In China, migrant workers are neither traditional peasants nor purely urban residents; instead, they are a marginal group residing between the urban and rural areas. They make great contributions to the development of the country, but the household registration system claims there is an essential distinction between migrant workers and urban residents. For a long time, this has isolated migrant workers in terms of social relationships and psychological, cultural, and political participation. Thus, migrant workers have become a vulnerable group whose members easily become trapped in poverty.

Social support networks have been an important factor in the survival of the poor. Therefore, in order to improve their living conditions, migrant workers need help with using social support effectively and also with managing their limited resources. Social support is a resource that individuals exchange with groups or other individuals through social relationships (Wang, 2004). It can be divided into formal and informal social networks. The formal social networks contain both governmental and regional organizations. Informal social networks include relationships that are built through blood ties, as well as geopolitical and private relationships (Lan & He, 2004). Based on related research (Zhang & Xing, 2007), social support is divided into five categories in this study: life, work, welfare, children’s education, and government policy.

Family resource management is the process of using resources in order to attain family goals through planning and taking the steps necessary for achieving those goals (Deacon & Firebaugh, 1988). It can be seen from different angles, including creating, converting, and selecting resources, as well as determining the best use for each resource (Goldsmith, 2005). In this study, we classified family resources according to four categories: emotion, finance, information, and services. Only those who know how to use and manage family resources can gain the maximum benefits of those resources. Individuals’ economic lives are deeply embedded in their social networks, and they can gain information, influence, trust, and other social resources through their social support networks (Granovetter, 1995); thus, social support has an important effect on migrant workers’ family resource management.

Following the studies outlined above, three hypotheses were tested:
Hypothesis 1: Migrant workers with different types of social support cognition will have different levels of family resource management.
Hypothesis 2: Migrant workers’ social support cognition type will be influenced by demographic variables.
Hypothesis 3: Migrant workers’ family resource management will be directly affected by their social support cognition type and by demographic variables.

Method

Participants

A total of 900 migrant workers from Suzhou were selected as participants by purposive sampling. They were asked to complete the Social Support Questionnaire (SSQ) and the Family Resource Management Questionnaire (FRMQ). A total of 501 valid responses were collected (65.23%). The characteristics of the participants were: male (61.5%), monthly income below US$235 (46.9%), unmarried (70.3%), worked in the manufacturing and service sectors (32.3% and 36.7%, respectively), and lived with family members (48.3%).

Instruments

The SSQ and FRMQ were developed by the authors of this study, based on an initial interview and related literature (Deacon & Firebaugh, 1988; Zhang & Xing, 2007). Both questionnaires were scored on a 5-point Likert scale, ranging from 1 = strongly disagree to 5 = strongly agree. The initial SSQ contained 40 items, and the initial FRMQ contained 36 items. We deleted five items from the SSQ and one item from the FRMQ because their composite reliability (CR) and item-total correlations were not significant. The remaining 35 items in the SSQ and 35 items in the FRMQ were analyzed using minimum likelihood extraction methods with a varimax rotation.

In the SSQ, five factors with eigenvalues of over 1 were identified: daily life, social security, children’s education, government policy, and social desirability. Loadings of less than .40 were eliminated, leaving 22 items. The five factors combined accounted for 59.13% of the variance. In the FRMQ, four factors with eigenvalues of over 1 were identified: information, service management, finance management, and emotion management. Loadings of less than .50 were eliminated, leaving 20 items. The four factors combined accounted for 56.28% of the variance. The values of Cronbach’s α for the SSQ subscales ranged between .71 and .88, and for the total scale was .87. The values for FRMQ were .67 to .86, and .81 for all the scores. The goodness-of-fit measures (Table 1) showed that both questionnaires achieved ideal reliability and validity.

Results

Social Support Cognition Types

A nonhierarchical cluster approach was used to cluster the migrant workers’ social support cognition types. We divided the coefficient into three groups using k means methods and then examined them using analysis of variance (ANOVA) and Scheffé multiple evaluations. Finally, we adopted discriminate analysis in order to verify the effectiveness of the cluster analysis. The results showed that the accuracy of the discriminate analysis was up to 95.2% (Wilks’ lambda = .836). Thus, we divided the migrant workers into three groups based on the degree of their cognition of their social support: moderate cognition of social support (Type I), high cognition of social support (Type II), and low cognition of social support (Type III).

The Relationship Between Social Support Cognition Types and Family Resource Management

There were significant differences between social support cognition types and family resource management (Wilks’ lambda = .836, p < .001), therefore, Hypothesis 1 was fully supported. Multivariate analysis of variance (MANOVA) results showed an overall main effect of social support types for the family resources management (F(2, 501) = 8.00, p < .001). The findings also revealed the significant differences in social support types for service management (F(2, 501) = 23.37, p < .001), finance management (F(2, 501) = 13.97, p < .001), and emotion management (F(2, 501) = 5.32, p < .01). However, the main effect of social support types for information management was not significant (F(2, 501) = 1.68, p > .05). Moreover, the results regarding the distribution of social support types in family resource management showed that all types of migrant workers reported the highest level of perception of finance management and the lowest level of perception of service management.

The Relationship Between Migrant Workers’ Demographic Variables and Social Support Cognition Types

In order to test Hypothesis 2, a chi-square test was conducted to analyze the relationship between demographic variables and social support types. The results revealed that there were significant differences among gender (χ² = 8.98, p < .05), monthly income (χ² = 14.55, p < .01), marital status (χ² = 16.34, p < .01), housemates (χ² = 30.86, p < .001), and social support cognition types. There was no significant difference between the occupation and social support cognition types of migrant workers (χ² = 12.80, p > .05). Therefore, Hypothesis 2 was only partially supported.

In terms of social support cognition, both male and female workers were mainly Type II. Migrant workers with a monthly income below US$235 were mainly Type III, but those with a monthly income above US$235 were mainly Type II. Both unmarried and married migrant workers were mainly Type II, while divorced workers were mainly Type III. Migrant workers who lived with their immediate family or other relatives were mainly Type II, but those who lived with workmates were mainly Type III.

Path Analysis of Demographic Variables and Social Support Cognition Types to Family Resource Management

Pearson’s correlation analyses were conducted in this study. Demographic variables and social support cognition types were coded as dummy variables that were used to quantify these attributes with the value 1 when an attribute was present and 0 when it was not present. Family resource management was significantly related to migrant workers’ having a monthly income below US$235 (r = .16, p < .01), having manufacturing and service occupations (r = .19, p < .01), living with workmates (r = -.15, p < .01), and social support cognition Types I and II (r = -.11, p < .05; r = .18, p < .01).

On the basis of the results of correlational analyses, we hypothesized that migrant workers’ monthly income (below US$235), occupation (manufacturing and service), housemates (workmates), and social support cognition Types I and II could significantly influence their family resource management. Using maximum likelihood estimates to fit structure models, the final model fits the observed data well: χ² = 22.11, χ²/df = 2.46 (p = .01), GFI = .99, CFI = .97, NFI = .96, RMSEA = .05. The signs associated with all paths were in the expected direction and all paths were significant, at p < .05. Therefore, Hypothesis 3 was supported.

These effects for all independent and dependent variables are summarized in Figure 1 in the final model. Specifically, migrant workers’ monthly income (below US$235), occupation (manufacturing), and housemates (workmates) had a positive influence on family resource management (estimate = .29, p < .001; estimate = .17, p < .05; estimate = .36, p < .001). In addition, social support cognition Type II positively affected family resource management (estimate = .18, p < .05).

Discussion

Differences Between Social Support Cognition Types and Family Resource Management

We found that there were significant differences in the family resource management of migrant workers with different social support cognition types. This finding illustrated that migrant workers with higher social support cognition had better resource management, especially with regard to finance management.

An individual’s cognition of social support will influence the way he or she uses resources. In China, the majority of migrant workers do not have the four basic types of insurance (endowment, medical, industrial injury, and unemployment) and are excluded from the social security system (Xiang & Chen, 2008). Migrant workers with a higher cognition of social support face fewer problems, have sound social insurance, and have sound support for their children’s education. Therefore, they have more resources for dealing with the needs of their families and managing unexpected crises. Thus, migrant workers with a higher degree of cognition of social support tend to have better family resource management.

The Relationship between Demographic Variables and Social Support Cognition Types

With regard to the relationship between social support cognition types and demographic variables, we found that there were significant differences among the demographic variables (gender, monthly income, marital status, and housemates) of migrant workers with different social support cognition types; this is consistent with previous research (Xu, 2003).

In terms of gender, there was little difference among male workers’ social support cognition types, while female workers were mainly in the high cognition (Type II) category. This may be because female migrant workers have fewer living demands than do male migrant workers, and female migrant workers may be more easily satisfied. In addition, Chinese women have traditionally focused on the family and Chinese men have traditionally focused on their careers, so women have fewer life stresses than do men (Liu & Xu, 2008). Furthermore, females were more active in seeking external support (China National Bureau of Statistics, 2007); hence, female migrant workers’ social support cognition was higher.

With respect to monthly income, migrant workers with a monthly income below US$235 were mainly in the Type III category, while those with a monthly income above US$235 were mainly in the Type II category. The migrant workers with higher incomes have better social support networks, therefore they are able to access more resources through contacts with other networks (Fischer, 1982). In short, the higher one’s monthly income is the better one’s social support will be.

In terms of marital status, unmarried and married migrant workers were mainly in the Type II category, while divorced workers were mainly in the Type III category. Although the value of traditional marriage has changed, divorce is still seen in a negative light. When people get divorced, their social networks tend to shrink, affecting their opportunities for social intercourse (Turner, 2006). Meanwhile, married migrant workers have less stress because the couple can share the family burden. Unmarried migrant workers are generally young, have wider social contacts, and can more easily get information and integrate into urban life. This means that their cognition of social support is high.

With regard to housemates, migrant workers who lived with their immediate family or other relatives were mainly in the Type II category, while those who lived with workmates were mainly in the Type III category. Family members and relatives who live with migrant workers play an important role in the workers’ network of emotional and material support, because they are able to take care of each other in daily life. Therefore, the social support cognition of these workers is high. However, migrant workers who live with workmates generally live in the suburbs, most often in a temporary work shed. Their accommodation is often crowded, humid, and dirty, as well as being poorly lighted and ventilated. But the government have not implemented policies that will improve their accommodation, therefore their cognition of social support is low.

The Effect of Demographic Variables and Social Support Cognition Types on Family Resource Management

Our focus in this article has been on showing that demographic variables and social support cognition types can significantly influence family resource management.

The results of path analysis showed that occupation (manufacturing), monthly income (below US$235), the worker’s housemates (workmates), and social support cognition type (Type II) were the most important positive predictors of family resource management. These findings illustrated that family resource management can be affected by economic situation, work, social culture, and social policy (Goldsmith, 2005).

In terms of occupation, manufacturing had a more direct effect on family resource management than did the others. In order to ensure the continuous development of the economy, the Chinese government has invested substantial resources in (and provided policy support to) the manufacturing sector, so migrant workers in this sector have access to more social resources than do migrant workers employed in other sectors. Thus, their sources of family resources were sufficient, which directly influenced their family resource management.

With regard to monthly income, earning less than US$235 had a direct effect on family resource management. In China, most migrant workers’ incomes were low, and they usually had to find ways to acquire more resources to improve their family life, such as working several jobs, studying new work skills in their spare time to increase their employment opportunities, and reducing household spending (Brown & Lichter, 2004). Hence, when the family income is low, migrant workers will choose reasonable and effective ways to manage their family resources.

With regard to housemates, living with workmates directly affected the family resource management of migrant workers, and they managed family resources better. Most groups of migrant workers live in shabby dormitories (China National Bureau of Statistics, 2007). They share these with housemates who come from different areas. This expands the scope of their social interactions, which, in turn, means they have more sources of information than do those who live in scattered housing. They also take care of, supervise, and learn how to manage family resources from each other. Thus, migrant workers who live with workmates make better use of their family resources than do other workers.

Compared to Types I and III, Type II social support cognition was the important positive predictor of family resource management. Social support is also an important family resource (Zhang & Ruan, 1999); it can maintain the basic functions of family, help workers deal with stressful events or crises, and give families necessary physical and mental support. In short, the higher one’s cognition of social support is, the better one’s resource management will be.

Our results in this study show that social support cognition and demographic variables were important to the family resource management of migrant workers, and directly influenced their quality of life. Therefore, migrant workers should learn how to make use of their family resources, create a plan for the use of such resources, and try either to increase resources or to use minimal resources to meet their needs. Well developed family resource management skills will improve quality of life.

This study had certain limitations in terms of sampling. The sample was not stratified according to the percentage of the population for the various categories of drivers (e.g., according to gender or age). This, therefore, limits the generalizability of the results to the whole population of migrant workers in China. Future researchers could use different samples of migrant workers, with more complex problem solving.

Brown, J. B., & Lichter, D. T. (2004). Poverty, welfare, and the livelihood strategies of nonmetropolitan single mothers. Rural Sociology, 69, 282-301. http://doi.org/grm

China National Bureau of Statistics. (2007). Analyzing the affecting factors of living quality of migrant workers in China [In Chinese]. Statistical Research, 24, 13-18.

Deacon, R. E., & Firebaugh, F. M. (1988). Family resource management: Principles and applications (2nd ed.). Boston, MA: Allyn & Bacon.

Fischer, C. S. (1982). To dwell among friends: Personal networks in town and city. Chicago, IL: University of Chicago Press.

Goldsmith, E. B. (2005). Resource management for individuals and families (3rd ed.). Belmont, CA: Thomson/Wadsworth.

Granovetter, M. (1995). Getting a job (Rev. ed.). Chicago, IL: University of Chicago Press.

Lan, L. Y., & He, S. (2004). The problems in and solutions to welfare system for peasant laborers in cities. Journal of Chongqing, University of Post and Telecommunications, 2, 97-100.

Liu, W. R., & Xu, A. Q. (2008). Economic stress in urban families: A social contextual perspective. Social Sciences, 2, 82-90.

Turner, H. A. (2006). Stress, social resources, and depression among never-married and divorced rural mothers. Rural Sociology, 71, 479-504. http://doi.org/grn

Wang, Y. F. (2004). Introduction of social support and psychological health [In Chinese]. Psychological Science, 27, 75-77.

Xiang, X., & Chen, Y. X. (2008). Research on social support system for migrant workers’ integration into the city. Social Work, 7, 33-35.

Xu, A. Q. (2003). The social assistance for disadvantaged group of single-parents. Jiangsu Social Sciences, 3, 62-72.

Zhang, W. H., & Ruan, D. Q. (1999). Social support networks of urban residents. Sociology Research, 3, 12-24.

Zhang, Y., & Xing, Z. J. (2007). Review on the social support and subjective cognition of happiness. Psychological Science, 30, 1436-1438.

This research was supported by the Soochow University substantial program

The research of migrant workers&rsquo

urbanization and social supports in south Jiangsu

211 project of leading academic discipline

subproject

Protecting the rights of the disadvantaged minority in harmonious society

Economics of ecoregions and social administration in the south of Jiangsu Province

subitem

Research on analyzing the school development in south Jiangsu.

I-Jun Chen, School of Education, Soochow University, No. 1, Wenjing Road, Suzhou, Jiangsu Province 215123, People’s Republic of China. Email: [email protected]