Financial Strain and Intimate Partner Violence Against Married Women in Postreform China: Evidence From Chengdu

Abstract

The primary goal of this study is to link both subjective and objective indicators of financial strain to two distinct dimensions of intimate partner violence (IPV) against women—the husband’s violent behavior and gendered control—in postreform China. The data for this study were drawn from a community survey conducted in Chengdu, the capital of Sichuan province, in 2017 (N = 340). By utilizing the family stress model and quantitative methods, the following results emerged from a series of multivariate statistical analyses: (a) among married women, self-perceived financial strain is significantly and positively associated with the risk of experiencing the husband’s perpetration of violent behavior and financial control; (b) low family income significantly elevates the likelihood of the husband’s exertion of personal and financial control over the wife, albeit the effect is weaker for financial control; and (c) unemployment of the husband significantly increases the likelihood of the husband’s exertion of financial controlling behavior against his wife. These results underscore the importance of gender and income inequalities in research on IPV against women in postreform China. These findings also cross-culturally substantiate the family stress model that has been utilized previously to examine the multifaceted associations between economic hardship and IPV in the U.S. Policy makers, academic researchers, and health practitioners are urged to recognize both subjective and objective financial strains as social and psychological determinants of IPV against women in postreform China.

Keywords

intimate partner violence controlling behavior financial strain family stress model postreform China

Introduction

Since the 1980s, a small but growing body of research on intimate partner violence (IPV) in the People’s Republic of China has emerged (Chan, 2011, 2014; Hou, Yu, Ting, Sze, & Fang, 2011; Lin, Sun, Liu, & Chen, 2018; Meng & Chan, 2000; Parish, Wang, Laumann, Pan, & Luo, 2004; Xu, 1997; Xu et al., 2005). Using a variety of community surveys, social scientists and public health scholars have documented that similar to industrialized and developing countries around the world, IPV against women has been prevalent in both the reforming and postreform eras of China. For example, based on a probability community survey conducted in 1987 in a major southwestern Chinese city, researchers reported that more than 50% of ever-married women experienced some form of IPV, either physical, psychological, or both, throughout the course of their marriages (Xu, 1997). Another study conducted among clinical patients in a southern Chinese city revealed a lower lifetime prevalence of both physical (38%) and sexual violence (16%; Xu et al., 2005). A more recent study from Beijing showed a higher lifetime prevalence of physical, psychological, and sexual IPV against women: 22%, 41.6%, and 23%, respectively (Hou et al., 2011). One of the few population-based and nationally representative studies estimated that about 34% of women experienced physical violence during their current relationship (Parish et al., 2004). Several studies, including those conducted in Hong Kong, identified patriarchal ideologies, women’s low contribution to household income, men’s low socioeconomic status, poverty, alcohol consumption, and drug use as risk factors of IPV against women (Chan, 2011, 2014; Lin et al., 2018; Parish et al., 2004). In addition, a small number of studies also documented adverse health outcomes of IPV against women, including compromised physical, mental, and sexual health (Parish et al., 2004; Xu, Campbell, & Zhu, 2001). Although these studies are informative, there has been no systematic research exploring the potential linkages between financial strain and IPV, particularly in the postreform era in the People’s Republic of China. Consequently, there are no concerted efforts in prior research to theorize these potential linkages. This study is developed to address these research limitations.

This study advances prior work in several distinct ways. First, this study adopts the definition of IPV against women from the World Health Organization (WHO) that encompasses a distinct dimension of IPV; that is, coercive or gendered control over women by their male intimate partners. According to the WHO definition, IPV is conceptualized as “behavior by an intimate partner or ex-partner that causes physical, sexual or psychological harm, including physical aggression, sexual coercion, psychological abuse and controlling behaviors” (WHO, 2017). To the best of our knowledge, few studies, if any, have examined these twin theoretical dimensions—violent behaviors or acts and gendered control—simultaneously in the context of Chinese families.

Second, this study utilizes the stress process model and its offshoot, the family stress model, to test several hypotheses by linking unemployment of the husband and family economic hardship to IPV against his wife in the context of postreform China. Over the past two decades, gender and income inequalities as well as poverty have increasingly become notable socioeconomic challenges in both urban and rural areas (Wu, 2004; Xie & Zhou, 2014). It has been argued that the transition from the formerly centralized economy to a market economy has resulted in a new stratification system that has disproportionately affected certain segments of the population, such as dislocated and laid-off urban workers, migrant workers, and rural residents (Xu, Zeng, Zheng, & Flatt, 2010). Prior research has shown that poverty-induced economic distress exerted profound impacts on the occurrence of IPV against women in contemporary China (Chan, 2011; Lin et al., 2018; Parish et al., 2004).

Third, this study extends prior research on IPV to investigate the role of self-perceived financial strain on IPV against women. Much of the extant research has largely overlooked potential linkages between subjective financial strain and IPV among women in general, and among Chinese women in particular. This study can help overcome this research oversight as researchers have already established that perceptions of economic hardship can lead to elevated levels of frustration and distress (Dunn et al., 2008; Glei, Goldman, & Weinstein, 2018), thus contributing to IPV.

In the pages that follow, we first elaborate on the definition of IPV. We then review and follow the logic of the family stress model to link financial strain, unemployment of the husband, low family income, and self-perceived financial strain to IPV sequentially. Based on the reviewed theoretical arguments and empirical literature, we develop four hypotheses that will be tested using the survey data from Chengdu, China.

Gendered Control as a Distinct Dimension of IPV

In prior research conducted in both the West and China, the construct of IPV, particularly against women, is often confined to violent behaviors, and measured by different versions of the Conflict Tactics Scales (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). These acts typically include psychological (verbal or emotional), physical, and sexual violence (Capaldi, Knoble, Shortt, & Kim, 2012; Xu et al., 2001). As a result, the male partner’s use of coercive or gendered control is routinely excluded or neglected (Anderson, 2005; Dobash, Dobash, Wilson, & Daly, 1992; Johnson, 1995, 2011). Although some studies have considered gendered controlling behaviors, more often than not, they have conceptualized these behaviors as risk factors, rather than a distinct dimension of IPV against women (Aizpurua, Copp, Ricarte, & Vázquez, 2021; Graham-Kevan, & Archer, 2008; Kishor & Johnson, 2004; Ogland, Xu, Bartkowski, & Ogland, 2014). Given these inconsistent and diverse conceptualizations, the present study expands on Michael Johnson’s typology of common couple violence, or situational couple violence, to explicitly embrace the male partner’s controlling behaviors as manifestations of gendered control (Johnson, 1995, 2006). This effort follows Anderson’s (1997) lead to integrate both violent behaviors (e.g., the common couple violence or situational couple violence) and gendered control (e.g., the husband’s economic and personal control over his wife) into one coherent and overarching conceptualization of IPV against women. Hence, unlike Schneider et al.’s (2016) study, violent behaviors and gendered control in this study are conceptualized as two distinct dimensions of the IPV construct. Therefore, they must be examined separately in postreform China.

In addition to maintaining parity to the WHO definition of IPV, this conceptualization is also consistent with a recent study that investigated the relationship between the Great Recession and IPV in the United States by utilizing the Fragile Families and Child Well-being Study (Schneider et al., 2016). A similar conceptual attempt is observed as well in another study conducted in southern China (Lin et al., 2018). Such an inclusive conceptualization can facilitate a better understanding of the relationship between financial strain and IPV against women in postreform China, given the increased and pervasive gender and income inequalities since the economic reform (He & Wu, 2017; Ji, Wu, Sun, & He, 2017; Xie & Zhou, 2014).

Financial Strain and IPV

Financial strain refers to the lack of financial resources to meet family needs, due to low income, job loss, temporary or chronic unemployment (Glei et al., 2018). In the family violence literature, financial strain is often interchangeable with such terms as “economic hardship” or “financial difficulties.” To link financial strain to IPV against women, this study follows the family stress model, which was developed initially to examine the relationship between economic hardship and child well-being (Conger, Conger, & Martin, 2010). Central to this framework is the notion that unemployment and economic hardship, that is, financial strain, can lead to economic distress, which may, in turn, lead to marital conflict, as well as a decline in the quality of parenting and child well-being. Theoretically, the family stress model is rooted in the stress process model, which suggests that adverse life events, such as job loss and acute or chronic financial strain, may manifest in undesirable outcomes, including depression and poor physical or sexual health (Parish et al., 2004; Pearlin, 1999). Scholars argue that in the context of family life, the extreme stress associated with financial difficulties is one of the primary mechanisms that undergirds the association between economic hardship and IPV against women (Fox, Benson, DeMaris, & VanWyk, 2002). In other words, the family stress model modifies the stress process model by explicitly linking economic stressors to IPV against women.

It has been argued that a family’s financial strain can undermine the psychological well-being of family members, which, in turn, contributes to other stressors. These stressors, such as relationship or parenting difficulties, create a situation wherein IPV is more likely to occur. As coping resources and the ability to manage financial crises diminish, the likelihood of IPV increases (Benson, Fox, DeMaris, & Van Wyk, 2003; Fox & Benson, 2006). Similarly, the frustration–aggression hypothesis suggests that aggression or violence is one of the several possible consequences of frustration, including frustration due to financial difficulties (DeWall, Anderson, & Bushman, 2011).

Recent studies conducted in the United States revealed that at the individual-level economic vulnerability, employment instability, and perceptions of economic strain were associated with IPV (Benson et al., 2003; Fox & Benson, 2006). This linkage was similarly observed for the husband’s loss of employment (Fox et al., 2002), low family income (Cunradi, Caetano, & Schafer, 2002), and economic hardships such as poverty (Golden, Perreira, & Durrance, 2013; Hardie & Lucas, 2010). To the best of our knowledge, however, no such systematic research has been conducted in postreform China.

Husband’s Unemployment and IPV

As a major contributor to financial strain, unemployment of the husband has long been recognized as a socially structured stress (Benson et al., 2003). Theoretically, there are three mechanisms underlying the associations between unemployment of the husband and the increased risk of IPV against his wife. First, in congruence with the family stress model, unemployment of the husband can be conceptualized as an acute or chronic stressor for both the husband and the family, thus escalating the risk of IPV against the wife. Second, a husband’s unemployment can undercut his breadwinner role, and compromise his sense of economic control and financial security. These adverse events may facilitate the tendency of the husband to exert greater economic and personal control over his wife. Third, unemployed husbands may simply stay at home more often than their employed peers, thus having more opportunities to interact negatively with their wives, which may result in IPV (Benson et al., 2003).

With these theoretical mechanisms in mind, unemployment of the husband may contribute to IPV against women in the context of Chinese families for two important reasons. First, during the time since the onset of the Chinese economic reform in the early 1980s, urban unemployment among workers in the state and collectively owned enterprises has increased (Xu et al., 2010). Second, privatization and marketization of the Chinese economy have exacerbated the gender earnings gap in the Chinese labor markets (He & Wu, 2017). Both of these changes—increased urban unemployment and the gendered earnings gap—have important implications for the potential effects of the husband’s unemployment on IPV against the wife in postreform China.

Low Family Income and IPV

In similar fashion to the husband’s unemployment, financial strain induced by low family income may compromise the ability of a couple to pay for immediate and essential expenses in their daily lives. These financial stressors may coalesce with other stressors such as parenting, unequal division of housework, and relationship maintenance, thereby collectively undermining the psychological well-being of both partners (Golden et al., 2013). The high levels of family stress triggered by economic hardship and experienced by the couple may generate a stressful context, making IPV against the wife more likely to occur. At the individual level, researchers have reported that financial difficulties are the most consistent and significant correlates of IPV against women in the United States (Rennison & Welchans, 2000). At the macro level, one recent study indicated that economic uncertainty as indicated by high unemployment rates increased IPV during difficult times, such as the Great Recession (Schneider et al., 2016). Regardless of the differences in the sociocultural context, it can be argued that couples experiencing economic distress due to low or unstable family income may encounter or engage in more frequent arguments and conflicts that can ultimately lead to an increased likelihood of IPV against women. However, prior research conducted in China has yielded inconsistent or insignificant findings across different locales in China, perhaps due largely to the heterogeneity of study sites and study subjects involving such high-risk populations as divorced women, women who utilized shelters, pregnant women, rural women, or older women (Hou et al., 2011; Lin et al., 2018; Xu et al., 2005). To address these research inconsistencies, additional attention should be given to the effects of diminished labor market opportunities and the decline in economic fortunes among those who are less educated and dislocated in the general population on IPV since the economic reform.

Self-Perceived Financial Strain and IPV

Although IPV has long been associated with financial strain (Golden et al., 2013), little scholarly attention has been given to the role of subjective or self-perceived financial stress on IPV. This omission is puzzling because subjective financial strain, defined as self-assessed or perceived financial difficulties and unemployment circumstances (Glei et al., 2018), has been included as a component of the family stress model (e.g., the family’s perceptions of the stressor). This subjective dimension can, arguably, have important implications for research on IPV against women as subjective evaluations also reflect gender and income disparities (Glei et al., 2018). In fact, much like the objective counterparts, such as the husband’s loss of employment and low family income, subjective financial strain is also reportedly associated with a host of negative health outcomes, including increased mortality (Glei et al., 2018; Szanton et al., 2008), decline in mental health (Dunn et al., 2008; Selenko & Batinic, 2011; Wilkinson, 2016) and physical health (Arber, Fenn, & Meadows, 2014; Shippee, Wilkinson, & Ferraro, 2012).

Given these theoretical arguments made in the health and family violence studies, there is sufficient reason to believe that subjective financial strain may increase the risk of IPV because of the shared social and psychological sources of distress, especially economic hardship. It can be argued that subjective financial strain may heighten the risk of IPV against women above and beyond the effects of objective indicators of financial strain. To date, however, no research has been conducted to assess such linkages between the subjective dimension of financial strain and IPV against women, either in the United States or in postreform China.

Hypotheses

Informed by the family stress model and the literature on IPV reviewed above, in the present study, we develop and test the following four research hypotheses.

Hypothesis 1: The wife’s self-perceived financial strain will be significantly and positively associated with her lifetime experience of the husband’s violent behaviors.

Hypothesis 2: Unemployment of the husband and low family income will be significantly and positively associated with the wife’s lifetime experience of the husband’s violent behaviors.

Hypothesis 3: The wife’s self-perceived financial strain will be significantly and positively associated with the likelihood of the husband’s financial and personal control.

Hypothesis 4: Unemployment of the husband and low family income will be significantly and positively associated with the likelihood of the husband’s financial and personal control.

Research Methodologies

Data

Data for this study were drawn from the Survey on Fertility Desire and Attitudes toward Life in Chengdu, conducted by Sichuan University in 2017. Located in southwestern China, Chengdu is the capital of Sichuan province and serves as its socioeconomic and cultural center. Administratively, Chengdu encompasses 11 urban districts, five county-level cities, and four counties that are largely rural areas. By 2016, the population size for the greater Chengdu area was 13.99 million, with 56% of its population residing in the urban communities (Chengdu Yearbook, 2017).

For the present study, a multistage sampling technique was combined with several nonrandom sampling strategies to select 800 targeted survey respondents. In the first stage, the research team from Sichuan University randomly selected seven urban districts (out of 11), one county-level city (out of five), and one rural county (out of four) from all Chengdu administrative areas based on their unique social and economic characteristics (Table 1). In the second stage, the research team randomly selected 18 streets from the nine chosen administrative areas. In the third stage, the research team randomly selected 51 urban and rural communities from the 18 streets. In the final stage, 642 respondents were reached for face-to-face interviews by way of quota sampling, simple random sampling, and/or convenience sampling, from the 51 communities with a response rate of 80.3%. Due to a limited research budget, a moderate sample size and a relatively small number of urban streets, rural villages, or communities were chosen following a series of predetermined quotas to ensure an adequate representation of different gender and age groups, occupational categories, and migrant status (i.e., migrant workers). As the survey was originally designed to study fertility desire, the average age of respondents was younger than that of the general population of the city, with only 6.7% of respondents being above 50 years of age versus 27% for the entirety of the city (Chengdu Yearbook, 2017). All interviews were conducted by professional interviewers from a well-established survey company, either at respondents’ homes or another location in their local communities.

Table 1.

Sample Distribution Across Nine Administrative Districts.

	Full Sample		Analytic Sample
	Frequency	Percent	Frequency	Percent
District 1	47	7.3	25	7.4
District 2	67	10.4	39	11.5
District 3	66	10.3	36	10.6
District 4	81	12.6	39	11.5
District 5	98	15.3	56	16.5
District 6	62	9.7	31	9.1
District 7	52	8.1	26	7.6
District 8	64	10.0	40	11.8
District 9	105	16.4	48	14.1
Total	642	100	340	100

To increase the likelihood for the respondents to provide honest answers, several survey sections that contain sensitive questions, such as IPV victimization and fertility behaviors, were filled out by the respondents. To ensure privacy, the interviewers were instructed to make sure no other individuals were present during the interview. As part of the research design, all respondents were asked if they were aware of resources that were available if they needed assistance, including community service and resource centers, police contact information, and the All-China Women’s Federation contact information in the city. Because this was a fertility desire survey, the interviewers were not specifically trained to handle the situations associated with IPV. The ethical standards and procedures for research with human subjects for the present study were approved by Sichuan University.

To test the hypotheses as outlined previously, this study utilized an analytic subsample of 340 married women (about 53% of the original sample). As displayed in Table 1, both the full sample and the subsample of married women from Chengdu approached even distributions across the nine chosen districts, with District 9 and District 5 having slightly higher numbers of respondents than the other districts. With very minor variations, the subsample of married women proportionately matched the full sample across the chosen districts. Therefore, the selection of the subsample of married women did not significantly deviate from the original sample distribution across the selected administrative districts or areas in the city.

Measures

Dependent variables

Consistent with the definition of IPV provided by WHO (2017), the present study used tripartite measures to operationalize IPV against women in Chengdu, China. The first dependent variable is a composite index of violent acts against the wife, consisting of four lifetime measures of physical, psychological, and sexual IPV, as experienced and reported by married Chengdu women. Respondents were asked if the husband has ever (a) hit or beat you, (b) insulted you or called you names, (c) ignored you for several days, and (d) forced you to have sex. The response categories are 0 = never, 1 = occasionally, 2 = sometimes, and 3 = often. These responses were summed to form a countlike composite variable of IPV against women, with Cronbach’s alpha = .60. The second dependent variable is a single questionnaire item asking respondents if their husbands had ever confined their personal freedom. The response categories were dummy-coded with 1 = yes and 0 = no. The third dependent variable is a single questionnaire item that asked respondents if their husbands had ever controlled them financially. The response categories were dummy-coded with 1 = yes and 0 = no. It is important to note that, though these two single-item measures were utilized to operationalize IPV by prior studies, together with other violent acts (e.g., Schneider et al., 2016), the construct of gendered control, i.e., financial and personal control, as delineated by WHO (2017), was operationalized separately in the present study. This operationalization allows the conceptual distinguishing of the two different dimensions of IPV—violent behaviors and gendered control—against married women in Chengdu, China.

Independent variables

One subjective and two objective measures were used to operationalize the construct of financial strain. To gauge subjective or self-perceived financial strain, respondents were asked, “In the past year, did you or your family encounter the following problems?” There were seven questionnaire items, (a) poor housing conditions and could not afford to build a house or purchase an apartment, (b) could not pay for children’s education, (c) could not afford health or medical care, (d) price was high to affect the standard of living, (e) income was low, (f) family members experienced unstable employment or unemployment, and (g) cost was high to maintain social networks or friendship ties (e.g., could not buy gifts for friends). All response categories were first dichotomized with 1 = yes and 0 = no, then averaged as an index variable, with Cronbach’s alpha = .717. Objective measures of financial strain included (a) monthly family income in Chinese Yuan (RMB) and (b) unemployment of the husband with 1 = yes and 0 = no.

Control variables

In this study, we controlled for a number of sociodemographic variables, akin to previous studies. Unemployment of the wife was dummy-coded with 1 = yes and 0 = no. Educational attainment of the husband and the wife was measured as levels of education, ranging from 1 = no formal education to 8 = postgraduate education. Both were treated as continuous variables. Respondent’s age was a continuous variable in years. Membership of the wife in the Communist Party to measure her political status, as well as urban residence, were dummy-coded as 1 = yes and 0 = no. Number of children and the measure of the husband’s marital-decision making power were also included. The husband’s decision-making power, with regard to family expenses, buying big-ticket items, buying a home or apartment, running the family business, family investment and children’s education, is an additive composite variable, with Cronbach’s alpha = .709. The descriptive statistics for all study variables are presented in Table 2.

Table 2.

Sample Characteristics.

Variables	n	Minimum	Maximum	M	SD
Dependent variables
Intimate partner violence	340	0	5	0.41	0.81
Personal control	340	0	1	0.04	0.18
Financial control	340	0	1	0.04	0.20
Independent variables
Perceived financial strain	340	0	1	0.33	0.27
Family monthly income	340	0	9	7.80	0.96
Husband unemployed	340	0	1	0.04	0.18
Control variables
Wife unemployed	340	0	1	0.20	0.40
Husband education	340	1	8	4.63	1.63
Wife education	340	1	8	4.34	1.67
Wife age	340	21	64	39.21	8.43
Wife party member	340	0	1	0.06	0.25
Wife urban registration	340	0	1	0.67	0.47
Number of children	340	0	4	1.07	0.65
Husband dominant decision maker	340	0	1	0.08	0.17

Analytic Strategies

As mentioned previously, the husband’s violent behaviors were constructed as a composite, countlike variable, ranging from 0 to 5 (“Intimate partner violence” in Table 2). This variable was analyzed using negative binomial regression models, as the variable was overdisbursed (Long & Freese, 2014); that is, the variance is greater than the mean (.66 > .41). Table 2 also shows that both the personal control and financial control variables were skewed, with a small number of respondents answering “yes,” suggesting that in the multivariate regression analysis, there would be many cells with low counts. As such, there would be a problem of small-sample bias in the regular maximum likelihood estimation of the conventional logistic regression models, which could sharply underestimate the probability of low frequency or rare events (King & Zeng, 2001). The degree of bias depends on how small the number is for respondents who answered “yes” or “no.” To bias-adjust regression coefficient estimates in our multivariate statistical modeling, Firth’s regression models for rare events were used to maximize a penalized likelihood function for the two variables pertaining to the husband’s controlling behaviors (Firth, 1993). Although the Firth’s regression models are not frequently utilized in social science research, the interpretation is similar to that of the conventional binary logistic regression models.

For each of the three dependent variables, four regression models were estimated. Model 1 tested the effects of the subjective measure of financial strain on (a) the husband’s violent acts and (b) the husband’s controlling behaviors against the wife. Models 2 and 3 explored the effects of objective measures of financial strain—family income and the husband’s unemployment—on (a) the husband’s violent acts and (b) the husband’s controlling behaviors against the wife. Model 4 was the full model that included all the financial strain measures. Statistical controls were included in all regression models to estimate the net effects of financial strain on the dependent variables. The multivariate statistical analyses were carried out in the Stata software, Version 15 (Stata Corporation, 2017).

Results

The negative binomial regression models displayed in Table 3 (Models 1 and 4) revealed a significant and positive effect of the wife’s self-perceived financial strain on IPV victimization (p < .001). That is, all else being equal, higher levels of the wife’s self-perceived financial strain increased the expected rate of violent victimization experienced by the wife. Although these results lend credence to Hypothesis 1, Hypothesis 2 is rejected, as none of the objective measures of financial strain—family income and the husband’s unemployment—are statistically significant in Models 2 through 4.

Table 3.

Negative Binomial Regression Models to Predict Violent Behaviors.

Variables	Model 1	Model 2	Model 3	Model 4
Perceived financial strain	1.595***			1.758***
Family monthly income		−0.032		0.070
Husband unemployed			0.449	0.903
Wife unemployed	−0.030	−0.028	−0.048	−0.067
Husband education	−0.095	−0.134	−0.132	−0.087
Wife education	0.026	0.012	0.011	0.022
Wife age	0.001	−0.007	−0.010	−0.004
Wife party member	−0.112	−0.074	−0.160	−0.293
Wife urban registration	0.112	0.158	0.139	0.068
Number of children	0.013	−0.004	0.031	0.085
Husband dominant decision maker	0.994*	1.167*	1.157*	0.950
Intercept	−1.401	−0.020	−0.198	−1.927
Likelihood ratio χ²	23.72**	7.33	7.80	26.43**
df	9	9	9	11
n	340	340	340	340

p < .05. **p < .01. ***p < .001.

Table 4 presented the net effects of both subjective and objective financial strain measures on the husband’s personal control over his wife. Although none of the Firth regression models were statistically significant (see all likelihood ratio chi-square statistics in Table 4), Models 2 and 4 indicated that the effect of family monthly income was significantly and negatively associated with the likelihood of the husband’s exertion of personal control (p < .05). In other words, low family income was positively associated with the likelihood of the husband’s exertion of personal control over his wife. These results provide weak and partial support for Hypothesis 4, whereas Hypothesis 3 is rejected, as the wife’s self-perceived financial strain is statistically insignificant in predicting the husband’s personal controlling behavior in Models 1 and 4.

Table 4.

Firth Logit Regression Models to Predict Personal Control.

Variables	Model 1	Model 2	Model 3	Model 4
Perceived financial strain	−0.402			−0.785
Family monthly income		−0.304*		−0.315*
Husband unemployed			−0.256	−0.372
Wife unemployed	0.598	0.487	0.646	0.550
Husband education	0.059	0.079	0.057	0.036
Wife education	−0.268	−0.242	−0.268	−0.235
Wife age	−0.038	−0.042	−0.032	−0.041
Wife party member	−0.373	−0.383	−0.193	−0.254
Wife urban registration	0.537	0.673	0.532	0.694
Number of children	−0.267	−0.402	−0.310	−0.437
Husband dominant decision maker	1.233	1.118	1.099	1.195
Intercept	−0.874	1.293	−1.204	1.834
Likelihood ratio χ²	4.74	8.74	4.90	9.14
df	9	9	9	11
n	340	340	340	340

p < .05. **p < .01. ***p < .001.

Turning to Table 5, it is observed that all else being equal, both the wife’s perceived financial strain and the husband’s unemployment significantly increased the likelihood of the husband’s exertion of financial control over his wife. These results are statistically significant in Models 1, 3, and 4, respectively, with all p values at or lower than the .05 level (p < .05 or p < .01). In addition, as shown in Model 2, while family monthly income was significantly and negatively (or low family income was positively) associated with the likelihood of the husband’s exertion of financial control (p < .05), this effect statistically diminished in Model 4. In light of these results, both Hypothesis 3 and Hypothesis 4 are partially supported.

Table 5.

Firth Logit Regression Models to Predict Financial Control.

	Model 1	Model 2	Model 3	Model 4
Perceived financial strain	1.198*			1.392*
Family monthly income		−0.233*		−0.272
Husband unemployed			2.318**	2.365**
Wife unemployed	−0.141	−0.150	−0.302	−0.330
Husband education	0.283	0.258	0.331*	0.361*
Wife education	−0.019	−0.029	−0.080	−0.017
Wife age	0.027	0.018	−0.004	0.016
Wife party member	−0.272	−0.274	−0.697	−0.804
Wife urban registration	−0.094	−0.087	−0.146	−0.142
Number of children	−0.768*	−0.780	−0.653	−0.512
Husband dominant decision maker	3.992***	4.132***	4.195***	3.934***
Intercept	−5.483*	−2.682	−3.949*	−3.552
Likelihood ratio χ²	20.61**	20.12**	21.82**	22.85**
df	9	9	9	11
n	340	340	340	340

p < .05. **p < .01. ***p < .001.

Conclusion and Discussion

The primary goal of this study was to link both subjective and objective indicators of financial strain to two distinct dimensions of IPV against women in postreform China, using a community survey conducted in Chengdu, the capital of Sichuan province, in 2017. Informed by the family stress model, our first hypothesis surmised that the wife’s self-perceived financial strain would be significantly and positively associated with the lifetime experience of the husband’s violence. This hypothesis was supported by our negative binomial regression results. However, it must be noted that it is unclear how the wife’s self-perceived financial strain could trigger IPV. We can only speculate that women who think they are financially challenged might be burdened by such overwhelming negative feelings, and that these feelings may, in turn, trigger the husband’s violent reactions. In other words, by definition, self-perceived financial strain involves a critical evaluation of the family’s economic circumstances. This critical self-evaluation may promote negative feelings toward familial responsibilities and exacerbate relationship conflicts, which can, in turn, increase the risk of IPV against women. It is also possible that the husband may develop similar perceptions of financial strain that may also lead to IPV against his wife. Unfortunately, the survey used in this study did not include responses from the husband pertaining to this matter. Nevertheless, it is imperative to recognize that the finding we reported here should not be interpreted as victim blaming. Rather, the linkages between the wife’s subjective perception of financial strain and IPV victimization should be understood in a broader social and familial context in which gender and social inequalities can trigger and exacerbate IPV against women.

Our second hypothesis predicted that unemployment of the husband and low family income would be significantly and positively associated with the husband’s violent acts against the wife. This hypothesis was not supported by our negative binomial regression results. Contrary to the family stress model, neither the husband’s loss of employment nor low family income was significantly associated with the husband’s perpetration of violent acts. This is contrary to our observation that the wife’s self-perceived financial strain was a strong predictor of the husband’s violent behaviors, as reported by married women in Chengdu. It is possible that the husband’s unemployment and/or low family income could only increase the husband’s tendency to exert gendered control, but not the perpetration of violent acts. This speculation is in line with the remaining findings of the study.

Our third hypothesis stated that the wife’s self-perceived financial strain would be significantly and positively associated with the likelihood of experiencing the husband’s financial and personal control. This hypothesis was partially supported by our analyses. As conjectured, the wife’s self-perceived financial strain significantly increased the likelihood of the husband’s controlling behavior pertaining to family finance, but not those pertaining to personal control. This finding is consistent with the patriarchal Chinese family system, in which the husband is expected to provide financial support for the family (Zuo, 2003). However, when Chinese wives who typically manage their family purse become increasingly aware of financial difficulties in the family, and communicate such difficulties to their husbands (Shu, Zhu, & Zhang, 2012; Zuo & Bian, 2005), their husbands may increase the demand for control over family finances, and restrict their wives’ access to financial resources. Once again, we want to remind the reader that self-perceived financial strain is a subjective dimension of economic distress (Glei et al., 2018). It reflects broader socioeconomic disparities as determinants of IPV. Therefore, our findings do not suggest or imply victim blaming.

Our final hypothesis stated that the husband’s unemployment and low family income would be significantly and positively associated with the likelihood of the husband’s financial and personal control. This hypothesis was partially supported by our regression results. In particular, we found that while low family income increased the likelihood of the husband’s exertion of personal and financial control, the husband’s unemployment only increased the likelihood of his exertion of financial control. These results are congruent with prior research findings from the United States in that when the husband was unable to perform the expected gender role of financial provider, the wife felt disappointed and the husband felt inadequate (Golden et al., 2013). These negative feelings can contribute to relationship conflicts, and heighten the risk for IPV (Benson et al., 2003). These findings support the idea that the lack of financial resources in the family can undermine the breadwinner role of Chengdu husbands, thus affecting their masculine identity, in similar fashion to what has been observed in the West (Fox et al., 2002; Golden et al., 2013). This feeling of inadequacy can lead the husband to resort to coercive financial control over the wife.

Like many of the previous studies, there are several caveats in the present research. First, even though the survey used in the current investigation was conducted recently, it was based on a nonrepresentative sample with regional coverage. As a result, the findings reported here should not be generalized to other locales in postreform China without careful consideration. A population-based study is recommended for future research. Second, the survey utilized in this study is cross-sectional. Therefore, no causal relationships between various financial strain measures and IPV against Chengdu women are implied. To establish such causal relationships with an appropriate temporal order, longitudinal or panel data will be required in the future. Third, the survey utilized in this study did not include multiple measures of the two dimensions of IPV against women, namely, violent behaviors and gendered control. Multiple items measuring physical, psychological, and sexual violence against women as featured in the revised Conflict Tactics Scales (CTS2) should be considered in future research (Straus et al., 1996). By the same token, multiple indicators of gendered control, as used in the Demographic and Health Surveys, should be considered in future research as well. These indicators include additional and specific controlling behaviors, such as prohibiting the wife from talking with other men, prohibiting the wife from meeting with friends, prohibiting the wife from having contact with her natal family, and limiting the wife’s physical mobility (United States Agency for International Development, 2014).

In closing, evidence from our Chengdu sample provides some support for application of the family stress model pertaining to IPV against women. We, therefore, conclude that the family stress model, which recognizes financial strain as an important family stressor and is responsible in part for the occurrence of the husband’s perpetration of violent behaviors and gendered control, is valid in a non-Western society. However, it is vital to point out that, although our results reveal several direct associations between both subjective and objective measures of family financial strain and the husband’s violent acts as well as gendered controlling behaviors against his wife, there might be mediating or indirect factors at work, such as marital conflict and psychological distress (e.g., depressive symptoms). Unfortunately, these potential mediating factors were not readily available in our Chengdu survey. We suggest the inclusion of these factors in future studies, so that a more robust test of the family stress model in the context of Chinese family life might be conducted. Nevertheless, our results underscore the importance of gender and income inequalities in research on IPV against women. We urge policy makers, academic researchers, and health practitioners to recognize both subjective and objective financial strains as social and psychological determinants of IPV in postreform China.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Sichuan University (SKYB201406 and 2018HHF-07).

ORCID iD

Xiaohe Xu

Author Biographies

Li Zheng, PhD, is assistant professor of sociology in the School of Public Administration at Sichuan University, China. Her research areas include gender, health inequalities, marriage, and family. Her current research examines the intersectionality of SES, gender, the household registration system (Hukou), and the health consequences in China.

Xiaohe Xu, PhD, is professor of sociology in the Department of Sociology at the University of Texas at San Antonio. His current research investigates intimate partner violence, mental health, and relationship quality in China and Southeast Asian countries.

Ting Xu, PhD, is associate professor of sociology in the School of Public Administration at Sichuan University, China. Her research focuses on education inequities in China and examines the diversity in families’ educational expectations across different sociocultural backgrounds.

Liu Yang, MA, is assistant professor at Tianfu College of SWUFE, China. Her research focuses on comparative studies of family policies. Her current work explores the relationship between fertility policies and gender inequalities.

Xiaoxia Gu, MA, is a sociology graduate student in the School of Public Administration at Sichuan University, China. Her research examines gender differences in work–family conflict and its impacts on health.

Lijuan Wang, MA, is a sociology graduate student in the School of Public Administration at Sichuan University, China. Her research examines family dynamics and the relationship between gender and health.

References

Aizpurua

Copp

Ricarte

J. J.

Vázquez

(2021). Controlling behaviors and intimate partner violence among women in Spain: An examination of individual, partner, and relationship risk factors for physical and psychological abuse. Journal of Interpersonal Violence, 36(1-2), 231-254. doi:10.1177/0886260517723744

Anderson

K. L.

(1997). Gender, status, and domestic violence: An integration of feminist and family violence approaches. Journal of Marriage and Family, 59, 655-669.

Anderson

K. L.

(2005). Theorizing gender in intimate partner violence research. Sex Roles, 52, 853-865. doi:10.1007/s11199-005-4204-x

Arber

Fenn

Meadows

(2014). Subjective financial well-being, income and health inequalities in mid and later life in Britain. Social Science & Medicine, 100, 12-20. doi:10.1016/j.socscimed.2013.10.016

Benson

M. L.

Fox

G. L.

DeMaris

Van Wyk

(2003). Neighborhood disadvantage, individual economic distress and violence against women in intimate relationships. Journal of Quantitative Criminology, 19, 207-235. doi:10.1023/A:102493020

Capaldi

D. M.

Knoble

N. B.

Shortt

J. W.

Kim

H. K.

(2012). A systematic review of risk factors for intimate partner violence. Partner Abuse, 3, 231-280. doi:10.1891/1946-6560.3.2.231

Chan

(2011). Co-occurrence of intimate partner violence and child abuse in Hong Kong Chinese families. Journal of Interpersonal Violence, 26, 1322-1342. doi:10.1177/0886260510369136

Chan

(2014). Assessing the risk of intimate partner violence in the Chinese population: The Chinese risk assessment tool for perpetrator (CRAT-P). Violence Against Women, 20, 500-516. doi:10.1177/1077801214535107

Chengdu Yearbook. (2017). Chengdu administrative divisions and population. Chengdu, China: Chengdu Yearbook. Retrieved from http://www.chengdu.gov.cn/chengdu/rscd/xzqhyrk.shtml

10.

Conger

R. D.

Conger

K. J.

Martin

M. J.

(2010). Socioeconomic status, family processes, and individual development. Journal of Marriage and Family, 72, 685-704. doi:10.1111/j.1741-3737.2010.00725.x

11.

Cunradi

C. B.

Caetano

Schafer

(2002). Socioeconomic predictors of intimate partner violence among white, black, and Hispanic couples in the United States. Journal of Family Violence, 17, 377-389. doi:10.1023/A:1020374617328

12.

DeWall

C. N.

Anderson

C. A.

Bushman

B. J.

(2011). The general aggression model: Theoretical extensions to violence. Psychology of Violence, 1, 245-258. doi:10.1037/a0023842

13.

Dobash

R. P.

Dobash

R. E.

Wilson

Daly

(1992). The myth of sexual symmetry in marital violence. Social Problems, 39, 71-91. doi:10.2307/3096914

14.

Dunn

Inskip

Kendrick

Oestmann

Barnett

Godfrey

Cooper

(2008). Does perceived financial strain predict depression among young women? Longitudinal findings from the Southampton Women’s Survey. Mental Health in Family Medicine, 5, 15-21.

15.

Firth

(1993). Bias reduction of maximum likelihood estimates. Biometrika, 80, 27-38. doi:10.2307/2336755

16.

Fox

G. L.

Benson

DeMaris

Van Wyk

(2002). Economic distress and intimate violence: Testing family stress and resources theories. Journal of Marriage and Family, 64, 793-807. doi:10.1111/j.1741-3737.2002.00793.x

17.

Fox

G. L.

Benson

M. L.

(2006). Household and neighborhood contexts of intimate partner violence. Public Health Reports, 121, 419-427. doi:10.1177/003335490612100410

18.

Glei

D. A.

Goldman

Weinstein

(2018). Perception has its own reality: Subjective versus objective measures of economic distress. Population and Development Review, 44, 695-722. doi:10.1111/padr.12183

19.

Golden

S. D.

Perreira

K. M.

Durrance

C. P.

(2013). Troubled times, troubled relationships: How economic resources, gender beliefs, and neighborhood disadvantage influence intimate partner violence. Journal of Interpersonal Violence, 28, 2134-2155. doi:10.1177/0886260512471083

20.

Graham-Kevan

Archer

(2008). Does controlling behavior predict physical aggression and violence to partners? Journal of Family Violence, 23, Article 539. doi:10.1007/s10896-008-9162-y

21.

Hardie

J. H.

Lucas

(2010). Economic factors and relationship quality among young couples: Comparing cohabitation and marriage. Journal of Marriage and Family, 72, 1141-1154. doi:10.1111/j.1741-3737.2010.00755.x

22.

(2017). Marketization, occupational segregation, and gender earnings inequality in urban China. Social Science Research, 65, 96-111. doi:10.1016/j.ssresearch.2016.12.001

23.

Hou

Ting

S. R.

Sze

Y. T.

Fang

(2011). The status and characteristics of couple violence in China. Journal of Family Violence, 26, 81-92. doi:10.1007/s10896-010-9343-3

24.

Sun

(2017). Unequal care, unequal work: Toward a more comprehensive understanding of gender inequality in post-reform urban China. Sex Roles, 77, 765-778. doi:10.1007/s11199-017-0751-1

25.

Johnson

M. P.

(1995). Patriarchal terrorism and common couple violence: Two forms of violence against women. Journal of Marriage and the Family, 57, 283-294.

26.

Johnson

M. P.

(2006). Conflict and control: Gender symmetry and asymmetry in domestic violence. Violence Against Women, 12, 1003-1018. doi:10.1177/1077801206293328

27.

Johnson

M. P.

(2011). Gender and types of intimate partner violence: A response to an anti-feminist literature review. Aggression and Violent Behavior, 16, 289-296. doi:10.1016/j.avb.2011.04.006

28.

King

Zeng

(2001). Logistic regression in rare events data. Political Analysis, 9, 137-163. doi:10.1093/oxfordjournals.pan.a004868

29.

Kishor

Johnson

(2004). Profiling domestic violence: A multicountry study. Calverton, MD: ORCMacro.

30.

Lin

Sun

I. Y.

Liu

Chen

(2018). Chinese women’s experience of intimate partner violence: Exploring factors affecting various types of IPV. Violence Against Women, 24, 66-84. doi:10.1177/1077801216671221

31.

Long

J. S.

Freese

(2014). Regression models for categorical dependent variables using Stata (3rd ed.). College Station, TX: Stata Press.

32.

Meng

Chan

(2000). Family violence in China: Past and present. Journal of Comparative Social Welfare, 16, 74-87. doi:10.1080/17486830008415784

33.

Ogland

E. G.

Bartkowski

J. P.

Ogland

C. P.

(2014). Intimate partner violence against married women in Uganda. Journal of Family Violence, 29, 869-879. doi:10.1007/s10896-014-9640-3

34.

Parish

W. L.

Wang

Laumann

E. O.

Pan

Luo

(2004). Intimate partner violence in China: National prevalence, risk factors and associated health problems. International Family Planning Perspectives, 30, 174-181. doi:10.1363/ifpp.30.174.04

35.

Pearlin

L. I.

(1999). The stress process revisited. In Aneshensel

C. S.

Phelan

J. C.

(Eds.), Handbook of the sociology of mental health (pp. 395-415). Boston, MA: Springer.

36.

Rennison

C. M.

Welchans

(2000). Intimate partner violence. Washington, DC: U.S. Department of Justice.

37.

Schneider

Harknett

McLanahan

(2016). Intimate partner violence in the Great Recession. Demography, 53, 471-505. doi:10.1007/s13524-016-0462-1

38.

Selenko

Batinic

(2011). Beyond debt: A moderator analysis of the relationship between perceived financial strain and mental health. Social Science & Medicine, 73, 1725-1732. doi:10.1016/j.socscimed.2011.09.022

39.

Shippee

T. P.

Wilkinson

L. R.

Ferraro

K. F.

(2012). Accumulated financial strain and women’s health over three decades. The Journals of Gerontology: Series B, Psychological Sciences and Social Sciences, 67, 585-594. doi:10.1093/geronb/gbs056

40.

Shu

Zhu

Zhang

(2012). Patriarchy, resources, and specialization: Marital decision-making power in urban China. Journal of Family Issues, 34, 885-917. doi:10.1177/0192513X12450001

41.

Stata Corporation. (2017). Stata statistical software: Release 15. College Station, TX: Author.

42.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

(1996). The Revised Conflict Tactics Scales (CTS2). Journal of Family Issues, 17, 283-316. doi:10.1177/019251396017003001

43.

Szanton

S. L.

Allen

J. K.

Thorpe

R. J.

Jr Seeman

Bandeen-Roche

Fried

L. P.

(2008). Effect of financial strain on mortality in community-dwelling older women. The Journals of Gerontology: Series B, Psychological Sciences and Social Sciences, 63, S369-S374. doi:10.1093/geronb/63.6.S369

44.

United States Agency for International Development. (2014). Domestic violence module: Demographic and health surveys methodology. Retrieved from https://dhsprogram.com/pubs/pdf/DHSQMP/DHS6_Module_Domestic_Violence_6Aug2014_DHSQMP.pdf

45.

Wilkinson

L. R.

(2016). Financial strain and mental health among older adults during the great recession. The Journals of Gerontology: Series B, Psychological Sciences and Social Sciences, 71, 745-754. doi:10.1093/geronb/gbw001

46.

World Health Organization. (2017). Violence against women. Retrieved from https://www.who.int/news-room/fact-sheets/detail/violence-against-women

47.

(2004). Urban poverty and marginalization under market transition: The case of Chinese cities. International Journal of Urban and Regional Research, 28, 401-423. doi:10.1111/j.0309-1317.2004.00526.x

48.

Xie

Zhou

(2014). Income inequality in today’s China. Proceedings of the National Academy of Sciences, 111, 6928-6933. doi:10.1073/pnas.1403158111

49.

(1997). The prevalence and determination of wife abuse in urban China. Journal of Comparative Family Studies, 28, 280-303.

50.

Campbell

Zhu

(2001). Intimate partner violence against Chinese women: The past, present, and future. Trauma, Violence, & Abuse, 2, 296-315. doi:10.1177/1524838001002004002

51.

Zeng

Zheng

Flatt

(2010). How does wives’ unemployment affect marriage in reforming urban China? Journal of Comparative Family Studies, 41, 717-734. doi:10.2307/41604400

52.

Zhu

O’Campo

Koenig

M. A.

Mock

Campbell

(2005). Prevalence of and risk factors for intimate partner violence in China. American Journal of Public Health, 95, 78-85. doi:10.2105/AJPH.2003.023978

53.

Zuo

(2003). From revolutionary comrades to gendered partners: Marital construction of breadwinning in urban China. Journal of Family Issues, 24, 314-337. doi:10.1177/0192513X02250888

54.

Zuo

Bian

(2005). Beyond resources and patriarchy: Marital construction of family decision-making power in post-Mao urban China. Journal of Comparative Family Studies, 36, 601-622.