the above relationship, and the moderating effect of their social integration on their subjective socio-economic status and subjective well-being. The mechanisms of mediating and moderating effects are as follows [32] (Figure 1 and Figure 2):

Figure 1: Conceptual framework of mediating effect

Figure 2: Conceptual framework of moderating effect

(1) Mechanism of mediating effect: In the influence of the independent variable X on the dependent variable Y, if X affects Y through variable M, then M is referred to as the mediating variable. In Figure 1, c represents the overall effect of X on Y. After adding the mediating variable M, c’ represents the direct effect of X on Y, and ab represents the mediating effect of M. Furthermore, in the mediating effect, there are partial mediating and complete mediating: on one hand, when X has a direct effect on Y, c’ is meaningful, and in this case, c = c’ + ab, indicating partial mediating effect of M; on the other hand, when X cannot directly influence Y, c’ is 0, and in this case, c = ab, indicating complete mediating effect of M.

(2) Mechanism of moderating effect: In the influence of the independent variable X on the dependent variable Y, if the relationship between Y and X is influenced by a third variable M, then M is considered as the moderating variable in the model. The moderating variable can be a categorical variable or a continuous variable, and it affects the strength and direction of the relationship between the dependent variable and the independent variable.

In this study, in the examination of the mediating effect, the objective socio-economic status of migrant populations is taken as the core independent variable of the model, including four indicators: average monthly family income (logarithmic), educational level, occupation type, and housing ownership. The subjective evaluation of the social status of migrant populations and the subjective evaluation of the degree of respect received are taken as two mediating variables, thereby constructing two causal pathways to conduct mediating effect tests for subjective socio-economic status. The subjective well-being level of migrant populations is taken as the dependent variable of the model (Figure 3). The mediating effect analysis is conducted using the stepwise method proposed by Baron and Kenny [33]: in the first step, control variables are included in the regression equation to test the direct impact of the objective socio-economic status of migrant populations on their subjective well-being (path A); in the second step, the impact of objective socio-economic status on the two mediating variables is tested separately (paths B and D); in the third step, the two mediating variables are included separately in the regression equation to test the significance of the two mediating effects (paths C and E). The model setting with mediating variables included is as follows:

Mijc=αXijc +βZijc +γc +μj +εij (2)

Yijc =αXijc + δMijc +βZijc +γc +μj +εij (3)

Where: the meanings of Yijc, Xijc, Zijc, γc, μj and εij are as mentioned above; Mijc represents the mediating variable, namely, the subjective evaluation of migrant populations regarding their own socio-economic status.

Figure 3: Conceptual framework of mediating effect of migrants’ self-rating socioeconomic status on subjective well-being

The study intends to adopt moderating effect analysis to reveal the moderating role of migrant populations’ social integration status in the relationship between socio-economic status and subjective well-being (Figure 4). The two dimensions of subjective socio-economic status of migrant populations are respectively regarded as explanatory variables in the moderating effect test model. The two dimensions of social integration (urban integration and local acceptance) are taken as moderating variables in the regression model, while the subjective well-being level of migrant populations is considered as the dependent variable in the regression model. In specific operations, the subjective evaluation of migrant populations’ socio-economic status and social integration status are subjected to interaction term processing and incorporated into the model analysis. The model setting with moderating variables included is as follows:

Yijc =αM0ijc + δM1ijc +βZijc +γc +μj +εij (4)

Where: the meanings of Yijc, Zijc, γc, μj and εij are as mentioned above; M0ijc represents the mediating variable in the mediating effect analysis, namely, the subjective evaluation of migrant populations regarding their own socio-economic status, set as the independent variable in this moderating effect model.

Figure 4: Conceptual framework of Moderating effect of migrants’ social integration on self-rating socioeconomic status and subjective well-being

3.3. Variable Selection and Measurement

The study intends to utilize the Satisfaction with Life Scale (SWLS) proposed by Diener et al. to measure the subjective well-being level of migrant populations [34] (Table 1). The questionnaire employs a Likert seven-point scale to record the respondents’ answers to the above questions.

Table 1: Satisfaction with Life Scale (SWLS)

The study comprehensively measures the objective socioeconomic status of the respondents through four core explanatory variables, including the respondent’s average monthly household income from the previous month (continuous variable), the respondent’s level of education (categorical variable), the respondent’s occupation type (categorical variable), and the respondent’s housing ownership status in the current place of residence (categorical variable). Two variables are selected to measure the subjective evaluation of socioeconomic status, namely “subjective evaluation of one’s own social status”1 and “subjective evaluation of the level of respect received”2, and only the self-evaluation scores of the respondents compared to their relatives, friends, and colleagues in their current place of residence are extracted. Two sets of questions measuring the degree of urban integration and local acceptance status are selected as moderating variables and included in the model. Control variables are categorized into four types: demographic characteristics, socioeconomic status, living conditions, and physical health conditions (Table 2). The results of multicollinearity tests show that the maximum variance inflation factor (VIF) for all explanatory variables is 2.38. Generally, if the VIF value is less than 3, it indicates that there is no serious multicollinearity problem in the regression model.

For the selection and measurement of moderating variables, the study first extracts question 518 from the questionnaire related to the social integration status of the respondents. Secondly, questions F, G, and H are extracted from the question 5183, and their scores are summed to obtain the degree of urban integration score for the floating population. The higher the score, the deeper the level of integration of the respondents into the city. Questions J, K, L, and M are extracted4, and their scores are summed to obtain the score for the local acceptance status of the floating population (the measurement of local acceptance status starts with the subjective attitude of the floating population). The higher the score, the more optimistic the situation of the respondents being accepted by local residents. Additionally, the physical health status is measured through six self-assessment questions5. The lower the score, the better the self-assessed physical health status of the respondents. Since the initial measurement criteria of the questionnaire are not conducive to drawing corresponding conclusions, this study adjusts the options of the questions so that higher scores indicate better self-assessed physical health status of the respondents.

4. Analysis Results

4.1. Basic Sample Characteristics and Descriptive Statistics of Variables

The descriptive statistics of variables are shown in Table 2. It can be observed that the average score of life satisfaction for the floating population is 21.59 points, indicating a relatively balanced state. More than half (55.15%) of the floating population have only received junior high school education or below, while the proportion of those with high school education (26.90%) and college education or above (17.95%) is much lower compared to those with junior high school education or below. The majority of the surveyed floating population (88.37%) are engaged in production and service occupations in agriculture, forestry, animal husbandry, and fishery. Regarding housing ownership, only 10.30% of the floating population have housing ownership in their current place of residence. In comparison with local relatives, friends, and colleagues, the assessment score of their own social status (4.61) among the floating population is below average, while the assessment score of the level of respect received (5.13) is average. In terms of social integration, both the degree of urban integration score (9.73) and the local acceptance status score (12.74) are above average, indicating a relatively good social integration status among the floating population. In terms of household registration, the majority of the floating population come from rural areas (83.61). Additionally, the floating population tends to reside in communities with a higher proportion of floating population (41.05%) or communities with a similar proportion of floating population and local residents (36.60%), indicating a higher degree of social isolation among the floating population. As for self-assessed health status, previous studies have indicated that income and education level only significantly affect the self-assessed health status of the middle and low-income groups. The data shows that the evaluation of the floating population’s physical health status is above average (23.16). Since the floating population is mostly composed of middle and low-income groups, it can be inferred that an increase in income and improvement in education level can significantly enhance the health status of the floating population.

Table 2: Model variables and sample means/ ratio

Next, the study further tabulated the subjective well-being scores of migrant populations in eight cities/districts (Table 3) and ranked them in descending order (Figure 5). It can be observed that overall, migrant populations in Qingdao, Shandong Province, have the highest levels of subjective well-being, while those in Zhengzhou, Henan Province, and Shenzhen, Guangdong Province, exhibit relatively lower levels of subjective well-being. Possible reasons for these differences may include variations in urban living costs, employment pressures and competitiveness, as well as differences in relevant policies targeting migrant populations among the surveyed cities, leading to differences in subjective well-being levels among migrant populations in different cities.

Table 3: Migrants’ subjective well-being among eight cities/ districts

Figure 5: Migrants’ subjective well-being in descending order among eight cities/ districts

4.2. Regression Results of Mediating/Moderating Effects

The study employed two sets of multilevel linear regression models to examine the mediating effects of two assumed mediator variables on the relationship between the objective socioeconomic status of the floating population and their subjective well-being. Model 1 represents the baseline model, including only the core explanatory variables and control variables. Models 2a and 2b represent, based on the baseline model, the addition of the subjective evaluation of the floating population’s social status as a mediator variable and subsequent regression, i.e., paths A, B, and C in Figure 1. Models 3a and 3b, on the basis of the baseline model, incorporate the subjective assessment of the floating population’s level of respect as a mediator variable and conduct stepwise regression, i.e., paths A, D, and E in Figure 1. The results are shown in Table 2.

Regarding the examination of the baseline model, controlling variables were included in the regression equation (Model 1) to test the direct impact of the floating population’s objective socioeconomic status on their subjective well-being. The results of Model 1 indicate that among various indicators measuring objective socioeconomic status, three variables, namely, monthly average household income (β=1.542, P<0.01), occupation as a civil servant, office worker, or related personnel and self-employed individual (β=0.408, P<0.01), and property ownership (β=1.014, P<0.01), are significantly positively correlated with the subjective well-being of the floating population.

Table 4: Stepwise regression results of mediating effects of subjective social status evaluation

Note: *P<0.1, **P<0.05, ***P<0.01

5. Multilevel Regression Analysis Results

5.1. Subjective Social-Economic Status of Migrants as Mediating Effects

In addressing the mediating effect model, this paper respectively takes the subjective evaluation of migrants’ status in society and the degree of respect they receive as the mediating variables, examining the influence of objective social-economic status on these two mediating variables (Model 2a and Model 3a). The results of Model 2a indicate that four variables, namely, average monthly household income (β=0.521, P<0.01), occupation as leaders or professionals in government agencies, party organizations, enterprises, and institutions (β=0.109, P<0.05), occupation as civil servants, employees, relevant personnel, or self-employed individuals (β=0.090, P<0.05), and property ownership (β=0.393, P<0.01), are significantly positively correlated with migrants’ subjective evaluation of their social status. Model 3a shows that average monthly household income (β=0.375, P<0.01), education level of junior college and above (β=0.099, P<0.05), occupation as civil servants, employees, relevant personnel, or self-employed individuals (β=0.066, P<0.1), and property ownership (β=0.257, P<0.01) are significantly positively correlated with migrants’ perceived level of respect.

Model 2b results show that three variables, namely average monthly household income (β=1.042, P<0.01), occupation as civil servants, employees, relevant personnel, or self-employed individuals (β=0.325, P<0.05), and property ownership (β=0.644, P<0.01), are significantly positively correlated with the subjective happiness level of migrants, and the regression coefficients of these three variables are lower than those in Model 1. Additionally, there is a significant positive correlation between the mediating variable (migrants’ subjective evaluation of their status in society) and subjective happiness (β=0.953, P<0.01). Thus, it can be seen that migrants’ subjective evaluation of their status in society can serve as a mediating variable in the effects of average monthly household income, occupation as civil servants, employees, relevant personnel, or self-employed individuals, and property ownership on subjective happiness. Model 3b results demonstrate that three variables, namely average monthly household income (β=1.299, P<0.01), occupation as civil servants, employees, relevant personnel, or self-employed individuals (β=0.365, P<0.05), and property ownership (β=0.851, P<0.01), are significantly positively correlated with the subjective happiness level of migrants, and the regression coefficients of these three variables are lower than those in Model 1. Furthermore, there is a significant positive correlation between the mediating variable (migrants’ subjective evaluation of the degree of respect they receive) and subjective happiness (β=0.643, P<0.01). Thus, similar to the results of Model 2b, migrants’ subjective evaluation of the degree of respect they receive can also serve as a mediating variable in the effects of average monthly household income, occupation as civil servants, employees, relevant personnel, or self-employed individuals, and property ownership on subjective happiness.

To further verify these conclusions, a Sobel test is planned to determine whether the subjective evaluation of social status plays a mediating role. The test results are presented in Table 3. It is found that in the effects of average monthly household income (Z=15.64, P<0.01), occupation as civil servants, employees, relevant personnel, or self-employed individuals (Z=4.471, P<0.01), and property ownership (Z=10.29, P<0.01) on subjective happiness, migrants’ subjective evaluation of their social status and the degree of respect they receive both act as mediating variables to some extent. Moreover, when the occupation is leaders or professionals in government agencies, party organizations, enterprises, and institutions as the core explanatory variable affecting subjective happiness, migrants’ subjective social-economic status serves as a mediating variable but exhibits a suppression or inconsistent mediating effect [35]. When the education level is junior college and above as the core explanatory variable affecting subjective happiness, migrants’ subjective evaluation of the degree of respect they receive acts as a complete mediating variable.

Table 5: Sobel test results of mediating effects of subjective social status evaluation

5.2. Moderating Effect of Urban Integration on Floating Population

Table 4 reflects the moderating effect of social integration on the subjective well-being of the floating population when the subjective socio-economic status of the floating population serves as the explanatory variable. The results of the baseline model 2b indicate a significant positive correlation between the subjective well-being level of the floating population and the subjective evaluation of social status (β=0.953, P<0.01). For each unit increase in subjective socio-economic status, the corresponding subjective well-being level increases by 0.953 points. Model 2c incorporates the interaction terms of urban integration, local acceptance status, and subjective evaluation of social status, analyzing the moderating effect of subjective socio-economic status on the subjective well-being of the floating population. The coefficient of the interaction term between urban integration and the subjective evaluation of social status is negative (-0.035) and significant at the 5% level, indicating that the more integrated the floating population is into their current residence, the less their subjective evaluation of social status affects their subjective well-being level. The interaction term coefficients between local acceptance status and the subjective evaluation of social status of the floating population are both insignificant, indicating that the subjective perception of being accepted by local residents does not significantly moderate the impact of subjective socio-economic status on happiness.

Model 3b is also based on the reference group of relatives, friends, and colleagues in the current place of residence as the baseline model. The results show a significant positive correlation between the subjective well-being level of the floating population and their subjective evaluation of being respected (β=0.643, P<0.01). For each unit increase in subjective socio-economic status of the floating population, the corresponding subjective well-being level increases by 0.643 points. Model 3c measures the moderating effect through the interaction terms of urban integration, local acceptance status, and subjective evaluation of being respected. The coefficient of the interaction term between urban integration and the subjective evaluation of being respected is negative (-0.037) and significant at the 5% level, indicating that the more integrated the floating population is into their current residence, the less their subjective evaluation of being respected affects their subjective well-being level. Regarding local acceptance status, the subjective perception of being accepted by local residents does not significantly moderate the impact of subjective socio-economic status on subjective well-being (β=0.022, P>0.1).

Table 6: Multilevel model results of moderating effects of social integration

Note: *P<0.1, **P<0.05, ***P<0.01

6. Conclusion and Discussion

6.1. Conclusion

Taking eight urban areas in China as case studies, this paper employs a multi-level linear regression model, mediating analysis, and moderating analysis to analyze the relationship between the objective socioeconomic status and subjective well-being of migrant populations in large Chinese cities. It measures the mediating effect of subjective socioeconomic status of migrant populations and examines the moderating effect of social integration of migrant destinations on the relationship between objective socioeconomic status and subjective well-being. The main conclusions are as follows:

1) In the relationship between objective socioeconomic status and subjective well-being, migrant populations play a significant mediating role compared to local relatives, friends, and colleagues regarding socioeconomic status. Specifically, indicators such as family monthly income, occupational types such as civil servants, office workers, relevant personnel, and self-employed individuals, as well as homeownership, are positively correlated with the subjective well-being of migrant populations. The subjective socioeconomic status of migrant populations partially mediates the above positive correlations. When the occupational type is a leader or professional in state organs, party and mass organizations, enterprises, or technical personnel, the subjective socioeconomic status of migrant populations mediates the above influences, but with a suppression effect.

2) The higher the degree of integration of migrant populations into the destination, the weaker the impact of their subjective evaluations of social status and respect from others on subjective well-being. In other words, the role of subjective evaluations of social status and respect from others in the changes of well-being becomes smaller as migrant populations gradually integrate into the destination. This may be because as migrant populations gradually integrate into the destination, they establish social connections with local residents in terms of lifestyle and values, and their sense of belonging and identity to the destination deepens. Additionally, as migrant populations integrate into the city, the social distance between them and the reference group gradually shortens, reducing their willingness to compare social status with local residents.

3) When the subjective perception of acceptance by local residents is used as a measure of social integration, there is no evidence to suggest that the above local acceptance indicators play a significant moderating role in the relationship between subjective socioeconomic status and subjective well-being. This may be because the local acceptance indicators used in the study are measured from the perspective of migrant populations themselves, rather than from the perspective of local residents, leading to significant subjectivity. In addition, compared to psychological factors, migrant populations may pay more attention to the impact of their economic livelihood on their well-being.

6.2. Discussion

This study incorporates social comparison and social integration into the research framework of subjective well-being of migrant populations and finds that social comparison of migrant populations mediates the relationship between socioeconomic status and subjective well-being. It also suggests that the social integration of migrant populations in the destination weakens the impact of subjective socioeconomic status on well-being. The data sources of this paper are mostly from major cities where migrant populations gather, and there is an urgent need to improve the well-being of migrant populations already settled in urban areas. Well-being reflects people’s good living conditions [36], and to improve residents’ well-being, attention should be paid to the social welfare and social integration issues of migrant populations. First, there is a need to further relax the household registration system for migrant populations [37] and promote the process of urban citizenship for migrant populations. Secondly, it is urgent to break down the differentiation and estrangement within urban communities and build trust among neighbors. Finally, governments and communities should improve public health services to better promote the social integration of migrant populations. Furthermore, it is necessary to strengthen investment in human capital for migrant populations, improve the income level of residents, especially migrant workers [29], reform the urban household registration system and housing policies, and further relax the conditions for settlement [30], thus promoting the urban citizenship and equal access to social welfare for migrant populations. Wang Guixin et al. pointed out that perceived stress has significant negative effects on individual health for both migrant populations and local residents [31]. While government departments need to properly handle the equal treatment and rights and interests of migrant populations and local residents, they also need to pay more attention to and build the psychological health of migrant populations. Although improving objective economic conditions such as income, occupation, and education can promote the improvement of subjective well-being levels, intrinsic factors such as subjective evaluation and psychological status of migrant populations should not be overlooked compared to such exogenous factors.

In addition, due to data and methodological limitations, this study also has its shortcomings. First, this study only measures the impact of the social status comparison between migrant populations and relatives, friends, and colleagues in the destination on subjective well-being. A single type of reference group may lead to biased research results. Migrant populations may have multiple reference groups, namely, both local residents in the destination and relatives and friends in the place of origin. Second, the evaluation of migrant populations’ social status and social integration is subjective, and the evaluation of migrant populations’ integration into the city by local residents has not been included in the research framework. Therefore, the objectivity and persuasiveness of the study need to be strengthened. Conducting analysis based on bidirectional evaluations would contribute to more scientifically grounded research conclusions.