Physical Exercise, Social Integration, and Urban Hukou Acquisition Decision-Making in China's Floating Population

This paper uses data from the 2014 and 2016 China Labor Force Survey to investigate the relationship between physical exercise, social integration and urban hukou acquisition decision-making in the oating population. We nd that physical exercise and social integration can directly promote the acquisition of urban hukou in oating populations, and physical exercise promotes the acquisition of urban hukou in oating populations by increasing their degree of social integration. So, social integration is the intermediary between the oating population's physical exercise and urban hukou acquisition decision-making. Specically, the mediating effect is most pronounced in the non-agricultural and female oating population. The results suggest that the construction of amenities for physical exercise and communal physical activities can not only promote the development of a harmonious society, but also effectively encourage the oating population to acquire urban hukou in their city of work and promote the development of the new urbanization of China. calculation of the path "physical exercise--social integration--decision-making of the oating population in acquiring urban hukou". Columns (1)-(3) verify and measure the path "participate in physical exercise--social integration--decision-making of the oating population in acquiring urban hukou". Column (1) shows the test results that do not include the intermediary factor (social integration). The coecient is positive and signicant. Column (2) shows the impact of participation in physical exercise on social integration; the coecient is positive and signicant, indicating that participation in physical exercise can indeed promote social integration. Column (3) shows the test results that include both the intermediary factor (social integration) and participation in physical exercise; the regression coecients are signicantly positive at the level of 1%, indicating that the higher the degree of social integration, the greater the willingness of the oating population to acquire urban hukou in the work city. And the coecient ( ϕ 1 ) is less than α 1 , indicating that social integration is a part of the mediating factors that affect the decision-making of the oating population in acquiring urban hukou through participating in physical exercise, and the mediating effect accounts for 1.95%. Furthermore, we used the Sobel test to verify the signicance of the mediation effect, and the Z statistics all passed a 5% statistical test, which again conrmed that the mediation effect was signicant. Similarly, in Columns (4)-(9), the paths "the minutes of physical exercise--social integration--decision-making of the oating population in acquiring urban hukou" and "the frequency of physical exercise--social integration--decision-making of the oating population in acquiring urban hukou " were tested respectively. It can be seen that social integration is a part of the mediating factor of the minutes of physical exercise and the frequency of physical exercise, with the mediating effects accounting for 1.56% and 1.80%, respectively. And the Sobel Z statistics all passed the statistical test at the level of 10%. To sum up, physical exercise can promote the degree of social integration of the oating population, and ultimately, increase the willingness of the oating population to acquire urban hukou in the work city. populations. Overall, the impact of physical exercise on is greater than on the male oating population. Similarly, Columns (7)-(12) show the impact of social integration on acquisition of urban hukou decision-making for the male and female oating populations. The differences between the two groups are relatively small; 0.006 and 0.01, respectively. Columns (13)-(18) show the effects of non-agricultural and agricultural hukou and physical exercise on acquisition of urban hukou decision-making. It can clearly be seen that the non-agricultural and agricultural hukou oating population has signicant coecients of 1% on the coecients of participation in physical exercise, minutes and frequency of physical exercise. In addition, for the non-agricultural hukou oating population, physical exercise has a greater impact on acquisition of urban hukou decision-making (0.065 V.S. 0.030). Furthermore, in Columns (19)-(24), the effects of social integration on acquisition of urban hukou decision-making are shown in the non-agricultural and agricultural hukou oating populations. It can be seen that the positive effect of social integration on acquisition of urban hukou decision-making for the non-agricultural hukou oating population is greater than that for the agricultural hukou oating population (0.018 V.S. 0.002). Overall, there is a signicant positive relationship between social integration and acquisition of urban hukou decision-making in the two groups. However, the positive impact of social integration and acquisition of urban hukou decision-making on the non-agricultural hukou oating population is greater than that on the agricultural hukou oating population.


Introduction
Physical exercise is not only an important way to improve physical tness, but also plays a role in expanding interpersonal communication, enhancing mutual trust, and promoting mutual exchange and integration of residents. In 2016, the China State Council successively promulgated the "plan to promote the acquiring of urban hukou of 100 Million non-registered urban dwellers in cities" Physical exercise plays a role in promoting physical and mental health 4,5 . In addition, physical exercise also has a positive impact on improving life satisfaction and subjective well-being 6-8 . Moljord et al. 9 found that the higher the frequency of physical exercise, the higher the individual's happiness level. Blomstrand et al. 10 follow-up survey of middle-aged women in the Netherlands also showed that there is a strong positive relationship between physical exercise during leisure time and subjective well-being. Edwards 11 found that regular physical exercise can not only signi cantly improve happiness, but also fortify individuals' stress resistance and mental health. Moreover, participating in physical exercise has a signi cant positive impact on improving self-discipline and self-con dence.
People who have participated extensively in outdoor physical exercise and those who have been involved in physical exercise for a longer period of time often have wider social networks, higher achievement levels, and a higher sense of belonging 12 . Federico et al. 13 research on Italian adults also found that groups who participated more frequently in physical exercise were also groups with higher education levels. Di Bartolomeo &Papa 14 found that, in the short-term, participants who have been exposed to physical exercise show more trust and are more prominently friendly. In addition, rather than individual-centered physical exercise, participation in collective physical exercise, such as team sports, is more likely to create wider social networks through mutual communication 15 . Similarly, Perks 16 and Brown et al. 17 found that trust is strengthened among members participating in physical exercise in the same community.
Speci cally, physical exercise has a positive impact on the degree of community integration 18 . This positive effect will be strengthened by an increase in the level of community interaction 19,20 . Some studies have found that the positive relationship between physical exercise and social integration is most signi cant among women 21 . At the same time, the degree of community integration plays a mediating role between participation in physical exercise and community safety perception 22 . Therefore, increasing the construction of sports facilities and promoting the integration of the community will effectively promote the stability and harmony of the community. integration has a positive relationship with the acquisition of urban hukou. That means that the higher the degree of social integration, the higher the willingness to acquire urban hukou among migrant workers in cities 25 . A sense of belonging and the degree of community integration of the oating populations are the key factors in their acquisition of urban hukou decision-making 26, 27 . When the oating populations have a higher level of acceptance of locals, a stronger sense of social identity, and a higher degree of social integration 28,29 , their willingness to acquire urban hukou in the long term is higher 30 . Furthermore, Sun, Li & Qi 31 found that the higher the social integration of migrant workers in small cities, the more they prefer to acquire urban hukou in cities.
Besides social integration, the characteristics of demography, family, and city also have an impact on the oating population's urban hukou decision-making 32 . For example, Gu et al. 33 found that the higher the education level, the higher the income level, and the more children in the family, the less likely is acquisition of urban hukou. Secondly, the migration distance also has a negative effect on the acquisition of urban hukou in the oating population. Zhu &Chen 34 research on the settlement decisions of migrants showed that women and highly educated migrants engaged in non-productive work are more willing to settle and acquire urban hukou in work cities in China. Therefore, there is a signi cant gender difference in the oating population in terms of acquiring urban hukou; women's willingness to acquire urban hukou is signi cantly higher than men's 35  the physical exercise of an individual in a city affect his or her choice of acquiring urban hukou in that city? The social effect of physical exercise has been shown to effectively improve community communication, community trust, community belonging, and the social integration of exercisers. So, does this social effect also exist in different groups, in particular in the oating population. Next, if physical exercise can promote the social integration of the oating population, and directly affects their acquisition of urban hukou, will physical exercise further in uence the acquisition behavior through the intermediary factor of social integration? To address these questions, this paper uses the latest China Labor Force Dynamics Survey data to explore the impact of physical exercise and social integration on the decision-making of the oating population in acquiring urban hukou. city have not been able to acquire the public services of the city until they acquire the city's urban household registration. Then, the clari cation of the relationship between physical exercise, social integration and the decision-making of the Chinese oating population in acquiring urban hukou will provide a new perspective for understanding the urban hukou decision-making of the Chinese oating population. At the same time, this study has important reference value for promoting the development of a new type of urbanization in China and the exploration of the laws around acquiring urban hukou.

Data
The data in this article comes from the China Labor Force Dynamics Survey (CLDS) conducted nationwide by Sun Yat-Sen University in 2014 and 2016. When the investigation was launched at Sun Yat-Sen University, the informed consent of the respondents was obtained. Therefore, we declare that the CLDS2014 and 2016 data used in this article have obtained the informed consent of the

Variables
Physical exercise (PE). In the CLDS, surveys were conducted on whether the respondent had taken physical exercise in the last month, and the duration and frequency of that exercise. Due to the abnormal values of more than 360 minutes per exercise and more than 14 instances of exercise per week in the study sample, referring to the practices of Killgore &Schwab 42 , we performed 1% sample tailing processing. Then, through calculation, we obtained the average duration (minutes) and frequency of physical exercise per month. We took the natural logarithm of the average monthly physical exercise duration and frequency.
Social integration. Regarding the degree of social integration, the trust and familiarity between the respondents and their neighbors were investigated in the CLDS. With reference to the practices of Kim et al. 43 , this paper used neighborhood trust and familiarity to characterize the respondents' social integration. Because the respondents answered the above two questions on a 5-point Likert scale, these two variables were summed to obtain the social integration index of the respondent. Again, we took the natural logarithm of it. In order to verify the robustness of the empirical results, we used these two variables as the representative variables of social integration.
Acquisition of urban hukou decision-making. In the CLDS, a survey was conducted of respondents' decision-making in acquiring urban hukou in the work city. Respondents to this question could choose, 'willing', 'unwilling', or 'uncertain'. This paper used this variable to represent the respondents' decision-making in acquiring urban hukou 44 . For the selection of uncertain samples, we classi ed them into the unwilling sample group in the basic regression model. However, in order to ensure that the research results in this article were more robust, we deleted the uncertain samples, and we analyzed the basial model again which can be regarded as the robustness analysis results.
Control variables. In order to accurately identify the impact of physical exercise and social integration on the respondents' decision to acquire urban hukou, this paper controlled the demographic and family characteristics of the respondents. Speci cally, the demographic characteristics mainly include: gender, age, education level, political status, type of hukou, health status, and work status. Family characteristics mainly include parents' educational level. Table 1 shows the descriptive statistics of all the variables.

Methods
Referring to the research by Zang et al. 45 and C. Wang et al. 46 , this paper constructed the following empirical model: (1) residence it = β 0 + β 1 social it + ∑β i X it + city ' Among them,residence it represents the willingness of individual i to acquire urban hukou at time t; physical it represents the physical exercise of individual i at time t; social it represents the degree of social integration of individual i at time t; X it represents the control variables, including: age, gender, etc.; δ i andγ i are unobservable cities and time xed effects; ϵ it and ξ it are random error terms.

Ethical considerations
No separate ethical approval was needed as the study is based on secondary data obtained from the 2014 and 2016 the China Labor Force Dynamics Survey (CLDS). CLDS data was collected following all the ethical clearance guidelines and norms. Its protocol was approved by the Institutional of Ethics Committee in China.
3 Results Columns (4)-(6) list the regression results that included control variables and unobservable factors of cities and time. Column (4) shows that the impact of participation in physical exercise on the decision-making of the oating population in acquiring urban hukou is 0.026, which is signi cant at the level of 1%. Its economic signi cance is that the probability of participating in physical exercise increases by one standard deviation, and the willingness of the oating population acquiring urban hukou in the work city increases by 0.0073 percentage points. Similarly, Columns (5) and (6) show that the impact of the minutes of physical exercises on the decisionmaking of the oating population in acquiring urban hukou is 0.008, and the impact of the frequency of physical exercise on the decision-making of the oating population in acquiring urban hukou is 0.007, both of which are signi cant at the level of 1%. The economic signi cance is that if the minutes and frequency of physical exercise increase by one standard deviation, the willingness of the oating population acquiring urban hukou in the work city will increase by 0.153 and 0.399 percentage points, respectively. This proves that physical exercise can promote the acquisition of urban hukou in the oating population in the work city. Note: *, **, and ** indicate signi cant statistical levels of 10%, 5%, and 1%, respectively; Robust standard error in parentheses. Table 3 presents the results of social integration analysis on the decision-making of the oating population in acquiring urban hukou. Columns (1)-(3) list the regression results that do not include the control variables and unobservable factors of cities and time. Columns (4)-(6) list the regression results that included control variables and unobservable factors of cities and time. Column (4) shows that the impact of social integration on the decision-making of the oating population in acquiring urban hukou is 0.008, which is signi cant at the level of 1%. Its economic signi cance is that if the degree of social integration increases by one standard deviation, the willingness of the oating population to acquire urban hukou in the work city will increase by 0.0001 percentage points. Similarly, Columns (5) and (6) show that the impact of the representative variable of neighborhood trust on the decision-making of the oating population in acquiring urban hukou is 0.014, and the impact of neighborhood familiarity is 0.011, both at the 1% signi cance level. Its economic signi cance is that if neighborhood trust and neighborhood familiarity increase by one standard deviation, the willingness of the oating population to acquire urban hukou in the work city will increase by 0.045 and 0.033 percentage points, respectively. This proves that the degree of social integration can promote the acquisition of urban hukou in the oating population in the work city.

Variables
Acquisition of urban hukou decision-making  Using the test of the mediation effect 47 , we veri ed and measured the mediating effect of physical exercise on the acquisition of urban hukou decision-making in the oating population through social integration. Equations (3)-(5) were used to test the path of "physical exercise--social integration--decision-making in the oating population in acquiring urban hukou". Speci cally, the rst step was to test the impact of physical exercise on the decision-making of the oating population in acquiring urban hukou, which is the coe cient α 1 in Equation (3). The second step was to test the impact of physical exercise on social integration, which is the coe cient β 1 in Equation (4). The third step was to test the impact of physical exercise and social integration on the decision-making of the oating population in acquiring urban hukou, that is, coe cients ϕ 1 and ϕ 2 in Equation (5). When the coe cient ϕ 1 is statistically signi cant, it indicates that there is a mediating effect. In the case where the coe cients β 1 and ϕ 2 are signi cant at the same time, if the coe cient ϕ 1 is signi cant and the impact of physical exercise on the decision-making of the oating population in acquiring urban hukou becomes smaller, it indicates that there is a partial mediation effect. On the contrary, if the coe cient is not signi cant, there is a complete mediation effect. Furthermore, if at least one of the coe cients β 1 and ϕ 2 is not signi cant, the Sobel test is required to determine the signi cance of the mediation effect (β 1 × ϕ 2 ). The meaning of other variables is consistent with Equation (1).
residence it = α 0 + α 1 physical it + ∑α i X it + city ' (3) social it = β 0 + β 1 physical it + ∑β i X it + city ' residence it = ϕ 0 + ϕ 1 physical it + ϕ 2 social it + ∑ϕ i X it + city ' i δ i + year ' i γ i + ξ it (5) Table 4 shows the results of the veri cation and calculation of the path "physical exercise--social integration--decision-making of the oating population in acquiring urban hukou". Columns (1)-(3) verify and measure the path "participate in physical exercise--social integration--decision-making of the oating population in acquiring urban hukou". Column (1) shows the test results that do not include the intermediary factor (social integration). The coe cient is positive and signi cant. Column (2) shows the impact of participation in physical exercise on social integration; the coe cient is positive and signi cant, indicating that participation in physical exercise can indeed promote social integration. Column (3) shows the test results that include both the intermediary factor (social integration) and participation in physical exercise; the regression coe cients are signi cantly positive at the level of 1%, indicating that the higher the degree of social integration, the greater the willingness of the oating population to acquire urban hukou in the work city. And the coe cient (ϕ 1 ) is less thanα 1 , indicating that social integration is a part of the mediating factors that affect the decision-making of the oating population in acquiring urban hukou through participating in physical exercise, and the mediating effect accounts for 1.95%. Furthermore, we used the Sobel test to verify the signi cance of the mediation effect, and the Z statistics all passed a 5% statistical test, which again con rmed that the mediation effect was signi cant. Similarly, in Columns (4)-(9), the paths "the minutes of physical exercise--social integration--decision-making of the oating population in acquiring urban hukou" and "the frequency of physical exercise--social integration--decision-making of the oating population in acquiring urban hukou " were tested respectively. It can be seen that social integration is a part of the mediating factor of the minutes of physical exercise and the frequency of physical exercise, with the mediating effects accounting for 1.56% and 1.80%, respectively. And the Sobel Z statistics all passed the statistical test at the level of 10%. To sum up, physical exercise can promote the degree of social integration of the oating population, and ultimately, increase the willingness of the oating population to acquire urban hukou in the work city. Note: *, **, and ** indicate signi cant statistical levels of 10%, 5%, and 1%, respectively; Robust standard error in parentheses; The proportion of the mediating effect of the Sobel Z statistic was obtained using the Sgmediation command in Stata15.

Heterogeneity analysis
In the above sections, we considered the oating population as individuals with exactly the same preferences for acquiring urban hukou, and obtained relevant analysis results through empirical model calculations. In this section, we introduce the heterogeneity of the oating population, focusing on the two dimensions of gender and the type of hukou and the impact of physical exercise and social integration on the acquisition of urban hukou decision-making of the oating population. Table 5   Note: *, **, and ** indicate statistical signi cance at 10%, 5%, and 1%, respectively; robust standard errors in parentheses. Table 6 documents the mediating effects of social integration in physical exercise and acquisition of urban hukou decision-making in the male and female, and non-agricultural and agricultural hukou oating populations. First of all, from the perspective of gender, the mediating effect of female oating population social integration is generally greater than that of males' (2.285%, 2.607%> 1.800%, 0.978%) in the path "participation in physical exercise/the frequency of physical exercise--social integration-acquisition of urban hukou decision-making". Secondly, from the perspective of the type of hukou, the mediating effect of social integration of the agricultural hukou oating population does not exist, that is, physical exercise will not increase the willingness of the agricultural hukou oating population to acquire urban hukou by promoting the improvement of social integration. In contrast, in the non-agricultural hukou oating population, the mediating effect of social integration is common in the path "participating in physical exercise/minutes of physical exercises /frequency of physical exercise--social integration--acquisition of urban hukou decision", with the mediating effect reaching 1.515%, 1.036% and 0.395%. At the same time, the Sobel Z statistics all passed a 10% statistical test, which again con rmed that the mediation effect was signi cant.

Robustness Checks
In the robustness analysis section, we deleted the "uncertain" answers, in terms of acquisition of urban hukou, form the sample. We reexamined the mediating effects of physical exercise and acquisition of urban hukou decision-making, social integration and acquisition of urban hukou decision-making, and physical exercise, social integration and acquisition of urban hukou decision-making. Table 7 reports the robustness results. The results show that the positive relationship between physical exercise, social integration, and acquisition of urban hukou decision-making still exists, and is signi cant at the 1% level. In addition, the coe cients of each variable are larger than those estimated above, which shows that we have previously underestimated the impact of physical exercise and social integration on the decision-making of the oating population in acquiring urban hukou. Similarly, the mediating effects of physical exercise, social integration, and decision-making of the oating population in acquiring urban hukou still exist, and are signi cant at the 10% level. Overall, this shows that our estimation results are very robust.

Discussion
Existing studies have basically explored the relationship between physical exercise and social integration, and social integration and acquisition of urban hukou decision-making. But the research samples have been relatively small and the representativeness relatively weak. In this paper, physical exercise, social integration, and acquisition of urban hukou decision-making were combined, and the nationally representative data of the China Labor Force Dynamics Survey were used to explore the mechanism of physical exercise, social integration, and acquisition of urban hukou decision-making in the oating population. First of all, the positive relationships between physical exercise and the decision-making of the oating population in acquiring urban hukou, and social integration and the decision-making of the oating population in acquiring urban hukou were veri ed. That is, when the oating population participates in physical exercise, it has a higher degree of social integration with urban residents and is more willing to acquire urban hukou in the working cities. Furthermore, the improvement of social integration will directly promote the acquisition of urban hukou by migrants in the work city, which is basically consistent with the conclusions of Cao et al. 48 .
Next, considering the in uence of control variables on the decision-making of the oating population in acquiring urban hukou, our research results are consistent with Hu et al. 36 , Zang et al. 45 , Xiang 35 , and L. Zhang &Tao 49 . There is a signi cant negative relationship between age and acquisition of urban hukou decision-making, which is consistent with Liu et al. 44 . That is, the older the oating population is, the less willing they are to acquire urban hukou in the work city. This may be mainly a symptom of the Chinese household registration system. China's household registration system sets requirements for education level, skill level, etc. The older oating population are often those with a lower level of education and lower skill levels. Therefore, this type of oating population is less willingness to acquire urban hukou in cities.
Secondly, the mediating effects of social integration through physical exercise and acquisition of urban hukou decision-making of the oating population are discussed. The study of Tonts 50 only emphasized that physical exercise is an important way to create rural community communication and maintain harmony in rural communities, but did not further analyze the impact of physical exercise on the acquisition of urban hukou in the oating population. Then, this article makes a very important academic contribution concerning the mediating effect of social integration. We link these three factors and verify their existence in China's oating population. On the one hand, this provides empirical evidence for the positive role of physical exercise in the construction of a harmonious socialist society; it provides new ideas for promoting China's new urbanization construction. Additionally, it enriches the research on the factors in uencing acquisition of urban hukou decision-making in the oating population. Also, studies by An et al. 51 and Gay &Trevarthen 52 point out that the environment of community exercise has an important impact on the exercise behavior of community residents, and that the richer the facilities for community exercise and the more continuous the participation of adults in the community, the higher the probability of physical exercise. It can be seen that it is not enough to just improve participation in physical exercise; the construction of community physical exercise facilities should be improved 29 , and an atmosphere of community participation in exercise should be fostered. Loading [MathJax]/jax/output/CommonHTML/jax.js Third, we compared the relationship between physical exercise, social integration, and acquisition of urban hukou decision-making in terms of gender and rural or urban hukou status. It was found that physical exercise has a greater impact on the decision-making in acquiring urban hukou among female migrant groups, and the mediating effect of social integration is also stronger among female groups. In addition, it is worth noting that the intermediary effect of social integration does not exist in the oating population with agricultural hukou, and the mechanism by which physical exercise affects the decision of the oating population in acquiring urban hukou through social integration is signi cant in the non-agricultural hukou oating population. One possible reason is that the proportion of the oating population with agricultural hukou participating in physical exercise is low. Therefore, the degree of social integration promoted by physical exercise is relatively low, and as a result, the mediating effect of social integration cannot be exerted.
Finally, although physical exercise has a positive effect on social integration and acquisition of urban hukou decision-making in the oating population, the reality of low awareness and low participation rates in physical exercise directly limits the role of physical exercise in the development of a harmonious society and new urbanization. Our data shows that only 28% of migrants participate in physical exercise (see Table 1).

Conclusions And Policy Implications
Based on the 2014 and 2016 China Labor Force Dynamics Survey data, this paper studies the relationship between physical exercise, social integration, and decision-making in the oating population in acquiring urban hukou, further examines the mediating effects of social integration, and validates the effect of physical exercise on decision-making of the oating population in acquiring urban hukou.
The results show that the participation of the oating population in physical exercise increases the frequency and duration of physical exercise, which promotes the willingness to acquire urban hukou in the work city. Meanwhile, social integration can also directly encourage the oating population to acquire urban hukou in the work city. Further mechanism analysis shows that physical exercise can be an important medium for promoting the integration of local and oating populations and social harmony, and it can increase the willingness of the oating population to acquire urban hukou in the work city by promoting the degree of social integration. And, the positive mediation effect is quite signi cant. Finally, we also nd that the mediating effect of physical exercise on the decision-making of the oating population in acquiring urban hukou through social integration is most signi cant in the non-agricultural hukou and female oating population. This paper has the following policy implications. (1) When promoting the oating populations in cities, government should increase awareness of physical exercise among gloating populations and guide them to participate in physical exercise, especially to increase the proportion of agricultural hukou holders participating in physical exercise. This will promote social integration and a healthy lifestyle.
(2) The government should increase the construction of community sports facilities, provide basic conditions for the oating population to participate in physical exercise, and promote public health. (3) Communities should hold collective sports activities. While strengthening the mutual exchanges among the residents of the community. This can also promote a harmonious community.
(4) Focus should be placed on sports activities with the non-agricultural hukou oating population as the mainstay, tapping the potential of the oating population with agricultural hukou to acquire urban hukou in cities. This can promote the urbanization of the non-urban hukou oating population in China.

Data availability
The datasets used and/or analyzed during the current study are not publicly available because of privacy issues but are available when you have applied. The application website is: http://css.sysu.edu.cn/Data.