Migration and marital instability among migrant workers in China: A gender perspective

Abstract

This paper presents an examination of how migration affects the marital stability of migrant workers, whether there is gender difference in migrant workers’ marital instability, and whether there is gender difference in the determinants of migrant workers’ marital stability. Poisson regression analysis of survey data related to Guangzhou rural–urban migrant workers showed that there is gender difference in marital instability, and that female migrant workers have a higher level of marital instability than male migrant workers. Migrating unaccompanied, inter-provincial migration, distance between the hometown of spouse from husband’s family’s location, and non-traditional gender views were found to be positively related to migrant workers’ marital instability. Some of these influences are mediated by marital satisfaction. In addition, the influence on marital instability of migrating unaccompanied, and distance between the hometown of spouse and husband’s family’s location were identified as more important for female migrant workers, while the influence of inter-provincial migration on marital instability is more important for male migrant workers. It was determined that there are gender differences in the influence of wife's income and gender views: the wife’s income relates positively to male migrant workers’ marital stability, but negatively to female migrant workers’ marital stability. The results show that migration has a stronger influence on female migrant workers’ perceptions of marital instability.

Keywords

Gender marital satisfaction marital stability migrant worker population migration

Introduction

Since 1978, when the Chinese government initiated the policy of market reform, studies have recorded increases in the divorce rate (Ye and Lin, 1998; Zeng, 1995; Zhang, 1997). The crude divorce rate increased from 0.30% in 1978 to 2.79% in 2015: overall the growth rate is about 15.75% per year.

This rapid rise in the divorce rate has attracted the attention of many scholars (Xu, 2012; Xu et al., 2013). Xu and Ye (2002) found that the divorce rate was higher in the relatively developed Eastern provinces of China compared to the underdeveloped inland provinces, and the divorce rate was higher in urban areas compared to rural areas. Xu and his colleagues (2013) also found that families with children were less likely to experience divorce.

Although our understanding of the topic of divorce has been increased considerably through these studies, most of them focus on the rate of divorce and its determinants. Divorce occurs when a marriage is legally dissolved after there has been lengthy separation and various legal procedures: divorce represents the consequences of marriage instability rather than the instability itself. However, many marriages remain intact even when couples feel highly dissatisfied with their marriage (Booth and White, 1980). Subsequently, some scholars began to study marital instability (Booth and White, 1980; Ye and Xu, 1999), a cognitive state in which individuals believe their marriage to be in trouble and consider whether to end it. It is often reflected in talking with friends about considering divorce (Booth and Edwards, 1985). Marriage instability may lead to divorce, but this is not the expected inevitable outcome (Booth et al., 1986). Ye and Xu (1999) suggested that marriage stability reflects the partners’ confidence that the marriage will last. A marriage can be unstable and end at any moment when a spouse begins to consider divorce. Therefore, studying partners’ perceptions of marriage instability offers a way to understand the dynamics and factors that are shaping the growing divorce rate in China.

Although there are more than 247 million migrant workers in China, and the majority have moved from rural areas to cities without local hukou status, marriage instability among these workers has been given only a little attention. The unique experiences of these workers have placed stress on their marriages. First, many migrant workers cannot afford to support their families at the new destination and therefore leave their partners behind, so that in effect they are married but separated. Second, the migrant population has changed the traditional division of gender roles and earning potential, thus altering the relationship and expectations between husbands and wives. Third, the migrant population has changed the traditional view of gender roles, which shapes the values of marriage and family among migrant workers and may create marital dissatisfaction. Although a few studies have found that the migrant population is associated with the divorce rate in rural areas (Du, 2010; Gao, 2011; Mo and Shi, 2015), these studies focus mainly on the situation back home, but not on the migrant workers in the city. In addition, the purpose of these studies is to describe the difference in risk of divorce between migrant workers and non-migrant workers, and to investigate the influence of migration status on divorce. The phenomenon of marriage instability of migrant workers in cities, and the effect of the migration experience on marital instability among migrant workers, are both unclear.

In addition, research seldom addresses gender differences in marriage instability in China. An overwhelming majority of migrant workers are male, which may have different implications for their experiences in the marriage market. In comparison, females may have an advantage in the marriage market among migrant workers. In addition, migration changes the traditional gender division of roles and distribution of resources between couples, which may be of more benefit to female migrant workers than to male migrant workers. The purpose of this article is to explore how factors suggested in the literature on marriage instability are related to migrant workers in China, and how they differ between males and females.

Theory of marital instability

Exchange theory is commonly applied to explain why some marriages remain intact and others dissolve. Exchange theorists suggest three factors that can be related to marital stability: the rewards of remaining in the current marriage; the barriers to divorce; and the alternatives to marriage (Becker, 1981; Becker et al., 1977; Levinger, 1965, 1976; Previti and Amato, 2003).

According to exchange theory, the attractiveness of a marital relationship is positively related to its perceived reward (Knoester and Booth, 2000), with reward being derived from marital resources such as love, goods, support, and security (Knoester and Booth, 2000). In most studies, reward is often operationalized by measures of marital quality or marital satisfaction (Glenn, 1991; Johnson et al.,1999). Less reward is associated with poorer marital quality and higher marital dissatisfaction, which lead to marital instability.

Exchange theorists further suggest that, when barriers to marital dissolution are strong and alternative opportunities are available, marital instability increases but unhappy unions may continue (Heaton and Albrecht, 1991; Levinger, 1976; Previti and Amato, 2003). Some possible barriers often discussed in the literature include financial security, holding traditional values, and having children (Knoester and Booth, 2000). Economic status can contribute to marital stability even if the marriage is in a state of disharmony. Low income not only renders spouses more dependent on each other, but also reduces the prospects for marital alternatives (Udry, 1981). In contrast, economic independence affords the unhappy spouse, and especially the wife, sufficient resources to end the relationship (Schoen et al., 2002). In addition, values such as traditional gender views can raise the cost of divorce and promote marriage stability. Traditional gender views reflect the degree of commitment to upholding the family. They also provide justification for household division of labour (Greenstein, 1996). Having children is an important barrier to divorce because parents have to consider their children’s well-being and arrangements for their care after marital dissolution (Hill, 1988).

The attractiveness of alternatives is the final factor suggested as affecting the stability of marriage (Knoester and Booth, 2000). White and Booth (1991) noted that married people who perceive a large number of opportunities for remarriage are more likely to divorce than others. A marriage market with gender imbalance provides numerous remarriage opportunities for the gender with the low ratio (South and Lloyd, 1995). Age at first marriage is seen as an important factor related to alternatives to divorce. Individuals who married at an early age have more opportunities for remarriage than those who married when older (South and Lloyd, 1995).

Gender differences in migration and marital instability

There is gender difference in the influence of migration on marital stability among migrants. Males and females are not only differentially vulnerable to various life problems (Simon, 1998; Kessler and McLeod, 1984), but also have differences in perceived control, self-assessment, and self-esteem (Mirowsky and Ross, 1989; Turner and Roszell, 1994). Migration-induced spousal separation is likely to disrupt spousal interaction. However, wives are usually more committed than husbands to existing marital relationships (Fuller and Edwards, 2004). Among migrant workers, therefore, migrating unaccompanied is a more important determinant of marital satisfaction and rewards for females than for males.

Second, there is gender difference in the influence of distance from hometown/village on migrant workers’ marital stability. In the context of a patriarchal culture, the wife is more likely to move in with the husband’s family, and these arrangements are more likely to deprive wives away from their hometowns of the social support and mediation beneficial to marriage, which may reduce the barrier to divorce.

Third, the influence of inter-provincial migration on martial instability can differ between males and females. Guttentag and Secord (1983) suggested that males and females have different attitudes to marriage. Females tend to establish stable and loyal marital relationships with their partners; males, however, often hope to establish free and unrestricted relationships with their partners. Inter-provincial migration offers more opportunities for male migrant workers to develop new social relationships that may lead them to be less committed to the spousal relationship within marriage.

Fourth, wives’ economic independence can have a stronger effect on marital stability for wives than for husbands. Kalmijn and Poortman (2006) found that wives’ employment and financial problems have a stronger relationship to divorce for females than for males. In addition, wives’ economic independence also motivates competition between spouses for family status, and wives may demand more decision-making power or press their husbands to share more housework. Because the status quo benefits men, the economic independence of wives may create more stress for husbands, and more decision-making power may lead to more happiness for wives. Wives’ economic independence therefore results in different rewards and barriers for wives and husbands.

Finally, the adoption of non-traditional gender values can have a stronger relationship to marital stability for women than for men. Amato and Booth (1995) found that less traditional attitudes have different effects on females and males. Wives with egalitarian attitudes have reported lower levels of satisfaction, and husbands with less egalitarian attitudes have reported higher levels of marital satisfaction (Heaton and Blake, 1999). In other words, the influence of gender views on marital rewards is different for men and women.

Chinese context and research hypotheses: migrant workers in China

Under the household registration system (hukou), Chinese migrant workers cannot afford to have the entire family move together to the city while they are looking for better economic opportunities. Many workers have to leave their families back home and move unaccompanied to cities for jobs. Migrating unaccompanied by definition means leaving their partners behind; and spousal separation not only undermines the frequency and quality of their communication, but also weakens the social and financial obligations and responsibilities of those who are far away (Frank and Wildsmith, 2005). In turn, the barrier to loosening the marital relationship is weakened and marital dissatisfaction increases. This migration pattern increases the risk of marital instability. Females who participate in the labour force not only gain more economic independence but also narrow the gap between spouses in family status. Such changes reduce the benefits of marriage for female migrant workers. In addition, through the workplace, they learn more about less traditional gender views: female migrant workers become more aware of gender inequality in the family and become less happy with their marriages (Amato and Booth, 1995).

The gender imbalance in migration has resulted in a female shortage and a male surplus in the marriage market, which not only contributes bargaining power to females but also increases the opportunities for alternatives to marriage for females. Thus a female migrant would be more likely than a male migrant to initiate divorce. Based on that premise, we suggest our first hypothesis:

Hypothesis 1: There is gender difference in marriage stability, and females have higher levels of marriage instability than males in the context of migration.

The unique migration experience shapes the decline of marital stability, and the effect can be different for males and females. In China, inter-provincial migration is common among migrant workers because of the regional imbalance in economic development. When migrant workers move farther away from their place of origin, they are also farther away from their original family network and hence the constraints deriving from that network become weaker. Because migrants face adjustment stress and loneliness, they actively seek social support and develop new relationships, sometimes intimate, which can lead to extramarital experiences (Booth et al., 1991). Given that males and females have different goals and attitudes in relation to marriage (Guttentag and Secord, 1983), females are more likely than males to establish stable and loyal marital relationships with their partners. Among inter-provincial migrants who are far away from their original family network and its supervision, therefore, male migrant workers are more likely than female migrants to develop extramarital relationships. These experiences increase the tendency for marital dissatisfaction, which may lead to marital instability among migrant workers. We thus suggest a second group of hypotheses:

Hypothesis 2a: Compared to intra-provincial migration, inter-provincial migration enhances marriage instability among migrant workers.

Hypothesis 2b: Among migrant workers, the influence of inter-provincial migration on marital instability is more obvious for males than for females.

Because many migrant workers must leave their family members, including spouses, at home and move to the cities unaccompanied for jobs, some of them may have fewer opportunities to interact with their spouse and fulfill their family role, which then reduces their marital satisfaction and stability. In contrast, females usually rely more on the relationship, financially and emotionally, which renders them more vulnerable to migrating unaccompanied. Thus, we suggest a third group of hypotheses:

Hypothesis 3a: Migrating unaccompanied increases marital instability among migrant workers.

Hypothesis 3b: Marriage satisfaction mediates between migrating unaccompanied and marital stability among migrant workers.

Hypothesis 3c: The influence of migrating unaccompanied on marital stability is more obvious for female migrants than for male migrants.

It has become common for younger migrants working far from home to marry spouses who also come from distant places, which may be far from each other. However, under the influence of a patriarchal culture in China, especially in rural areas, this type of marriage requires the wife to move into her husband’s family household, and thus far away from her own hometown, which may have some negative effects on the marriage, especially on the female’s marriage. Distant from their hometowns, and lacking social support, females may experience more dissatisfaction with their marriages, which may lead to higher levels of marital instability. Therefore, we suggest a fourth group of hypotheses:

Hypothesis 4a: A great distance between the hometowns of spouses exacerbates the level of marriage instability among migrant workers.

Hypothesis 4b: Marriage satisfaction mediates between the great distance between hometowns of spouses and marital stability among migrant workers.

Hypothesis 4c: The influence on marital stability of a great distance between the hometowns of spouses is more obvious for female migrant workers than for male migrant workers.

As females participate in the labour market, they achieve more economic independence, which not only undermines their dependence on their husbands and families but also shapes the power relations between spouses. Some studies have shown that wives’ economic independence may challenge the family status of their husbands and hence result in family conflict, which undermines the quality and satisfaction of the marriage, and may lead to marital instability. Therefore, we suggest a fifth group of hypotheses:

Hypothesis 5a: Wives’ economic contribution has a negative effect on marital stability among migrant workers.

Hypothesis 5b: Marriage satisfaction mediates between the wives’ economic contribution and marital stability among migrant workers.

Hypothesis 5c: The influence that wives’ economic contribution has on marital stability is more obvious for female migrant workers than for male migrant workers.

Migrant workers are exposed to non-traditional gender views as they move to the city; wives who are exposed to these views may become aware of gender inequality and less happy with their marriages (Amato and Booth, 1995). Their satisfaction with their marriage may be affected, which may lead to lower levels of marital stability. In addition, non-traditional gender views render female migrant workers more aware of the inequality in the division of gender roles, and male migrant workers more likely to accept the equal division of gender roles. The situation contributes different effects to marital stability. We therefore suggest a final group of hypotheses:

Hypothesis 6a: Non-traditional gender views decrease the level of marriage stability among migrant workers.

Hypothesis 6b: Marriage satisfaction mediates between the gender view and marital stability among migrant workers.

Hypothesis 6c: There is gender difference in the influence of gender views on marital stability among migrant workers.

Data, variables and method

Data

Data for this study come from a survey of rural–urban migrants conducted by Xi’an Jiaotong University, Shaanxi Normal University and Sun Yat-Sen University from late December 2015 to March 2016 in Guangzhou, the capital of Guangdong Province. As ‘the South Gate of China’, Guangzhou is one of the most populous cities in Chinese Mainland and a key receiving city for rural–urban migrants in China.

Because there was no sampling frame available for our migrant survey, a loose quota sampling method was adopted. To increase our sample’s representativeness and diversity, the quota sampling attempts to cover major migrant-receiving districts and all typical industries in which rural–urban migrants engage. The survey was conducted in the five districts in Guangzhou that had the largest number of migrants: Baiyun, Tianhe, Panyu, Haizhu, and Huangpu. According to data released in 2016 by the Guangzhou Statistics Bureau, 79.4% of migrants in Guangzhou lived in these five districts at the end of 2015. The smallest sampling unit of the survey is the street/town with a quota of 100. The samples of the survey were rural–urban migrants over 15 years old with non-Guangzhou agricultural hukou. All participants were interviewed face-to-face by trained interviewers. The total sample size was 1621 after ineligible participants were excluded.

Variables

The dependent variable in the study is marital instability. Booth and Edwards (1985) operationalized the propensity to divorce by focusing on two aspects: cognitive state, such as thinking that the marriage is in trouble; and, second, action, such as talking with friends and relatives about divorce. We adopted the approach suggested by Booth and Edwards (1985), and used five items to assess marriage instability. The questions indicate whether the person has had such thoughts or actions in the last year (see Table 1). Response categories ranged from 1 = ‘always’ to 4 = ‘never’. We changed the response of all items into the same direction. Marital instability scores are the sum of the responses to all items, ranging from 5 to 20. Higher scores indicate a higher level of marital instability.

Table 1.

Marital instability index.

Item	Response
Have you ever thought your marriage might be in trouble?	1–4
Have you ever talked with family members, or friends about problems in your marriage?	1–4
How often have you thought that you don’t like living together with your spouse?	1–4
Have you ever thought of getting a divorce?	1–4
Have you ever discussed a divorce with members of your family or friends?	1–4

The key independent variables included in the analyses are:

Gender;

Migration distance, a binary variable to indicate whether the migration is inter-provincial or intra-provincial;

Migration pattern, which indicates whether the migrant moved with spouse or unaccompanied;

Distance between hometowns of husband and wife, indicated by four variables: within county, different county, different city, and different province; and

Wife’s income, made up of female migrants’ average monthly income and male migrants’ wives’ average monthly income.

Gender views are derived from an eight-item scale which indicates whether the respondent agrees the following.

A man’s main task in life is to earn a substantial salary, and a woman’s main task in life is to take care of family.

A husband should share equally in household chores if his wife works full time.

A husband should earn more than his wife.

A woman should put family first.

Women can have the same careers as men.

Compared to boys, girls should learn to do housework when they are young.

It is much better for everyone in the family for the wife to stay at home when the family needs someone to do so.

Woman would feel the happiest when she take care the family and children.

Respondents were asked whether they ‘strongly agreed’ (= score of 1), ‘agreed’ (2), ‘neither agreed nor disagreed’ (3), ‘disagreed’ (4), or ‘strongly disagreed’ (5) with the above statements. Gender view scores are the sum of the responses to all items, ranging from 8 to 40. Higher scores indicate higher levels of non-traditional gender views.

Marital satisfaction is an index based on responses to eight items.

I am satisfied with my marriage.

I am satisfied with my spouse’s role of wife or husband.

I am satisfied with my marital relationship.

I am satisfied with my spouse’s breadwinning ability.

I am satisfied with my spouse’s care of the family.

I am satisfied with my spouse’s companionship.

I am satisfied with my spouse’s loyalty in marriage.

I am satisfied with my sexual relationship with my spouse.

Respondents were asked whether they considered these aspects of marriage, according to these values: 1 = ‘strongly unsatisfactory’, 2 = ‘unsatisfactory’, 3 = ‘all right’, 4 = ‘satisfactory’, or 5 = ‘strongly satisfactory’. Marital satisfaction scores are the sum of the responses to all items, ranging from 8 to 40. Higher scores indicate higher levels of marital satisfaction.

We included the following set of control variables: birth cohort, age at first marriage, duration of marriage, age at first migration, and number of children: ‘number of children’ is a variable that indicates 0 (no child), 1 (one child), and 2 (more than one child). Birth cohorts include three categories: born in 1980 or later; born 1970–1980; and born before 1970.

We used Poisson Regression Analysis to analyze the data. The research was conducted in three steps. First, we described marital stability and the main variables by gender. Second, we analyzed the influence of the independent variables on marital stability. Third, we considered whether there were gender differences with these independent variables.

Results

Gender difference in marital instability, and main variables

Table 2 shows the means of all the variables included in the analysis by gender. Marital instability is higher for females, and the gender difference in marital instability is statistically significant (p < 0.001). More specifically, the results show that some 14.15 % of migrant workers have thought about divorce in the past year; with 17 % of females, and 10.24 % of males, having thought of divorce. About 8.35% of migrant workers had discussed divorce with their families or friends in the past year, with about 10.75% of females and only 5.71% of males reporting that they had discussed divorce. The results show that immigrant workers have a high rate of marital instability, and the rate is higher among females than males.

Table 2.

The means of all variables and their gender differences.

Variable	Male	Female	Difference
Marital instability	5.89	6.28	0.39 ***
Migration distance
Inter-provincial migration	0.61	0.58	0.03
Intra-provincial migration	0.39	0.42	−0.03
Migration pattern
Migrating unaccompanied	0.33	0.15	0.18 ***
Migration with spouse	0.67	0.85	−0.18***
Distance of hometown between husband and wife
Within county	0.66	0.65	0.01
Different county	0.12	0.09	0.03⁺
Different city	0.11	0.16	−0.05*
Different province	0.11	0.10	0.01
Wives’ income (log)	6.99	8.03	−1.04***
Wives’ income	2219.31	3591.32	−1372.01***
Gender view	21.13	23.29	−2.16***
Marital satisfaction	30.82	29.8	1.02**
Birth cohort
Born after 1980	0.58	0.6	−0.02
Born in 1970s	0.23	0.23	0
Born before 1970	0.19	0.16	0.13⁺
Age at first migration	24.59	23.01	1.58***
Marriage duration	15.8	14.5	1.3*
Age at first migration	21.85	22.58	−0.73⁺
Number of Children
0	0.06	0.09	−0.03
1	0.40	0.41	−0.01
2 and more	0.54	0.50	0.04⁺

Note. ⁺p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001.

There is gender difference in migrating unaccompanied, wife’s income, gender views, and marital satisfaction. Males are significantly more likely than females to migrate unaccompanied. Migrant females have higher incomes than male migrants’ wives. Meanwhile, female migrants also report significantly higher levels of non-traditional gender views than male migrant workers, and they report lower levels of marital satisfaction than male migrants.

Determinants of marital instability

Table 3 presents the regression results of the ways in which suggested factors are related to the marital instability of migrant workers. Model 1 reveals that female migrant workers are more likely than male migrant workers to have high levels of marital instability (r = 0.36; p < 0.001). Even when we add the control variables to the equation in the next few models, the significant relationship remains (r = 0.33; p < 0.001). The results support Hypothesis 1.

Table 3.

The determinants of marital instability among migrant workers.

	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Gender (female = 1)	0.36***	0.40***	0.48***	0.48***	0.42***	0.33***
	0.06	0.06	0.07	0.07	0.08	0.08
Birth cohort^b
Born after 1980		0.25	0.35⁺	0.29	0.29	0.34
		0.20	0.20	0.21	0.21	0.22
Born in 1970s		−0.001	0.06	0.05	0.05	0.06
		0.13	0.13	0.13	0.13	0.14
Age at first marriage		0.01	0.02	0.01	0.01	0.01
		0.01	0.01	0.011	0.01	0.01
Marriage duration		−0.01	−0.003	−0.004	−0.002	−0.006
		0.01	0.009	0.009	0.009	0.01
Age at first migration		0.005	0.003	0.002	0.002	0.003
		0.004	0.004	0.004	0.004	0.004
Number of children^a
1		0.43***	0.37**	0.35**	0.35**	0.30**
		0.12	0.12	0.13	0.13	0.13
2 and more		0.22⁺	0.13	0.10	0.11	0.03
		0.13	0.13	0.14	0.14	0.14
Migration distance (inter-provincial migration = 1)			0.12⁺	0.16*	0.17*	0.29***
			0.07	0.07	0.07	0.07
Migration pattern (migrating unaccompanied = 1)			0.37***	0.39***	0.41***	0.30***
			0.07	0.07	0.07	0.07
Distance of hometown between husband and wife^c
Different county			0.18*	0.21*	0.23*	0.24*
			0.09	0.10	0.10	0.10
Different city			−0.15	−0.11	−0.12	−0.17
			0.10	0.10	0.10	0.10
Different province			0.24**	0.28**	0.26**	0.21*
			0.10	0.10	0.10	0.096
Wives’ income (log)				0.023	0.024	0.038
				0.03	0.03	0.03
Gender view					0.03***	0.017*
					0.008	0.008
Marital satisfaction						−0.1***
						0.005
Cons	−0.11*	−0.79⁺	−1.22**	−1.27*	−1.99***	1.09*
	0.047	0.43	0.45	0.52	0.55	0.59
N	1066	1030	1014	951	948	921
LR chi² (13)	***	***	***	***	***	***

Note. ^areference group is 0 (no child); ^breference group is born before 1970; ^creference group is within county. ⁺p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; the data in the parentheses are standard errors.

Migration has a positive influence on migrant workers’ marital instability. There is positive association between migration distance and marital instability. Model 6 reveals that after all variables are controlled, cross-province migration has a significant positive effect on marital instability (r = 0.29; p < 0.001). The results support Hypothesis 2a.

Migration pattern has a significant association with migrant workers’ marital instability. Even when other variables are added to the equation, the effects of migrating unaccompanied on marital stability are still significant at the 0.001 level. However, after marital satisfaction is added to the equation, the coefficient is reduced from 0.41 (Model 5) to 0.30 (Model 6). The results reveal that there is a mediation effect of marital satisfaction (p < 0.1). The change suggests that marital satisfaction may modify the effect of migrating unaccompanied on marital instability. Some migrant workers who migrate unaccompanied may have low levels of marital satisfaction, which leads to high levels of marital instability. These results support Hypotheses 3a and 3b. Spouses coming from different hometowns is positively related to marital instability among migrant workers (p < 0.05), and this relationship is partly due to marital satisfaction. Model 6 shows that when we add marital satisfaction and other variables to the equation, the coefficient is reduced from 0.26 to 0.21 at the level of 0.05. The results show that a longer distance between hometowns of spouses poses some risk to the marital stability of migrant workers. The results support Hypothesis 4a: however, the results also do not support the mediation effect of marital satisfaction (p < 0.1) and thus do not support Hypothesis 4 b.

We did not find significant association between wife’s income and marital stability of migrant workers. The results do not support Hypotheses 5a and 5b.

Gender views are positively related to marital instability, and the relationship is partly shaped by marital satisfaction. The results of Model 5 and Model 6 show that gender views are significantly and positively related to marital instability (r = 0.017). After controlling for marital satisfaction, the correlation decreases from 0.03 to 0.017, and the statistical significance decreases from the 0.001 level to the 0.05 level. The results reveal that non-traditional gender views lower the level of migrant workers’ marital satisfaction and then exacerbate the level of marital instability. The results support Hypotheses 6a and 6b.

Gender difference in the determinants of marital instability among migrant workers

Table 4 presents the results for determinants of marital instability for males and females separately; they reveal a significant gender difference in the determinants of marital instability among migrant workers.

Table 4.

Gender difference in determinants of marital instability among migrant workers.

	Male		Female
	Model 7	Model 8	Model 9	Model 10
Birth cohort^b
Born after 1980	−0.15	−0.23	0.75**	0.77**
	0.34	0.37	0.27	0.27
Born in 1970s	0.11	0.077	0.007	−0.015
	0.21	0.24	0.17	0.18
Age at first marriage	0.016	0.014	0.008	−0.004
	0.017	0.018	0.015	0.02
Marriage duration	−0.017	−0.027⁺	0.01	0.004
	0.015	0.017	0.01	0.01
Age at first migration	−0.008	−0.009	0.013⁺	0.015*
	0.006	0.007	0.006	0.006
Number of children^a
1	0.36⁺	0.47*	0.43**	0.36*
	0.22	0.23	0.155	0.16
2 and more	0.14	0.31	0.16	0.08
	0.23	0.25	0.17	0.17
Migration distance (inter-	0.39**	0.52***	0.07	0.21*
provincial migration = 1)	0.125	0.13	0.086	0.09
Migration pattern (migrating	0.07	0.18	0.69***	0.41***
unaccompanied = 1)	0.11	0.12	0.095	0.10
Distance of hometown between husband and wife^c
Different county	0.004	−0.03	0.38**	0.36**
	0.16	0.17	0.12	0.13
Different city	−0.09	−0.23	−0.10	−0.12
	0.17	0.20	0.12	0.13
Different province	−0.009	0.03	0.43***	0.30*
	0.16	0.16	0.12	0.12
Wives’ income (log)	0.06	0.11***	−0.17***	−0.16**
	0.04	0.04	0.05	0.055
Gender view	0.05***	0.03*	0.02*	0.01
	0.014	0.015	0.01	0.01
Marital satisfactory		−0.10***		−0.09***
		0.008		0.006
Cons	−2.17*	0.93	−0.45	2.55*
	0.88	0.96	0.75	0.80
N	466	452	482	469
LR chi² (13)	***	***	***	***
Pseudo R²	0.041	0.161	0.075	0.176

Note. ^areference group is 0(no child); ^breference group is born before 1970; ^creference group is within county. ⁺p < 0.1; *p < 0.05; **p < 0.01; ***p < 0.001; the data in the bracket are standard errors.

Looking first at the influence of migration on marital instability for males, only migration distance has a positive relationship with marital instability at the 0.001 level. For females, migration pattern and spouses coming from different hometowns are significantly related to marital instability. Cross-provinces migration leads to higher levels of marital instability for males only. Conversely, migrating unaccompanied, and spouses coming from different counties and provinces, are associated with more marital instability for females only. The results support Hypotheses 2b, 3c and 4c. The wife’s income appears to be an important determinant of marital instability for both males and females when only either males or females are included in the analysis. Model 8 reveals that wives’ income is positively related to marital instability for male migrant workers (r = 0.11; p < 0.001) and for female migrant workers (r = −0.16; p < 0.01), although the impacts are opposite, which does not support Hypothesis 5c.

The relationship of gender views and marital instability differs for male and female migrant workers. Models 7 and 8 reveal that gender views have a direct influence on male migrant workers even when controlling for marital satisfaction. The decrease of coefficient and confidence interval also suggests that marital satisfaction reduces the influence of gender views on males’ marital instability. However, the results of Models 9 and 10 show that when marital satisfaction is controlled, the statistical significance of the association between gender views and marital instability disappears for female migrant workers, revealing that non-traditional gender views lower the marital satisfaction of female migrant workers and thus weaken their marital stability. The four-step mediation test reveals that there is a mediation effect of marital satisfaction (p < 0.1). These findings show that there is gender difference in the influence of gender views on marital instability, which partly supports Hypothesis 6c.

Finally, the results suggest that marital satisfaction is a key contributor to marital stability for both men and women. Marital satisfaction lowers the level of marital instability for both female and male migrant workers.

Conclusion

Although research is beginning to address the relationship between migration and marital stability, most of the studies focus on the residents at the place of origin and then explore the effects of migration and non-migration on the rate of divorce. There has been little research to address the impact of the migration experience on marital instability among migrant workers. At the same time, studies on migrant workers’ marital instability lack a gender perspective, so we still do not know whether there are gender differences in marital instability and its determinants among migrant workers. This study extends work on this topic by looking at the influence of the migration experience and its related factors on marital instability, exploring gender differences in marital stability and its determinants. Our results suggest that the migration experience and related factors are significantly related to marital instability, and that there are gender differences in marital instability and its determinants.

The rewards received from marriage, the barriers to divorce, and the alternatives to marriage all have significant relationships with marital instability. In the case of migrant workers in China, we expected and found that migration distance, migrating unaccompanied, distance between hometowns of husband and wife, wife’s income, gender views, and marital satisfaction are also related to marital instability. The relationships of these variables with marital stability are different for men and women.

We suggest that the findings have major implications. First, our study shows the value of using the exchange perspective to understand marital instability among migrant workers in China. Marriage can be unstable when husband and wife find the relationship less rewarding. Migration increases the availability of alternative relationships and weakens the barriers, so that marital instability among migrant workers increases. If this is the case, more attention should be given to the marriages of migrant workers. Second, the findings suggest that the analysis should consider in full the unique experiences of migrant workers. The rapid urban development in China has created opportunities for migrants from rural areas. However, the hukou system shapes their experiences in the city. Migrating unaccompanied and leaving wives behind changes and weakens the barriers to marital instability. Studies should consider fully the unique context of migrant workers at home and in the city in order to understand better their marital instability. Finally, our study has shown clearly that the ways in which the suggested factors are related to marital stability differ between males and females. The findings suggest that future studies should develop further a perspective to understand the gender differences in full.

Footnotes

Funding

This research received a grant from the National Social Science Foundation of China (No. 14CRK020) and the China Postdoctoral Science Foundation (grant number 2017M610619).

References

Amato

Booth

(1995) Changes in gender role attitudes and perceived marital quality. American Sociological Review 60(1): 58–66.

Becker

(1981) A Treatise on the Family (enlarged edition), Cambridge, MA: Harvard University Press.

Becker

Landes

Michael

(1977) An economic analysis of marital instability. Journal of Political Economy 85(6): 1141–1187.

Booth

Edwards

(1985) Age at marriage and marital instability. Journal of Marriage and Family 47(1): 67–75.

Booth

White

(1980) Thinking about divorce. Journal of Marriage and the Family 42(3): 605–616.

Booth

Edwards

Johnson

(1991) Social integration and divorce. Social Forces 70(1): 207–224.

Booth

Johnson

White

Edwards

(1986) Divorce and marital instability over the life course. Journal of Family Issues 7(4): 421–442.

(2010) The impact of migration on union dissolution. Comparative Economic & Social Systems 151(5): 105–112. (in Chinese).

Frank

Wildsmith

(2005) The grass widows of Mexico: Migration and union dissolution in a binational context. Social Forces 83(3): 919–947.

10.

Fuller

Edwards

(2004) Gender differences in the psychological well-being of married men and women: An Asian case. The Sociological Quarterly 45(2): 355–378.

11.

Gao

(2011) Rural divorce rate and migrant rural workers: A panel data research study based on Chinese 2003–2009 rural data. World Economy 183(10): 55–69. (in Chinese).

12.

Glenn

(1991) Quantitative research on marital quality in the 1980s: A critical review. In: Booth

(ed.) Contemporary Families: Looking Forward, Looking Back, Minneapolis, MN: National Council on Family Relations, pp. 28–41.

13.

Greenstein

(1996) Gender ideology and perceptions of the fairness of the division of household labor: Effects on marital quality. Social Forces 74(3): 1029–1042.

14.

Guttentag

Secord

(1983) Too Many Women? The Sex Ratio Question, Beverly Hills, CA: Sage Publications.

15.

Heaton

Albrecht

(1991) Stable unhappy marriages. Journal of Marriage and Family 53(3): 747–758.

16.

Heaton

Blake

(1999) Gender differences in determinants of marital disruption. Journal of Family Issues 20(1): 25–45.

17.

Hill

(1988) Marital stability and spouses' shared time: A multidisciplinary hypothesis. Journal of Family Issues 9(4): 427–451.

18.

Johnson

Caughlin

Huston

(1999) The tripartite nature of marital commitment: Personal, moral, and structural reasons to stay married. Journal of Marriage and the Family 61: 160–177.

19.

Kalmijn

Poortman

A-R

(2006) His or her divorce? The gendered nature of divorce and its determinants. European Sociological Review 22(2): 201–214.

20.

Kessler

McLeod

(1984) Sex differences in vulnerability to undesirable life events. American Sociological Review 49(5): 620–631.

21.

Knoester

Booth

(2000) Barriers to divorce: When are they effective? When are they not? Journal of Family Issues 21(1): 78–99.

22.

Levinger

(1965) Marital cohesiveness and dissolution: An integrative review. Journal of Marriage and Family 27(1): 19–28.

23.

Levinger

(1976) A social psychological perspective on marital dissolution. Journal of Social Issues 32(1): 21–47.

24.

Mirowsky

Ross

(1989) Social Causes of Psychological Distress, New York: Aldine de Gruyter.

25.

Shi

(2015) Influence of rural mobility on the divorce rate. Chinese Journal of Population Science 35(5): 104–112. (in Chinese).

26.

Previti

Amato

(2003) Why stay married? Rewards, barriers, and marital stability. Journal of Marriage and Family 65(3): 561–573.

27.

Schoen

Astone

Kim

Rothert

Standish

(2002) Women’s employment, marital happiness, and divorce. Social Forces 81(2): 643–662.

28.

Simon

(1998) Assessing sex differences in vulnerability among employed parents: The importance of marital status. Journal of Health and Social Behavior 39(1): 38–54.

29.

South

Lloyd

(1995) Spousal alternatives and marital dissolution. American Sociological Review 60(1): 21–35.

30.

Turner

Roszell

(1994) Psychosocial resources and the stress process. In: Avison

Gotlib

(eds) Stress and Mental Health: Contemporary Issues and Prospects for the Future, New York: Plenum Press, pp. 179–212.

31.

Udry

(1981) Marital alternatives and marital disruption. Journal of Marriage and Family 43(4): 889–897.

32.

White

Booth

(1991) Divorce over the life course: The role of marital happiness. Journal of Family Issues 12(1): 5–21.

33.

(2012) The influencing factors of divorce risk: A comprehensive model. Sociological Studies 27(2): 109–125. (in Chinese).

34.

(2002) An analysis of China's provincial differences in divorce rates. Population Research 4: 28–35. (in Chinese).

35.

Qiu

(2013) The impact of children on divorce risk. Sociological Studies 28(4): 26–48. (in Chinese).

36.

Lin

(1998) An analysis of China’s divorce situation and its reason. Population and Economic 3: 22–28. (in Chinese).

37.

(1999) The Chinese marital stability and its determinants. Chinese Journal of Population Science 7: 7–12. (in Chinese).

38.

Zhang

(1997) The contemporary Chinese divorce situation. Population Research 6: 26–31. (in Chinese).

39.

Zeng

(1995) A Study of China’s Divorce in 1980s, Beijing: Peking University Press. (in Chinese).