Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Quality of Life of Rural–Urban Migrant Children in China: A Cross-Sectional Study

Quality of Life of Rural–Urban Migrant Children in China: A Cross-Sectional Study Abstract This study aims to examine the quality of life of Chinese migrant children and to explore whether social capital mediates the relationship between a child’s migration status and quality of life. A cross-sectional survey was conducted in a district of Shanghai, China. Based on a sample consists of 1,577 Chinese children in fourth to sixth grade, we used multivariable linear regression to examine the associations and mechanisms among migration, social capital and quality of life. Migrant children reported lower scores for quality of life than their urban counterparts. However, we identified higher levels of neighbourhood social cohesion, trust in their school and neighbourhood, and higher level of maternal autonomy support can mediate the damaging effects of migration status on quality of life. The results indicate that hukou restrictions put migrant children at a disadvantageous position and negatively affect their quality of life. While their migration status is unlikely to change soon, we found that higher levels of certain social capital can alleviate the negative effects of migration status on children’s quality of life, and thus can be utilised for welfare programme design. Quality of life, ecological systems theory, social capital, migrant children, China Introduction Contexts of internal migration in Mainland China Since the economic reform in the late 1970s, fluid labour movement, initially from rural to urban regions, has been one of the main drivers of China’s economic development. As defined by China’s household registration system (hukou), internal migrants are those who live somewhere other than their registered birthplace. By the end of 2013, the migrant population had reached 245 million, accounting for one-sixth of the national population (Hu, 2014). Among this number were 35.81 million children who migrated with their parents or were born after their parents’ arrival in cities (Duan et al., 2013). Many studies have paid attention to international migrants, as they are expected to confront difficulties due to differences of race, geopolitical distance and political systems (Mallee and Pieke, 2014). However, movement within a country’s borders also can be difficult. The government viewed migrants as temporary residents in urban areas and rejected their access to public services in host cities. Migration experience is a risk factor for child development because migrant children encounter an inter-generational acculturation gap (Chan, 2015), lack of educational opportunities (Xiong, 2015) and maladjustment in urban environments (Cheng and Selden, 1994). Children without a local hukou cannot access publicly funded schools, so most of them attend lower-quality migrant schools. Some local governments have made efforts to improve migrant children’s schooling. Take Shanghai, for instance, which allocated a certain amount of money to children in migrant schools and implemented a certain quota for the enrollment of migrant children in public schools (Qian and Walker, 2015). Even with local initiatives that attempt to ameliorate inequality in educational resources between migrant and urban children, the best schools with high-quality resources and teachers are still inaccessible to migrant children (Lai et al., 2014). Thus, this study selected migrant schools of different quality and those public schools open to migrant children to ensure a variety of schools and students in the sample. In addition, migrant families are not eligible to buy or rent subsidised apartments, so they are usually concentrated in the outskirts of cities. Minhang district of Shanghai was chosen as the survey site in view of its location at the southern edge of the city and its large proportion of migrant population. In addition to poor schooling experience and living conditions, scholars have found migrant children were less likely to utilise public health services compared to local urban children (Sun et al., 2016). These disadvantages that are attached to migration status have negative impacts on the lives of migrant children. Migration as a challenge to quality of life Migration is often viewed as a stressful event, as it breaks the equilibrium between the person and the environment, especially when it involves tremendous social costs but with uncertain economic benefits (Hwang et al., 2007). Most migrants inevitably experience life changes and readjustment challenges during their relocation. Children are often victims of such changes over which they have limited power; therefore, any household change can influence the resources available for migrant children. Hwang et al. (2007) argued that involuntary migrants were prone to stress when they had little control over some strong external forces. In addition, migration can not only directly, but also indirectly, elevate health problems by changing the intangible resources (i.e. social capital) embedded in social relations. Although migration does not have definitely negative effects on child quality of life, it depends on what changes occur to them and how much control migrant children have. Quality of life is a person’s subjective evaluation about the cultural, social and environmental context (WHO, 1997). Quality of life examines the difficulties of children in physical, emotional, social and schooling functioning in daily lives, which can identify the specific health care needs of children. It is not hard to imagine children are vulnerable to many threats caused by migration. For instance, internal migrant children within China may face difficulties resulting from family separation, inadequate parenting and restricted access to public benefits (Cheng and Selden, 1994). In addition, scholars have investigated some outcomes of migrant children, which usually focused on limited dimensions of quality of life. Educationists focused on schooling and academic performances of migrant children (Hu and Szente, 2010; Gong et al., 2015), psychologists addressed their mental health outcomes (Wong et al., 2010; Shi and Wang, 2010) and paediatricians highlighted the physical health outcomes of migrant children (Lu et al., 2008; Ji et al., 2016). However, few studies have attempted to simultaneously investigate the four aspects of quality of life among migrant children. In this study, we apply ecological systems theory and social capital theory to explore the association among migration status, social capital and children’s quality of life. Findings from this study have the potential to illuminate intervention strategies and shed light on policy making. Conceptual frameworks Ecological systems theory As Finkelhor and Hashima (2001) argued, individual disadvantage not only arises from his or her physical and mental weakness, but also depend on environmental factors (Liu, 2012). Ecological systems theory is useful to identify a series of ecological changes that happen to migrant children and link them to the quality of life of migrant children. Specifically, family, school, peers and neighbourhood have the most immediate influence on children’s lives (Bronfenbrenner, 1994). Given the living conditions and community environment in which migrant children spend their lives, it is likely that these environmental factors pose several risks to migrant children’s quality of life. Migrant children often experience a series of ecological changes that do not take place in local urban children. For instance, migration is related to the changes of family structure, social status and family wealth in the microsystem, as many migrant children have to face difficulties of family separation and inadequate parenting (Smith-Greenaway and Madhavan, 2015; Wu and Zhang, 2015). Meanwhile, the single-parent family is often associated with deprivation of economic resources of children, which further impacts the physical, emotional and behaviour health of children (Conger and Donnellan, 2007). In addition, changes also occur at the mesosystem level, as the residential mobility may negatively impact the relationships between migrant children and their parents, peers and neighbourhoods. The ecological systems theory provides a framework that depicts the important social contexts of migrant children. It is difficult to consider all relations and influences on the numerous contexts. Nevertheless, this study focuses its interest on the individual, family, school, peer and neighbourhood levels. All of the changes related to migration are likely to influence the quality of life of individuals. Social capital theory Although the ecological systems theory highlights the potential negative impacts of migration on migrant children’s quality of life, individuals are not passive receivers of their environmental influences. According to a meta-analytical article of the health outcomes of migrant children, migrant children from migrant schools presented sickness or deficit compared to their local urban counterparts, while migrant children from public school did not (Sun et al., 2016). Therefore, we need to increase the focus on the resiliency of migrant children, which can positively respond to the harsh ecological changes. This study takes a strength-based approach, which can not only identify the strengths of migrant children, but also suggest community-based practices. Social capital has been given special attention because it argues that individuals can actively resort to the surrounding environment to achieve certain goals. The social capital within a family indicates that children can depend on their parent’s financial and human capital to achieve certain goals, such as the presence of parents and attention paid to children by their parents (Coleman, 1988). In addition, social capital in the community indicates that children’s development is also influenced by their families’ relationship with other significant adults in the community (Coleman, 1988). Studies have identified social capital as an important factor for child health and quality of life (Drukker et al., 2003; Morrow, 2004) and many countries have incorporated this idea into child and adolescent health-promotion programmes (Rocco and Suhrcke, 2012; Rushton and Kraft, 2014). According to a systematic review conducted by McPherson et al. (2014), research has proved the link between positive relations with parents, peers, teachers and neighbours and better health outcomes, subjective well-being and life satisfaction (Lau and Li, 2011). The social capital of Chinese migrant children is rarely studied. Some studies have employed social support or parental social capital as a proxy due to a lack of measures for migrant children’s social capital (Wu et al., 2010, 2014), but they cannot capture the core of social capital, which is mutually beneficial resources for certain goals. Other studies have failed to capture all of children’s important relationships (Lacobucci and Duhachek, 2003; Lau and Li, 2011). Social capital deserves more attention and better measures because migrant children’s perceived social capital relates to their health and well-being. In addition, it is uncertain whether migrant children and their families can build beneficial social relations to facilitate good life quality. Coleman (1988) reasoned that social capital is generated through a relational closure (i.e. a family or a community). A change of residence can break this closure and reduce a family’s level of social capital (LeSage and Ha, 2012). However, Lau and Li’s study found that children in Shenzhen without local hukou had close teacher–student relationships, and that their parents had close bonds with their school, which positively linked to subjective well-being (2011). These inconsistent findings regarding migrant children and their families’ social capital is puzzling. Yet, the concept of social capital is an important one to study because they are amenable factors that can be utilised in designing programmes that promote the quality of life among migrant children. While previous studies have sought to identify the association between the disadvantages of migrant children and their negative health outcomes, little has yet examined the protective influence of social capital on quality of life among Chinese migrant children. The current study proposes a conceptual framework of combining the ecological systems theory and the social capital theory, which is shown in Figure 1. Despite the locating level in the ecological systems, these factors can be summarised in four clusters: (i) socio-demographic characteristics, (ii) tangible resources available for a child, (iii) social capital and (iv) quality of life of a child. Given the relationships among migration status, social capital and quality of life, it is reasonable to propose that migration status and social capital are key determinants of children’s quality of life. In addition, social capital is related to both migration status and children’s quality of life, so we speculate that social capital may mediate the relationship between migration status and child quality of life. Figure 1 View largeDownload slide Conceptual framework. Circles indicate different levels of ecological systems. Figure 1 View largeDownload slide Conceptual framework. Circles indicate different levels of ecological systems. Furthermore, this study tends to answer the following research questions: (i) Is the quality of life of migrant children lower than that of local urban children in urban China? (ii) Is the migration status and social capital a major determinant of the quality of life of children? (iii) Does social capital in migrant children mediate the relationship between migration status and child quality of life? Methods Study sample This study chose Minhang district of Shanghai, China, as the study site, which is the major neighbourhood of the migrant population. In order to include enough migrant child participants, we administered a school-based survey. The survey was conducted between October and November 2015. We contacted all principals of migrant schools in Minhang district and eight out of fourteen agreed to participate. In addition, four public schools open to migrant children agreed to participate. The schools are located in six towns of Minhang district and the quality of the migrant schools varies in terms of the conditions of facility, the composition of migrant/local students and the location of the neighbourhood. The quality and reputation of the four public schools are ordinary, which guarantees that the socio-economic backgrounds of the children do not differ excessively from the migrant schools. Our sample included all eligible children in twelve participating schools, who were enrolled in fourth to sixth grades. With the principals’ support and teachers’ facilitation, we had a response rate of 100 per cent: 1,322 migrant children (542 female and 781 male), 26.17 per cent of whom attended public schools. In addition, 255 local children at the same public schools were recruited as a comparison group (104 male and 151 female). The surveys were conducted in the classroom setting and instructions (briefing and consensus seeking) were given by a researcher. On average, the students took 35 minutes to complete the questionnaire. This study was approved by the Human Research Ethics Committee at the University of Hong Kong (EA1506011). Measures Main outcome variable The dependent variable was the quality of life of migrant children, measured by Pediatric Quality of Life Generic Core Scales for children aged eight to twelve (PedsQL 4.0 Child Self-Report) (Yang et al., 2011). PedQL has been widely used to examine child health-related outcomes through self-report (Varni et al., 2001). The PedsQL 4.0 Generic Core Scales have shown high construct validity and reliability (Cronhach’s alpha = 0.88) (Varni et al., 2001). The Chinese version also has good reliability (Cronbach’s alpha = 0.86) and construct validity (Yang et al., 2011). Cronbach’s alpha for this sample was 0.87. It consists of twenty-three items referring to four subscales of quality of life: physical functioning (eight items), emotional functioning (five items), social functioning (five items) and school functioning (five items). Items were scored on a five-point Likert-type scale, indicating the frequency of problematic functioning. The scores were reverse coded, ranging from 0 and 100, so that a higher score represents a better quality of life. Independent variables The key independent variables of this study are the migration status and social capital of the children. A child was defined as a migrant child if he/she had a rural and non-local hukou (migrant child = 1, local child = 0). The measurement of the children’s social capital consists of two parts: the Social Capital Questionnaire for Adolescent Students (SCQ-AS; Paiva et al., 2014) and parental autonomy support subscales of Perceptions of Parents Scales (child POPS; Grolnick et al., 1991). Social Capital Questionnaire is the first instrument to detect a child’s social capital with a three-point scale (ranging from 1 = disagree to 3 = agree). The twelve-item scales demonstrated satisfactory reliability (Cronbach’s alpha = 0.71) and construct validity in the original version (Paiva et al., 2014). We supplemented two more indicators of parental autonomy and control by using a twelve-item subscale of POPS. Participants rated parental autonomy support on twelve items (six for mothers and six for fathers) according to four descriptions of types of parent and then chose the most appropriate one for each item (scores ranged from 1 to 4). The sum of these six items indicates the degree of maternal/paternal autonomy support; higher scores indicate more autonomy support, ranging from 6 to 24. Both subscales in the original version have satisfactory reliability (Cronbach’s alpha = 0.53 and 0.67) and validity and are found to be linked to children’s mental health outcomes (Grolnick et al., 1991). Two authors translated the English version of the scales into Chinese and then a professional translator back-translated it. Based on the 1,577 participants in this study, the Cronbach’s alpha coefficients for the scales of school cohesion, school friendships, neighbourhood cohesion, trust, maternal autonomy support and paternal autonomy support were 0.44, 0.60, 0.83, 0.56, 0.46 and 0.58, respectively. We then calculated the sum of each subscale to indicate six dimensions of the social capital of a child. Other independent variables We included several socio-demographic covariates of child health outcomes: (i) child age, child gender (female = 1), child ethnicity (ethnic minority = 1); (ii) public-school attendance (attending public school = 1) and school transfer (one transfer or more = 1) related to schooling experience; (iii) socio-economic status (SES), family structure (single-parent family = 1, two parents live together = 1, the presence of siblings = 1); and (iv) tangible resources (homeownership = 1, availability of private room for a child = 1, participation in extra-curricular course = 1). Age, SES and the number of siblings is a continuous variable and the others were all coded as dummy variables. Migrant children are likely to experience multiple transfers due to the instability of their parents’ work. Residential instability is viewed as a risk factor for child health, so we controlled for the frequency of previous school transfers. SES often links to children’s access to resources and quality of life (Rajmil et al., 2014). Based on Straus and Douglas’s (2004) study, SES was computed as a sum of the educational level of the father and mother (scores ranged from 1 to 8 for both variables) and household income level (scores ranged from 1 to 12). The Cronbach’s alpha for the resulting scale was 0.64. In addition, housing conditions (homeownership and availability of private room for a child) can represent the consumption ability of a household to some extent, which is also related to the financial and material resources. It is also related to the physical environment of child activity and the child’s privacy. Extra-curricular course attendance reflects the resources that benefit a child’s social and emotional development out of the classroom. Data analysis The purpose of this study is to explore the determinants and potential mediators of children’s quality of life. First, we summarised the dependent variable, independent variables and control variables by the children’s migration status and computed the group difference between migrant and local children with t-test (for continuous variables) and chi-square test (for categorical variables). Second, we employed multivariate regression to investigate the extent to which migration status is related to children’s quality of life. Finally, we employed the Baron-Kenny approach and bootstrapping to identify and estimate the indirect effect of social capital factors. It is noteworthy that the Sobel test is a conservative way with low power to test the indirect effect; however, bootstrapping is highly recommended as an increasingly popular method (Creedon and Hayes, 2015). We conducted non-parametric resampled residual bootstrapping of mediation with 2,000 replicates to investigate the mediation effects of six factors of social capital respectively on the association of migration status with quality of life. All analyses were run in Stata 12. Results Descriptive analysis Table 1 demonstrates the mean, standard deviation and the analysis results of the group variance. Significant differences were found between migrant children (82.40) and local urban children (84.03) in reported scores of quality of life. Regarding the socio-demographic backgrounds, the two groups had lots in common, including the proportion of minority group, two parents live together and single-parent family. The average age of migrant children participants was 11.11, while that of local urban children was 10.93. It is noteworthy that the migrant children showed no difference from local children in reported school cohesion and school friendships, while they reported higher levels of neighbourhood cohesion and trust in neighbourhood and school. The three indicators of tangible resources showed that migrant children lived in poorer conditions and were less likely to participate in extra-curricular activities/courses compared to local urban children. Table 1 Distribution of sample characteristics Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 *p < 0.05; **p < 0.01; ***p < 0.001. View Large Table 1 Distribution of sample characteristics Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 *p < 0.05; **p < 0.01; ***p < 0.001. View Large Association of quality of life with migration status and social capital Table 2 shows the association of quality of life with migration status based on five models, adjusting for different variables in four blocks. The variance inflation factor (VIF) test showed no indication of multicollinearity in the present data. In Model 1, the coefficient for migration status was statistically significant. Migrant children’s quality-of-life score was 1.63 lower than that of their local counterparts. The influence of migration status was not constant when including three individual-level demographics in Model 2. In Model 3, the coefficient for SES, single-parent family and the number of siblings was statistically significant. One unit increase in SES is associated with a 0.14 increase in quality of life, holding all else constant. One more sibling at a household is associated with a 1.12 lower score in quality of life of a child, holding all else constant. Model 4 includes three more variables of tangible resources, and only the coefficient for the availability of a private room was statistically significant. Children with a private room at home on average reported a 1.44 higher score of quality of life than those without one. Next, when including social capital of the child in Model 5, the effects of number of siblings and private room for a child were both overridden. However, all of the six social capital variables showed a positively significant effect on the quality of life of children. Table 2 Association of migration status with quality of life of child, adjusting for socio-demographic variables, tangible resources and social capital variable blocks Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 *p < 0.05; **p < 0.01; ***p < 0.001. aReference group: local urban children; breference group: female child; creference group: Han Chinese; dchild from migrant school; ereference group: child has no transfer experience; freference group: two-parent family; greference group: parents live apart; hreference group: child has no sibling; ireference group: rent/dorm/other; jreference group: has no private room for a child; kreference group: no participation in extra-curricular activities/courses. View Large Table 2 Association of migration status with quality of life of child, adjusting for socio-demographic variables, tangible resources and social capital variable blocks Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 *p < 0.05; **p < 0.01; ***p < 0.001. aReference group: local urban children; breference group: female child; creference group: Han Chinese; dchild from migrant school; ereference group: child has no transfer experience; freference group: two-parent family; greference group: parents live apart; hreference group: child has no sibling; ireference group: rent/dorm/other; jreference group: has no private room for a child; kreference group: no participation in extra-curricular activities/courses. View Large Exploring the mediation effect of social capital According to the suggestions of Judd and Kenny (1981) for identifying a mediator, we estimated the following three regression equations: (i) regressing the social capital variables on the quality of life, (ii) regressing the quality of life on the migration status and (iii) regressing the quality of life on both the migration status and the social capital. Then, we found that the six social capital variables were all positively associated with the quality of life at a 95 per cent confidence level. Migration status was negatively associated with the quality of life (β = –1.63, p < 0.05). However, the effect of migration status on the quality of life was not significant when social capital variables were controlled. Therefore, social capital functions as a mediator, as it meets the above three conditions. In line with our previous hypotheses, we investigated whether migration status has any significant indirect effects on the quality of life via social capital with bootstrapping analysis. According to Table 3, neighbourhood cohesion and trust partially mediate the effect of migration status on quality of life, while maternal autonomy support can fully mediate it. Other dimensions of social capital do not have a mediation effect (Figures 2–7). Figure 2 View largeDownload slide Model of migration status effects on the quality of life with school cohesion as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 2 View largeDownload slide Model of migration status effects on the quality of life with school cohesion as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 3 View largeDownload slide Model of migration status effects on the quality of life with school friendship as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 3 View largeDownload slide Model of migration status effects on the quality of life with school friendship as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 4 View largeDownload slide Model of migration status effects on the quality of life with neighbourhood cohesion as a mediator . *p < 0.05; **p < 0.01; ***p < 0.001. Figure 4 View largeDownload slide Model of migration status effects on the quality of life with neighbourhood cohesion as a mediator . *p < 0.05; **p < 0.01; ***p < 0.001. Figure 5 View largeDownload slide Model of migration status effects on the quality of life with trust (in school and neighbourhood) as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 5 View largeDownload slide Model of migration status effects on the quality of life with trust (in school and neighbourhood) as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 6 View largeDownload slide Model of migration status effects on the quality of life with maternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 6 View largeDownload slide Model of migration status effects on the quality of life with maternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 7 View largeDownload slide Model of migration status effects on the quality of life with paternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 7 View largeDownload slide Model of migration status effects on the quality of life with paternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Table 3 Standardised direct, indirect and total effects of different domains of social capital on the association between migration status and quality of life Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Control variables: ethnic minority, frequency of school transfer and single-parent family. *p < 0.05; **p < 0.01; ***p < 0.001. View Large Table 3 Standardised direct, indirect and total effects of different domains of social capital on the association between migration status and quality of life Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Control variables: ethnic minority, frequency of school transfer and single-parent family. *p < 0.05; **p < 0.01; ***p < 0.001. View Large Discussion In general, migrant children reported lower quality-of-life scores, were more likely to have school transfer, were less likely to live in a higher SES family and were less likely to report higher levels of maternal autonomy support relative to children with local hukou. We found that frequent school transfer, low SES, single-parent family structure and lack of social capital are all negatively associated with quality of life. This indicates that migration status lays the foundations for many risk factors for quality of life. Direct hukou reform would eliminate many of the systematic barriers that migrant children face, such as authorising migrants’ free access to justice, education and health care. The future of hukou reform is gloomy. Take Guangdong’s policy, for instance; it allows eligible migrant children to take the college entrance exam (gaokao), but only 9,570 applicants were approved (Li, 2016). To some extent, this policy provides an opportunity to continue in education, but a large portion of migrant children have discontinued their education before high school. It is uncertain how much migrant children can benefit from this reform. The results showed that the coefficient for migration status was not significant while adding other variables, meaning that migration status does not have a determinate effect on quality of life. In this instance, the current study continues the probe into social approaches to improving the quality of life of migrant children. The results have several implications for research and practice. This study allowed young adolescents to report social capital rather than using proxies, which adds value to its exploration of adolescent-perceived social capital in broad contexts and verification of social capital’s relationship to migration status and quality of life for Chinese migrant children. Social capital is not only directly related to children’s quality of life, but also clarifies certain circumstances in which migration status does not have a definite connection with quality of life. Critically, we found that migrant children were more likely to report higher levels of neighbourhood cohesion and trust of school and neighbourhood than their local counterparts, all of which improve their quality-of-life score. It is not surprising that migrant children have higher levels of social capital in these two domains in this study. First, we observed that most migrants from the same hometown are likely to live in the same community and their children may go to the same schools nearby, so it is easy for them to build contacts. Second, migrant children’s higher levels of certain social capital can also be attributed to the Shanghai government’s investment to improve the schooling of migrant children. School can play a determining role in constructing trustful and cohesive relationships for migrant children and families. For instance, Scotland’s schools employed a social capital programme to prevent children who lacked opportunities being socially excluded in fierce competition (Flint, 2011 ). On the one hand, the programme promoted a ‘shared identity’ and ‘shared goal’ for all stakeholders. On the other hand, it helped children at risk to know about their environment and turn to the right people for help when there was a need. The negative effect of a single-parent family is constantly significant. It is not surprising because children living with a single parent receive less attention compared to those living with both parents. Therefore, programmes for improving child quality of life should pay special attention to children living with a single parent. However, lack of maternal autonomy support may lead to greater disadvantages for migrant children. According to self-determination theory, children can benefit from parents’ autonomy support (To et al., 2016). Many countries have recognised the importance of parenting in child development and applied early interventions, such as the Head Start programme in the USA. Traditional parenting skills in China are rather different and are more authoritative. It emphasises on disciplining children using appropriate punishments, but urban families (with urban hukou) are more likely than rural families to appreciate autonomy support for adolescents (Chen and Li, 2012). Previous studies have often found lower support for autonomy to be related to lower SES, meaning that migrant children are more vulnerable to low autonomy support (To et al., 2016). Therefore, programmes should target vulnerable children and provide more family-based interventions, such as offering childcare and teaching parents positive interaction and communication skills, which may buffer higher quality of life. Findings of this study should be interpreted with the following caveats. First, due to the limited data collection, we only employed a purposeful sampling method in one district of a city with participants of a narrow age group. Samples generated from this sampling strategy cannot represent the population of migrant children as a whole. Therefore, strict design in future studies is needed to demonstrate more dynamics with a representative sample. Second, the cross-sectional nature of this study made it not possible to conclude on any causative relationship. Future studies should consider using a cohort study design, which could allow collection of longitudinal data by tracking the lives of migrant children to address the dynamic changes along with the ongoing systematic reform; this could update our knowledge about this group and also provide substantial evidence for policy making. Third, some subscales of the instrument of social capital used in this study did not have high internal consistency, which can be attributed to the different characteristics of the sample. The sample of this study (n = 1,577) was much bigger than that in the original validation study (n = 101). According to previous studies, sample size has an adverse relationship with the Cronbach’s alpha coefficient (Lacobucci and Duhachek, 2003; Javali et al., 2011). Other than that, the number of items also influences the calculation of internal consistency. Most indicators were only measured by three or four items, which may cause an underestimation of reliability (Tavakol and Dennick, 2011). Therefore, we need to be cautious about Cronbach’s alpha values in order to avoid mistakes in our research. However, authors have consensus about the appropriate content and the length of the scale for measuring adolescents’ social capital. In addition, findings of this study have reference value for future studies. First, the difficulties and risks facing Chinese internal migrant children have a great deal in common with those in Europe and South Asia. For example, these migrant children often suffer from low SES, instability of family structure and inappropriate support from parents (Sun et al., 2016). Thus, the conceptual framework can be employed in future studies in other countries. Second, it suggests that the disadvantaged children can overcome difficulties by resorting to social capital and improve their quality of life. Therefore, social capital could be a protective factor for other at-risk children as well. Examinations of social capital in other at-risk children will advance the application of social capital in child studies. Conclusion This study is one of the few that focuses on the quality of life among migrant children in urban China and the associated risk and protective factors to quality of life. We find that, on average, migrant children in urban settings have lower quality of life. In addition, our findings suggest that migrant children can resort to social capital to mitigate the negative impact of migration status on their quality of life. Our findings have practical implications to policy makers who are concerned about resource distribution and the well-being of migrant children and their families in urban settings. This study provides evidence to suggest an alternative intervention approach for practitioners and a new direction to policy making on caring for migrant children in urban China. Acknowledgements We thank Bin Fan, Ph.D., for co-ordinating participants in Shanghai. Special thanks go to Yang Liu and Xiaonuan Sun, who facilitated the data collection. The project was granted ethical approval by the Human Research Ethics Committee of The University of Hong Kong in June 2015 (reference number: EA1506011). We have obtained participants’ consent to use all the figures for publication, online and in print, and in perpetuity. Funding There is no funding source. Conflict of interest statement. None declared. References Bronfenbrenner U. ( 1994 ) ‘ Ecological models of human development ’, Readings on the Development of Children , 2 ( 1 ), pp. 37 – 43 . WorldCat Chan K. L. ( 2015 ) ‘ Migration in China: Changing China, changing the world ’, in Chan K. L. (ed.), Chinese Migration and Families-At-Risks , Newcastle upon Tyne, Scholars Publishing Ltd . Google Preview WorldCat COPAC Chen X. , Li D. ( 2012 ) ‘ Parental encouragement of initiative-taking and adjustment in Chinese children from rural, urban, and urbanized families ’, Journal of Family Psychology , 26 ( 6 ), pp. 927 – 36 . Google Scholar Crossref Search ADS PubMed WorldCat Cheng T. , Selden M. ( 1994 ) ‘ The origins and social consequences of China’s hukou system ’, The China Quarterly , 139 , pp. 644 – 68 . Google Scholar Crossref Search ADS WorldCat Coleman J. S. ( 1988 ) ‘ Social capital in the creation of human capital ’, American Journal of Sociology , 94 , pp. S95 – 120 . Google Scholar Crossref Search ADS WorldCat Conger R. D. , Donnellan M. B. ( 2007 ) ‘ An interactionist perspective on the socioeconomic context of human development ’, Annual Review of Psychology , 58 ( 1 ), pp. 175 – 99 . Google Scholar Crossref Search ADS PubMed WorldCat Creedon P. S. , Hayes A. F. ( 2015 ) ‘ Small sample mediation analysis: How far can we push the Bootstrap ’, in Annual Conference of the Association for Psychological Science , New York, NY . Google Preview WorldCat COPAC Drukker M. , Kaplan C. , Feron F. , Van Os J. ( 2003 ) ‘ Children’s health-related quality of life, neighbourhood socio-economic deprivation and social capital: A contextual analysis ’, Social Science and Medicine , 57 ( 5 ), pp. 825 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat Duan C. , Lv L. , Wang Z. , Guo J. ( 2013 ) ‘ The survival and development status of floating children in China: An analysis of the Sixth Population Census data ’, South China Population , 28 , pp. 44 – 55 . WorldCat Finkelhor D. , Hashima P. Y. ( 2001 ) ‘ The victimization of children and youth: A comprehensive Overview ’, in White S. O. (ed.), Handbook of Youth and Justice , New York, NY , Plenum Publishers . Google Preview WorldCat COPAC Flint N. ( 2011 ) Schools, Communities and Social Capital: Building Blocks in the ‘Big Society’ (Research Associate Full Report). National College for Leadership of Schools and Children’s Services (NCSL), available online at http://dera.ioe.ac.uk/2981/ (accessed 17 September 2014). Gong X. , Zhang H. , Yao H. ( 2015 ) ‘ The determinants of compulsory education performance of migrant children in Beijing: An analysis of two cohorts ’, International Journal of Educational Development , 45 , pp. 1 – 15 . Google Scholar Crossref Search ADS WorldCat Grolnick W. S. , Ryan R. M. , Deci E. L. ( 1991 ) ‘ Inner resources for school achievement: Motivational mediators of children’s perceptions of their parents ’, Journal of Educational Psychology , 83 ( 4 ), pp. 508 – 17 . Google Scholar Crossref Search ADS WorldCat Hu H. ( 2014 ) ‘ Report on the development of Chinese migrant population ’, Xinhua News , available online at http://www.news.cn/ (accessed 25 October 2018). Hu B. Y. , Szente J. ( 2010 ) ‘ Education of young Chinese migrant children: Challenges and prospects ’, Early Childhood Education Journal , 37 ( 6 ), pp. 477 – 82 . Google Scholar Crossref Search ADS WorldCat Hwang S. , Xi J. , Cao Y. , Feng X. , Qiao X. ( 2007 ) ‘ Anticipation of migration and psychological stress and the Three Gorges Dam project, China ’, Social Science & Medicine (1982) , 65 ( 5 ), pp. 1012 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat Javali S. B. , Gudaganavar N. V. , Raj S. M. ( 2011 ) ‘ Effect of varying sample size in estimation of coefficients of internal consistency ’, WebmedCentral Biostatistics , 2 , pp. WMC001649 . WorldCat Ji Y. , Zhang Y. , Wang Y. , Chang C. ( 2016 ) ‘ Association between family factors and children’s oral health behaviors: A cross‐sectional comparative study of permanent resident and migrant children in large cities in China ’, Community Dentistry and Oral Epidemiology , 44 ( 1 ), pp. 92 – 100 . Google Scholar Crossref Search ADS PubMed WorldCat Judd C. M. , Kenny D. A. ( 1981 ) ‘ Process analysis: Estimating mediation in evaluation research ’, Evaluation Review , 5 ( 5 ), pp. 602 – 19 . Google Scholar Crossref Search ADS WorldCat Lacobucci D. , Duhachek A. ( 2003 ) ‘ Advancing alpha: Measuring reliability with confidence ’, Journal of Consumer Psychology , 13 ( 4 ), pp. 478 – 87 . Google Scholar Crossref Search ADS WorldCat Lau M. , Li W. ( 2011 ) ‘ The extent of family and school social capital promoting positive subjective well-being among primary school children in Shenzhen, China ’, Children and Youth Services Review , 33 ( 9 ), pp. 1573 – 82 . Google Scholar Crossref Search ADS WorldCat Lai F. , Liu C. , Luo R. , Zhang L. , Ma X. , Bai Y. , Sharbono B. , Rozelle S. ( 2014 ) ‘ The education of China’s migrant children: The missing link in China’s education system ’, International Journal of Educational Development , 37 , pp. 68 – 77 . Google Scholar Crossref Search ADS WorldCat LeSage J. P. , Ha C. L. ( 2012 ) ‘ The impact of migration on social capital: Do migrants take their bowling balls with them? ’, Growth and Change , 43 ( 1 ), pp. 1 – 26 . Google Scholar Crossref Search ADS WorldCat Li G. ( 2016 ) ‘ Guangdong fully open college entrance examination for non-local students, no influence on admission rate of Guangdong local students [in Chinese] ’, Guangming Online , available online at http://news.gmw.cn/2016–05/25/content_20246774.htm/ (accessed 25 October 2018). Liu T. ( 2012 ) ‘ Prevalence and risk factors of child victimization in China ’, doctoral thesis, The University of Hong Kong, Hong Kong, China, available online at https://hub.hku.hk/handle/10722/174510 (accessed 24 October 2018). Lu M. , Zhang J. , Ma J. , Li B. , Quan H. ( 2008 ) ‘ Child health insurance coverage: A survey among temporary and permanent residents in Shanghai ’, BMC Health Serv Res , 8 , pp. 238 . Google Scholar Crossref Search ADS PubMed WorldCat Mallee H. , Pieke F. N. ( 2014 ) Internal and International Migration: Chinese Perspectives , New York, NY , Taylor & Francis . Google Preview WorldCat COPAC McPherson K. E. , Kerr S. , McGee E. , Morgan A. , Cheater F. M. , McLean J. , Egan J. ( 2014 ) ‘ The association between social capital and mental health and behavioural problems in children and adolescents: An integrative systematic review ’, BMC Psychology , 2 , pp. 7 . Google Scholar Crossref Search ADS PubMed WorldCat Morrow V. ( 2004 ) ‘ Children’s social capital: Implications for health and well-being ’, Health Education , 104 ( 4 ), pp. 211 – 25 . Google Scholar Crossref Search ADS WorldCat Paiva P. C. P. , de Paiva H. N. , de Oliveira Filho P. M. , Lamounier J. A. , e Ferreira E. F. , Ferreira R. C. , Kawachi I. , Zarzar P. M. ( 2014 ) ‘ Development and validation of a social capital questionnaire for adolescent students (SCQ-AS) ’, PloS One , 9 ( 8 ), pp. e103785 . Google Scholar Crossref Search ADS PubMed WorldCat Qian H. , Walker A. ( 2015 ) ‘ The education of migrant children in Shanghai: The battle for equity ’, International Journal of Educational Development , 44 , pp. 74 – 81 . Google Scholar Crossref Search ADS WorldCat Rajmil L. , Herdman M. , Ravens-Sieberer U. , Erhart M. , Alonso J. ( 2014 ) ‘ Socioeconomic inequalities in mental health and health-related quality of life (HRQOL) in children and adolescents from 11 European countries ’, International Journal of Public Health , 59 ( 1 ), pp. 95 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Rocco L. , Suhrcke M. ( 2012 ) ‘ Is social capital good for health? A European perspective ’, WHO Regional Office for Europe Copenhagen , available online at http://www.euro.who.int/data/assets/pdffile/0005/170078/Is-Social-Capital-good-for-your-health.pdf? ua=1/ (accessed 26 October 2018). Rushton F. E. , Kraft C. ( 2014 ) ‘ Building brains, forging futures: The pediatrician’s role ’, International Journal of Pediatrics and Adolescent Medicine , 1 ( 1 ), pp. 3 – 7 . Google Scholar Crossref Search ADS WorldCat Shi X. , Wang Y. ( 2010 ) ‘ Migrant children’s loneliness and its influencing factors ’, Journal of Hunan Agricultural University (Social Sciences) , 11 ( 4 ), pp. 32 – 7 . WorldCat Smith-Greenaway E. , Madhavan S. ( 2015 ) ‘ Maternal migration and child health: An analysis of disruption and adaptation processes in Benin ’, Social Science Research , 54 , pp. 146 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat Straus M. A. , Douglas E. M. ( 2004 ) ‘ A short form of the Revised Conflict Tactics Scales, and typologies for severity and mutuality ’, Violence and Victims , 19 ( 5 ), pp. 507 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat Sun X. , Chen M. , Chan K. L. ( 2016 ) ‘ A meta-analysis of the impacts of internal migration on child health outcomes in China ’, BMC Public Health , 16 , pp. 66 . Google Scholar Crossref Search ADS PubMed WorldCat Tavakol M. , Dennick R. ( 2011 ) ‘ Making sense of Cronbach’s alpha ’, International Journal of Medical Education , 2 , pp. 53 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat To S. , Helwig C. C. , Yang S. ( 2016 ) ‘ Predictors of children's rights attitudes and psychological well-being among rural and urban mainland Chinese adolescents ’, Social Development , 26 ( 1 ), pp. 185 – 203 . Google Scholar Crossref Search ADS WorldCat Varni J. W. , Seid M. , Kurtin P. S. ( 2001 ) ‘ PedsQL™ 4.0: Reliability and validity of the Pediatric Quality of Life Inventory™ Version 4.0 Generic Core Scales in healthy and patient populations ’, Medical Care , 39 ( 8 ), pp. 800 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat WHO. ( 1997 ) WHOQOL: Measuring Quality of Life , Geneva , World Health Organization , p. 1 . WorldCat COPAC Wong D. F. K. , Chang Y. , He X. , Wu Q. ( 2010 ) ‘ The protective functions of relationships, social support and self-esteem in the life satisfaction of children of migrant workers in Shanghai, China ’, Int J Soc Psychiatry , 56 ( 2 ), pp. 143 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Wu Q. , Palinkas L. A. , He X. ( 2010 ) ‘ An ecological examination of social capital effects on the academic achievement of Chinese migrant children ’, British Journal of Social Work , 40 ( 8 ), pp. 2578 – 97 . Google Scholar Crossref Search ADS WorldCat Wu Q. , Tsang B. , Ming H. ( 2014 ) ‘ Social capital, family support, resilience and educational outcomes of Chinese migrant children ’, British Journal of Social Work , 44 ( 3 ), pp. 636 – 56 . Google Scholar Crossref Search ADS WorldCat Wu X. , Zhang Z. ( 2015 ) ‘ Population migration and children’s school enrollments in China, 1990–2005 ’, Social Science Research , 53 , pp. 177 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat Xiong Y. ( 2015 ) ‘ The broken ladder: Why education provides no upward mobility for migrant children in China ’, The China Quarterly , 221 , pp. 161 – 84 . Google Scholar Crossref Search ADS WorldCat Yang X. , Xiao N. , Yan J. ( 2011 ) ‘ The PedsQL in pediatric cerebral palsy: reliability and validity of the Chinese version pediatric quality of life inventory 4.0 generic core scales and 3.0 cerebral palsy module ’, Quality of Life Research , 20 ( 2 ), pp. 243 – 52 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2018. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The British Journal of Social Work Oxford University Press

Quality of Life of Rural–Urban Migrant Children in China: A Cross-Sectional Study

Download PDF
20 pages

Loading next page...
 
/lp/oxford-university-press/quality-of-life-of-rural-urban-migrant-children-in-china-a-cross-MYEK7aSuOp

References (49)

Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved.
ISSN
0045-3102
eISSN
1468-263X
DOI
10.1093/bjsw/bcy095
Publisher site
See Article on Publisher Site

Abstract

Abstract This study aims to examine the quality of life of Chinese migrant children and to explore whether social capital mediates the relationship between a child’s migration status and quality of life. A cross-sectional survey was conducted in a district of Shanghai, China. Based on a sample consists of 1,577 Chinese children in fourth to sixth grade, we used multivariable linear regression to examine the associations and mechanisms among migration, social capital and quality of life. Migrant children reported lower scores for quality of life than their urban counterparts. However, we identified higher levels of neighbourhood social cohesion, trust in their school and neighbourhood, and higher level of maternal autonomy support can mediate the damaging effects of migration status on quality of life. The results indicate that hukou restrictions put migrant children at a disadvantageous position and negatively affect their quality of life. While their migration status is unlikely to change soon, we found that higher levels of certain social capital can alleviate the negative effects of migration status on children’s quality of life, and thus can be utilised for welfare programme design. Quality of life, ecological systems theory, social capital, migrant children, China Introduction Contexts of internal migration in Mainland China Since the economic reform in the late 1970s, fluid labour movement, initially from rural to urban regions, has been one of the main drivers of China’s economic development. As defined by China’s household registration system (hukou), internal migrants are those who live somewhere other than their registered birthplace. By the end of 2013, the migrant population had reached 245 million, accounting for one-sixth of the national population (Hu, 2014). Among this number were 35.81 million children who migrated with their parents or were born after their parents’ arrival in cities (Duan et al., 2013). Many studies have paid attention to international migrants, as they are expected to confront difficulties due to differences of race, geopolitical distance and political systems (Mallee and Pieke, 2014). However, movement within a country’s borders also can be difficult. The government viewed migrants as temporary residents in urban areas and rejected their access to public services in host cities. Migration experience is a risk factor for child development because migrant children encounter an inter-generational acculturation gap (Chan, 2015), lack of educational opportunities (Xiong, 2015) and maladjustment in urban environments (Cheng and Selden, 1994). Children without a local hukou cannot access publicly funded schools, so most of them attend lower-quality migrant schools. Some local governments have made efforts to improve migrant children’s schooling. Take Shanghai, for instance, which allocated a certain amount of money to children in migrant schools and implemented a certain quota for the enrollment of migrant children in public schools (Qian and Walker, 2015). Even with local initiatives that attempt to ameliorate inequality in educational resources between migrant and urban children, the best schools with high-quality resources and teachers are still inaccessible to migrant children (Lai et al., 2014). Thus, this study selected migrant schools of different quality and those public schools open to migrant children to ensure a variety of schools and students in the sample. In addition, migrant families are not eligible to buy or rent subsidised apartments, so they are usually concentrated in the outskirts of cities. Minhang district of Shanghai was chosen as the survey site in view of its location at the southern edge of the city and its large proportion of migrant population. In addition to poor schooling experience and living conditions, scholars have found migrant children were less likely to utilise public health services compared to local urban children (Sun et al., 2016). These disadvantages that are attached to migration status have negative impacts on the lives of migrant children. Migration as a challenge to quality of life Migration is often viewed as a stressful event, as it breaks the equilibrium between the person and the environment, especially when it involves tremendous social costs but with uncertain economic benefits (Hwang et al., 2007). Most migrants inevitably experience life changes and readjustment challenges during their relocation. Children are often victims of such changes over which they have limited power; therefore, any household change can influence the resources available for migrant children. Hwang et al. (2007) argued that involuntary migrants were prone to stress when they had little control over some strong external forces. In addition, migration can not only directly, but also indirectly, elevate health problems by changing the intangible resources (i.e. social capital) embedded in social relations. Although migration does not have definitely negative effects on child quality of life, it depends on what changes occur to them and how much control migrant children have. Quality of life is a person’s subjective evaluation about the cultural, social and environmental context (WHO, 1997). Quality of life examines the difficulties of children in physical, emotional, social and schooling functioning in daily lives, which can identify the specific health care needs of children. It is not hard to imagine children are vulnerable to many threats caused by migration. For instance, internal migrant children within China may face difficulties resulting from family separation, inadequate parenting and restricted access to public benefits (Cheng and Selden, 1994). In addition, scholars have investigated some outcomes of migrant children, which usually focused on limited dimensions of quality of life. Educationists focused on schooling and academic performances of migrant children (Hu and Szente, 2010; Gong et al., 2015), psychologists addressed their mental health outcomes (Wong et al., 2010; Shi and Wang, 2010) and paediatricians highlighted the physical health outcomes of migrant children (Lu et al., 2008; Ji et al., 2016). However, few studies have attempted to simultaneously investigate the four aspects of quality of life among migrant children. In this study, we apply ecological systems theory and social capital theory to explore the association among migration status, social capital and children’s quality of life. Findings from this study have the potential to illuminate intervention strategies and shed light on policy making. Conceptual frameworks Ecological systems theory As Finkelhor and Hashima (2001) argued, individual disadvantage not only arises from his or her physical and mental weakness, but also depend on environmental factors (Liu, 2012). Ecological systems theory is useful to identify a series of ecological changes that happen to migrant children and link them to the quality of life of migrant children. Specifically, family, school, peers and neighbourhood have the most immediate influence on children’s lives (Bronfenbrenner, 1994). Given the living conditions and community environment in which migrant children spend their lives, it is likely that these environmental factors pose several risks to migrant children’s quality of life. Migrant children often experience a series of ecological changes that do not take place in local urban children. For instance, migration is related to the changes of family structure, social status and family wealth in the microsystem, as many migrant children have to face difficulties of family separation and inadequate parenting (Smith-Greenaway and Madhavan, 2015; Wu and Zhang, 2015). Meanwhile, the single-parent family is often associated with deprivation of economic resources of children, which further impacts the physical, emotional and behaviour health of children (Conger and Donnellan, 2007). In addition, changes also occur at the mesosystem level, as the residential mobility may negatively impact the relationships between migrant children and their parents, peers and neighbourhoods. The ecological systems theory provides a framework that depicts the important social contexts of migrant children. It is difficult to consider all relations and influences on the numerous contexts. Nevertheless, this study focuses its interest on the individual, family, school, peer and neighbourhood levels. All of the changes related to migration are likely to influence the quality of life of individuals. Social capital theory Although the ecological systems theory highlights the potential negative impacts of migration on migrant children’s quality of life, individuals are not passive receivers of their environmental influences. According to a meta-analytical article of the health outcomes of migrant children, migrant children from migrant schools presented sickness or deficit compared to their local urban counterparts, while migrant children from public school did not (Sun et al., 2016). Therefore, we need to increase the focus on the resiliency of migrant children, which can positively respond to the harsh ecological changes. This study takes a strength-based approach, which can not only identify the strengths of migrant children, but also suggest community-based practices. Social capital has been given special attention because it argues that individuals can actively resort to the surrounding environment to achieve certain goals. The social capital within a family indicates that children can depend on their parent’s financial and human capital to achieve certain goals, such as the presence of parents and attention paid to children by their parents (Coleman, 1988). In addition, social capital in the community indicates that children’s development is also influenced by their families’ relationship with other significant adults in the community (Coleman, 1988). Studies have identified social capital as an important factor for child health and quality of life (Drukker et al., 2003; Morrow, 2004) and many countries have incorporated this idea into child and adolescent health-promotion programmes (Rocco and Suhrcke, 2012; Rushton and Kraft, 2014). According to a systematic review conducted by McPherson et al. (2014), research has proved the link between positive relations with parents, peers, teachers and neighbours and better health outcomes, subjective well-being and life satisfaction (Lau and Li, 2011). The social capital of Chinese migrant children is rarely studied. Some studies have employed social support or parental social capital as a proxy due to a lack of measures for migrant children’s social capital (Wu et al., 2010, 2014), but they cannot capture the core of social capital, which is mutually beneficial resources for certain goals. Other studies have failed to capture all of children’s important relationships (Lacobucci and Duhachek, 2003; Lau and Li, 2011). Social capital deserves more attention and better measures because migrant children’s perceived social capital relates to their health and well-being. In addition, it is uncertain whether migrant children and their families can build beneficial social relations to facilitate good life quality. Coleman (1988) reasoned that social capital is generated through a relational closure (i.e. a family or a community). A change of residence can break this closure and reduce a family’s level of social capital (LeSage and Ha, 2012). However, Lau and Li’s study found that children in Shenzhen without local hukou had close teacher–student relationships, and that their parents had close bonds with their school, which positively linked to subjective well-being (2011). These inconsistent findings regarding migrant children and their families’ social capital is puzzling. Yet, the concept of social capital is an important one to study because they are amenable factors that can be utilised in designing programmes that promote the quality of life among migrant children. While previous studies have sought to identify the association between the disadvantages of migrant children and their negative health outcomes, little has yet examined the protective influence of social capital on quality of life among Chinese migrant children. The current study proposes a conceptual framework of combining the ecological systems theory and the social capital theory, which is shown in Figure 1. Despite the locating level in the ecological systems, these factors can be summarised in four clusters: (i) socio-demographic characteristics, (ii) tangible resources available for a child, (iii) social capital and (iv) quality of life of a child. Given the relationships among migration status, social capital and quality of life, it is reasonable to propose that migration status and social capital are key determinants of children’s quality of life. In addition, social capital is related to both migration status and children’s quality of life, so we speculate that social capital may mediate the relationship between migration status and child quality of life. Figure 1 View largeDownload slide Conceptual framework. Circles indicate different levels of ecological systems. Figure 1 View largeDownload slide Conceptual framework. Circles indicate different levels of ecological systems. Furthermore, this study tends to answer the following research questions: (i) Is the quality of life of migrant children lower than that of local urban children in urban China? (ii) Is the migration status and social capital a major determinant of the quality of life of children? (iii) Does social capital in migrant children mediate the relationship between migration status and child quality of life? Methods Study sample This study chose Minhang district of Shanghai, China, as the study site, which is the major neighbourhood of the migrant population. In order to include enough migrant child participants, we administered a school-based survey. The survey was conducted between October and November 2015. We contacted all principals of migrant schools in Minhang district and eight out of fourteen agreed to participate. In addition, four public schools open to migrant children agreed to participate. The schools are located in six towns of Minhang district and the quality of the migrant schools varies in terms of the conditions of facility, the composition of migrant/local students and the location of the neighbourhood. The quality and reputation of the four public schools are ordinary, which guarantees that the socio-economic backgrounds of the children do not differ excessively from the migrant schools. Our sample included all eligible children in twelve participating schools, who were enrolled in fourth to sixth grades. With the principals’ support and teachers’ facilitation, we had a response rate of 100 per cent: 1,322 migrant children (542 female and 781 male), 26.17 per cent of whom attended public schools. In addition, 255 local children at the same public schools were recruited as a comparison group (104 male and 151 female). The surveys were conducted in the classroom setting and instructions (briefing and consensus seeking) were given by a researcher. On average, the students took 35 minutes to complete the questionnaire. This study was approved by the Human Research Ethics Committee at the University of Hong Kong (EA1506011). Measures Main outcome variable The dependent variable was the quality of life of migrant children, measured by Pediatric Quality of Life Generic Core Scales for children aged eight to twelve (PedsQL 4.0 Child Self-Report) (Yang et al., 2011). PedQL has been widely used to examine child health-related outcomes through self-report (Varni et al., 2001). The PedsQL 4.0 Generic Core Scales have shown high construct validity and reliability (Cronhach’s alpha = 0.88) (Varni et al., 2001). The Chinese version also has good reliability (Cronbach’s alpha = 0.86) and construct validity (Yang et al., 2011). Cronbach’s alpha for this sample was 0.87. It consists of twenty-three items referring to four subscales of quality of life: physical functioning (eight items), emotional functioning (five items), social functioning (five items) and school functioning (five items). Items were scored on a five-point Likert-type scale, indicating the frequency of problematic functioning. The scores were reverse coded, ranging from 0 and 100, so that a higher score represents a better quality of life. Independent variables The key independent variables of this study are the migration status and social capital of the children. A child was defined as a migrant child if he/she had a rural and non-local hukou (migrant child = 1, local child = 0). The measurement of the children’s social capital consists of two parts: the Social Capital Questionnaire for Adolescent Students (SCQ-AS; Paiva et al., 2014) and parental autonomy support subscales of Perceptions of Parents Scales (child POPS; Grolnick et al., 1991). Social Capital Questionnaire is the first instrument to detect a child’s social capital with a three-point scale (ranging from 1 = disagree to 3 = agree). The twelve-item scales demonstrated satisfactory reliability (Cronbach’s alpha = 0.71) and construct validity in the original version (Paiva et al., 2014). We supplemented two more indicators of parental autonomy and control by using a twelve-item subscale of POPS. Participants rated parental autonomy support on twelve items (six for mothers and six for fathers) according to four descriptions of types of parent and then chose the most appropriate one for each item (scores ranged from 1 to 4). The sum of these six items indicates the degree of maternal/paternal autonomy support; higher scores indicate more autonomy support, ranging from 6 to 24. Both subscales in the original version have satisfactory reliability (Cronbach’s alpha = 0.53 and 0.67) and validity and are found to be linked to children’s mental health outcomes (Grolnick et al., 1991). Two authors translated the English version of the scales into Chinese and then a professional translator back-translated it. Based on the 1,577 participants in this study, the Cronbach’s alpha coefficients for the scales of school cohesion, school friendships, neighbourhood cohesion, trust, maternal autonomy support and paternal autonomy support were 0.44, 0.60, 0.83, 0.56, 0.46 and 0.58, respectively. We then calculated the sum of each subscale to indicate six dimensions of the social capital of a child. Other independent variables We included several socio-demographic covariates of child health outcomes: (i) child age, child gender (female = 1), child ethnicity (ethnic minority = 1); (ii) public-school attendance (attending public school = 1) and school transfer (one transfer or more = 1) related to schooling experience; (iii) socio-economic status (SES), family structure (single-parent family = 1, two parents live together = 1, the presence of siblings = 1); and (iv) tangible resources (homeownership = 1, availability of private room for a child = 1, participation in extra-curricular course = 1). Age, SES and the number of siblings is a continuous variable and the others were all coded as dummy variables. Migrant children are likely to experience multiple transfers due to the instability of their parents’ work. Residential instability is viewed as a risk factor for child health, so we controlled for the frequency of previous school transfers. SES often links to children’s access to resources and quality of life (Rajmil et al., 2014). Based on Straus and Douglas’s (2004) study, SES was computed as a sum of the educational level of the father and mother (scores ranged from 1 to 8 for both variables) and household income level (scores ranged from 1 to 12). The Cronbach’s alpha for the resulting scale was 0.64. In addition, housing conditions (homeownership and availability of private room for a child) can represent the consumption ability of a household to some extent, which is also related to the financial and material resources. It is also related to the physical environment of child activity and the child’s privacy. Extra-curricular course attendance reflects the resources that benefit a child’s social and emotional development out of the classroom. Data analysis The purpose of this study is to explore the determinants and potential mediators of children’s quality of life. First, we summarised the dependent variable, independent variables and control variables by the children’s migration status and computed the group difference between migrant and local children with t-test (for continuous variables) and chi-square test (for categorical variables). Second, we employed multivariate regression to investigate the extent to which migration status is related to children’s quality of life. Finally, we employed the Baron-Kenny approach and bootstrapping to identify and estimate the indirect effect of social capital factors. It is noteworthy that the Sobel test is a conservative way with low power to test the indirect effect; however, bootstrapping is highly recommended as an increasingly popular method (Creedon and Hayes, 2015). We conducted non-parametric resampled residual bootstrapping of mediation with 2,000 replicates to investigate the mediation effects of six factors of social capital respectively on the association of migration status with quality of life. All analyses were run in Stata 12. Results Descriptive analysis Table 1 demonstrates the mean, standard deviation and the analysis results of the group variance. Significant differences were found between migrant children (82.40) and local urban children (84.03) in reported scores of quality of life. Regarding the socio-demographic backgrounds, the two groups had lots in common, including the proportion of minority group, two parents live together and single-parent family. The average age of migrant children participants was 11.11, while that of local urban children was 10.93. It is noteworthy that the migrant children showed no difference from local children in reported school cohesion and school friendships, while they reported higher levels of neighbourhood cohesion and trust in neighbourhood and school. The three indicators of tangible resources showed that migrant children lived in poorer conditions and were less likely to participate in extra-curricular activities/courses compared to local urban children. Table 1 Distribution of sample characteristics Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 *p < 0.05; **p < 0.01; ***p < 0.001. View Large Table 1 Distribution of sample characteristics Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 Migrant children Local urban children Difference- test/chi2 test Mean (SD) Mean (SD) Dependent variable 82.40 (10.75) 84.03 (10.36) * Independent variable  Social capital   (1) School cohesion 11.11 (1.18) 11.22 (1.06)   (2) School friendship 8.54 (1.03) 8.65 (0.83)   (3) Neighbourhood social cohesion 5.14 (1.26) 4.89 (1.32) *   (4) Trust: neighbourhood/school 7.96 (1.28) 7.61 (1.22) *   (5) Paternal autonomy support 16.11 (3.25) 16.54 (2.75) *   (6) Maternal autonomy support 15.92 (3.17) 16.44 (2.54) * Control variables  Socio-demographics:   Age 11.11 (0.78) 10.93 (0.91) **   Female, % 542 (40.97%) 151 (59.22%) ***   Ethnic minority, % 38 (2.87%) 7 (2.75%)   Public school attendance, % 346 (26.17%) 255 (100%) ***   Transfer frequency, %: ***    Never transfer 996 (75.28%) 215 (84.32%)   One transfer 263 (19.73%) 10 (3.92%)    Several transfer 66 (4.99) 30 (11.76%)   Family structure, %:    Single parent 34 (2.57%) 5 (1.96%)    Two parents live together 1,207 (91.30%) 239 (93.73%)    Presence of sibling 519 (39.41%) 18 (7.09%) ***    Number of siblings per household 0.64 (0.69) 0.09 (0.28) ***   SES 14.69 (3.87) 20.70 (4.38) *  Tangible resources, %:   Homeownership 187 (14.22%) 202 (79.22%) ***   Private room for a child 543 (41.07%) 200 (78.43%) ***   Participation in extra-curricular course 569 (43.04%) 174 (68.24%) *** N 1,322 255 *p < 0.05; **p < 0.01; ***p < 0.001. View Large Association of quality of life with migration status and social capital Table 2 shows the association of quality of life with migration status based on five models, adjusting for different variables in four blocks. The variance inflation factor (VIF) test showed no indication of multicollinearity in the present data. In Model 1, the coefficient for migration status was statistically significant. Migrant children’s quality-of-life score was 1.63 lower than that of their local counterparts. The influence of migration status was not constant when including three individual-level demographics in Model 2. In Model 3, the coefficient for SES, single-parent family and the number of siblings was statistically significant. One unit increase in SES is associated with a 0.14 increase in quality of life, holding all else constant. One more sibling at a household is associated with a 1.12 lower score in quality of life of a child, holding all else constant. Model 4 includes three more variables of tangible resources, and only the coefficient for the availability of a private room was statistically significant. Children with a private room at home on average reported a 1.44 higher score of quality of life than those without one. Next, when including social capital of the child in Model 5, the effects of number of siblings and private room for a child were both overridden. However, all of the six social capital variables showed a positively significant effect on the quality of life of children. Table 2 Association of migration status with quality of life of child, adjusting for socio-demographic variables, tangible resources and social capital variable blocks Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 *p < 0.05; **p < 0.01; ***p < 0.001. aReference group: local urban children; breference group: female child; creference group: Han Chinese; dchild from migrant school; ereference group: child has no transfer experience; freference group: two-parent family; greference group: parents live apart; hreference group: child has no sibling; ireference group: rent/dorm/other; jreference group: has no private room for a child; kreference group: no participation in extra-curricular activities/courses. View Large Table 2 Association of migration status with quality of life of child, adjusting for socio-demographic variables, tangible resources and social capital variable blocks Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 Model 1: Migration status Model 2: Individual socio-demographics Model 3: Household and schoolsocio-demographics Model 4: Tangible resources Model 5: Social capital b (95% CI) b (95% CI) b (95% CI) b (95% CI) b (95% CI) Socio-demographics  Migration statusa –1.63* (–3.06, –0.20) –1.43 (–2.88, 0.20) 0.91 (–0.98, 2.79) 0.83 (–1.13, 2.79) –0.46 (–2.29, 1.36)  Age –0.49 (–1.15, 0.17) –0.09 (–0.76, 0.58) –0.15 (–0.83, 0.52) –0.33 (–0.96, 0.30)  Genderb 0.54 (–0.52, 1.61) 0.74 (–0.35, 1.82) 0.68 (–0.40, 1.76) –0.62 (1.64, 0.41)  Ethnicityc –3.16* (–6.32, –0.01) –2.61 (–5.77, 0.55) –2.59 (–5.75, 0.57) –2.11 (–5.04, 0.82)  SES 0.14* (0.00, 0.28) 0.08 (–0.06, 0.23) –0.01 (–0.15, 0.13)  Type of schoold 1.03 (–0.30, 2.36) 0.58 (–0.83, 2.00) 0.25 (–1.07, 1.58)  Frequency of school transfere –0.95* (–1.86, –0.03) –0.83 (–1.76, 0.09) –0.35 (–1.21, 0.51)  Single-parent familyf –6.61*** (–10.18, –3.04) –6.55*** (–10.12, –2.98) –5.02** (–8.34, –1.70)  Parents live togetherg 0.66 (–1.33, 2.65) 0.72 (–1.27, 2.71) 0.32 (–1.53, 2.18)  Presence of siblingh –0.07 (–1.50, 1.36) –0.04 (–1.46, 1.39) 0.14 (–1.18, 1.47)  Number of siblings –1.12* (–2.14, –0.10) –1.18* (–2.20, –0.17) –0.90 (–1.85, 0.04) Tangible resources  Homeownershipi –0.31 (–1.98, 1.37) –0.15 (–1.70, 1.40)  Private room for a childj 1.44* (0.22, 2.65) 1.00 (–0.14, 2.14)  Participate in extra-curricular activities/coursesk 0.89 (–0.22, 2.00) 0.54 (–0.49, 1.57) Social capital  School cohesion 0.99*** (0.52, 1.46)  Friendship 1.34*** (0.79, 1.89)  Neighbourhood social cohesion 0.57** (0.16, 0.99)  Trust: school and neighbourhood 1.04*** (0.61, 1.47)  Maternal autonomy support 0.45*** (0.27, 0.62)  Paternal autonomy support 0.26** (0.09, 0.43) Model summary Number of observations 1,577 1,577 1,577 1,577 1,577 Model F 4.56 2.97 4.31 4.01 16.18 Df 1 4 11 14 20 ΔR2 0.0046 0.0222 0.0055 0.1387 R2 0.0029 0.0075 0.0297 0.0352 0.1739 *p < 0.05; **p < 0.01; ***p < 0.001. aReference group: local urban children; breference group: female child; creference group: Han Chinese; dchild from migrant school; ereference group: child has no transfer experience; freference group: two-parent family; greference group: parents live apart; hreference group: child has no sibling; ireference group: rent/dorm/other; jreference group: has no private room for a child; kreference group: no participation in extra-curricular activities/courses. View Large Exploring the mediation effect of social capital According to the suggestions of Judd and Kenny (1981) for identifying a mediator, we estimated the following three regression equations: (i) regressing the social capital variables on the quality of life, (ii) regressing the quality of life on the migration status and (iii) regressing the quality of life on both the migration status and the social capital. Then, we found that the six social capital variables were all positively associated with the quality of life at a 95 per cent confidence level. Migration status was negatively associated with the quality of life (β = –1.63, p < 0.05). However, the effect of migration status on the quality of life was not significant when social capital variables were controlled. Therefore, social capital functions as a mediator, as it meets the above three conditions. In line with our previous hypotheses, we investigated whether migration status has any significant indirect effects on the quality of life via social capital with bootstrapping analysis. According to Table 3, neighbourhood cohesion and trust partially mediate the effect of migration status on quality of life, while maternal autonomy support can fully mediate it. Other dimensions of social capital do not have a mediation effect (Figures 2–7). Figure 2 View largeDownload slide Model of migration status effects on the quality of life with school cohesion as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 2 View largeDownload slide Model of migration status effects on the quality of life with school cohesion as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 3 View largeDownload slide Model of migration status effects on the quality of life with school friendship as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 3 View largeDownload slide Model of migration status effects on the quality of life with school friendship as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 4 View largeDownload slide Model of migration status effects on the quality of life with neighbourhood cohesion as a mediator . *p < 0.05; **p < 0.01; ***p < 0.001. Figure 4 View largeDownload slide Model of migration status effects on the quality of life with neighbourhood cohesion as a mediator . *p < 0.05; **p < 0.01; ***p < 0.001. Figure 5 View largeDownload slide Model of migration status effects on the quality of life with trust (in school and neighbourhood) as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 5 View largeDownload slide Model of migration status effects on the quality of life with trust (in school and neighbourhood) as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 6 View largeDownload slide Model of migration status effects on the quality of life with maternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 6 View largeDownload slide Model of migration status effects on the quality of life with maternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 7 View largeDownload slide Model of migration status effects on the quality of life with paternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Figure 7 View largeDownload slide Model of migration status effects on the quality of life with paternal autonomy support as a mediator. *p < 0.05; **p < 0.01; ***p < 0.001. Table 3 Standardised direct, indirect and total effects of different domains of social capital on the association between migration status and quality of life Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Control variables: ethnic minority, frequency of school transfer and single-parent family. *p < 0.05; **p < 0.01; ***p < 0.001. View Large Table 3 Standardised direct, indirect and total effects of different domains of social capital on the association between migration status and quality of life Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Direct effect Indirect effect Total effect School cohesion –1.30 –0.25 –1.55 School friendship –1.27 –0.28 –1.55 Neighbourhood cohesion –1.96** 0.41* –1.55 Trust (school/neighbourhood) –2.29** 0.73** –1.55 Maternal autonomy support –1.14 –0.41* –1.55 Paternal autonomy support –1.28 –0.27 –1.55* Control variables: ethnic minority, frequency of school transfer and single-parent family. *p < 0.05; **p < 0.01; ***p < 0.001. View Large Discussion In general, migrant children reported lower quality-of-life scores, were more likely to have school transfer, were less likely to live in a higher SES family and were less likely to report higher levels of maternal autonomy support relative to children with local hukou. We found that frequent school transfer, low SES, single-parent family structure and lack of social capital are all negatively associated with quality of life. This indicates that migration status lays the foundations for many risk factors for quality of life. Direct hukou reform would eliminate many of the systematic barriers that migrant children face, such as authorising migrants’ free access to justice, education and health care. The future of hukou reform is gloomy. Take Guangdong’s policy, for instance; it allows eligible migrant children to take the college entrance exam (gaokao), but only 9,570 applicants were approved (Li, 2016). To some extent, this policy provides an opportunity to continue in education, but a large portion of migrant children have discontinued their education before high school. It is uncertain how much migrant children can benefit from this reform. The results showed that the coefficient for migration status was not significant while adding other variables, meaning that migration status does not have a determinate effect on quality of life. In this instance, the current study continues the probe into social approaches to improving the quality of life of migrant children. The results have several implications for research and practice. This study allowed young adolescents to report social capital rather than using proxies, which adds value to its exploration of adolescent-perceived social capital in broad contexts and verification of social capital’s relationship to migration status and quality of life for Chinese migrant children. Social capital is not only directly related to children’s quality of life, but also clarifies certain circumstances in which migration status does not have a definite connection with quality of life. Critically, we found that migrant children were more likely to report higher levels of neighbourhood cohesion and trust of school and neighbourhood than their local counterparts, all of which improve their quality-of-life score. It is not surprising that migrant children have higher levels of social capital in these two domains in this study. First, we observed that most migrants from the same hometown are likely to live in the same community and their children may go to the same schools nearby, so it is easy for them to build contacts. Second, migrant children’s higher levels of certain social capital can also be attributed to the Shanghai government’s investment to improve the schooling of migrant children. School can play a determining role in constructing trustful and cohesive relationships for migrant children and families. For instance, Scotland’s schools employed a social capital programme to prevent children who lacked opportunities being socially excluded in fierce competition (Flint, 2011 ). On the one hand, the programme promoted a ‘shared identity’ and ‘shared goal’ for all stakeholders. On the other hand, it helped children at risk to know about their environment and turn to the right people for help when there was a need. The negative effect of a single-parent family is constantly significant. It is not surprising because children living with a single parent receive less attention compared to those living with both parents. Therefore, programmes for improving child quality of life should pay special attention to children living with a single parent. However, lack of maternal autonomy support may lead to greater disadvantages for migrant children. According to self-determination theory, children can benefit from parents’ autonomy support (To et al., 2016). Many countries have recognised the importance of parenting in child development and applied early interventions, such as the Head Start programme in the USA. Traditional parenting skills in China are rather different and are more authoritative. It emphasises on disciplining children using appropriate punishments, but urban families (with urban hukou) are more likely than rural families to appreciate autonomy support for adolescents (Chen and Li, 2012). Previous studies have often found lower support for autonomy to be related to lower SES, meaning that migrant children are more vulnerable to low autonomy support (To et al., 2016). Therefore, programmes should target vulnerable children and provide more family-based interventions, such as offering childcare and teaching parents positive interaction and communication skills, which may buffer higher quality of life. Findings of this study should be interpreted with the following caveats. First, due to the limited data collection, we only employed a purposeful sampling method in one district of a city with participants of a narrow age group. Samples generated from this sampling strategy cannot represent the population of migrant children as a whole. Therefore, strict design in future studies is needed to demonstrate more dynamics with a representative sample. Second, the cross-sectional nature of this study made it not possible to conclude on any causative relationship. Future studies should consider using a cohort study design, which could allow collection of longitudinal data by tracking the lives of migrant children to address the dynamic changes along with the ongoing systematic reform; this could update our knowledge about this group and also provide substantial evidence for policy making. Third, some subscales of the instrument of social capital used in this study did not have high internal consistency, which can be attributed to the different characteristics of the sample. The sample of this study (n = 1,577) was much bigger than that in the original validation study (n = 101). According to previous studies, sample size has an adverse relationship with the Cronbach’s alpha coefficient (Lacobucci and Duhachek, 2003; Javali et al., 2011). Other than that, the number of items also influences the calculation of internal consistency. Most indicators were only measured by three or four items, which may cause an underestimation of reliability (Tavakol and Dennick, 2011). Therefore, we need to be cautious about Cronbach’s alpha values in order to avoid mistakes in our research. However, authors have consensus about the appropriate content and the length of the scale for measuring adolescents’ social capital. In addition, findings of this study have reference value for future studies. First, the difficulties and risks facing Chinese internal migrant children have a great deal in common with those in Europe and South Asia. For example, these migrant children often suffer from low SES, instability of family structure and inappropriate support from parents (Sun et al., 2016). Thus, the conceptual framework can be employed in future studies in other countries. Second, it suggests that the disadvantaged children can overcome difficulties by resorting to social capital and improve their quality of life. Therefore, social capital could be a protective factor for other at-risk children as well. Examinations of social capital in other at-risk children will advance the application of social capital in child studies. Conclusion This study is one of the few that focuses on the quality of life among migrant children in urban China and the associated risk and protective factors to quality of life. We find that, on average, migrant children in urban settings have lower quality of life. In addition, our findings suggest that migrant children can resort to social capital to mitigate the negative impact of migration status on their quality of life. Our findings have practical implications to policy makers who are concerned about resource distribution and the well-being of migrant children and their families in urban settings. This study provides evidence to suggest an alternative intervention approach for practitioners and a new direction to policy making on caring for migrant children in urban China. Acknowledgements We thank Bin Fan, Ph.D., for co-ordinating participants in Shanghai. Special thanks go to Yang Liu and Xiaonuan Sun, who facilitated the data collection. The project was granted ethical approval by the Human Research Ethics Committee of The University of Hong Kong in June 2015 (reference number: EA1506011). We have obtained participants’ consent to use all the figures for publication, online and in print, and in perpetuity. Funding There is no funding source. Conflict of interest statement. None declared. References Bronfenbrenner U. ( 1994 ) ‘ Ecological models of human development ’, Readings on the Development of Children , 2 ( 1 ), pp. 37 – 43 . WorldCat Chan K. L. ( 2015 ) ‘ Migration in China: Changing China, changing the world ’, in Chan K. L. (ed.), Chinese Migration and Families-At-Risks , Newcastle upon Tyne, Scholars Publishing Ltd . Google Preview WorldCat COPAC Chen X. , Li D. ( 2012 ) ‘ Parental encouragement of initiative-taking and adjustment in Chinese children from rural, urban, and urbanized families ’, Journal of Family Psychology , 26 ( 6 ), pp. 927 – 36 . Google Scholar Crossref Search ADS PubMed WorldCat Cheng T. , Selden M. ( 1994 ) ‘ The origins and social consequences of China’s hukou system ’, The China Quarterly , 139 , pp. 644 – 68 . Google Scholar Crossref Search ADS WorldCat Coleman J. S. ( 1988 ) ‘ Social capital in the creation of human capital ’, American Journal of Sociology , 94 , pp. S95 – 120 . Google Scholar Crossref Search ADS WorldCat Conger R. D. , Donnellan M. B. ( 2007 ) ‘ An interactionist perspective on the socioeconomic context of human development ’, Annual Review of Psychology , 58 ( 1 ), pp. 175 – 99 . Google Scholar Crossref Search ADS PubMed WorldCat Creedon P. S. , Hayes A. F. ( 2015 ) ‘ Small sample mediation analysis: How far can we push the Bootstrap ’, in Annual Conference of the Association for Psychological Science , New York, NY . Google Preview WorldCat COPAC Drukker M. , Kaplan C. , Feron F. , Van Os J. ( 2003 ) ‘ Children’s health-related quality of life, neighbourhood socio-economic deprivation and social capital: A contextual analysis ’, Social Science and Medicine , 57 ( 5 ), pp. 825 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat Duan C. , Lv L. , Wang Z. , Guo J. ( 2013 ) ‘ The survival and development status of floating children in China: An analysis of the Sixth Population Census data ’, South China Population , 28 , pp. 44 – 55 . WorldCat Finkelhor D. , Hashima P. Y. ( 2001 ) ‘ The victimization of children and youth: A comprehensive Overview ’, in White S. O. (ed.), Handbook of Youth and Justice , New York, NY , Plenum Publishers . Google Preview WorldCat COPAC Flint N. ( 2011 ) Schools, Communities and Social Capital: Building Blocks in the ‘Big Society’ (Research Associate Full Report). National College for Leadership of Schools and Children’s Services (NCSL), available online at http://dera.ioe.ac.uk/2981/ (accessed 17 September 2014). Gong X. , Zhang H. , Yao H. ( 2015 ) ‘ The determinants of compulsory education performance of migrant children in Beijing: An analysis of two cohorts ’, International Journal of Educational Development , 45 , pp. 1 – 15 . Google Scholar Crossref Search ADS WorldCat Grolnick W. S. , Ryan R. M. , Deci E. L. ( 1991 ) ‘ Inner resources for school achievement: Motivational mediators of children’s perceptions of their parents ’, Journal of Educational Psychology , 83 ( 4 ), pp. 508 – 17 . Google Scholar Crossref Search ADS WorldCat Hu H. ( 2014 ) ‘ Report on the development of Chinese migrant population ’, Xinhua News , available online at http://www.news.cn/ (accessed 25 October 2018). Hu B. Y. , Szente J. ( 2010 ) ‘ Education of young Chinese migrant children: Challenges and prospects ’, Early Childhood Education Journal , 37 ( 6 ), pp. 477 – 82 . Google Scholar Crossref Search ADS WorldCat Hwang S. , Xi J. , Cao Y. , Feng X. , Qiao X. ( 2007 ) ‘ Anticipation of migration and psychological stress and the Three Gorges Dam project, China ’, Social Science & Medicine (1982) , 65 ( 5 ), pp. 1012 – 24 . Google Scholar Crossref Search ADS PubMed WorldCat Javali S. B. , Gudaganavar N. V. , Raj S. M. ( 2011 ) ‘ Effect of varying sample size in estimation of coefficients of internal consistency ’, WebmedCentral Biostatistics , 2 , pp. WMC001649 . WorldCat Ji Y. , Zhang Y. , Wang Y. , Chang C. ( 2016 ) ‘ Association between family factors and children’s oral health behaviors: A cross‐sectional comparative study of permanent resident and migrant children in large cities in China ’, Community Dentistry and Oral Epidemiology , 44 ( 1 ), pp. 92 – 100 . Google Scholar Crossref Search ADS PubMed WorldCat Judd C. M. , Kenny D. A. ( 1981 ) ‘ Process analysis: Estimating mediation in evaluation research ’, Evaluation Review , 5 ( 5 ), pp. 602 – 19 . Google Scholar Crossref Search ADS WorldCat Lacobucci D. , Duhachek A. ( 2003 ) ‘ Advancing alpha: Measuring reliability with confidence ’, Journal of Consumer Psychology , 13 ( 4 ), pp. 478 – 87 . Google Scholar Crossref Search ADS WorldCat Lau M. , Li W. ( 2011 ) ‘ The extent of family and school social capital promoting positive subjective well-being among primary school children in Shenzhen, China ’, Children and Youth Services Review , 33 ( 9 ), pp. 1573 – 82 . Google Scholar Crossref Search ADS WorldCat Lai F. , Liu C. , Luo R. , Zhang L. , Ma X. , Bai Y. , Sharbono B. , Rozelle S. ( 2014 ) ‘ The education of China’s migrant children: The missing link in China’s education system ’, International Journal of Educational Development , 37 , pp. 68 – 77 . Google Scholar Crossref Search ADS WorldCat LeSage J. P. , Ha C. L. ( 2012 ) ‘ The impact of migration on social capital: Do migrants take their bowling balls with them? ’, Growth and Change , 43 ( 1 ), pp. 1 – 26 . Google Scholar Crossref Search ADS WorldCat Li G. ( 2016 ) ‘ Guangdong fully open college entrance examination for non-local students, no influence on admission rate of Guangdong local students [in Chinese] ’, Guangming Online , available online at http://news.gmw.cn/2016–05/25/content_20246774.htm/ (accessed 25 October 2018). Liu T. ( 2012 ) ‘ Prevalence and risk factors of child victimization in China ’, doctoral thesis, The University of Hong Kong, Hong Kong, China, available online at https://hub.hku.hk/handle/10722/174510 (accessed 24 October 2018). Lu M. , Zhang J. , Ma J. , Li B. , Quan H. ( 2008 ) ‘ Child health insurance coverage: A survey among temporary and permanent residents in Shanghai ’, BMC Health Serv Res , 8 , pp. 238 . Google Scholar Crossref Search ADS PubMed WorldCat Mallee H. , Pieke F. N. ( 2014 ) Internal and International Migration: Chinese Perspectives , New York, NY , Taylor & Francis . Google Preview WorldCat COPAC McPherson K. E. , Kerr S. , McGee E. , Morgan A. , Cheater F. M. , McLean J. , Egan J. ( 2014 ) ‘ The association between social capital and mental health and behavioural problems in children and adolescents: An integrative systematic review ’, BMC Psychology , 2 , pp. 7 . Google Scholar Crossref Search ADS PubMed WorldCat Morrow V. ( 2004 ) ‘ Children’s social capital: Implications for health and well-being ’, Health Education , 104 ( 4 ), pp. 211 – 25 . Google Scholar Crossref Search ADS WorldCat Paiva P. C. P. , de Paiva H. N. , de Oliveira Filho P. M. , Lamounier J. A. , e Ferreira E. F. , Ferreira R. C. , Kawachi I. , Zarzar P. M. ( 2014 ) ‘ Development and validation of a social capital questionnaire for adolescent students (SCQ-AS) ’, PloS One , 9 ( 8 ), pp. e103785 . Google Scholar Crossref Search ADS PubMed WorldCat Qian H. , Walker A. ( 2015 ) ‘ The education of migrant children in Shanghai: The battle for equity ’, International Journal of Educational Development , 44 , pp. 74 – 81 . Google Scholar Crossref Search ADS WorldCat Rajmil L. , Herdman M. , Ravens-Sieberer U. , Erhart M. , Alonso J. ( 2014 ) ‘ Socioeconomic inequalities in mental health and health-related quality of life (HRQOL) in children and adolescents from 11 European countries ’, International Journal of Public Health , 59 ( 1 ), pp. 95 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Rocco L. , Suhrcke M. ( 2012 ) ‘ Is social capital good for health? A European perspective ’, WHO Regional Office for Europe Copenhagen , available online at http://www.euro.who.int/data/assets/pdffile/0005/170078/Is-Social-Capital-good-for-your-health.pdf? ua=1/ (accessed 26 October 2018). Rushton F. E. , Kraft C. ( 2014 ) ‘ Building brains, forging futures: The pediatrician’s role ’, International Journal of Pediatrics and Adolescent Medicine , 1 ( 1 ), pp. 3 – 7 . Google Scholar Crossref Search ADS WorldCat Shi X. , Wang Y. ( 2010 ) ‘ Migrant children’s loneliness and its influencing factors ’, Journal of Hunan Agricultural University (Social Sciences) , 11 ( 4 ), pp. 32 – 7 . WorldCat Smith-Greenaway E. , Madhavan S. ( 2015 ) ‘ Maternal migration and child health: An analysis of disruption and adaptation processes in Benin ’, Social Science Research , 54 , pp. 146 – 58 . Google Scholar Crossref Search ADS PubMed WorldCat Straus M. A. , Douglas E. M. ( 2004 ) ‘ A short form of the Revised Conflict Tactics Scales, and typologies for severity and mutuality ’, Violence and Victims , 19 ( 5 ), pp. 507 – 20 . Google Scholar Crossref Search ADS PubMed WorldCat Sun X. , Chen M. , Chan K. L. ( 2016 ) ‘ A meta-analysis of the impacts of internal migration on child health outcomes in China ’, BMC Public Health , 16 , pp. 66 . Google Scholar Crossref Search ADS PubMed WorldCat Tavakol M. , Dennick R. ( 2011 ) ‘ Making sense of Cronbach’s alpha ’, International Journal of Medical Education , 2 , pp. 53 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat To S. , Helwig C. C. , Yang S. ( 2016 ) ‘ Predictors of children's rights attitudes and psychological well-being among rural and urban mainland Chinese adolescents ’, Social Development , 26 ( 1 ), pp. 185 – 203 . Google Scholar Crossref Search ADS WorldCat Varni J. W. , Seid M. , Kurtin P. S. ( 2001 ) ‘ PedsQL™ 4.0: Reliability and validity of the Pediatric Quality of Life Inventory™ Version 4.0 Generic Core Scales in healthy and patient populations ’, Medical Care , 39 ( 8 ), pp. 800 – 12 . Google Scholar Crossref Search ADS PubMed WorldCat WHO. ( 1997 ) WHOQOL: Measuring Quality of Life , Geneva , World Health Organization , p. 1 . WorldCat COPAC Wong D. F. K. , Chang Y. , He X. , Wu Q. ( 2010 ) ‘ The protective functions of relationships, social support and self-esteem in the life satisfaction of children of migrant workers in Shanghai, China ’, Int J Soc Psychiatry , 56 ( 2 ), pp. 143 – 57 . Google Scholar Crossref Search ADS PubMed WorldCat Wu Q. , Palinkas L. A. , He X. ( 2010 ) ‘ An ecological examination of social capital effects on the academic achievement of Chinese migrant children ’, British Journal of Social Work , 40 ( 8 ), pp. 2578 – 97 . Google Scholar Crossref Search ADS WorldCat Wu Q. , Tsang B. , Ming H. ( 2014 ) ‘ Social capital, family support, resilience and educational outcomes of Chinese migrant children ’, British Journal of Social Work , 44 ( 3 ), pp. 636 – 56 . Google Scholar Crossref Search ADS WorldCat Wu X. , Zhang Z. ( 2015 ) ‘ Population migration and children’s school enrollments in China, 1990–2005 ’, Social Science Research , 53 , pp. 177 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat Xiong Y. ( 2015 ) ‘ The broken ladder: Why education provides no upward mobility for migrant children in China ’, The China Quarterly , 221 , pp. 161 – 84 . Google Scholar Crossref Search ADS WorldCat Yang X. , Xiao N. , Yan J. ( 2011 ) ‘ The PedsQL in pediatric cerebral palsy: reliability and validity of the Chinese version pediatric quality of life inventory 4.0 generic core scales and 3.0 cerebral palsy module ’, Quality of Life Research , 20 ( 2 ), pp. 243 – 52 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2018. Published by Oxford University Press on behalf of The British Association of Social Workers. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

The British Journal of Social Work – Oxford University Press

Published: Jul 1, 2019

There are no references for this article.