Cambodian Remarried Women Are at Risk for Domestic Violence

Abstract

Divorce rates continue to rise, especially in urban centers, which in turn contributes to increasing numbers of women who remarry. While remarriage is one of the only options for survival for divorced women, especially those from low socioeconomic status, remarriage also brings with it increased stressors of financial strain and the strain of blended families. This study tested the hypothesis that remarried women compared with first-time married and divorced women are at increased risk for domestic violence. The sample was drawn from the Cambodia Demographic and Health Survey, consisting of 1,560 women with the average age of 31.64. Results showed that 20% of women reported emotional violence and 14%, physical violence. Based on hierarchical multiple regressions, this study found that remarried women were more likely to experience physical and emotional abuse than women in their first marriage or women who remain divorced/not in union. Further interaction analyses showed that domestic violence varies depending on place of residence, number of children younger than 5 years, partners’ education, and wealth index. Rural residents who were in poorest and poorer groups and urban residents in their poorer and middle groups of their wealth index showed high risk of domestic violence. Remarried women with two or three children younger than 5 years showed highest risk of domestic violence. Additional three-way interaction analysis revealed that remarried women residing in rural/urban areas with a spouse having no education and/or primary level of education were at highest risk of domestic violence. The study lends support to the structural role of the power of male dominance on women’s social and emotional well-being. This study suggested that to reduce men’s perceived domination, structural intervention that includes implementing gender-responsive curriculum in formal education, as well as strengthening domestic violence law enforcement would help reduce domestic violence against women.

Keywords

domestic violence emotional violence physical violence Cambodian remarried women divorced

Despite policies and laws intended to prevent violence against women, and particularly domestic violence ( Korean Women’s Development Institute and The Asia Foundation Cambodia (2014), 22% of women report being victims of domestic violence in Cambodia (National Institute of Public Health, National Institute of Statistics, & ORC Macro, 2006). More than half of Cambodians report knowing an abusive husband (Ministry of Women’s Affairs, 2010), indicating that the prevalence of domestic abuse could be higher than reported. Recent research shows that both physical and social marital resources predict domestic violence among Cambodian married women (Eng, Li, Mulsow, & Fischer, 2010; Yount & Carrera, 2006). However, little is known about the role of marital status (i.e., married, remarried, and divorced) in the occurrence of domestic violence. In a male dominated culture such as Cambodia, divorced women are often stigmatized as failures as wives and mothers, and often are blamed for the dissolution of the marriage (Fulu et al., 2013). Because of the stigma associated with divorce, remarried women may be seen as less desirable partners, and may enter into subsequent marriages with not only a higher risk of abuse, but also limited resources and support due to their marginalized status. The purpose of this study is to examine the relationship between marital status and domestic violence. The hypothesis is that being remarried, as opposed to being either divorced/not in union or married, will be associated with increased emotional and physical violence.

Divorce and Remarriage and Domestic Violence in Cambodia

While the divorce rate in Cambodia remains relatively low when compared with the United States and other Western countries, increasing numbers of Cambodian women are affected by divorce. According to the Cambodian Inter-Censal Population Survey (National Institute of Statistics, 2013), 7% of marriages end in divorce and divorce and remarriage are on the rise, particularly in urban areas. The arduous divorce process, which requires that couples appear in court 3 to 5 times over the course of several years to be granted a divorce (Van der Keur, 2014), is often avoided, and couples part ways without obtaining a legal dissolution of the marriage. This may lead to the possibility that the dissolution of unions is much higher than the officially reported divorce rate.

In addition to the legal and bureaucratic hurdles, the social, cultural, and economic difficulties that effect divorced women could explain the low divorce rate. Divorce is highly stigmatized in Cambodian culture as the dissolution of marriage is often seen as a woman’s failure as a wife and a mother (Ledgerwood, 1996). Unequal employment opportunities and participation in unpaid domestic work means that women in Cambodia are often financially dependent on their husbands. The latest Demographic and Health Survey data showed that 36% of married women in rural areas are paid employees, compared with 89% of their counterparts in urban areas (National Institute of Statistics, Directorate General for Health, & ICF Macro, 2011). This dependence could deter women from seeking divorce (Van der Keur, 2014). Cultural factors, such as the emphasis on harmony and reconciliation in many Cambodian marriages (Brickell, 2014), could lead to an expectation to mediate and reconcile marital differences, even in situations of domestic abuse. In addition, the low divorce rate could also be a further reflection of certain dominant cultural expectations, which discourage women from speaking out against their husbands, especially in public. Despite the relatively low official divorce rates, the rising incidence of divorce in Cambodia indicates that it is increasingly important to understand the challenges that remarried women face.

There is a lack of research and information regarding remarriage in Cambodia. Eight percent of women in the 2005 Cambodia Demographic and Health survey were remarried; however, the 2013 census does not assess remarriage. Despite little information regarding remarriage in Cambodia, it is clear that there are differing expectations for divorced men and women. While it is expected, and accepted, that divorced men will remarry (Brickell & Chant, 2010), women have reported being forbidden to remarry by the local council as punishment for their divorce, and as a way of coercing women to remain married regardless of their circumstances (Naren & Melamed, 2005). Remarriage can be seen as a way to lessen the stigma of divorce and bring a woman back into the socially acceptable role as a wife. As described by Brickell (2014), remarriage can be a “tool to regain status, acceptance, and self-confidence at the same time as familial stability and happiness” (p. 33). A pressing need to remarry, whether through economic dependence or for the reconciliation of social status, could result in the acceptance of unions that would have been deemed undesirable as a first marriage (Brickell, 2014; Brickell, Prak, & Poch, 2014).

In addition to the contextually specific challenges that Cambodian women face, there are also broader challenges that are important in understanding remarriage and domestic violence in Cambodia. Research indicates that the same characteristics of individuals that led to domestic violence in first marriages continue into subsequent relationships (Kalmuss & Seltzer, 1986). Despite interventions such as counseling and overt recognition of past experiences with domestic violence, women who remarry are more likely to enter into abusive situations than women who do not remarry (Kalmuss & Seltzer, 1986).

Many times in remarriage, one or both parties enter the relationship with added stressors, such as children, financial issues, and possibly a complicated relationship with the previous spouse. The role of children in remarriage can cause significant complications, and the impact of stresses induced by these factors can lead to the increase of domestic violence in newly formed families (Horne, 1993). Yount and Carrera (2006) found that women with a lower household standard of living were more likely to experience domestic violence. This is in accordance with other findings (Smith, 1990; Yount, 2005) that support two mechanisms addressing the relationship between marital resources and domestic violence. Men who lack economic resources may use physical violence to control their partners (Felson & Messner, 2000; Goode, 1971) and men who live in households with limited resources could experience increased levels of stress, leading to domestic violence (Johnson, 1995).

The added stress of caring for young children in marriage creates an economic and social dependency for women making them more likely to tolerate domestic violence (Yount & Carrera, 2006). While these stressors, coupled with traditional beliefs regarding male dominance, could increase the likelihood of domestic violence for first-time married women, the effects could be more critical for remarried women as they have already brought with them marginalized status and have already been deemed less desirable partners. Therefore, they enter into subsequent relationships with a higher level of compromise compared with their first-time married counterparts.

Theoretical Framework

Domestic violence is likely to be perpetuated and tolerated in societies that uphold beliefs of male dominance (Coleman & Straus, 1986; Dobash & Dobash, 1979; Dobash & Dobash, 1998; Johnson, 1995; Straus, 1973; Yodanis, 2004) and research indicates that increased sex role egalitarianism is related to fewer incidents of domestic violence (Bernard, Bernard, & Bernard, 1985; Coleman, 1980; Coleman & Straus, 1986; Crossman, Stitch, & Bender, 1990). Remarried women in Cambodia face traditional beliefs about male dominance and are at risk of being viewed as less valuable due to the stigmas associated with previous marriages. These factors indicate that remarried Cambodian women could be more likely to experience domestic violence.

Ho (1990) described sociocultural factors of Asian societies that can lead to domestic violence, which include the prioritization of males and the perpetuation of traditional sex roles. Traditional attitudes toward remarriage reinforce male dominance by limiting the power and resources of women in coping with domestic violence, with Cambodian culture imposing strict rules regarding the subservience of women. The cultural tradition of silencing Cambodian women when faced with dominant men contributes to the tolerance of domestic violence.

The Chbab Srey, or rules for girls, is a popular 19th-century poem that outlines proper behavior for women in Cambodia. Chbab Srey promotes silence and submission as key characteristics of proper Cambodian women and is regarded as an important part of Cambodian culture (Walsh, 2007). Parts of the Chbab Srey are included in school curriculum and perpetuate this culture of silence and submission (Grace & Eng, 2015). These social norms are pervasive across communities (Jack & Astbury, 2014; LICADHO, 2007), and serve to regulate women, restrict their movement, and reduce their occupational choice and control (NGO-CEDAW & CAMBOW, 2011). In addition, Chbab Srey insists that women “keep the fire in the home,” indicating that domestic issues are not to be discussed outside of the family and decreasing a woman’s likelihood to seek divorce or outside help with domestic violence (Brickell, 2014).

The Current Study

In view of the role of physical and social marital resources in domestic violence, as well as the sociocultural factors associated with the male-dominant Cambodian society, it is important to examine the role of marital status in domestic violence against Cambodian women. Divorced and remarried women could be perceived as transgressing cultural norms regarding sexuality. Considering the value of chaste and virtuous wives in dominant strains of Cambodian culture and the link between the perceived value of wives and spousal abuse (Ho, 1990), women who remarry could be more likely to experience domestic violence. There is minimal research that investigates the impact of marital status on domestic violence against Cambodian women. This study attempts to address this gap in understanding.

Method

Participants

The data used in this study were drawn from the 2005 Cambodia Demographic and Health Survey (National Institute of Public Health, National Institute of Statistics, & ORC Macro, 2006). The sample in this study was nationally representative, consisting of 16,823 women and 6,731 men, ages 15 to 49 years. For the purpose of this study, only ever-married women were selected, leaving a sample size of 8,290. Those who were currently first-time married consisted of 88%, remarried 7%, and divorced 5%. As married women represented 88% of the total, original sample, we randomly selected only 8% of them for the purpose of this present study so that the three groups of women (married, remarried, and divorced/not in union) were proportional. After we selected 8% of married women, the final sample used in this study is 1,560. These 1,560 women included 585 married (38%), 391 divorced/not in union (25%), and 584 remarried (37%) women. Random sampling of this 8% of married women was done using STATA command “sample 8 if marital==3” where 3 represents being married. Among this sample, only 300 women had a complete response to physical and emotional violence. This present study used both of these samples, keeping the 1,560 for analyses independent of domestic violence variables, and the 300 for regression analyses with domestic violence as outcome variables.

Table 1 presented characteristics of participants in the study by marital status. Twenty percent and 14% of the women reported, respectively, emotional and physical violence. The average age of the respondents was 31.64 years (SD = 7.28). The majority of the respondents resided in rural areas (80%). The respondents’ education levels (the wives) were significantly lower compared with the levels of their partners (husbands), t(1, 1448) = −14.92, p < .001; specifically, while 29% of the husbands held a secondary or higher level of education, only 11% of the wives did so. More number of wives reported having no education (31%) compared with their husbands (20%). Most respondents reported having one (46%) or two children (39%) younger than the age of 5 years and more than half of them were currently employed (61%).

Table 1.

Frequency Distribution of Marital Status by Associated Characteristics.

	Remarried% (n)	Married% (n)	Divorced% (n)	Total% (n)
Education of partner
No education	8 (122)	7 (103)	5 (69)	20 (294)
Primary education	20 (295)	20 (290)	10 (146)	50 (731)
Secondary and above	10 (139)	12 (181)	7 (104)	29 (424)
Working status
Not working	16 (250)	16 (244)	7 (109)	39 (603)
Working	21 (333)	22 (341)	18 (282)	61 (956)
Education of the respondents
No education	12 (186)	11 (167)	8 (130)	31 (483)
Primary education	22 (344)	21 (330)	15 (227)	58 (901)
Secondary and above	3 (54)	6 (88)	2 (34)	11 (176)
Place of residence
Urban	7 (111)	9 (134)	4 (67)	20 (312)
Rural	30 (473)	29 (451)	21 (324)	80 (1,248)
Contraceptive use
Yes	15 (230)	15 (236)	1 (12)	31 (478)
No	23 (354)	22 (349)	24 (379)	69 (1,082)
Wealth
Poor/poorest	22 (343)	20 (314)	15 (231)	57 (888)
Middle/above	15 (241)	17 (271)	10 (160)	43 (672)
Father alive
Yes	13 (45)	23 (81)	12 (43)	47 (169)
No	26 (93)	17 (60)	10 (37)	53 (190)
Mother alive
Yes	24 (87)	31 (110)	18 (63)	73 (260)
No	13 (48)	9 (31)	5 (17)	27 (96)
Number of children younger than 5 years
None	2 (29)	1 (18)	1 (23)	4 (70)
One child	16 (249)	16 (243)	14 (221)	46 (713)
Two or more	20 (306)	21 (324)	9 (147)	50 (777)
Total	37 (584)	38 (585)	25 (391)	1,560

Procedures

The 2005 Cambodia Demographic and Health Survey was conducted by the Institute of Public Health, Ministry of Health, and the National Institute of Statistics of the Ministry of Planning with the primary objective of collecting a nationally representative sample of Cambodian adult men and women. The survey consisted of the Household, Man, and Woman questionnaires and a full pretest of the questionnaires was conducted in May 2005. Twenty-four women and 23 men were trained to implement the surveys within selected villages. Team members collected data via individual interviews at participants’ homes from September 9, 2005 to March 7, 2006. A total of 15,046 households were selected for sampling and 14,243 households completed the survey, yielding a 98% completion rate. Questionnaire data were entered using CSPro, a program developed jointly by the U.S. Census Bureau, the ORC Macro MEASURE DHS program, and Serpro S.A. All questionnaire data were entered twice and underwent internal verification and secondary editing to minimize errors.

Permission to use the 2005 Cambodia DHS data was obtained from the DHS Program. Upon receiving permission to use the data, the authors went through the codebook and selected variables of interest for the current study. Responses to each of the variables were checked for missing values and frequency distributions. The responses to physical and emotional violence were originally coded as number of times the women were abused in the past 12 months. As the responses to the “never” occurred was extremely high, we grouped the responses to two groups, “never” and “one or more times.”

Measures

Dependent variables

Two main dependent variables used in this study were emotional and physical violence. The two scales that DHS employed were a shortened version of the Conflict Tactics Scale by Straus (1990). Physical violence is measured by 10 indicators such as, “Spouse ever slapped” and “Spouse ever twisted her arm or pulled her hair.” Cronbach’s alpha is .81. Emotional violence is measured by three indicators, including “Spouse ever humiliated her?” and “Spouse ever threatened her with harm?” Cronbach’s alpha is .80. Responses for each item ranged from 0 to 1 in which “0” indicated “no,” and “1,” “yes”; thus, the higher the scores, the more the emotional/physical violence.

Independent and control variables

Marital status

This variable was created based on two other variables asking respondents their current marital status with the response including (a) married, (b) widowed, (c) divorced, and (d) not living together, and their number of unions, with the response including (a) once and (b) more than once. The creation of the remarried variable was based on those whose number of unions was more than one and who were currently married. We combined the number of widowed, divorced, and not living together responses into one category designated as “divorced/not in union.” Although widowed, separated, and divorced are not the same, they indicate a common theme: not being with a spouse due to certain reasons. Our primary aim in this study is to examine domestic violence among those who are officially married, and it is expected that those who are not in union are less likely to be abused compared with those who are in union. More specifically, those who are remarried will be at a higher risk of domestic violence compared with those who are first-time married. Thus, a variable of marital status was created with three categories: first-time married, remarried, and divorced/not in union.

Egalitarian variables

Six variables representing egalitarian behaviors and attitudes were employed as control variables for the study’s analysis: education of partner and of the respondents (no education, primary, and secondary/above), employment status (yes/no), gender role attitudes, region (rural/urban), and contraceptive use (yes/no). The gender role attitudes scale consisted of seven items asking respondents their opinions if they agree (coded 1) or disagree (coded 0) on the following items: family decision should be made by men, husband should not help with household chores, married women should not be allowed to work, wife does not have the right to express opinion, wife should tolerate being beaten to keep family together, better to educate son rather than daughter, acceptable for man to have sex outside of his marriage. Cronbach’s alpha for this scale is .50. A low alpha value for the gender role attitude scale might have been due to the dichotomous response to the questions, agree (1) and disagree (0). As Cronbach’s alpha relies on internal consistency or the calculation of the correlations among all the items (Cronbach, 1951), then the value of alpha might have been affected (low) for a measure with dichotomous items.

Family resources measures

Family resources included variables such as wealth index (divided into five quintiles—poorest, poorer, middle, rich, richest), father/mother alive (yes/no), and actual number of children younger than 5 years.

Data Analysis

The primary focus of this study was to examine the role of marital status, that is, being remarried compared with being married and divorced/not in union, in domestic violence. First, we conducted bivariate correlation analysis (Table 2) to examine relationships among all of the studied variables. This correlation analysis specifically allows for the examination of factors associated with marital status in a single matrix. Furthermore, to examine the role of marital status in domestic violence, we conducted two separate, three-step hierarchical regression models, one for emotional violence (Table 3) and the other for physical violence (Table 4). For both regression models, the first step added egalitarian measures, followed by the addition of family resources, and finally, the third step, marital status. To examine the unique contribution of each step added to the models, changes in F test and R² were also used and added to the tables. As marital status variable is categorical, we dummy coded it and used first-time married as a reference group. Follow-up interaction effect analyses (both two- and three-way ANOVA) were then conducted to further explore relationships among marital status and egalitarian variables and family resources. To achieve this, marginsplot analysis was used to produce interaction plots—only significant effects were reported.

Table 2.

Bivariate Correlations Among All Studied Variables.

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Physical violence	—
2. Emotional violence	.72***	—
3. Partner education	−.13*	−.14*	—
4. Employment (0 = not working)	−.07	−.14*	.00	—
5. Gender role attitudes	.05	.08	−.10	−.03	—
6. Education of respondents	−.05	−.05	.42***	.02	−.21***	—
7. Region (1 = urban, 2 = rural)	−.11*	−.06	−.17***	−.01	.11*	−.13***	—
8. Contraceptive use (0 = no)	.09	.17**	.14***	.03	−.02	.12***	−.08**	—
9. Wealth	−.06	−.08	.43***	.09***	−.17**	.37***	−.34***	.17***	—
10. Father alive	−.01	−.02	.18**	.10	−.06	.03	−.02	−.02	.15**	—
11. Mother alive	.06	−.02	.12*	−.08	−.06	.03	−.03	−.07	.15**	.29***	—
12. Children younger than 5 years	.16**	.19**	−.07**	−.09***	−.08	−.06*	.06*	−.01	−.11***	.05	.00	—
13. Divorced (reference: Married)	−.10	−.15*	.02	.13***	−.05	−.05	.04	−.35***	−.03	.07	.07	−.13***	—
14. Remarried (reference: Married)	.18***	.20***	−.07*	−.07**	.03	−.04	.02	.15***	−.02	−.23***	−.15**	.04	−.45***

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

Table 3.

Hierarchical Multiple Regression Predicting Emotional Domestic Violence From Egalitarian Measures, Family Resources, and Current Marital Status.

	Model 1	Model 2	Model 3
	β/SE	β/SE	β/SE
Step 1: Egalitarian measures
Education of partner (reference: No education)
Primary level	−.19* (−.05)	−.18* (−.05)	−.15 (−.05)
Secondary and above	−.22* (−.06)	−.21* (−.06)	−.19* (−.06)
Employment status (0 = not working)	−.12* (−.03)	−.10 (−.04)	−.09 (−.04)
Gender role attitudes	.07 (−.11)	.10 (−.11)	.10 (−.11)
Education of respondents (reference: No education)
Primary level	.09 (−.04)	.12 (−.05)	.14 (−.05)
Secondary and above	−.05 (−.07)	−.02 (−.07)	.00 (−.07)
Region (1 = urban, 2 = rural)	−.09 (−.05)	−.12 (−.05)	−.12* (−.05)
Contraceptive use (0 = no, 1 = some types)	.19** (−.04)	.18** (−.04)	.15* (−.04)
Step 2: Family resources
Wealth (reference: Poorest)
Poorer		−.03 (−.05)	−.05 (−.05)
Middle		.00 (−.06)	.00 (−.06)
Rich		−.05 (−.06)	−.08 (−.06)
Richest		−.03 (−.07)	−.02 (−.07)
Father alive		.02 (−.04)	.05 (−.04)
Mother alive		.03 (−.04)	.05 (−.04)
Number of children younger than 5 years		.18** (−.02)	.17** (−.02)
Step 3: Marital status (reference: Being married)
Being divorced			01 (−.05)
Being remarried			.17** (−.04)
R² Change (total)	.10	.03 (.14)	02 (.16)
F for change in R²	3.99***	1.44 (ns)	3.66*
n	283	280	280

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

Table 4.

Hierarchical Multiple Regression Predicting Physical Domestic Violence From Egalitarian Measures, Family Resources, and Current Marital Status.

	Model 1	Model 2	Model 3
	β/SE	β/SE	β/SE
Step 1: Egalitarian measures
Education of partner (reference: No education)
Primary level	−.07 (−.01)	−.08 (−.02)	−.04 (−.02)
Secondary and above	−.19* (−.02)	−.21* (−.02)	−.19* (−.02)
Employment status (0 = not working)	−.07 (−.01)	−.03 (−.01)	−.03 (−.01)
Gender role attitudes	.03 (−.03)	.06 (−.04)	.06 (−.04)
Education of respondents (reference: No education)
Primary level	.07 (−.01)	.13 (−.01)	.14 (−.01)
Secondary and above	−.05 (−.02)	.02 (−.02)	.04 (−.02)
Region (1 = urban, 2 = rural)	−.15* (−.01)	−.18** (−.02)	−.19** (−.02)
Contraceptive use (0 = no, 1 = some types)	.10 (−.01)	.08 (−.01)	.06 (−.01)
Step 2: Family resources
Wealth (reference: Poorest)
Poorer		.06 (−.02)	.03 (−.02)
Middle		.05 (−.02)	.05 (−.02)
Rich		−.04 (−.02)	−.08 (−.02)
Richest		−.04 (−.02)	−.03 (−.02)
Father alive		.03 (−.01)	.08 (−.01)
Mother alive		.10 (−.01)	.12 (−.01)
Number of children younger than 5 years		.18** (−.01)	.16* (−.01)
Step 3: Marital status (reference: Being married)
Being divorced			.04 (−.02)
Being remarried			.21** (−.01)
R² change (total)	.07	.05 (.12)	.03 (.15)
F for change in R²	2.56*	2.09*	5.24**
N	283	280	280

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

Results

Table 2 presents results based on bivariate correlation analysis among all of the studied variables. This allows us to examine bivariate relationships between domestic violence and other variables, and also provides an overview of the associated characteristics of marital status in a single matrix. Remarried (as compared with first-time married) women were more likely to report higher levels of both physical (r = .18, p < .001) and emotional violence (r = .20, p < .001). Remarried (as compared with first-time married) respondents were also more likely to have a partner with low education (r = −.07, p < .05), not be employed (r = −.07, p < .01), use some type of contraception methods (r = .15, p < .001), not have a living father (r = −.23, p < .001), and not have a living mother (r = −.15, p < .01). Table 1 also shows that physical and emotional violence were statistically significant (r = .72, p < .001), indicating that those who reported higher physical violence tended to also report higher emotional violence.

Figure 1 shows bar graphs of mean values between emotional and physical violence by marital status, using ANOVA analysis. For emotional violence, the highest mean score was found among remarried respondents (M = .21), followed by first-time married (M = .11) and divorced (M = .05) women, F(2, 297) = 7.08, p < .01, η² = .05. Similar significant results were also found in physical violence, F(2, 297) = 5.17, p < .01, η² = .03.

Figure 1.

Mean comparison of emotional and physical domestic violence by marital status.

Table 3 shows results based on hierarchical regression predicting emotional violence from marital status and controlling for egalitarian and family resources variables. Step 1 for the egalitarian model was significant, F(8, 274) = 3.99, p < .001, R² = .10, suggesting that this model explained 10% of the variance in the emotional violence variable. Family resources variables were added to Step 2, and the change in F test was not significant, suggesting that family resources variables did not uniquely contribute to the prediction of emotional violence. Finally, adding marital status in Step 3 increased R² by .02. The overall model was significant, F(16, 263) = 3.10, p < .001, and the change in F test was also significant, F(1, 263) = 6.59, p < .05, suggesting that marital status uniquely contributed to the prediction of emotional violence. Overall, the results suggested that being remarried, as compared with being first-time married, was more likely to result in women being emotionally abused by their spouses, even after controlling for potential confounding variables such as egalitarian and family resources variables. Additional factors predicting emotional violence included a partner with low education (β = −.19, p < .05), residing in an urban region (β = −.12, p < .05), using some types of contraceptive methods (β = .15, p < .05), and number of children younger than 5 years (β = .17, p < .01).

Similar to emotional violence, Table 4 showed the results of physical violence in three models. Model 1, where egalitarian variables were added, was significant, F(8, 274) = 2.56, p < .05, R² = .07, suggesting that this model collectively explained 7% of the variance in the physical violence variable. Adding Step 2, family resources variables, increased R² by .05 with a significant F change, F(7, 264) = 2.09, p < .05, suggesting that the family resources variables uniquely contributed to the prediction of physical violence. Finally, adding Step 3, marital status variable, increased R² by .03 with a significant F change, F(2, 262) = 5.24, p < .01. Overall, each model contributed to the prediction of physical violence. Specific variables predictive of physical violence included education of the partner (β = −.19, p < .05), place of residence (β = −.19, p < .05), number of children younger than 5 years (β = .16, p < .05), and marital status (β = .21, p < .01). Specifically, women who reported higher levels of physical violence were more likely to have a partner with no education, reside in urban areas, have a higher number of children younger than 5 years, and be remarried, with the strongest effect being remarried and residing in urban areas.

Overall, the results suggest that remarried adult women in Cambodia were at increased risk for both physical and emotional forms of domestic violence. Having a partner with a low level of education, residing in urban areas, having a higher number of children younger than 5 years, and being remarried were all predictive of both emotional and physical violence. The use of some types of contraceptive methods was predictive only for emotional violence, but not physical violence, suggesting that women’s knowledge of family planning was indicative of egalitarian behavior, which was perceived as threatening to the husbands who in turn emotionally abused their wives. Results from correlation analysis (Table 2) showed that contraceptive use was associated with women’s own education (r = .12, p < .001), a variable related to egalitarian behavior.

Interaction Effects Analysis

Interaction effects analysis was conducted between domestic violence and all of the predictor variables. For the emotional violence outcome, only two interaction effects were found—Marital × Children Younger Than 5 Years and Place of Residence × Wealth Index. Similarly, these two interaction terms were also found for physical violence. These four interaction effects findings were illustrated in Figures 2 to 5. Figure 2 showed that the association between being remarried and emotional violence varied depending on the number of children younger than 5 years, F(7, 286) = 3.72, p < .001, R² = .19, n = 300. Specifically, remarried women were at a higher risk of emotional violence when they had between one (95% confidence interval [CI] = [.02, .17]), two (95% CI = [.20, .35]), and three (95% CI = [.48, .80]) children younger than 5 years in the family, with three children at the highest level of risk. First-time married women having one (95% CI = [.01, .16]) and two (95% CI = [.07, .20]) children younger than 5 years were also shown to be at risk for emotional violence, but the effects were not as strong as compared with their remarried women counterparts.

Figure 2.

Interaction effects of marital status by children younger than 5 years on emotional violence.

Figure 3.

Interaction effects of place of residence by wealth on emotional violence.

Figure 4.

Interaction effects of marital status by children younger than 5 years on physical violence.

Figure 5.

Interaction effects of place of residence by wealth on physical violence.

Physical violence (Figure 4) showed a similar pattern of interaction effects compared with emotional violence (Figure 2) with a higher risk found for remarried women having two (95% CI = [.04, .09]) and three (95% CI = [.13, .23]) children younger than 5 years—and again those with three children were at the highest level of risk. In addition, first-time married women having two children younger than 5 years (95% CI = [.00, .05]) were also prone to physical violence, though the effect was modest.

Figure 3 illustrates the interaction effects between place of residence by wealth (measured by wealth index) on emotional violence, indicating that while women in urban areas reported a higher risk of emotional violence, the incidents varied depending on their reported wealth. Specifically, urban residents were at a higher risk of emotional violence if they were in the poorer (95% CI = [.15, .47]) and middle (95% CI = [.05, .62]) groups of their wealth index. For rural residents, only those who were in the poorest (95% CI = [.10, .22]), poorer (95% CI = [.04, .18]), and middle (95% CI = [.01, .20]) groups of their wealth index showed a significant level of emotional violence risk, with the poorest group at the highest level of risk. However, there was no significant difference between rural and urban residents regarding emotional violence risk for those women in the richer and richest groups of their wealth index.

Similar patterns were also found for physical violence (Figure 5) revealing that urban residents were at a higher risk for those in the poorer (95% CI = [.11, .21]) and middle (95% CI = [.07, .25]) groups of their wealth index. For rural residents, highest risk was found only among the poorest group (95% CI = [.01, .05]). A richer group of residents in rural areas (95% CI = [.01, .06]) also reported a high level of physical violence risk. However, both residents in the urban and rural areas, who were in the richest group of their wealth index, showed no significant difference regarding physical violence.

Discussion

The primary purpose of the present study was to examine the role of marital status in domestic violence. Based on the findings of this study, the hypothesis was supported in that remarried women reported a higher incidence of domestic violence than women who were first-time married or women who were currently divorced/not in union. Along with the stigma and cultural norms that place shame and failure on women who divorce, the option of remarriage makes women more likely to experience domestic violence. This exposure to violence places women at risk for significant marginalization and decreased quality of life.

Because divorce (among women) is stigmatized and deemed not valuable, remarriage perhaps provides a better alternative solution to gaining social acceptance. As shown in the data of this study, the divorce rate was 4.72% and the remarriage rate was 7.04%, suggesting that remarriage is encouraged. Yet, as shown in the findings of this study, women who remain divorced were the least likely to experience domestic violence compared with both remarried women and those in first marriages. While this would appear a natural assumption, given that divorced women do not have a spouse, the findings indicate that divorced women also have more autonomy and are more likely to be employed regardless of their geographic location or family support (e.g., father and mother still living). Divorced women are at an advantage in terms of their ability to build social networks outside of the home as they have more exposure to building social capital and resources to support themselves and their children. As divorce becomes more acceptable, women may feel less inclined to enter into subsequent marriages out of the need to adhere to social norms. Further research is needed to examine the social reforms that indicate more progressive ideas toward divorce in Cambodia.

While perhaps social norms are becoming more progressive in terms of accepting divorced women, this trend tends to be more prevalent in women with strong egalitarian beliefs and family resources. As shown in this study, women entering into their subsequent marriages were more likely to be traditional, less egalitarian, and have low family resources. Specifically, low partner education, unemployment, and not having a living father and/or mother were all indicators of domestic violence in remarried women. The data showed that remarried women tended to be older (M = 35, SD = 6.55) than first-time married women (M = 29, SD = 6.74) and divorced women (M = 30, SD = 7.13), thus increasing the likelihood of having a deceased parent. This low resources factor might put pressure on women to enter into their subsequent marriages out of a need for support and survival.

Considering the role of physical marital resources and social marital resources in domestic violence and the influence of male-dominant beliefs in spousal abuse, it is important to examine the role of marital status in domestic violence against Cambodian women. Second marriages can include added stressors such as children, financial problems, and complicated relationships with a previous spouse. The economic costs of domestic violence are tremendous and have widespread effects both inside and outside the home. Women suffer a range of health problems, are more likely to miss work, and have a higher incidence of murder and suicide (Korean Women’s Development Institute and The Asia Foundation Cambodia (2014)).

While the current study has shed light on some significant trends related to domestic violence and marital status, there is a need for additional research to explore these trends in more depth. Social stigma against divorced and remarried women is deeply embedded in Cambodian culture, and laws are in place to make divorce both cost-prohibitive and taxing on women (United Nations Development Program [UNDP], 2009). While it is unlikely that these laws will change in the near future, results from this study indicate that there are other ways to approach issues of domestic violence within current marriages to promote respect for women and denounce the idea that emotional and physical violence in marriages is acceptable. Men with higher education are significantly less likely to abuse their wives. This trend suggests that social and education interventions, with both men and women, can help to reduce the prevalence of domestic violence. These interventions would not only reduce the prevalence of domestic violence, but could also reduce the prevalence of divorce as women would not feel the need to divorce to save themselves from violent situations within the home.

Domestic violence against women also has significant ramifications for their children. Women who experienced abuse by their spouse reported that their children missed school 20 times or more because of the violence in the home (UNDP, 2009). This may lead to lower educational attainment, thus contributing to the cycles of domestic violence seen in adults who are products of a violent family environment (UNDP, 2009). As this study shows, education is a key factor in promoting more progressive views toward eliminating domestic violence. While the Law on the Prevention of Domestic Violence and the Protection of the Victims in 2005 (UNDP, 2009) has shown some positive impacts on attitudinal changes and decreased incidence of domestic violence, strengthening educational attainment among children would make the laws even more effective. Law enforcement is perhaps too late as a preventive measure against domestic violence. Improving educational access and quality could be the best way to prevent domestic violence.

Place of residence played a key role in both physical and emotional violence, showing specifically that those residing in urban areas tended to report higher incidence of violence compared with those residing in rural areas. Follow-up interaction analysis between place of residence and domestic violence, with all of the potential predictor variables, showed that place of residence had a significant interaction effect with wealth on domestic violence. This suggests that the risk of domestic violence varied depending on how wealthy the women and their families were. Urban residents only showed high risk for both emotional and physical violence when they were in the poorer and middle groups of their wealth index. The difference was not found for both rural and urban residents if they were in the richest group of their wealth index. It was also found that residents in rural areas faced the highest risk of violence if they were in the poorest group of their wealth index.

The results of this study revealed that while wealth played an important role in domestic violence, the actual problem could lie in the level of educational attainment. At the bivariate correlation analysis, wealth was positively associated with the educational attainment of the respondents (r = .37, p < .001) and educational attainment of their spouse (r = .43, p < .001). Importantly, the findings of this study showed that women with low levels of education tended to have a spouse with similar levels of education. Though there was no relationship between women’s education and domestic violence, there was a relationship between their spouses’ education and level of education, which suggests that poorly educated women were most likely to be at risk for domestic violence. Additional analysis (Figure 6) using a three-way interaction term among place of residence, education of partners, and marital status, F(4, 265) = 2.41, p < .05, R² = .19, n = 283, demonstrated that remarried women with a spouse having no education (95% CI = [.03, .10]) or and primary level of education (95% CI = [.02, .07]) and who were residing in rural areas were most likely to be at risk for physical violence. In addition, remarried women in urban areas whose partner had a primary level of education also showed a high risk of physical violence (95% CI = [.19, .32]). Similar patterns were also found for emotional violence (Figure 7).

Figure 6.

Three-way interaction among place of residence, partner education, and marital status on physical violence.

Figure 7.

Three-way interaction among place of residence, partner education, and marital status on emotional violence.

Conclusion

Educational initiatives, in terms of access and quality, provide the opportunity to untangle the complexity of social and cultural norms that are deeply embedded in Cambodian culture. The Cambodian Millennium Development Goals specifically target the reduction of all forms of violence against women and children (UNDP, 2009). While the goal was to increase the population’s awareness of the illegality of violence by 2015, little research has been conducted recently to measure such progress. One of the limitations of this study resides in the fact that current statistics do not factor in variables related to remarriage and domestic violence. While more recent data indicates decreases in domestic violence against women overall, the current study provides evidence of a significant need to target remarried women as they are more likely to experience domestic violence than women in general.

The current study has developed a critical view of domestic violence against remarried women in Cambodia. It has shed light on the importance of preventative measures to educate men and women about domestic violence, especially in the case of remarriage, low resources (wealth and number of children younger than 5 years), and low educational attainment among the women’s partners. Stigmatization of remarried women in the Cambodian context is deeply embedded in cultural norms, yet through progressive ideas about, and education for, the remediation of violent behaviors toward women, there are opportunities for significant decreases in domestic violence. Although policy recommendations include legal aid, access to protection from abusive situations, and other social supports (Korean Women’s Development Institute and The Asia Foundation Cambodia (2014), further research is necessary to understand if, and how, these actions are being taken, and the impact that they have on domestic violence against remarried women.

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) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Sothy Eng is professor of practice at Lehigh University’s Comparative and International Education program. His research focuses on the role of educational constructs and sociocultural factors in student learning achievement (e.g., science, technology, engineering, and mathematics [STEM], dropout, school readiness, college attendance), youth empowerment, gender equity, domestic violence, and health behaviors. His overall research seeks to generate a more comprehensive understanding of the role of individuals and structures (e.g., national policy, programming, curriculum) as related to educational development. Through a social capital lens, he examines the ways in which educational promotion is enhanced through relational activities and perceptions between and among actors and the social entities.

Whitney Szmodis is a PhD candidate in teaching, learning, and technology at Lehigh University, Bethlehem, PA. Her area of research focuses on the gender disparity in STEM education and the role that sociocultural factors play in girls’ aspirations to their future careers.

Kelly Grace is a PhD student of comparative and international education at Lehigh University, Bethlehem, PA. Her research interests include gender equity in international education development.

References

Bernard

J. L.

Bernard

S. L.

Bernard

M. L.

(1985). Courtship violence and sex-typing. Family Relations, 34, 573-576.

Brickell

(2014). “Plates in a basket will rattle”: Marital dissolution and home “unmaking” in contemporary Cambodia. Geoforum, 51, 262-272.

Brickell

Chant

(2010). “The unbearable heaviness of being” reflections on female altruism in Cambodia, Philippines, The Gambia and Costa Rica. Progress in Development Studies, 10, 145-159.

Brickell

Prak

Poch

(2014). Domestic violence law: The gap between legislation and practice in Cambodia and what can be done about it (DFID research report). Economic and Social Research Council. Retrieved from http://www.katherinebrickell.com/katherinebrickell/wp-content/uploads/2014/01/DV-Law-Prelim-Report-2014.pdf

Coleman

D. H.

Straus

M. A.

(1986). Marital power, conflict, and violence in a nationally representative sample of American couples. Violence and Victims, 1(2), 141-157.

Coleman

K. H.

(1980). Conjugal violence: What 33 men report. Journal of Marital and Family Therapy, 6, 207-213.

Cronbach

L. J.

(1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297-334.

Crossman

R. K.

Stitch

S. M.

Bender

M. M.

(1990). Sex role egalitarianism and marital violence. Sex Roles, 22, 293-304.

Dobash

R. E.

Dobash

R. P.

(1979). Violence against wives: A case against the patriarchy. New York, NY: Free Press.

10.

Dobash

R. E.

Dobash

R. P.

(1998). Rethinking violence against women. Thousand Oaks, CA: Sage.

11.

Eng

Mulsow

Fischer

(2010). Domestic violence against women in Cambodia: Husband’s control, frequency of spousal discussion, and domestic violence reported by Cambodian women. Journal of Family Violence, 25, 237-246.

12.

Felson

R. B.

Messner

S. F.

(2000). The control motive in intimate partner violence. Social Psychology Quarterly, 63, 86-94.

13.

Fulu

Warner

Miedema

Jewkes

Roselli

Lang

(2013). Why do some men use violence against women and how can we prevent it? Quantitative findings from the United Nations Multi-Country Study on Men and Violence in Asia and the Pacific. UNDP, UNFPA, UN Women and UNV.

14.

Goode

W. J.

(1971). Force and violence in the family. Journal of Marriage and Family, 33, 624-636.

15.

Grace

Eng

(2015, June). There is no place for “Chbab Srey” in Cambodian schools. The Cambodia Daily. Retrieved from https://www.cambodiadaily.com/opinion/there-is-no-place-for-chbab-srey-in-cambodian-schools-85230/

16.

C. K.

(1990). An analysis of domestic violence in Asian American communities: A multicultural approach to counseling. Women & Therapy, 9, 129-150.

17.

Horne

J. E.

(1993). The Brady Bunch and other fictions: How courts decide child custody disputes involving remarried parents. Stanford Law Review, 45, 2073-2142.

18.

Jack

D. C.

Astbury

(2014). Overcoming discrimination, persecution, and violence against women. In Mustakova-Possardt

Lyubansky

Basseches

Oxenberg

(Eds.), Toward a socially responsible psychology for a global era (pp. 207-229). New York, NY: Springer.

19.

Johnson

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

20.

Kalmuss

Seltzer

J. A.

(1986). Continuity of marital behavior in remarriage: The case of spouse abuse. Journal of Marriage and Family, 48, 113-120.

21.

Korean Women’s Development Institute and The Asia Foundation Cambodia. (2014). Strengthening National Policy Research Capacities in the Field of Gender Equality in Cambodia. Seoul: Korean Women’s Development Institute.

22.

Ledgerwood

. (1996). Politics and gender: Negotiating concepts of the ideal woman in present day Cambodia. Asia Pacific Viewpoint, 32(136), 139-152.

23.

LICADHO. (2007). Violence against women in Cambodia 2006. Phnom Penh, Cambodia: LICADHO. Retrieved from https://www.licadho-cambodia.org/reports/files/105LICADHOReportViolenceWoman2006.pdf

24.

Ministry of Women’s Affairs. (2010). Violence against women 2009 follow up survey: Final study report. Phnom Penh: Cambodian Ministry of Women’s Affairs. Retrieved from http://www.kh.undp.org/content/dam/cambodia/docs/DemoGov/Violence_Against_Women_2009_Follow-Up_Survey-Executive_Summary_Eng.pdf

25.

Naren

Melamed

(2005, November). Divorce system gives many women few options. Cambodia Daily. Retrieved from https://www.cambodiadaily.com/archives/land-title-owners-likelier-to-escape-poverty-50797

26.

National Institute of Public Health, National Institute of Statistics, & ORC Macro. (2006). Cambodia demographic and health survey 2005. Phnom Penh, Cambodia: Author.

27.

National Institute of Statistics. (2013). Cambodia Inter-Censal Population Survey 2013: Final report. Retrieved from http://countryoffice.unfpa.org/cambodia/drive/CIPS_Report_English_Final.pdf

28.

National Institute of Statistics, Directorate General for Health, & ICF Macro. (2011). Cambodia Demographic and Health Survey 2010. Phnom Penh, Cambodia: Author.

29.

Cambodian NGO Committee on CEDAW (NGO-CEDAW) and Cambodian Committee for Women (CAMBOW). (2011). Implementation of the convention on the elimination of all forms of discrimination against women in Cambodia, 2010. Retrieved from http://tbinternet.ohchr.org/Treaties/CEDAW/Shared%20Documents/KHM/INT_CEDAW_NGO_KHM_13348_E.pdf

30.

Smith

M. D.

(1990). Sociodemographic risk factors in wife abuse: Results from a survey of Toronto women. Canadian Journal of Sociology/Cahiers canadiens de sociologie, 15, 39-58.

31.

Straus

M. A.

(1973). A general systems theory approach to a theory of violence between family members. Social Science Information, 12, 105-125.

32.

Straus

M. A.

(1990). The conflict tactics scales and its critics: An evaluation and new data on validity and reliability. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 49-73). New Brunswick, NJ: Transaction.

33.

United Nations Development Program. (2009). Violence against women: 2009 follow-up survey. Retrieved from http://www.kh.undp.org/content/dam/cambodia/docs/DemoGov/Violence_Against_Women_2009_Follow-Up_Survey-Executive_Summary_Eng.pdf

34.

Van Der Keur

. (2014). Legal and gender issues of marriage and divorce in Cambodia. Cambodia Law and Policy Journal, 2, 1-22. Retrieved from http://cambodialpj.org/wp-content/uploads/2014/05/DCCAM_CLPJ_Dorine.pdf

35.

Walsh

(2007). Project on women’s rights in Cambodia: LICADHO–UQAM partnership. Retrieved from http://www.ieim.uqam.ca/IMG/pdf/Walsh_Cambodia_women-2.pdf

36.

Yodanis

C. L.

(2004). Gender inequality, violence against women, and fear: A cross-national test of the feminist theory of violence against women. Journal of Interpersonal Violence, 19, 655-675.

37.

Yount

K. M.

(2005). Resources, family organization, and domestic violence against married women in Minya, Egypt. Journal of Marriage and Family, 67, 579-596.

38.

Yount

K. M.

Carrera

(2006). Domestic violence against married women in Cambodia. Social Forces, 85, 355-387.