Is Attitudinal Acceptance of Violence a Risk Factor? An Analysis of Domestic Violence Against Women in Pakistan

Abstract

Violence against women (VAW) is a grave problem in Pakistan, and women from all socioeconomic groups are vulnerable to domestic violence in varying degrees. It is argued that patriarchal definition of gender roles may reinforce the internalized inferiority of women. So, it may not be a mere coincidence that a large number of women in Pakistan justify VAW for various reasons. The objectives of this article are threefold: (a) to identify the drivers of VAW, (b) to see if women’s attitudinal acceptance of violence is causally linked with observed violence against women, and (c) to see if attitudinal acceptance of violence mediates between the socioeconomic status of women and observed violence. We used data from the Pakistan Demographic and Health Survey (PDHS) 2012-13. The sample consisted of 3,265 ever married women aged between 15 and 49 years who were interviewed for domestic violence. We used multivariate logit regression analysis to identify the drivers of VAW and used the Karlson-Holm-Breen (KHB) method for mediation analysis. We found that women’s attitudinal acceptance of violence, their childhood experience of violence in their parental household, the education of both husband and wife, and some occupation types significantly predicted their experience of spousal violence. In addition, we found that women’s attitudinal acceptance of violence mediated the relationships between socioeconomic factors (education and wealth status) and VAW. The significance of the study lies in the fact that it highlights the need to modify the perceptions of violence through change in educational policy. Among multiple other factors, an increase in the economic status of women is an effective hedge against the risk of spousal violence.

Keywords

battered women domestic violence domestic violence and cultural contexts perceptions of domestic violence predicting domestic violence

Violence against women (VAW) is a serious breach of human rights and a major public health issue worldwide. The term VAW is often used interchangeably with domestic violence, intimate partner violence, and spousal violence (Hindin, Kishor, & Ansara, 2008). The United Nations (1993) describes VAW as any act of gender-based violence resulting in physical, sexual, or psychological harm/suffering to women. Not only does VAW violate basic human rights, but may also result in physical and mental health problems (Campbell, 2002; M. Ellsberg et al., 2008).

VAW is one of the most pervasive human rights violations, denying women equality, security, dignity, self-worth, and their right to enjoy fundamental freedoms. It takes place in all societies regardless of ethnicities, social and economic backgrounds, cultures, religions, and geographical borders. On average, one third of the population of women globally face human rights violation at the hands of some family member (L. Heise, M. Ellsberg, & Gottemoeller, 1999). Although it is a global issue, its prevalence varies substantially (Bott, Morrison, & M. Ellsberg, 2005; Garcia-Moreno, L. Heise, Jansen, M. Ellsberg, & Watts, 2005; Hindin et al., 2008). The prevalence rates of VAW are relatively higher than the human development index (HDI), and the gender-related development index (GDI) is lower than the HDI in all South Asian countries. In terms of gender equality, Pakistan ranks 147th among 188 countries, with a Gender Inequality Index (GII) of 0.55.¹ GII measures the loss of human development caused by gender inequality. Gender inequality is measured by three factors: reproductive health, empowerment, and labor market participation. The GII value of the highest ranked countries in recent years is generally less than 0.1 while the lowest ranked countries generally score above 0.7.

An estimate of the data from 56 countries showed higher rates of VAW in the Eastern Mediterranean (37%) and South-East Asia (38%) compared with the United States of America (30%), Europe (25%), and the Western Pacific (25%; Thomson, Bah, Rubanzana, & Mutesa, 2015). Shown in that study is that within a region or country, incidents of VAW may differ extensively, depending on national histories, policies, cultural identities, resources, and other factors. Popular media across the globe also portray gender inequality, devaluation of women, and direct VAW (Kohlman et al., 2014), which are reinforced by prevailing community perceptions of gender differences (Sugarman & Frankel, 1996; Uthman, Lawoko, & Moradi, 2009). Domestic stressors such as financial stress, poverty, or alcohol abuse may further increase the risk of VAW (Sugarman & Frankel, 1996; Wallace, Haerpfer, & Abbott, 2008).

Pakistan is a predominantly patriarchic society, and gender inequality partly stems from cultural values reinforcing conventional male roles as head of the family, in charge of finances, and decision-maker. Despite the presence of legal and religious inheritance laws, cultural standards reinforce deprivation of women from inherited property. A study based on an urban sample from Pakistan reported that 58% of the women experienced physical violence at least once in their life, 55% experienced sexual violence, and as much as 84% suffered from psychological abuse (Ali, Asad, Mogren, & Krantz, 2011). A national household survey in Pakistan on domestic violence reported that one third of 23,430 women reported having experienced some type of physical violence (Shaikh, 2016).

Although VAW is the most pervasive and well-recognized public health issue, it is kept secret inside the walls of the home. To better understand the extent, nature, risks, and effects of the problem, many studies have been carried out in developed countries. However, considering the diversity in cultural and social contexts, empirical studies are needed from developing countries as well. Pakistani nationally representative contextual data on VAW are very rare. Therefore, the first aim of the study is to identify the prevalence and correlates of VAW based on nationally representative data taken from the Demographic Health Survey, wave 2012-13.

Attitudinal Acceptance of VAW

Several sociodemographic and behavioral risk factors and correlates of VAW have been identified in diverse studies. Among these, attitudinal acceptance of VAW has been considered an important determinant (Aslam, Zaheer, & Shafique, 2015; Gage, 2005; Jesmin, 2015b). Women may accept and justify VAW in light of their training of conventional gender roles, cultural and religious norms, financial and emotional dependency, and so forth (Begum, Donta, Nair, & Prakasam, 2015; Koenig, Stephenson, Ahmed, Jejeebhoy, & Campbell, 2006). Consequently, they may opt to continue suffering from gender violence, and, in turn, neither report nor conform to any endeavor working against VAW or human rights violation (Nnadi, 2012). Attitudinal acceptance favoring the climate of VAW is likely to exacerbate the issue further.

Children being reared in this climate of domestic violence may suffer from severe psychosocial and mental health problems (Bancroft, Silverman, & Ritchie, 2011). The effects may be particularly important for a female child being reared in a patriarchal culture with a climate accepting of VAW. A mother from such a culture may vertically transmit “attitudinal acceptance of violence” to the girl child who, in turn, may continue accepting VAW as part of the conventional gender roles and imbalanced distribution of power in marital relationships. Social learning theory explains how a permissive attitude toward VAW is learned from culture and family (Bandura & Walters, 1977). Parents, and particularly mothers, are likely to be an inﬂuential source of such learned attitudes and behaviors. In addition, the intergenerational transmission theory of violence also supports the same that such an attitude and the resulting behavior is passed on to generations after generations in cultures where gender inequalities are deeply embedded (Ehrensaft et al., 2003). The patriarchal culture of Pakistan may well fit such a criterion.

However, a very limited number of studies on the attitude toward VAW are extant in Pakistan despite the clear understanding of cultural norms discriminating against women and a variety of popular media portraits of gender inequality. Among the few available studies from Pakistan are those that studied men’s attitudes toward wife-beating (Fikree, Razzak, & Durocher, 2005), women’s views on religious teachings about spousal violence (Shaikh, 2016), and barriers to reporting VAW (Anderson et al., 2003). The current study aims to assess attitudinal acceptance of violence as a predictor of VAW in the cultural context of Pakistan based on a nationally representative sample taken from DHS-wave 2013-14.

Attitudinal Acceptance as a Mediator of Socioeconomic Status-VAW Relationship

Because of the negative repercussions of VAW on physical and mental health of women and children, numerous studies have started to identify factors associated with VAW. Among these, socioeconomic status (SES) and acceptance of violence are considered very important. Even if previous literature suggests SES as a predictor of VAW (Nelson & Lund, 2017; Ribeiro et al., 2017; Vyas & L. Heise, 2016), the findings on the link between SES and VAW are inconclusive for several reasons. Some of the reasons that stand out more conspicuously are measurement error caused by failure to adjust for important predictors or socioeconomic homogeneity of the samples (Evans, Davies, & DiLillo, 2008). However, the majority of the findings favor high SES as a protective factor and low SES as a risk factor against VAW (Devries et al., 2014).

Several indicators associated with low SES may work as antecedents of VAW. For example, previous literature shows that women with less education and low income are at greatest risk of spousal violence (Okuda et al., 2015). A likely explanation is that education increases marital satisfaction, which then outweighs attractive alternatives (Amato & Booth, 2001). Furthermore, rising rates of educational homogamy concentrate resources in educated couples, raising the quality of these marriages (Schwartz & Mare, 2005).

However, a question remains as to which factors are likely to explain the SES-VAW link. It is suggested that attitudinal acceptance is likely to explain this link. Consistent with the mediation hypothesis, the literature suggests that women from lower SES are more likely to accept VAW as part of family adjustment. Previous literature based on data from different cultural and ethnic backgrounds show that socioeconomic factors such as education level, wealth status, and social disadvantage are correlated with a person’s attitudes toward domestic violence (Flood & Pease, 2009; Markowitz, 2003). It is quite likely that women from lower SES may have frequent chances of witnessing VAW in the context of financial strains in the family, and frequent exposures make them desensitized to the harmful consequences, thus, they may accept VAW as part of family adjustment. The empirical evidence from previous literature supports the same that girls witnessing violence were more likely to experience violence later in life, and acceptance of violence mediated the link between SES and domestic violence (Uthman, Moradi, & Lawoko, 2011; Vung & Krantz, 2009).

Theoretical support to understand the direct and indirect relations between SES, attitudinal acceptance of violence, and VAW can be provided by an integrated ecological framework. This proposal is based on different authors’ work and proposes “an integrated ecological framework” with a view to studying and getting to know VAW (Casique & Furegato, 2006). Because explanations for VAW may range from individual, to psychosocial, to sociocultural factors, the ecological model not only incorporates all these factors in a single model but also explains how some factors explain, interact, or strengthen the link of others with violence (Lawson, 2015). The model explains factors at four levels in the form of overlapping rings: individual, family, communitarian, and sociocultural, as shown in Figure 1.

Figure 1.

Ecological model to understand violence.

Some alternative theoretical approaches also exist that explain the relationship between the attitudinal acceptance of violence and the actual experience of violence from different dimensions. The social norms approach developed by Berkowitz (2005) explains the link between acceptance of violence and VAW by assuming that people generally make mistaken perceptions of others’ attitudes and behaviors. Prevalence of unhealthy behaviors such as tolerance of violence is usually overestimated, which may, in turn, affect individual’s violent behavior in two ways: (a) by justifying and thereby increasing the tendency of violent behavior, and (b) by increasing the likelihood of an individual remaining silent about any discomfort caused by such behavior, thereby reinforcing its social tolerance. Furthermore, a family violence model developed by Boudouris (1982) describes family as a dynamic organization of interdependent constituents where attitudes and behaviors of one family member (e.g., violent man) are affected by attitudes, responses, and feedback of other members. Thus, violent behaviors of men may be strengthened by the feedback from women in the form of accepting attitudes toward violence. Goode’s resource theory explains that all social systems depend to some extent on power or force, to use it as one of several resources. The use of force occurs when a person lacks other resources such as income or education (Renfrew, 1997). Social learning theory explains attitudinal acceptance of VAW in terms of cultural and family dynamics (L. L. Heise, 1998).

Although, VAW is the most pervasive and well-recognized global public health issue, it is commonly kept secret inside the walls of the home. To better understand the extent, nature, risks, and effects of the problem, many studies have been carried out in developed countries, and research is growing in developing countries as well, including Pakistan. Considering the diversity in cultural, ethnic, and social contexts, empirical studies are needed from the patriarchal society of Pakistan, which ranks low on the global gender-gap rating among Asian countries (World Economic Forum, 2014). Given the significant diversity in educational and wealth status of Pakistani women across class, region, and rural and urban setting (Ali, Mogren, & Krantz, 2013), it would be interesting to assess the role of SES in VAW in a nationally representative sample from diverse backgrounds.

Here, we restate the main objectives of the present study: (a) to identify the correlates of VAW in Pakistan, (b) to assess whether attitudinal acceptance of violence is associated with VAW, and (c) to assess whether attitudinal acceptance of violence mediates the link between SES and VAW.

Method

Participants

This study uses data from the Pakistan Demographic and Health Survey (PDHS) 2012-13, which consists of a nationally representative sample of 13,558 women aged between 15 and 49 years who are currently married or have been married at some stage of their lives. The module on domestic violence asked the women if their current husband subjected them to violence during the 12 months before the survey was undertaken.

PDHS data were conducted by the National Institute of Population Studies, Pakistan, and ICF International, United States. The reasons for using data from the Demographic and Health Survey (DHS) include random selection of the sample, nationally representative data, carefully constructed survey instruments, and data collection at both individual and household levels.

The current study used data from the third wave of national surveys carried out between October 2012 and March 2013. The sample was recruited using a two-stage sampling strategy: first, stratified; and then, a random sampling strategy. The data were collected from four provinces including Punjab, Sindh, Balochistan, and Khyber Pakhtunkhwa and from Gilgit Baltistan while excluding Azad Jammu and Kashmir, Federally Administered Tribal Areas (FATA), and the restricted military and protected areas.

At the first stratified sampling stage, urban and rural areas from four provinces and Gilgit Baltistan were divided into smaller units called enumeration blocks. These blocks consisted of 200 to 250 households on average, which were further divided into high-, middle-, and low-income groups. Then, at the second stage, 14,569 households from stratified blocks were identified. Inclusion criteria included being in the age range between 15 and 49 years and being married at the time of data collection or had been married at some stage of their lives. One third of the identified sample was further selected using a systematic random sampling technique for the domestic violence module. However, all the selected women for the domestic violence module could not be interviewed for various reasons (e.g., presence of husband or some other family member at the time of data collection). The final sample included 3,526 women who responded to the domestic violence module.

Outcome Variable

The outcome variable in this study is experience of physical violence. The Conflict Tactics scale first constructed by Straus (1979) was adapted in developing the Domestic Violence module of the DHS. Violence here refers to the less severe violence that can take any one or more of the following forms: (a) ever been pushed, shook, or had something thrown at by husband/partner; (b) ever been slapped by husband/partner; (c) ever been punched with fist or hit by something harmful by husband/partner; and (d) ever had arm twisted or hair pulled by husband/partner. The response format was yes = 1 or no = 0. Women reporting to have experienced at least one of these behaviors were the subjects of physical violence. The current study uses two indicators of violence: less severe violence and severe violence. Out of the sample of 3,526 women who were asked if they suffered domestic violence (less severe), 26.2% women (N = 899) affirmed that they suffered domestic violence. No violence is coded as “0,” and experience of violence is coded as “1.”

Predictor Variables

The major variable of interest in this study is the attitudinal acceptance of violence, which reflects on respondents’ perceptions of violence. This factor is evaluated in several previous studies (Douki, Nacef, Belhadj, Bouasker, & Ghachem, 2003; L. Heise et al., 1999; Henning, Jones, & Holdford, 2005; Oyediran & Isiugo-Abanihe, 2005; Sokoloff & Dupont, 2005). The PDHS domestic violence module includes multiple questions regarding the perceptions of the respondents regarding spousal violence. Each respondent was asked if she feels that her husband is justified in physically abusing her in a set of circumstances, viz. if she goes out without telling husband, neglects the children, argues with husband, refuses to have sex with husband, or she burns food. We have used these five binary variables to construct an ordinal variable that is likely to take the minimum value of 0 if a woman feels that her husband is not justified in abusing her no matter what the circumstances are. The maximum value of this variable is likely to be 5, which indicates that a woman feels that her husband is completely justified in abusing her in all the given circumstances. Table 1 shows that a woman is most likely to condone spousal violence if she neglects children (35.9%) and least likely to condone violence because of burning food (17.95%).

Table 1.

Women’s perception of domestic violence (N=3265).

Husband Is Justified in Beating Wife If	No (%)	Yes (%)
Wife goes out without telling husband	67.81	32.19
Wife neglects the children	66.74	33.26
Wife argues with husband	64.10	35.90
Wife refuses to have sex with husband	66.98	33.02
Wife burns the food	82.05	17.95

The majority of the women did not condone VAW no matter what the circumstances are (55%). Nearly 8% of the women believe that the husband is justified in beating his wife for at least one reason out of the five given reasons. However, around 15% of the women believed that a husband is justified in subjecting his wife to violent behavior in all five situations (Figure 2).

Figure 2.

Attitudinal acceptance of violence on an ordinal scale (0-5).

The other risk factors are childhood experience of women in their parental household. Each respondent was asked if her father ever beat her mother. Based on this question, we included a dichotomous variable whether her father abused her mother in her childhood (yes/no). A number of previous studies have used the retrospective report of violence in childhood as a crucial predictor of experience of domestic violence in adulthood (Abramsky et al., 2011; Hotaling & Sugarman, 1986; Martin et al., 2002). Some additional risk factors based on the previous literature are woman’s age and her husband’s age (Abramsky et al., 2011; Garcia-Moreno et al., 2005; Hindin et al., 2008), age at first cohabitation (Garcia-Moreno et al., 2005), number of total children ever born, current pregnancy status (Bacchus, Mezey, & Bewley, 2006; Khosla, Dua, Devi, & Sud, 2005), sex of household head, her education and the husband’s education (Abramsky et al., 2011; Koenig, Ahmed, Hossain, & Khorshed Alam Mozumder, 2003; Magdol et al., 1997; Sokoloff & Dupont, 2005), her wealth status (M. C. Ellsberg, Pena, Herrera, Liljestrand, & Winkvist, 1999; Farmer & Tiefenthaler, 1997; Jewkes, 2002), occupation (Lenze & Klasen, 2017), and residential (urban/rural) status (Abramsky et al., 2011).

Hypothesized Mediator

It is argued that education and wealth status of woman crucially determine her attitudinal acceptance of violence. The attitudinal acceptance of violence, in turn, determines her experience of domestic violence (Abramsky et al., 2011; Jewkes, 2002; Koenig et al., 2003; Martin et al., 2002). In other words, the attitudinal acceptance of violence mediates the relationship between the socioeconomic factors, such as education and wealth status of women, and their experience of domestic violence. Figure 3 shows this relationship.

Figure 3.

Attitudinal acceptance of violence mediating between education/wealth status of women and the spousal violence.

Estimation Strategy

Bivariate analysis

We used bivariate analysis to estimate the strength of association between domestic violence and the attitudinal acceptance of violence and a set of additional demographic, socioeconomic, and spatial risk factors. Bivariate analysis is important because it gives useful insights about the individual risk factors. Its limitation is that it does not give any information regarding the causal direction. For this purpose, we use regression analysis.

Multivariate logistic regression analysis

The dependent variable in our logit regression model is domestic violence, which involves a binary outcome, that is, a woman either experienced spousal violence, or she did not experience spousal violence. We have, therefore, chosen the logit regression model to analyze our data.

Karlson-Holm-Breen (KHB) method for a mediation analysis

We use the KHB method for a mediation analysis (Breen, Karlson, & Holm, 2013). Contrary to other decomposition methods, the KHB method gives “unbiased decompositions, decomposes effects of both discrete and continuous variables, and provides analytically derived statistical tests for many models of the GLM family” (Kohler & Karlson, 2015, p. 1). We used the user written khb command in Stata (v.14) to do the mediation analysis.

Results

Bivariate Analysis

The sociodemographic, economic, and reproductive factors associated with domestic violence are presented in Table 2. The factors that show a strong association with a woman’s experience of spousal violence are her attitudinal acceptance of violence (p < .001), her childhood experience in her parental household (p < .001), her age at first cohabitation (p < .01), number of children ever born (p < .001), education (p < .001), husband’s education (p < .001), wealth index (p < .001), occupation (p < .05), and residential type (p < .01).

Table 2.

Bivariate analysis.

Experienced Less Severe Violence	No (%)	Yes (%)	Total (%)	F	p
Husband justified in beating wife (# of reasons)
0 (n = 1,804)	65.6	43.6	59.9	17.9	<.001
1 (n = 242)	6.6	5.9	6.4
2 (n = 195)	4.3	6.1	4.8
3 (n = 230)	6.4	9.4	7.2
4 (n = 319)	6.9	14.9	9
5 (n = 475)	10.2	20.1	12.8
Total (n = 3,265)	100	100	100
Respondent’s father ever beat her mother
No (n = 2,492)	87.8	53.3	78.8	274.4	<.001
Yes (n = 773)	12.2	46.7	21.2
Total (n = 3,265)	100	100	100
Total children ever born
0-5 (n = 2,414)	76.2	67.4	73.9	8.1	<.001
6-10 (n = 800)	22.3	31	24.6
>10 (n = 51)	1.5	1.7	1.5
Total (n = 3,265)	100	100	100
Currently pregnant
No or unsure (n = 2,919)	90.1	88.7	89.7	0.8	.362
Yes (n = 346)	9.9	11.3	10.3
Total (n = 3,265)	100	100	100
Woman’s educational attainment
Primary (n = 2,238)	68.3	82.7	72.1	21.1	<.001
Secondary (n = 604)	19.6	13.2	17.9
Higher (n = 423)	12.1	4.1	10
Total (n = 3,265)	100	100	100
Husband’s educational attainment
Primary (n = 1,428)	45.6	57.3	48.6	15.1	<.001
Secondary (n = 1,055)	34.6	32.6	34.1
Higher (n = 782)	19.8	10.2	17.3
Total (n = 3,265)	100	100	100
Wealth index
Poor (n = 1,195)	34.7	45.2	37.4	16.4	<.001
Middle (n = 597)	17.7	23.9	19.3
Rich (n = 1,473)	47.6	30.9	43.2
Total (n = 3,265)	100	100	100
Respondent’s occupation (grouped)
Not working (n = 2,484)	70.1	65.4	68.9	2.3	<.05
Professional/technical/managerial (n = 111)	3.5	1.5	3
Clerical (n = 1)	0	0.2	0.1
Sales (n = 20)	0.8	0.9	0.8
Agricultural—self-employed (n = 1)	0	0.2	0.1
Agricultural—employee (n = 210)	10.5	13.3	11.2
Household and domestic (n = 28)	0.6	1.7	0.9
Services (n = 209)	6.9	7	6.9
Skilled manual (n = 45)	1.5	2.9	1.9
Unskilled manual (n = 156)	6.1	6.8	6.3
Total (n = 3,265)	100	100	100
Type of place of residence
Urban (n = 1,542)	36.6	28.1	34.4	8.3	<.01
Rural (n = 1,723)	63.4	71.9	65.6
Total (n = 3,265)	100	100	100

Note. The p value is based on the design-based F test, which is a corrected weighted Pearson χ² statistic.

The women who reported that their father never beat their mother were more likely to report experience of spousal violence (53.3%) than the women who reported that their fathers beat their mothers (46.7%; Table 2). The prevalence of spousal violence was highest among the women who were between 35 and 39 years old (22%), followed by the 30 to 34 years of age category (19.9%). The majority of the women who reported spousal violence were between 15 and 19 years old (59.3%). Nearly one third of the women reporting spousal violence experience were those whose husbands were 31 to 40 years old.

Nearly two thirds of the women who had given birth to five or fewer children (67.4%) reported violence. The women who were not pregnant were nearly eight times more likely to report spousal violence (88.7%) than the women who were pregnant (11.3%). As the educational attainment of a women increases, she becomes less likely to experience violence. A large majority of the women who had attained up to primary-level education reported spousal violence (82.7%) but only 4.1% of the women with a higher level of education reported spousal violence. Conversely, 10% of the women reported spousal violence when their husbands had attained a higher level of education. This indicates that a woman’s education is a better predictor of her experience of spousal violence. The women who do not work are considerably more likely to report spousal violence (65.4%) than the women who had professional, technical, or managerial background (1.5%). Women in rural areas are also considerably more likely to report spousal violence (71.9%) than the women in the urban areas (28.1%).

Multiple Logistic Regression Analysis

Table 3 gives the estimates of multiple logistic regression analysis. After controlling for the independent factors given in Table 2, the attitudinal acceptance of violence, the behavior of the respondent’s father to her mother, husband’s educational attainment, and some occupational types turned out to significantly affect a woman’s likelihood of experiencing spousal violence. The women who believe that a husband is justified in beating his wife in certain circumstances are more likely to experience violence than the women who believe that a husband is not justified in beating his wife in any circumstances. The women who believed that the husband was justified in beating his wife for three or more reasons (see Table 1 for the details of the reasons) are significantly more likely to experience violence than the women who believed that a husband is not justified in beating his wife in either no circumstances or in a couple of circumstances.

Table 3.

Multivariate logistic regression.

Experiences Violence (Less Severe) by Husband	OR	95% CI
Husband justified in beating wife for number of reasons
0	1.00	[1.00, 1.00]
1	1.05	[0.74, 1.49]
2	1.34	[0.93, 1.93]
3	1.43*	[1.01, 2.02]
4	1.96***	[1.49, 2.58]
5	1.48**	[1.14, 1.93]
Respondent’s father ever beat her mother
No	1.00	[1.00, 1.00]
Yes	6.17***	[5.03, 7.57]
Total children ever born
0-5	1.00	[1.00, 1.00]
6-10	1.16	[0.94, 1.43]
>10	1.41	[0.77, 2.57]
Currently pregnant
No or unsure	1.00	[1.00, 1.00]
Yes	1.07	[0.82, 1.39]
Woman’s educational attainment
Primary	1.00	[1.00, 1.00]
Secondary	0.93	[0.72, 1.21]
Higher	0.69	[0.46, 1.03]
Husband’s educational attainment
Primary	1.00	[1.00, 1.00]
Secondary	0.96	[0.77, 1.19]
Higher	0.74*	[0.56, 0.98]
Wealth index
Poor	1.00	[1.00, 1.00]
Middle	1.27	[0.98, 1.64]
Rich	0.95	[0.71, 1.27]
Women’s occupation (Grouped)
Not working	1.00	[1.00, 1.00]
Professional/technical/managerial	1.02	[0.56, 1.85]
Clerical	1.00	[1.00, 1.00]
Sales	1.14	[0.52, 2.50]
Agricultural—self-employed	1.00	[1.00, 1.00]
Agricultural—employee	1.62*	[1.12, 2.34]
Household and domestic	3.33**	[1.47, 7.55]
Services	1.13	[0.79, 1.61]
Skilled manual	1.88	[0.98, 3.59]
Unskilled manual	1.53*	[1.02, 2.29]
Residence type
Urban	1.00	[1.00, 1.00]
Rural	1.04	[0.82, 1.32]
Constant	0.17***	[0.12, 0.24]
Observations	3263
χ²	421.87
Hosmer-Lemeshow χ² (8)	4.51
Prob > χ²	0.81

Note. Exponentiated coefficients; 95% confidence intervals in brackets. OR = Odds Ratio; CI = Confidence Interval.

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

The behavior of the respondent’s father vis-à-vis her mother significantly affects the odds of her experience of spousal violence. The women whose fathers beat their mothers are more than 6 times more likely to experience spousal violence. While the woman’s education does not serve as a deterrent to spousal violence, the husband’s education is found to be a protective factor against violence, odds ratio (OR) = 0.74, 95% confidence interval (CI) = [0.56, 0.98]. Counterintuitively, the wealth status of a woman does not affect the odds of her being subjected to spousal violence in the multivariate logit model. Women with certain occupational types are more likely to experience spousal violence. Women working in the agriculture sector are 1.6 times more likely to experience spousal violence than the women who do not do any paid job. Domestic workers are about 3.3 times more likely to experience spousal violence than the women who do not do any paid work.

Mediation Analysis

The direct and indirect pathways between education and experience of spousal violence mediated by attitudinal acceptance of violence are given in Table 4. The estimates along with the ORs and 95% CIs are shown for each pathway while controlling for respondent’s wealth status, residential status (urban/rural), and the information about the fact that the father of the respondent beat her mother and her husband’s education.

Table 4.

Mediating role of women’s education (Total).

Effect	β	SE (β)	Z	p Value	OR	(95% CI)
Women’s Education
Total effect	−.347	0.12	−2.9	.004	0.707	[0.56, 0.89]
Direct effect	−.274	0.121	−2.272	.023	0.76	[0.60, 0.96]
Indirect effect	−.0729	0.032	−2.255	.024	0.93	[0.87, 0.99]

Note. OR = Odds Ratio; CI = Confidence Interval.

Considering education as a continuous variable, we see a statistically significant mediating effect of attitudinal acceptance of violence between the association of education and experience of spousal violence. The mediating effect (indirect effect) was 21% of the total effect of education on the observed violence (Table 4). The odds that a woman with a given level of education will experience violence are 8%² higher than the women with a higher level of education.

When the educational attainment of women was categorized into tertiles to represent primary, middle, and higher education with the primary level as the reference group and adjusting for the potential confounders (respondent’s occupation, pregnancy status, residence, and her experience of parental violence), the mediating effect (indirect effect) was 30.77% and 18.09% of the total effect of women’s education on spousal violence in the middle and higher education category, respectively, compared with the women with primary education (Table 5). The odds that a woman with a given level of wealth status will experience violence are 7% higher than the women with higher level of wealth status. Increase in the level of education decreased the risk of spousal violence, and this relationship is statistically significant in both the middle and higher education categories.

Table 5.

Mediation of women’s education (disaggregated by primary, middle and higher levels).

Effect	β	SE (β)	Z	p Value	OR	(95% CI)
Middle education
Total effect	−.284	0.179	−1.584	.113	0.753	[0.53, 1.07]
Direct effect	−.197	0.181	−1.088	.276	0.821	[0.58, 1.17]
Indirect effect	−.087	0.042	−2.102	.036	0.916	[0.84, 0.99]
Higher education
Total effect	−.763	0.259	−2.952	.003	0.466	[0.28, 0.77]
Direct effect	−.625	0.26	−2.408	.016	0.535	[0.32, 0.89]
Indirect effect	−.138	0.046	−2.998	.003	0.871	[0.80, 0.95]

Note. OR = Odds Ratio; CI = Confidence Interval.

Similarly, attitudinal acceptance of violence mediates between the wealth status of a woman and her experience of spousal violence. The mediating effect of attitudinal acceptance of violence between wealth status and experience of violence (indirect effect) was 26% of the total effect of wealth status on women’s experience of spousal violence, OR = 0.933; 95% CI = [0.87, 1.00] (Table 6).

Table 6.

Mediating role of women’s wealth status.

Effect	β	SE (β)	Z	p Value	OR	(95% CI)
Wealth
Total effect	−.266	0.076	−3.509		0.766	[0.66, 0.89]
Direct effect	−.197	0.077	−2.545	.011	0.821	[0.71, 0.96]
Indirect effect	−.0694	0.038	−1.838	.066	0.933	[0.87, 1.00]

Note. OR = Odds Ratio; CI = Confidence Interval.

Categorizing the wealth status of women into tertiles to represent poor, middle, and rich wealth groups with the poor group as the reference category and adjusting for the potential confounders (respondent’s occupation, pregnancy status, residence, and her experience of parental violence), we found the mediating effect (indirect effect) to be 36% and 24% of the total effect of women’s wealth status on spousal violence in the middle income and rich category, respectively, compared with the indirect impact in the poor women’s category (Table 7). It needs to be highlighted that though the increase in wealth reduces the risk of spousal violence in both middle- and high-wealth tertiles, the significant impact is found only in the high-wealth tertile.

Table 7.

Mediating role of women’s wealth status (disaggregated by low, middle and high wealth tertiles).

Effect	β	SE (β)	Z	p Value	OR	(95% CI)
Middle-wealth tertile
Total effect	.149	0.154	0.965	.335	1.16	[0.86, 1.57]
Direct effect	.203	0.155	1.313	.189	1.225	[0.90, 1.66]
Indirect effect	−.054	0.047	−1.166	.243	0.947	[0.86, 1.04]
High-wealth tertile
Total effect	−.584	0.157	−3.727		0.557	[0.41, 0.76]
Direct effect	−.446	0.16	−2.79	.005	0.64	[0.47, 0.88]
Indirect effect	−.138	0.052	−2.678	.007	0.871	[0.79, 0.96]

Note. OR = Odds Ratio; CI = Confidence Interval.

Discussion

Attitudinal acceptance of violence significantly affects women’s experience of spousal violence. The women who believe that a husband is justified in beating a wife in certain circumstances are more likely to experience violence than the women who believe that a husband is not justified in beating his wife in any circumstances. There are several explanations for this relationship. Women may accept and justify VAW in the light of their training of conventional gender roles, cultural and religious norms, financial and emotional dependency, and so forth (Begum et al., 2015; Koenig et al., 2006). Consequently, they may opt to continue suffering from gender violence and, in turn, neither report nor conform to any endeavor working against VAW or human rights violation (Nnadi, 2012). Attitudinal acceptance favoring the climate of VAW is likely to exacerbate the issue further. Women are socialized since their early childhood to accept the norms of femininity, domestication, conformity, and familism; their sense of subordination is embedded in the family roles and gender-role socialization. Spousal violence can also be explained in terms of “social structural theory,” which predicts that experience of violence may stem from socialization of women to be reliant on men for their economic survival even in abusive relationships (Pulerwitz et al., 2015). Although empirical support for the finding is present from different cultures, the current findings present the extension of previous findings to the developing, patriarchal society of Pakistan.

Women whose fathers beat their mothers are more likely to experience spousal violence. The intergenerational transmission of the attitudinal acceptance of violence possibly explains this result (Koenig et al., 2006). Women who frequently saw their fathers abusing their mothers in their childhood tend to internalize that abuse as a norm and are more likely to remain in a relationship with an abusive partner (Begum et al., 2015). The men who witness parent-to-parent violence in their childhood are more likely to believe in the right of husbands to control their wives and use violence for this purpose (Martin et al., 2002). The association between violence in wife’s family of origin and spousal violence is well documented in literature (Abramsky et al., 2011; Hotaling & Sugarman, 1986).

Both husband’s and wife’s education significantly reduce the probability of spousal violence, though the impact of the wife’s education is statistically significant at a slightly larger confidence level (p < .1). The impact of women’s education on reducing domestic violence is consistent with previous studies (Abramsky et al., 2011; Koenig et al., 2003; Sokoloff & Dupont, 2005). It is important to note that our study finds only the effect of higher education of the husband and wife on reducing the risk of spousal violence while the secondary education has no significant impact on violence. This situation partially conforms with a U-shaped relationship between spousal violence and education, where women at the lowest and highest levels of education experience spousal violence significantly less than women with middle-level education. The inverse relationship between woman’s education and her experience of spousal violence may be rooted in the dynamics of Pakistan’s labor market. Women with higher education are more likely to earn more, which hedges them against spousal violence. As higher education is also associated with higher income groups, there is a reason to believe that highly educated women are more likely to be married off to more educated and wealthier spouses, which, in turn, decreases the odds of spousal violence. This association can be borne out by a very high rate of consanguine marriages in Pakistan (51.8%; Zakar, Zakar, & Aqil, 2014). It might also be surmised that the average household size mediates the relationship between education and violence. It may be hypothesized that better educated women, often from higher SES, live in households with more people including domestic servants, which serves as a deterrent against violence. However, the data show that the average household size of the rich households is smaller than the poor households (4.84 persons in the top-income quintile vs. 8.06 persons in bottom-income quintile; Household Integrated Economic Survey, 2017). Second, the majority of the 8.5 million domestic servants in Pakistan are women and young boys³ who are not expected to serve as a deterrent against violence. It is quite likely that smaller household size may serve as a protective factor against aggression and violence, given the empirical evidence that larger household size and crowded places strain the relationships due to deprivation of privacy and competition for limited resources to meet the basic needs that, in turn, affect the person’s ability to cope with frustrations (Marshy, 1999) and lead to aggression and violence (Fatima & Sheikh, 2014).

Although pregnancy is found to be associated with interpartner violence in many previous studies (D’Silva et al., 2018; Finnbogadóttir & Dykes, 2016; Orpin, Papadopoulos, & Puthussery, 2020; Persily & Abdulla, 2017), pregnancy has a positive but insignificant association with spousal violence in this study. One possible reason for the imprecision in the estimation of the effect of pregnancy on violence is the relatively small percentage of women (10.3%) who were pregnant at the time of interview. Previous literature has identified possible reasons behind increased risk of spousal violence during the times of pregnancy. Financial worries about the baby, decreased emotional and sexual availability of women during pregnancy, and possessiveness in the partner are some of the reasons behind the increased risk of spousal violence (Bacchus et al., 2006).

Similarly, the high-income status of women does not reduce the odds of violence against them. These findings go against the theoretical prediction that an increase in the income of a woman or other type of financial gains from outside marriage such as family support or public support decrease the level of spousal violence because such support raises the women’s “threat level” (Farmer & Tiefenthaler, 1997). Association between spousal violence and wealth status of a woman is found to be tenuous at best (see M. Ellsberg et al., 2008, for example).

Compared with the women who do not do any paid job, women working in some professions are more likely to suffer spousal violence. Skilled women doing manual jobs, household and domestic workers, and women working in the agricultural sector are significantly more likely to experience spousal violence than the women who do not do any paid work. On the contrary, there is no evidence to suggest that women doing some professional, technical, or managerial job are less likely to suffer from spousal violence. One possible reason behind this apparent paradox is that women with low bargaining power, measured by their low-wage profession, face increased risk of spousal violence because their husbands, who are presumably poor themselves, use violence as a tool to keep them in check in anticipation of their increased bargaining power (Heath, 2014).

The mediation analysis shows that women’s attitudinal acceptance of violence significantly mediates between her education and wealth status and her experience of spousal violence. Dividing the education and wealth into tertiles again indicates a reduction in the risk of spousal violence. This risk is mediated significantly by attitudinal acceptance of violence. The most consistent protective effect against spousal violence was observed when women’s wealth status was used as an independent factor.

An indirect significant relationship between both middle education and higher education (relative to primary education, which is the reference category) with spousal violence shows that both middle and higher education are effective protective factors against spousal violence, though the impact of middle education in reducing the odds of spousal violence is smaller, OR = 0.92; 95% CI = [0.84, 0.99], compared with higher education, OR = 0.87; 95% CI = [0.80, 0.95]. Literature has found ample evidence to suggest that only higher education hedges against spousal violence (Antai, 2011; Jesmin, 2015b; Peltzer & Pengpid, 2014; Sambisa, Angeles, Lance, Naved, & Thornton, 2011) because it guarantees better job opportunities with better wages (Heath & Jayachandran, 2016; Nazar & Chaudhry, 2017) and higher empowerment in terms of greater control of fertility decisions (Memon & Jonker, 2018). Although consistent with the previous literature (Flood & Pease, 2009; Uthman et al., 2011), the finding extends the previous findings from developed countries to a developing, patriarchal, and the lowest ranked country in terms of gender gap. To the best of our knowledge, this is the first study to evaluate the mediating role of attitudinal acceptance of violence in the relation between SES and VAW in Pakistan’s context.

An indirect insignificant relationship between middle-wealth tertile (relative to the low-wealth tertile, which is the reference category) and spousal violence, OR = 0.95; 95% CI = [0.86-1.04], and an indirect significant relationship between high-wealth tertile (relative to the low-wealth tertile) and spousal violence, OR = 0.87; 95% CI = [0.79, 0.96], shows that middle-wealth status of women is not an effective protective factor against spousal violence. Literature has found ample evidence to suggest that only high-wealth status hedges against spousal violence (Jesmin, 2015a; Sambisa et al., 2011). On a different note, some studies have found better education and higher wealth status positively associated with spousal violence because it is viewed as threatening patriarchal norms, which can be neutralized through violent means (Weitzman, 2014).

The current study has methodological strengths over the previous studies. First, given the serious repercussions of VAW, very little attention has been paid by earlier researchers to assess attitudinal acceptance of violence in relation to VAW and whether this factor explains the link between SES and VAW in the Pakistani context. The current study addresses this issue systematically. Second, the study used data from a nationally representative sample from Pakistan, increasing the generalizability of the finding to all Pakistanis. Third, in contrast to the previous studies, which have used small samples that were usually recruited through convenience sampling techniques, the current study used a large random and representative sample of Pakistani females. Nevertheless, the cross-sectional nature of the study limits our ability to draw causal inferences regarding mediational findings. It is quite likely that frequent experiences of VAW may have caused attitudinal acceptance of violence, and changing attitudes can break the vicious circle in either direction. However, as Holmbeck (1997) notes, a mediating relationship from a predictor to a criterion through a mediator may not necessarily be causal, and it can also be said for indirect effects.

Conclusion

This study has found that women’s attitudinal acceptance of violence, their childhood experience of violence in their parental household, their education, and some occupation types significantly affect their experience of spousal violence. When a woman believes that a husband is not at all justified in beating his wife, she is significantly less likely to experience spousal violence. Various ecological models and “social structural theory” are used to explain the attitudinal acceptance of VAW. Woman’s experience of parent-to-parent violence also significantly increases the odds that she will herself experience violence. Contrary to the evidence in existing literature, woman’s age, wealth status, residential status (urban/rural), and pregnancy status are not significant predictors of her experience of violence in the multivariate regression analysis.

In addition, we found that woman’s attitudinal acceptance of violence mediates between socioeconomic factors (education and wealth status) and her experience of spousal violence. An increase in the years of education and/or wealth make women less accommodating of spousal violence, which, in turn, result in their refusal to accept violence, which is associated with a smaller likelihood of their experience of spousal violence. The significance of the study lies in the fact that it highlights the need to modify the attitudinal acceptance of violence through change in educational policy. Among multiple other factors, an increase in the economic status of women is an effective hedge against the risk of being subjected to domestic violence.

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.

ORCID iD

Rafi Amir-ud-Din

Notes

Author Biographies

Rafi Amir-ud-Din is a PhD in economics and an assistant professor at COMSATS University Islamabad, Lahore Campus. His research interests include health economics, poverty and well-being and the economic context of gender relations. He teaches graduate and undergraduate courses in economics and supervises graduate students in diverse topics related to his research interests. He has won two prestigious Higher Education of Pakistan funded projects in 2018, namely, the Thematic Research Grant Program (TRGP) and National Research Program for Universities (NRPU). He is a certified professional trainer for Stata at the Centre of Excellence for Research Management Sciences (CERM) at CUI, Lahore Campus. He is a member of the Pakistan Society of Development Economists, Islamabad, and Pakistan Society of Applied Economists, Applied Economic Research Center, Karachi. He is also a visiting fellow of Sustainable Development Policy Initiative, Islamabad. He peer reviews for a number of journals.

Shameem Fatima is associated with COMSATS University Islamabad, Lahore Campus, as an assistant professor of psychology. Having a PhD in Applied Psychology, she has been actively involved in research and teaching, thereby supervising students’ research work at the Master’s and MPhil levels. Her research interests and areas include parent-adolescent relations, applied psychology, educational psychology, issues related to transgender population, and violence against women and children.

Shazia Aziz is an assistant professor of English at COMSATS University Islamabad, Lahore Campus, teaching graduate and undergraduate classes. Her research interests include but are not limited to computer-mediated communication, Internet linguistics, critical discourse analysis, business communication, enhanced language training (ELT), literature, and sociolinguistics.

References

Abramsky

Watts

C. H.

Garcia-Moreno

Devries

Kiss

Ellsberg

. . . Heise

(2011). What factors are associated with recent intimate partner violence? Findings from the WHO multi-country study on women’s health and domestic violence. BMC Public Health, 11(1), Article 109.

Ali

T. S.

Asad

Mogren

Krantz

(2011). Intimate partner violence in urban Pakistan: Prevalence, frequency, and risk factors. International Journal of Women’s Health, 3, 105-115.

Ali

T. S.

Mogren

Krantz

(2013). Intimate partner violence and mental health effects: A population-based study among married women in Karachi, Pakistan. International Journal of Behavioral Medicine, 20, 131-139.

Amato

P. R.

Booth

(2001). The legacy of parents’ marital discord: Consequences for children’s marital quality. Journal of Personality and Social Psychology, 81, 627-638.

Anderson

M. A.

Gillig

P. M.

Sitaker

McCloskey

Malloy

Grigsby

(2003). “Why doesn’t she just leave?”: A descriptive study of victim reported impediments to her safety. Journal of Family Violence, 18, 151-155.

Antai

(2011). Traumatic physical health consequences of intimate partner violence against women: What is the role of community-level factors? BMC Women’s Health, 11, Article 56. doi:10.1186/1472-6874-11-56

Aslam

S. K.

Zaheer

Shafique

(2015). Is spousal violence being “vertically transmitted” through victims? Findings from the Pakistan demographic and health survey 2012-13. PLoS ONE, 10(6), e0129790.

Bacchus

Mezey

Bewley

(2006). A qualitative exploration of the nature of domestic violence in pregnancy. Violence Against Women, 12, 588-604.

Bancroft

Silverman

J. G.

Ritchie

(2011). Impeding recovery: the batterer as parent postseparation. In The batterer as parent: Addressing the impact of domestic violence on family dynamics (pp. 123-162). Thousand Oaks, CA: Sage.

10.

Bandura

Walters

R. H.

(1977). Social learning theory (Vol. 1). Englewood Cliffs, NJ: Prentice Hall.

11.

Begum

Donta

Nair

Prakasam

C. P.

(2015). Socio-demographic factors associated with domestic violence in urban slums, Mumbai, Maharashtra, India. Indian Journal of Medical Research, 141, 783-738.

12.

Berkowitz

A. D.

(2005). An overview of the social norms approach. In Lederman

L. C.

Stewart

L. P.

(Eds.), Changing the culture of college drinking: A socially situated health communication campaign (pp. 193-214). New York, NY: Hampton Press, Inc.

13.

Bott

Morrison

Ellsberg

(June, 2005). Preventing and responding to gender-based violence in middle and low-income countries: A global review and analysis. World Bank Policy Research Working Paper 3618. Washington, DC: World Bank. Retrieved from http://documents.worldbank.org/curated/en/852691468140377416/pdf/wps3618.pdf

14.

Boudouris

(1982). Behind Closed Doors: Violence in the American Family. Contemporary Sociology, 11(4), 432-433. doi:10.2307/2068814

15.

Breen

Karlson

K. B.

Holm

(2013). Total, direct, and indirect effects in logit and probit models. Sociological Methods & Research, 42, 164-191.

16.

Campbell

J. C.

(2002). Health consequences of intimate partner violence. The Lancet, 359, 1331-1336.

17.

Casique

L. C.

Furegato

A. R. F.

(2006). Violence against women: Theoretical reflections. Revista Latino-Americana de Enfermagem, 14, 950-956.

18.

D’Silva

Frey

Kumar

Mohanraj

Manhart

L. E.

Kaysen

. . . Rao

(2018). Sociocultural and structural perpetuators of domestic violence in pregnancy: A qualitative look at what South Indian women believe needs to change. Health Care for Women International, 39, 243-260.

19.

Devries

K. M.

Child

J. C.

Bacchus

L. J.

Mak

Falder

Graham

. . . Heise

(2014). Intimate partner violence victimization and alcohol consumption in women: A systematic review and meta-analysis. Addiction, 109, 379-391.

20.

Douki

Nacef

Belhadj

Bouasker

Ghachem

(2003). Violence against women in Arab and Islamic countries. Archives of Women’s Mental Health, 6, 165-171.

21.

Ehrensaft

M. K.

Cohen

Brown

Smailes

Chen

Johnson

J. G.

(2003). Intergenerational transmission of partner violence: A 20-year prospective study. Journal of Consulting and Clinical Psychology, 71, 741-753.

22.

Ellsberg

Jansen

H. A. F. M.

Heise

Watts

C. H.

Garcia-Moreno

& WHO Multi-country Study on Women’s Health and Domestic Violence against Women Study Team. (2008). Intimate partner violence and women’s physical and mental health in the WHO multi-country study on women’s health and domestic violence: An observational study. The Lancet, 371, 1165-1172.

23.

Ellsberg

M. C.

Pena

Herrera

Liljestrand

Winkvist

(1999). Wife abuse among women of childbearing age in Nicaragua. American Journal of Public Health, 89, 241-244.

24.

Evans

S. E.

Davies

DiLillo

(2008). Exposure to domestic violence: A meta-analysis of child and adolescent outcomes. Aggression and Violent Behavior, 13, 131-140.

25.

Farmer

Tiefenthaler

(1997). An economic analysis of domestic violence. Review of Social Economy, 55, 337-358.

26.

Fatima

Sheikh

(2014). Socioeconomic status and adolescent aggression: The role of executive functioning as a mediator. The American Journal of Psychology, 127, 419-430.

27.

Fikree

F. F.

Razzak

J. A.

Durocher

(2005). Attitudes of Pakistani men to domestic violence: A study from Karachi, Pakistan. The Journal of Men’s Health & Gender, 2, 49-58.

28.

Finnbogadóttir

Dykes

A.-K.

(2016). Increasing prevalence and incidence of domestic violence during the pregnancy and one and a half year postpartum, as well as risk factors: A longitudinal cohort study in Southern Sweden. BMC Pregnancy and Childbirth, 16(1), Article 327.

29.

Flood

Pease

(2009). Factors influencing attitudes to violence against women. Trauma, Violence, & Abuse, 10, 125-142.

30.

Gage

A. J.

(2005). Women’s experience of intimate partner violence in Haiti. Social Science & Medicine, 61, 343-364.

31.

Garcia-Moreno

Heise

Jansen

H. A. F. M.

Ellsberg

Watts

(2005). Violence against women. Science, 310, 1282-1283.

32.

Gelles

R. J.

Straus

M. A.

(1979). Violence in the American Family. Journal of Social Issues, 35(2), 15-39. doi:10.1111/j.1540-4560.1979.tb00799.x

33.

Heath

(2014). Women’s access to labor market opportunities, control of household resources, and domestic violence: Evidence from Bangladesh. World Development, 57, 32-46.

34.

Heath

Jayachandran

(2016). The causes and consequences of increased female education and labor force participation in developing countries. National Bureau of Economic Research. Retrieved from http://www.nber.org/papers/w22766.pdf

35.

Heise

Ellsberg

Gottemoeller

(1999). Ending violence against women. Population Reports, 27(4), 1-1.

36.

Heise

L. L.

(1998). Violence against women: An integrated, ecological framework. Violence Against Women, 4, 262-290.

37.

Henning

Jones

A. R.

Holdford

(2005). “I didn’t do it, but if I did I had a good reason”: Minimization, denial, and attributions of blame among male and female domestic violence offenders. Journal of Family Violence, 20, 131-139.

38.

Household Integrated Economic Survey. (2017). Household Integrated Economic Survey (HIES) 2015-16. Islamabad: Pakistan Bureau of Statistics, Government of Pakistan.

39.

Hindin

M. J.

Kishor

Ansara

D. L.

(2008). Intimate partner violence among couples in 10 DHS countries: Predictors and health outcomes. Calverton, MA: DHS Program Office. Retrieved from http://dhsprogram.com/pubs/pdf/AS18/AS18.pdf

40.

Holmbeck

G. N.

(1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599-610.

41.

Hotaling

G. T.

Sugarman

D. B.

(1986). An analysis of risk markers in husband to wife violence: The current state of knowledge. Violence & Victims, 1, 101-124.

42.

Jesmin

S. S.

(2015a). Married women’s justification of intimate partner violence in Bangladesh: Examining community norm and individual-level risk factors. Violence & Victims, 30, 984-1003. doi:10.1891/0886-6708.VV-D-14-00066

43.

Jesmin

S. S.

(2015b). Social determinants of married women’s attitudinal acceptance of intimate partner violence. Journal of Interpersonal Violence, 32(21), 3226-3244. doi:10.1177/0886260515597436

44.

Jewkes

(2002). Intimate partner violence: Causes and prevention. The Lancet, 359, 1423-1429.

45.

Khosla

Dua

Devi

Sud

(2005). Domestic violence in pregnancy in North Indian women. Indian Journal of Medical Sciences, 59, 195-199.

46.

Koenig

M. A.

Ahmed

Hossain

M. B.

Khorshed Alam Mozumder

A. B.

(2003). Women’s status and domestic violence in rural Bangladesh: Individual- and community-level effects. Demography, 40, 269-288.

47.

Koenig

M. A.

Stephenson

Ahmed

Jejeebhoy

S. J.

Campbell

(2006). Individual and contextual determinants of domestic violence in North India. American Journal of Public Health, 96, 132-138.

48.

Kohler

Karlson

(2015). KHB: Stata module to decompose total effects into direct and indirect via KHB-method. Statistical Software Components. Retrieved from https://EconPapers.repec.org/RePEc:boc:bocode:s457215

49.

Kohlman

Baig

Balice

DiRubbo

Plancencia

Skale

. . . Aquino

(2014). Contribution of media to the normalization and perpetuation of domestic violence. Austin Journal of Psychiatry and Behavioral Sciences, 1(4), 1018.

50.

Krug

E. G.

Mercy

J. A.

Dahlberg

L. L.

Zwi

A. B.

(2002). The world report on violence and health. The Lancet, 360(9339), 1083-1088. doi:10.1016/s0140-6736(02)11133-0

51.

Lawson

D. M.

(2015). Family violence: Explanations and evidence-based clinical practice. Alexandria, VA: American Counseling Association.

52.

Lenze

Klasen

(2017). Does women’s labor force participation reduce domestic violence? Evidence from Jordan. Feminist Economics, 23, 1-29.

53.

Magdol

Moffitt

T. E.

Caspi

Newman

D. L.

Fagan

Silva

P. A.

(1997). Gender differences in partner violence in a birth cohort of 21-year-olds: Bridging the gap between clinical and epidemiological approaches. Journal of Consulting and Clinical Psychology, 65, 68-78.

54.

Markowitz

F. E.

(2003). Socioeconomic disadvantage and violence: Recent research on culture and neighborhood control as explanatory mechanisms. Aggression and Violent Behavior, 8, 145-154.

55.

Marshy

(1999). Book Review: Perceptions of Israeli Arabs: territoriality and identity. Ecumene, 6(1), 115-116. doi:10.1177/096746089900600107

56.

Martin

S. L.

Moracco

K. E.

Garro

Tsui

A. O.

Kupper

L. L.

Chase

J. L.

Campbell

J. C.

(2002). Domestic violence across generations: Findings from northern India. International Journal of Epidemiology, 31, 560-572.

57.

Memon

Jonker

(2018). Educational level and family planning among Pakistani women: A prospective explorative knowledge, attitude and practice study. Middle East Fertility Society Journal. Advance online publication, 23(4), 464-467.

58.

Nazar

Chaudhry

I. S.

(2017). The return on investment for education in Pakistan. Pakistan Journal of Commerce and Social Sciences, 11, 1069-1083.

59.

Nelson

J. R.

Lund

E. M.

(2017). Socioeconomic status and geographical rural settings’ contribution to oppression of women with disabilities who experience gender violence. In Religion, disability, and interpersonal violence (pp. 149-162). Cham, Switzerland: Springer.

60.

Nnadi

(2012). An insight into violence against women as human rights violation in Nigeria: A critique. Journal of Law & Politics, 5, 48-56.

61.

Okuda

Olfson

Wang

Rubio

J. M.

Blanco

(2015). Correlates of intimate partner violence perpetration: Results from a National Epidemiologic Survey. Journal of Traumatic Stress, 28, 49-56.

62.

Orpin

Papadopoulos

Puthussery

(2020). The prevalence of domestic violence among pregnant women in Nigeria: A systematic review. Trauma, Violence, & Abuse. Advance online publication, 21(1), 3-15.

63.

Oyediran

K. A.

Isiugo-Abanihe

U. C.

(2005). Perceptions of Nigerian women on domestic violence: Evidence from 2003 Nigeria Demographic and Health Survey. African Journal of Reproductive Health, 9, 38-53.

64.

Peltzer

Pengpid

(2014). Female genital mutilation and intimate partner violence in the Ivory Coast. BMC Women’s Health, 14, Article 13. doi:10.1186/1472-6874-14-13

65.

Persily

C. A.

Abdulla

(2017). Domestic violence and pregnancy in rural West Virginia. Online Journal of Rural Nursing and Health Care, 1(3), 10-25.

66.

Pulerwitz

Hughes

Mehta

Kidanu

Verani

Tewolde

(2015). Changing gender norms and reducing intimate partner violence: Results from a quasi-experimental intervention study with young men in Ethiopia. American Journal of Public Health, 105, 135-137.

67.

Renfrew

J. W.

(1997). Aggression and its causes: A biopsychosocial approach. Oxford, UK: Oxford University Press.

68.

Ribeiro

M. R. C.

da Silva

A. A. M.

de Britto

M. T. S.S.

Batista

R. F. L.

Ribeiro

C. C. C.

Schraiber

L. B.

. . . Barbieri

M. A.

(2017). Effects of socioeconomic status and social support on violence against pregnant women: A structural equation modeling analysis. PLoS ONE, 12(1), e0170469.

69.

Sambisa

Angeles

Lance

P. M.

Naved

R. T.

Thornton

(2011). Prevalence and correlates of physical spousal violence against women in slum and nonslum areas of urban Bangladesh. Journal of Interpersonal Violence, 26, 2592-2618. doi:10.1177/0886260510388282

70.

Schwartz

C. R.

Mare

R. D.

(2005). Trends in educational assortative marriage from 1940 to 2003. Demography, 42, 621-646.

71.

Shaikh

M. A.

(2016). Domestic violence in consanguineous marriages—Findings from Pakistan Demographic and Health Survey 2012-13. JPMA: The Journal of the Pakistan Medical Association, 66, 1319-1323.

72.

Sokoloff

N. J.

Dupont

(2005). Domestic violence at the intersections of race, class, and gender: Challenges and contributions to understanding violence against marginalized women in diverse communities. Violence Against Women, 11, 38-64.

73.

Straus

M. A.

(1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT) scales. Journal of Marriage and Family, 41, 75-88.

74.

Sugarman

D. B.

Frankel

S. L.

(1996). Patriarchal ideology and wife-assault: A meta-analytic review. Journal of Family Violence, 11, 13-40.

75.

Thomson

D. R.

Bah

A. B.

Rubanzana

W. G.

Mutesa

(2015). Correlates of intimate partner violence against women during a time of rapid social transition in Rwanda: Analysis of the 2005 and 2010 demographic and health surveys. BMC Women’s Health, 15(1), Article 96.

76.

United Nations. (1993). General Assembly resolution 67/97, Declaration on the Elimination of Violence against Women, A/RES/48/104 (20 December 1993). Retrieved from http://www.un.org/documents/ga/res/48/a48r104.htm

77.

Uthman

O. A.

Lawoko

Moradi

(2009). Factors associated with attitudes towards intimate partner violence against women: A comparative analysis of 17 sub-Saharan countries. BMC International Health and Human Rights, 9(1), Article 14.

78.

Uthman

O. A.

Moradi

Lawoko

(2011). Are individual and community acceptance and witnessing of intimate partner violence related to its occurrence? Multilevel structural equation model. PLoS ONE, 6(12), e27738.

79.

Vung

N. D.

Krantz

(2009). Childhood experiences of interparental violence as a risk factor for intimate partner violence: A population-based study from northern Vietnam. Journal of Epidemiology and Community Health, 63, 708-714.

80.

Vyas

Heise

(2016). How do area-level socioeconomic status and gender norms affect partner violence against women? Evidence from Tanzania. International Journal of Public Health, 61, 971-980.

81.

Wallace

Haerpfer

Abbott

(2008). Women in Rwandan politics and society. International Journal of Sociology, 38, 111-125.

82.

Weitzman

(2014). Women’s and men’s relative status and intimate partner violence in India. Population and Development Review, 40, 55-75.

83.

World Economic Forum. (2014). The global gender gap report 2014. Geneva, Switzerland: Author.

84.

Zakar

M. Z.

Aqil

(2014). Adverse birth outcomes associated with consanguineous marriage in Pakistan. International Journal of Gynecology & Obstetrics, 127, 211-212.