Abstract
Background:
There are approximately 47.5 million female migrant workers living in major cities in China. Numerous studies have documented the marginalized living conditions confronting migrant workers in cities, such as employment difficulties, unjustifiably demanding working conditions, lack of medical insurance and social provision, poor housing conditions, unfavourable educational arrangements for migrant children, and discrimination by urban residents. In addition, female migrant workers may suffer from discrimination, exploitation and oppression.
Aim:
This study aimed to explore the difficulties and perceived meaningfulness of migration and their effect on the mental health status of female migrant workers in Shanghai, Kunshan, Dongguan and Shenzhen.
Methods:
A total of 959 female migrant workers from 12 factories completed the questionnaires, which included the Brief Symptom Inventory, the Migration Stress Scale and the Meaning in Migration Scale.
Results:
The findings indicate that 24% of female migrant workers could be classified as having poor mental health and the percentage in Shenzhen (35%) was far greater than in the three other cities in China. ‘Financial and employment-related difficulties’, ‘cultural differences’, gender-specific stressors and ‘better future for self and children’ significantly accounted for the mental health outcomes of female migrant workers.
Conclusion:
Recommendations for policy change and service initiatives targeted at improving the mental health of female migrant workers are discussed.
Background
The last three decades have witnessed an unprecedented large-scale rural–urban migration in China. According to the Second National Agricultural Census, approximately 132 million migrant workers are currently living in major cities in China, 36% of whom are female (National Bureau of Statistics, 2008). Numerous studies have documented the marginalized living conditions confronting migrant workers in cities, such as employment difficulties, unjustifiably demanding working conditions, lack of medical insurance and social provision, poor housing conditions, unfavourable educational arrangements for migrant children, and discrimination by urban residents (Park & Wang, 2007; Solinger, 1999; Wong, Chang, & He, 2007).
In the past few years, attention has shifted towards understanding specific groups of migrant workers and their families in China, such as the left-behind children in rural areas and migrant workers in specific types of job, such as those working in factories or construction sites. One specific group that has caught the attention of researchers is that of female migrant workers in Chinese factories. Several studies have suggested that female migrant workers suffer from discrimination, exploitation and oppression in the context of globalization and industrialism (Lee, 1998; Pun, 2006; Tan, 2000). In contrast, other studies have revealed that female migrant workers may benefit from migration, with advantages such as greater autonomy, freedom from the oppression of rural patriarchy and greater exposure to modern living (Gaetano, 2004; Wong, He, Leung, Lau, & Chang, 2008).
Similarly, there is contradictory evidence concerning the mental health of female migrant workers in China. Li et al. (2007) compared the mental health of three groups of people in Hangzhou, China: 4,453 migrant workers, 1,957 local residents in Hangzhou, and 1,909 rural residents in Western Zhejiang Province. Their findings suggested that male migrant workers had better mental health outcomes than female migrant workers. In contrast, a survey of 475 migrant workers recruited from various communities of migrant workers in Shanghai conducted by Wong et al. (2008) found that female migrant workers had better mental health than male migrant workers; 25% of male migrant workers and 6% of female migrant workers could be classified as mentally unhealthy. However, a survey conducted by Wen and Wang (2009) with 1,026 migrant workers in Shanghai indicated that migrant workers experienced a sense of loneliness and women were more likely to feel lonely than men.
The above literature has highlighted the inconclusive findings concerning the study of female migrant workers in China. First, it is not certain whether female migrant workers are experiencing good or poor mental health. Second, it is difficult to conclude whether migration has brought benefits or losses for female migrant workers. From our point of view, these contradictory findings may be related to a number of issues. First, the difference in the findings may be linked to the use of different outcome measures. For example, the studies conducted by Li et al. (2007) and Wong et al. (2008) used two different mental health instruments: the mental health subscale of the Short Form-36 and the Brief Symptom Inventory (BSI), respectively. Second, the difference might also result from the fact that the samples of migrant workers came from various sources, including communities, factories, construction sites and service industries. Third, there are regional differences in the type of migrant workers that are likely to be recruited in places such as Shanghai and Hangzhou. Indeed, due to chain migration, certain cities might recruit migrant workers predominantly from certain regions of China. Given such possibilities, this study attempts to explore whether regional differences might affect the mental health of female migrant workers working in factories in four cities in China: Shanghai, Kunshan, Dongguan and Shenzhen.
Stress and coping framework in the study of migration and mental health
This study adopts a stress and coping framework in understanding the migration experience of female migrant workers in China. During the migration process, migrants experience difficulties that constitute sources of stress that may lead to adverse health outcomes. In the literature, migrants are reported to experience at least four major sources of stress. First, migrants suffer difficulties in managing aspects of daily living such as financial and job-related difficulties (Nicassio, Solomon, & McCullough, 1986; Thompson, Hartel, Manderson, Woelz-Stirling, & Kelaher, 2002; Wong et al., 2008), poor living conditions (Papadopoulos, Lees, Lay, & Gebrehiwot, 2004; Wong et al., 2007) and discrimination (Aroian, Norris, Patsdaughter, & Tran, 1998; Yeh et al., 2003). Second, migrants are confronted with losses such as the loss of social networks, employment, social status and familiar living environment (Aroian et al., 1998; Bhugra, 2004). Third, they encounter cultural differences such as language differences, value conflicts and lifestyle changes (Bhugra, 2004; Vedder & Virta, 2005). Fourth, they face unfulfilled expectations such as feeling disillusioned with the reality of living in the host country (Thompson et al., 2002; Ward & Searle, 1991). Finally, female migrants are found to be suffering from patriarchal control at home and the workplace (Gaetano & Jacka, 2004) and at times, experience sexual harassment and discrimination (Puri & Cleland, 2007). Many studies have suggested that migrants who experience these sources of stress have poor mental health (Bhugra, 2004; Thompson et al., 2002; Ward & Searle, 1991). Adopting the framework of migration stress proposed by Wong and Lee (2003), this study operationalizes migration stress to include five types of stressor: gender-specific stressors; difficulty in managing aspects of daily living; losses; cultural differences; and unfulfilled expectations. It is hypothesized that female migrant workers who experience a higher level of migration stress have poorer mental health.
However, migration stress exerts differential effects on individual migrants. One possible factor accounting for this difference is that individual migrants may ascribe different meaning to migration. Those who perceive migration as bringing beneficial effects to themselves and others will perceive migration more favourably and are more able to withstand the stress associated with the migration experience, thus leading to better mental health outcomes. Watkins et al. (2003) found that prospective Vietnamese migrants who believed that there were better educational, socioeconomic and employment opportunities in Australia had better mental health than those who did not. However, a study conducted by Ruback, Pandey, Begum, and Tariq (2004) on internal migrants in India revealed that those who believed in fate as a reason for migration had significantly poorer mental health. Wong and He (2008) maintained that female migrant workers in Shanghai who proposed ‘more financial and material gains’ as a reason for migration had better mental health. In summary, migrants who hold different reasons for migration may have different levels of mental health, irrespective of the intensity of stress they experience. This study hypothesizes that those female migrant workers who ascribe positive meaning to migration will have less migration stress and therefore better mental health outcomes.
Based on the literature review, the objectives of this study were as follows:
To identify the types of migration stress confronting female migrant workers in four regions in China.
To explore regional differences in migration stress, meaning of migration and mental health outcomes among female migrant workers in the four regions in China.
To examine the effect of meaning of migration on the relationship between migration stress and the mental health of female migrant workers in the four regions in China.
Methodology
Research site
This study adopted a survey method to explore the mental health status, migration stress and meaning of migration experienced by female migrant workers in Shanghai, Kunshan, Dongguan and Shenzhen, China. These cities are the major destinations for rural–urban migration and are located in the two main economic regions, the Yangtze River Delta (Shanghai and Kunshan) and the Pearl River Delta (Shenzhen and Dongguan). According to the Monitoring Report of National Bureau of Statistics (2010), there were 28.16 million and 32.82 million migrant workers in these two main economic regions in 2009, representing 19.4% and 22.6% of the total population of migrant workers in China, respectively. Among them, 34.9% are women and 39.1% work in the factories. There were nearly 1 million migrant workers in Kunshan in 2009; among them 46% are women and around 80% work in manufacturing factories (Kunshan Bureau of Statistics, 2010). In Dongguan, there are 4.13 million migrant workers; more than 50% are women and 82% work in manufacturing sectors (Dongguan Bureau of Statistics, 2010). Shanghai hosts about 5.6 million migrant workers (Shanghai Bureau of Statistics, 2010); nearly half of them are women and roughly 30% work in industries. Shenzhen has around 6 million migrant workers; over 40% are women and 80% have jobs in different industries (Shenzhen Bureau of Statistics, 2010).
Sampling and data collection
‘Migrant workers’ was defined in this study as those who held an officially endorsed temporary resident permit to work in the cities in China. To be eligible as a participant, a migrant worker had to: (1) be 18 years old or above; (2) have a rural resident status (but officially allowed to work in the city); (3) be working in textile or electronic factories; and (4) have lived in the city for at least three months. In this study, we used a multi-stage cluster sampling procedure to recruit research participants. In stage one, we chose the above four cities. In stage two, we selected three factories randomly from each city. In stage three, we contacted the management teams of these factories to ask for their permission to conduct the study. In each factory, 100 workers were randomly selected to participate in the survey. All potential participants were informed that they had the right to decline the survey and that their job would not compromised. Finally, verbal consent was obtained from each participant and they were then asked to complete a self-administered questionnaire. It is necessary to note that despite regional differences in spoken Chinese dialects, all Chinese share the same written language. Therefore, there was no need to make any adjustments in the questionnaires. This study was endorsed by the Ethics Committee of the East China University of Science and Technology.
We managed to recruit 959 participants for this study. Our final sample consisted of 261 (27%) female migrant workers from Shanghai, 266 (28%) from Kunshan, 224 (23%) from Shenzhen and 208 (22%) from Dongguan. Over 200 questionnaires were deemed invalid because more than 80% of the items in these questionnaires were either unanswered or were uniformly provided with the same answers. Further analyses found no significant difference in the demographic profiles of the 959 participants and those whose questionnaires were discarded.
Instruments
The questionnaire used in this study included socio-demographic characteristics, the Migration Stress Scale, the Brief Symptom Inventory and the Meaning in Migration Scale.
Migration Stress Scale (MSS)
In this study, ‘migration stress’ referred to the severity of stress experienced by female migrant workers during the process of adapting to lives in the cities. The MSS that was used here was adapted from that of Wong and He (2008). The original scale aimed to measure the severity of stress experienced by migrant workers, and was self-constructed upon a careful review of the relevant literature on migration and stress. The scale had 33 items and four dimensions: financial and employment difficulties (14 items), cultural differences (seven items), lack of social life (seven items), and interpersonal tensions and conflicts (five items). In this study, six items were added to the scale to provide an additional measure of gender-specific stresses experienced by female migrant workers. The items were generated from pilot interviews with 15 female migrant workers. These items reflected the female migrant workers’ experience of sexual harassment, gender discrimination and threats to personal safety. Hence, our MSS was expanded to 39 items. This was a four-point scale, ranging from 1 = ‘no stress’ to 4 = ‘a lot of stress’. Cronbach’s α for the full scale was .92, and .85 (financial and employment-related difficulties), .77 (cultural differences), .82 (lack of social life), .71 (interpersonal tensions and conflicts) and .64 (gender-specific stress) for the subscales.
Brief Symptom Inventory (BSI)
The BSI is designed to evaluate the mental health status of general and clinical populations aged 13 or above (Derogatis & Melisaratos, 1983). It includes 53 items consisting of nine dimensions: depression, somatization, anxiety, hostility, obsession–compulsion, interpersonal sensitivity, phobic anxiety, paranoid ideation and psychoticism. The Chinese version of the BSI was translated and used by Cheng, Leong, and Geist (1993). In their study, they found a high internal consistency of the BSI in a sample of 719 psychiatric outpatients, with Cronbach’s α coefficients on the nine dimensions ranging from .71 to .80. They also performed a two-week interval test-retest reliability on a sample of 60 non-patients. The coefficients ranged from .68 to .91. Convergent validity between the BSI and the Minnesota Multiphasic Personality Inventory (MMPI) was demonstrated with a sample of 209 symptomatic volunteers. Overall, the re-analysis showed maximal correlations between the BSI and MMPI for interpersonal sensitivity, depression and anxiety. Correlations for the other dimensions were smaller in magnitude yet similar in pattern. In all, the BSI dimensions and MMPI scales seemed to converge highly. In this study, the Cronbach’s α for the scale was .96.
Meaning in Migration Scale (MMS)
The MMS was adopted from Wong and He (2008) to assess participants’ reasons for migrating to the cities. In the original study, a research team member in Shanghai interviewed 25 migrant workers and generated a list of reasons for migrating to Shanghai from rural areas. An exploratory factor analysis was conducted and a principle component analysis was performed on these items, which was followed by a Varimax rotation. Sixteen items that had a factor loading greater than .3 were selected to form the present scale. A four-factor solution was discovered that accounted for 69.5% of the total variance. These four factors were: personal aspiration and achievement (six items); better future for self and children (four items); better life than in the villages (four items); and more financial and material gains (two items). In that study, Cronbach’s α for the full scale was .87 and for the four subscales was .81, .71, .62 and .67, respectively. The scale is a four-point scale, measuring the degree of agreement with the reasons for coming and living in the cities (1 = ‘totally disagree’, 4 = ‘totally agree’). The higher the score, the greater the positive meaning ascribed by the respondents. In the present study, Cronbach’s α was .79.
Data analysis
All data were managed and analysed in SPSS (version 18.0). Descriptive statistics were calculated to identify socio-demographic characteristics and variations in mental health status, meaning in migration and migration stress among female migrant workers. Using Ritsner, Ponizovsky, Kurs, & Modia’s (2000) threshold, female migrant workers with a Global Severity Index greater than .78 were classified as mentally unhealthy. Finally, hierarchical regression analyses were performed to examine the effects of socio-demographic variables, migration stress and meaning in migration on the mental health of female migrant workers in China.
Results
Table 1 shows that 67.5% of the participants were aged 25 or under, and the majority were single (64.1%). A large proportion of the female migrant workers had middle or high school education (52.2% and 36.5%, respectively), and only 7.9% had tertiary education. More than 60% of female migrant workers worked on the factories’ assembly lines. The average monthly salary was RMB$1,337 and they worked an average of 9.65 hours per day. Approximately 61% lived in accommodation provided by the companies. In summary, most of the female migrant workers in this sample were young, single, had a relatively low income and a junior secondary school education, and worked on the factory assembly line for nearly 10 hours a day. The socio-demographic characteristics of this sample were similar to those found in other studies (Wen & Wang, 2009; Wong & He, 2008).
Socio-demographic characteristics of female migrant workers in four cities in China.
Comparing the data from the four cities, those from Shanghai were much younger, with 79.2% aged 25 or under, and a much larger percentage had a college education. Female migrant workers in the Pearl River Delta were more likely to work longer hours and live in accommodation provided by the companies. Their monthly salary was slightly higher than in Shanghai and Kunshan.
Table 2 shows that ‘financial and employment-related difficulties’ were the most severe sources of stress for female migrant workers. There was no significant difference in the level of stress experienced by migrant workers in the four cities, except for ‘cultural differences’. The BSI scores for the overall scale and for most of the subscales reflect the trend that female migrant workers living in Shenzhen and Dongguan had poorer mental health than their counterparts in Shanghai and Kunshan.
Means (and standard deviations) of scores on the MMS, BSI and MSS in four cities in China.
Based on the GSI thresholds developed by Ritsner et al. (2000), 202 of the participants (24%) were regarded as mentally unhealthy. The percentages of mentally unhealthy female migrant workers in Shanghai, Kunshan, Shenzhen and Dongguan were 18.4% (n = 44), 23.3% (n = 47), 35% (n = 71) and 21.1% (n = 40), respectively. The Shenzhen sample clearly had the highest rate of participants classified as mentally unhealthy.
The hierarchical regression reported in Table 3 indicates that age, educational level and daily working hours significantly predicted the outcome in mental health in model 1. In model 2, financial and employment-related difficulties, cultural differences and gender-specific stressors significantly predicted the poor mental health found in the sample. In model 3, when migration stress was controlled for, meaning in migration did not exert a significant effect on mental health except for the subscale of ‘better future for self and children’.
Hierarchical regression (BSI as dependent variable).
p ≤ .1, ** p ≤ .05, *** p ≤ .01
Discussion
This study has identified that 24% of female migrant workers could be regarded as mentally unhealthy, a much higher figure than the 8% found in a previous study conducted by Wong et al.(2008). The discrepancy in these findings may be explained in terms of the different sample characteristics found in the two studies. Whereas this sample included exclusively factory workers, the participants in Wong et al.’s study were recruited from a greater variety of sources: the communities where migrant workers lived, service industries, construction sites and factories in Shanghai. Indeed, the working conditions of migrant workers in the factories have always been a cause for concern for researchers and policymakers. Migrant workers in the factories have to work long hours – almost 10 hours per day and usually with overtime, including weekends. As many of them live on the factory premises, they have little time to get out and enjoy leisure time or a social life. A few may also experience sexual harassment and discrimination by their superiors in the factories (Gaetano & Jacka, 2004; Li, Stanton, Fang, & Lin, 2006; Pun, 2006; Roberts, 2002; Wong et al., 2008; Wong & He, 2008). In contrast, as Wen and Wang (2009) suggested, migrant workers who live independently, particularly those living with their spouse and/or children in Shanghai, expressed positive feelings of attachment and less loneliness in Shanghai. Thus, it is not surprising to find that our sample of female migrant workers had a higher rate of being mentally unhealthy than some other studies.
It is alarming to find that a much higher percentage of female migrant workers (35%) were considered mentally unhealthy in Shenzhen than in other regions. Specific regional circumstances may explain such differences. First, the cost of living in Shenzhen is relatively high whereas the average salary of migrant workers is relatively low (compared with other cities in China such as Shanghai) (Lee, 1998). Second, the nature and structure of enterprises create a considerable strain on migrant workers. Compared with enterprises in Shanghai, which are mostly large-scale state-owned, financed with foreign capital, or run by the townships, the enterprises in Shenzhen are mostly smaller in scale and run by Hong Kong and Taiwanese investors (Wan & Liu, 2007). There is a tendency for larger-scale state-owned enterprises and those financed with foreign capital to be more observant of government regulations than smaller enterprises (Xin & Frances, 1998). Indeed, numerous incidents of unjust treatment of factory workers in the Pearl River Delta have been reported (Thireau & Hua, 2003). For example, it is common for migrant workers in factories in Shenzhen to report overdue payment and withholding of salary. Thus, it is not surprising that migrant workers in Shenzhen have poorer mental health than their counterparts in other regions of China.
In general, ‘financial and employment-related difficulties’ significantly contribute to the poor mental health of female migrant workers. This is understandable because many migrant workers move from the poverty-stricken central parts of China to the cities because of the financial rewards (Wong et al., 2008). The living environments and facilities in rural villages are quite primitive, and migrant workers work hard so that they can send money back to their villages to build their own houses and to improve the living standard of their families (Croll & Huang, 1997). Thus, financial issues are a major concern of migrant workers.
The ‘cultural differences’ was a significant factor influencing the mental health of the migrant workers in the four regions. This finding echoes those of previous studies (Vedder & Virta, 2005). According to Bhugra (2004), migration involves a process of dealing with cultural differences, such as language difference, value conflicts and lifestyle changes. Although they are internal migrants, female migrant workers in this study still have to face certain adaption issues. They speak with heavy accents, which can be easily identified by the local residents in the cities, and they dress and act differently (Gaetano & Jacka, 2004). While these differences may create certain adjustment difficulties for some female migrant workers, they may also constitute a source of discrimination by the city residents. Understandably, female migrant workers may feel frustrated with these changes (Wong et al, 2007). Consequently, their mental health can be affected.
This study also found that gender-specific stressors contributed significantly to the mental health of female migrant workers. Gender-specific stressors included aggressive attitudes and behaviour manifested by factory managers and male colleagues, and covert or overt discrimination displayed by male workers. There is an abundance of studies suggesting that females who are exposed to such situations have poor mental health (e.g. Gaetano & Jacka, 2004). Thus, it is not surprising to find the same results in our sample of female migrant workers.
Meaning in migration did not strongly influence the mental health of female migrant workers. The only factor found to predict mental health of female migrant workers was ‘better future for self and children’. Many female migrant workers move to the cities in search of a better future and many consider such a move as a way of liberating themselves from the rather dominant patriarchal rural communities (Pun, 2006; Wong & He, 2008). Those who have children would like to bring them to the cities so that they can receive a better education. Thus, female migrant workers who ascribed such reasons for migrating to the cities could have better mental health than those who did not.
Limitations
Because the participants were recruited from factories in only four cities in China, the findings of this study cannot be generalized to all female migrant workers in factories in China. Neither can the findings be extended to other groups of migrant workers. The use of self-report questionnaires has some limitations, including problems with memory and the desire of participants to present themselves in a positive light. Finally, the scales in this study had not undergone a rigorous validation process. Future research should attempt to validate these instruments before inferences can be drawn from their findings.
Conclusion
Female migrant workers constitute an important workforce in China. How well they adjust to life in the city may have detrimental effects on the self and society. Thus, it is important for the Chinese government to actively facilitate the adaptation of this group of migrant workers in the cities. It is also essential for the various levels of government to consider developing and/or streamlining the mechanisms for providing adequate medical, housing and educational benefits for migrant workers in the cities.
It is also necessary for the government to develop services that address the mental health needs of female migrant workers, particularly those in Shenzhen. First, from a prevention point of view, it would be useful to provide female migrant workers with psycho-education on mental health and mental illness so that early detection of mental health problems can be achieved. Second, counselling services should be available for those who may be suffering from mental health problems. Third, it is necessary to provide programmes so that female migrant workers can better understand the subtle social and cultural etiquettes of people in the cities. It is also useful to organize activities involving the participation of local residents and female migrant workers. Such encounters may foster better understanding between the two groups. Fourth, mechanisms must be developed to protect female migrant workers from harassment and discrimination, and gender-sensitive policy and service should be developed to protect the well-being of female migrant workers. Finally, at the macro level, the government should: (1) provide funding for the development of a kind of a social club for migrant workers, which should be conveniently located in the community where a sizeable number of migrant workers reside, to provide information and counselling on a variety of issues including financial and employment difficulties, mental health issues and other adjustment problems; and (2) monitor the industries to ensure that they follow the labour laws regarding migrant workers’ working conditions.
Footnotes
Acknowledgements
This research was funded by the Humanities and Social Sciences Foundation of the Ministry of Education of China.