Domestic Violence in India

Abstract

This article assesses the prevalence and risk factors of domestic violence in India. The study uses the 2005-2006 India National Family Health Survey-III (NFHS-III) and focuses on the 69,484 ever-married women ages 15 to 49 from all regions, who were administered the domestic violence module. The results show that 31% of respondents experienced physical violence in the past 12 months before the survey; the corresponding figure for sexual violence was 8.3%. The multivariate logistic regression results show key determinants of physical and sexual violence. Some of the most salient findings are that urban residence, household wealth, affiliation with Christian religious denominations, wife’s age at marriage and education are associated with lower risk of physical and sexual violence. In contrast, being employed and being the wife of a man who drank alcohol increased the odds of experiencing both physical and sexual violence. Moreover, respondents who believed that wife-beating was justified under certain circumstances were more likely to experience domestic violence. These results and significant regional differences observed in this study suggest that gender role conditioning and cultural norms both contribute to domestic violence. Interventions, therefore, need to go beyond the institutional and legal levels to include cultural capital, which addresses partner and relationship issues.

Keywords

domestic violence ever married women physical violence sexual violence India

Introduction

Acts of physical, psychological, and sexual violence against women can be found in all societies around the world. Despite the variation in levels and types of domestic violence, the widespread nature of the violation of the basic human rights has been addressed by different social researchers in all parts of the world as an increasing human rights and public health concern (Ellsberg, Heise, Pena, Agurto, & Winkvist, 2001; Heise, Pitanguy, & Germain, 1994) and recognized as a global hidden epidemic by the World Health Organization (WHO, 2002, 2006). The need to address the issue and pay more attention to possible social policy implementation is important because domestic violence has detrimental impact on the individual and society. Domestic violence is not only associated with health burdens, but also intergenerational effects as well as sociodemographic consequences of physical and sexual abuse (WHO, 2005).

Women who experience violent acts in their households, the environment that is supposed to be the safest place, not only suffer from health-related issues, but also carry emotional and psychological burden. From a societal perspective, national economies in countries where domestic violence is prevalent experience detrimental effects through a decrease in female labor participation, increasing disability and medical expenses (Ahmed, Koenig, & Stephenson, 2006; Campbell, 2002; Jejeebhoy, 1998a; Martin, Tsui, Maitra, & Marinshaw, 1999; Singh, Mahapatra, & Datta, 2008). This study examines the prevalence and risk factors of physical and sexual violence among ever-married women in India.

Situating Domestic Violence in India

Similar to other countries in the world, violence against wives in India is deeply rooted in cultural norms of patriarchy, hierarchy, and multigenerational families, where female obedience and modesty is controlled through abusive behavior and accepted not only by men, but also by women; and fundamental to the existence of violence against women is the notion of power.

Legal provision

In India, the Dowry Prohibition Act was passed in 1961. It defines dowry as: “any property or valuable security given or agreed to be given either directly or indirectly by one party to a marriage to the other party to the marriage or by the parents of either party to a marriage or by any other person, to either party to the marriage or to any other person at or before [or any other time after the marriage] in connection with the marriage of the said parties’. As per this definition, gifts of jewelry, clothes and cash traditionally given by the groom’s family would also be covered by the anti-dowry law and hence declared illegal” (Kishwar, 2005, p. 1).

However, it was not until 1983 that India introduced its first legal provisions in relation to marital violence by passing amendments to the Indian Penal Code while introducing Section 498-A and Section 304 B. Section 498A introduced in 1983 made “cruelty” (mental and physical) to the wife by the husband or his relatives a cognizable, nonbailable offense. The law specifically addresses four types of cruelty: “(a). conduct that is likely to drive a woman to suicide; (b). conduct which is likely to cause grave injury to the life, limb or health of the woman; (c). harassment with the purpose of forcing the woman or her relatives to give some property; and (d). harassment because the woman or her relatives are unable to yield to demands for more money or does not give some property” (Kishwar, 2005, pp. 24-25). In 1986, Section 304B (dowry death) was introduced. This provision made it possible to prosecute the husband and in-laws of a woman; “if she died as a result of burns or any other injury within seven years of marriage, under suspicious circumstances and if it could be shown that she was subjected to cruelty or harassment by the husband and/or in-laws in relation to demand of dowry” (Jaising, 2009, p. 51). Following long debates and revisions, the law known as the Protection of Women from Domestic Violence Act was passed and took effect in 2006. For the first time, the Act invoked constitutional rights of women. It is unique because it has both civil and criminal law provisions. The Act broadens the definition of domestic violence to include physical, sexual, emotional, verbal, and economic abuse (Jaising, 2009). Despite the recent law, domestic violence remains a prevalent issue among most socioeconomic and cultural subgroups in India. Many women are taught not only to accept but also to rationalize domestic violence and keep their experiences to themselves.

Domestic violence has been an important theme in the discourse and activism of the women’s movement in India. The understanding of domestic violence has gone through different phases. Suneetha and Nagaraj (2006) examined three critical phases in the understanding of the domestic violence discourse in India: dowry paradigm, domestic violence, and development/empowerment. In the 1970s the dowry paradigm/dowry lens determined the early understanding of violence in the family. “The public discourse was dominated by analyses of dowry, its patterns, spread, effects, impact etc” (Suneetha & Nagraj, 2006, p. 4355). However, Kishwar (1988) and Agnes (1992) argue that the exclusive focus on dowry undermined the interplay of power in the family. The 1990s saw a shift to a much broader focus on domestic violence. This broader understanding of violence invoked critical issues of power relations in the family. The focus therefore shifted from dowry to wife-beating and overall domestic violence (ICRW, 1999; Suneetha & Nagaraj, 2006; Vindhya, 2002). In late 1990’s, violence became a development related issue that is, “violence as an obstacle in the realization of women’s rights, their participation in the economy/market as workers and consumers and their ability to be the managers of the household” (Suneetha & Nagaraj, 2006, p. 4356). These studies argued that toward the late 1990s wife-beating and domestic violence in general took shape as a development issue and the discussion often focused on the effects and costs of such violence. SocioCultural Contexts

Krishnaraj (2007) aptly remarks that violence as a coercive instrument to uphold cultural codes of honor may be visible or invisible. In 1997, the International Center for Research on Women (ICRW) began a series of studies on domestic violence in India using population-based surveys. The studies ranged from trends in domestic violence using household data from 346 married women in rural Gujarat (Visaria, 1999); hospital records in Maharahstra (Jaswal, 1999); and court records in Bangalore (Elizabeth, 1999). These studies show that domestic violence is widespread in all communities in India irrespective of sociodemographic identities like caste (an ascribed ranking of groups, unique to India), class, religion, age, and education. The 2010 National Crime Records Bureau documents the gravity of domestic violence against women in India; cruelty by husband and relatives accounted for 44%, Dowry Prohibition Act and dowry death combined accounted for 6.3%. Together these variables accounted for 50.3% of crimes committed against women.

A 2005 United Nations Populations fund report showed that approximately two thirds of married women in India were victims of domestic violence. In addition, the report also asserted that one incident of violence against a woman translates into seven working days that a woman loses (cited in Indian Express, 2005). Other studies have extensively focused on the health impact of domestic violence, which may include maternal mortality, gynecological infections, induced abortions, and unwanted pregnancies (Asling-Monemi, Pena, Ellsberg, & Persson, 2003; Cokkinides, Coker, & Sanderson, 1999; Heise et al., 1994; Jejeebhoy, 1998a; Kapadia-Kundu, Dyalchand, & Kulkarni, 2004; Parker, McFarlane, & Soeken, 1994; WHO, 2005). Furthermore, the WHO multicountry report also showed that violence against women is associated with sexually transmitted infections (WHO, 2005), which may also include HIV/AIDS.

Another important dimension in the understanding of domestic violence is provided by the differentiation between family and household. This enables a shift from a restricted focus on the consequences of domestic violence to include issues like health and economic productivity. Although the family is a crucial site of contestation between power and powerlessness in India, Karlekar (1998) differentiates between the family and the household. The household is a physical structure, localized and a specific constellation of emotional and economic relationships among its core members. The family is more amorphous, seen as larger kinship groupings, spread over time and space. The household, then, is “the operational unit” that “provides the ground, so as to speak, for a working out of family ideologies around specific roles and expectations” (cited in Ahmed-Ghosh, 2004, p. 58). Early marriage and motherhood not only affect the health of young girls, but also the intersection of culture, male authority, female dependence on elderly authority, and preference for male children make them vulnerable to violence (Bruce & Clark, 2003; Cherukuri, 2008; Madhurima, 1996). Given the foregoing, this study aims to examine the levels and factors associated with physical and sexual violence using a nationally representative sample of ever married women in India.

Theoretical Framework and Past Research

Patriarchy and gender roles

Although no framework can capture all the complex factors that affect domestic violence, in this article, we try to explain domestic violence using patriarchy and rigid gender roles lens. Krishnaraj (2007) conceptualizes patriarchal domination as male dominance through superior rights, privileges, authority, and power. Structurally, this socializes women and men into differential gender roles, often resulting in powerlessness among women. This powerlessness may lead to violence against women (Visaria, 2000). In fact, Heise and colleagues noted that violence is an extension of a continuum of beliefs that grants men the right to control women’s behavior (Heise et al., 1994). Patriarchal domination prescribes rigid gender roles, and the affect of predefined gender roles on physical violence against women has been established, and gender roles have been identified as the pivot around which power relations between women and men are manifested (Verma & Collumbien, 2003).

Agarwal (1995) also argues that young married women in India often enter households that have clearly defined gender hierarchies and roles (Kapadia-Kundu et al., 2007). That is, when a woman enters a family, she tends to be at the lowest hierarchy in her new family. Physical violence against women by intimate partners is an extreme form of discrimination and is inextricably linked to gender role demands and expectations (Barzelatto, 1998); the same could be said for sexual violence. The concept of “rigid” gender roles and norms has been used in a model explaining the “causation of intimate partner violence” that refers to “disciplining” women when they do not fulfill gender role expectations (Jewkes, Levin, & Penn-Kekana, 2002). The study shows that patriarchal domination along with rigid gender roles creates a space for domestic violence against women. Specifically, factors like education, economic security, which are often challenges faced by women in patriarchal societies and rigid gender roles manifestation in the household, like the size of the household, the woman’s age at marriage and first child birth explain the occurrence of physical and sexual violence against married women in India.

Past Research

Although previous research affirms the efficacy of some of these factors, which have been known to influence domestic violence in the literature, studies on domestic violence have found contradictory results among the household level variables. For instance, a multicountry study by Kishor and Johnson (2004) found contradictory results for the type of place of residence; women living in urban areas had increased risk of experiencing domestic violence than their rural counterparts except for Egypt, India, and Haiti where rural women were at greater risk of experiencing violence. The study was based on data from the Demographic and Health Surveys (DHS) in nine developing countries: Cambodia, Colombia, Dominican Republic, Egypt, Haiti, India, Nicaragua, Peru, and Zambia. Also, using the 1998-1999 Indian National Family Health Survey (INFHS) Ackerson and Subramanian (2008) found that rural women were at higher risk of experiencing domestic violence. On the one hand, the anonymity of urban living could generally be associated with a higher risk of violence. On the other hand, rural living coupled with poverty and gendered traditions may exacerbate domestic violence and even condone it.

Furthermore, although the wealth index is not necessarily a causal factor, it has been generally assumed in the literature that women from the poorest quintiles are at increased risk of domestic violence (Djamba & Kimuna, 2008; Kimuna & Djamba, 2008). Also, research has shown that the odds of experiencing violence have been shown to increase with the number of children a woman has (Ellsberg, Peña, Herrera, Liljestrand, & Winkvist, 2000; Martin et. al., 1999). These studies noted that in most countries women with no children have the lowest odds of experiencing violence relative to women with five or more children. Although the direction of causality may be at issue, the question is, does increase in number of children lead to increased risk of violence or is it the other way? Furthermore, increase in the number of children tends to reduce the level of resources per capita (e.g., income), which may increase the level of stress and predispose the husband toward violence (Martin et al., 1999). On the other hand, women who are subjected to violence may have less control over their sexuality and childbearing behavior, which could increase the number of children born to such women.

Other factors that have been associated with domestic violence have included wife’s characteristics. Age has been cited by Fernandez (1997) regarding the fact that older women are less likely to experience violence than their younger counterparts. Part of the reasoning here has to do with increased social status and economic resources. But it is also true that the longer a woman is married, the greater the increase in the exposure to the risk of violence. Thus, findings have been inconclusive. Furthermore, previous studies showed that younger women (less than 19 years) in India are more likely to experience physical violence than their older counterparts (Kapadia-Kundu et al., 2004; Madhurima, 1996). This is partly because young married women in India often enter households that have clearly defined gender hierarchies and roles (Agarwal, 1995). We will revisit this issue by examining the effect of age at first marriage on physical and sexual violence in the NFHS-III sample.

Other studies have shown that rates of ever-experiencing violence can be expected to rise with marital duration; because a longer marriage provides a longer period of exposure (Djamba & Kimuna, 2008; Kimuna & Djamba, 2008). However, marital duration can also be considered a proxy for compatibility in a marriage, particularly in cultures where divorce is legal and socially accepted. In such a case, the experience of violence, both ever and current is likely to be negatively associated with marital duration.

Furthermore, education has been shown to empower women. Malhotra and Mather, (1997) argued that women with more education tend to have more resources that they can draw on in confronting abusive situations than women with no education. Also, women who have ever been married to husbands with secondary or postsecondary education have lower odds of experiencing domestic violence. Education seems to be a protective factor to domestic violence. It is also assumed that women who are engaged in paid employment tend to have more say in domestic matters than women who are not in the labor market (see discussion in García, 2000). In contrast, some of the literature indicates that where agricultural land is inherited exclusively by sons, women are more likely to be culturally devalued (Dyson & Moore, 1983; Kimuna & Djamba, 2008), and hence at a higher risk of experiencing domestic violence.

Other important factors in the literature have been the justification of wife-beating and partner’s alcohol consumption. In a study by Jejeebhoy (1998b), wife-beating was found to be condoned in traditional societies and largely regarded as a consequence of a man’s right to inflict physical punishment on the spouse. The pervasiveness of violence in these communities leads to its tolerance as normative behavior. Jejeebhoy (1998b) found that 93% of women from a sample in rural Tamil Nadu, India believe that a man might be justified in beating his wife in certain circumstances (such as the neglect of household duties, disobedience, or use of alcohol). In a study of south Indian villages, Rao (1997) observed that although communities could occasionally intervene to stop wife-beating, there was generally a high tolerance for such behavior that usually prevented intervention. Further studies have found that among all variables hypothesized to influence the likelihood of domestic violence, a partner’s alcohol use has the most consistent relationship to domestic violence (Coker, Smith, McKeown, & King, 2000; Djamba & Kimuna, 2008; Kimuna & Djamba, 2008). In this study we examine some variables used in previous research and add other variables to examine how aspects of gender role affect domestic violence against ever married women in India.

Data and Method

Data

Data for this study were drawn from the 2005-2006 INFHS coordinated by the International Institute for Population Sciences. This is a nationally representative probability sample of all women age 15 to 49 (N = 124,385); from which 83,703 women were administered the gender based violence module. We excluded never married women through the duration of marriage variable and obtained the 69,484 sample of ever married women used here to examine the factors associated with physical and sexual violence.

The NFHS-III interviewed never married as well as ever married women ages 15 to 49 and all men ages 15 to 54. The sample covered 99% of India’s population living in 29 states. As a part of the Woman’s Questionnaire, a module of questions on domestic violence was included. The module collected information on sexual and physical violence perpetrated by husbands on their wives as well as violence by other household members (IIPS, 2007). In this article, we focus exclusively on ever married women of reproductive age (15-49), who were administered the domestic violence module. The NFHS-III administered the module on violence to only one eligible woman from each selected household. The selection was random and it yielded 69,484 eligible respondents. At the start of every interview, informed consent for the survey was obtained from the respondents and protection of their anonymity and confidentiality were guaranteed. To prepare for a possibility of abused women asking for help, a list of organizations that provide help was constructed. All interviewers who administered the module on domestic violence were provided special training (detailed information on sampling procedure and IRB approval for the India DHS can be found in IIPS, 2007).

Dependent variables

Our dependent variables are physical violence and sexual violence. The 2005-2006 India NFHS-III collected detailed information on the different forms of violence. The domestic violence module used questions constructed from the Conflict Tactics Scale (Straus, 1990) to measure physical and sexual violence. The questions included: Does/did your husband ever do any of the following things to you in the past 12 months—(a) Push you, shake you, or throw something at you? (b) Slap you? (c) Punch you with his fist or with something that could hurt you? (d) Kick you or drag you? (e) Try to strangle you or burn you? (f) Threaten or attack you with a knife, gun, or any other weapon? (g) Twist your arm, pull your hair? Sexual violence questions were does/did your husband ever do the following things to you: (h) Physically force you to have sexual intercourse with him even when you did not want to? (i) Force you to perform any sexual acts you did not want to? Questions “a to g” were measures for physical violence; and questions “h and i” measured sexual violence.

Independent variables

Our analysis focuses on three sets of independent variables: (a) Household characteristics; (b) wife’s characteristics; and (c) Partner’s characteristics. For household characteristics, type of place of residence was categorized into urban and rural. Household size was measured by the mean number of all household members. It is assumed that the number of people in the household would be a deterrent factor of domestic violence. The wealth index is measured by five categories ranging from the poorest to the richest. Four variables were created to measure religion. One identified Hindus, the second identified Muslims; the third identified Christians, and the fourth indexed other religions (Sikh, Buddhist, Jain, and those without any religious affiliations).

For wife’s characteristics, age was measured at 5-year categories from 15 to 19 to 44 to 49 years. Marital duration was grouped in categories that ranged from 0 to 4 to 30+. Similar to the husband’s occupation, wife’s occupation was grouped into not working, sales, and clerical; professional, technical, and managerial; domestic and services; agro-self employed, and skilled, unskilled manual. Attitude toward justification of wife-beating was measured by responses to five statements indicating whether such behavior is justified if the wife (a) goes out without telling the husband; (b) neglects children; (c) argues; (d) refuses to have sex, and (e) burns food. We expect that positive responses to these statements will be associated with lower risks of physical and sexual violence. Similar to wife’s education, husband’s education was categorized as follows: no education, primary, secondary, and higher. Husband’s alcohol drinking was measured at a nominal level (ever use vs. never use); therefore, the frequency of alcohol consumption could not be assessed. These variables’ descriptive statistics are listed in Table 1 and our multivariate logistic regression was calculated. We ran four logistic models to assess the relative affect of the household variables (Model I); without wealth index (Model II); with wife’s characteristics (Model III) and with husband’s sociodemographic characteristics (Model IV). But, first, we present descriptive results.

Table 1.

Percentage Distribution of Respondents by Selected Characteristics, India 2005-2006, NFHS-III

	Number	Percent/Average
Household variables
Place of residence
Urban	30,522	43.9
Rural	38,962	56.1
Household size (Mean)	69,484	5.23
Wealth index
Poorest	9,734	14.0
Poor	11,117	16.0
Middle	13,551	19.5
Richer	16,051	23.1
Richest	19,031	27.4
Region
South	13,560	19.5
North	12,400	17.8
Central	12,400	17.8
East	10,432	15.0
Northeast	11,626	16.7
West	9,066	13.0
Religion
Hindu	51,660	74.4
Muslim	8,597	12.4
Christian	5,714	8.2
Other	3,421	4.9
Number of living children
No children	6,699	9.6
1 to 2	33,544	48.3
3 to 4	22,726	32.7
5+	6,515	9.4
Wife’s characteristics
Age group
15 to 19	3,031	4.4
20 to 24	10,733	15.4
25 to 29	14,982	21.6
30 to 34	14,410	20.7
35 to 39	11,837	17.0
40 to 44	8,487	12.2
45 to 49	6,004	8.6
Age at first marriage
1<18	34,946	50.3
18 to 20	19,133	27.5
21+	15,405	22.2
Duration of marriage
0 to 4	11,197	16.1
5 to 9	14,461	20.8
10 to 14	14,221	20.5
15 to 19	12,107	17.4
20 to 24	8,769	12.6
25 to 29	6,003	8.6
30+	2,726	3.9
Education
No education	27,555	39.7
Primary	10,741	15.5
Secondary	25,148	36.2
Higher	6,035	8.7
Occupation
Not working	40,076	57.7
Sales, clerical	2,637	3.8
Professional, technical, managerial	2,374	3.4
Domestic, services	2,797	4.0
Agro-self employed	15,047	21.7
Skilled, unskilled manual	6,501	9.4
Wife-beating justified
Goes out without telling husband	20,444	29.6
Neglects children	25,504	36.9
Argues	20,280	29.5
Refuses sex	9,502	14.0
Burns food	13,105	19.0
Husband’s characteristics
Age (Mean)	65,406	37.4
Education
No education	15,895	23.1
Primary education	10,773	15.6
Secondary	32,494	47.2
Higher	9,700	14.1
Occupation
Not working	1,248	1.8
Sales, clerical	13,354	19.3
Professional, technical, managerial	6,501	9.4
Domestic, services	4,573	6.6
Agro-self employed	17,581	25.4
Skilled, unskilled manual	26,015	37.6
Drinks alcohol
No	43,513	62.7
Yes	25,902	37.3
Total
N	69,484	100.0

Note: Total percent may not add up to 100 because of rounding.

Results

Descriptive Results

Household characteristics

The analysis is based on a total of 69,484 ever married women administered the domestic violence module (rural 38,962 and urban 30,522) from the 2005-2006 India NFHS-III couple data. Table 1 shows percent distribution of respondents age 15 to 49 by selected characteristics. The majority of respondents were living in rural area (56.1%). The average number of people in a household was 5.23. The percent distribution for the wealth index increased with the level of wealth ranging from 14% for the poorest category to 27% for the richest category. The geographical distribution of the sample shows that all regions were represented, although there were more respondents from the South (19.5%) and fewer from the West (13%). All other regions were somewhere between. The Hindu religion has the highest proportion of respondents in the sample (74.4%). Whereas 12.4% of respondents were Muslims, Christian respondents were 8.2%. The “other” category had 4.9%. Furthermore, almost half (48.3%) of the ever married women in the sample had one or two children and over one third (32.7%) had three 3 or four children. Childlessness was not very prominent, nearly 10% (9.6%) of women reported having had no children.

Wife’s characteristics

Data in Table 1 show that most respondents (21.6%) were in the age group 25 to 29, followed by age group 30 to 34 (20.7%). The age group with the smallest number of respondents was 15 to 19 (4.4%). Furthermore, 50.3% of ever married women first got married before the age of 18 years. Almost 28% (27.5%) of respondents got married at age 18 to 20 years and slightly more than 22% (22.2%) first got married when they were older than 20 years. Women’s marital duration is quite equally distributed among the groups. Slightly more than 16%s (16.1%) had been married for less than 5 years; almost 21% (20.8%) had been married for 5 to 9 years; 20.5% for 10 to 14 years, and 17.4% for 15 to 19 years. Women who had been married for 20 to 24 years comprised of nearly 13% (12.6%); 8.6% had been married for 25 to 29 years and almost 4% (3.9%) had been married for 30 years or more.

Table 1 also shows that almost 40% (39.7%) of the respondents had no education; slightly more than 36% (36.2%) had a secondary education, 15.5% had a primary education, and almost 9% (8.7%) had higher education. The majority of the ever married women in the sample (57.7%) were not working; the next highest proportion (21.7%) was working in agriculture, and the lowest percentage (3.4%) was in professional category. The proportions of respondents who justified wife-beating for reasons such as going out without telling the husband; neglecting children, arguing, refusing sex and burning food varied across categories. However, the highest proportion of respondents (36.9%) justified being beaten for neglecting children followed by 29.6% and 29.5% respectively, who justified going out without telling husband and arguing with husband as a reason for wife-beating. Nineteen percent justified wife-beating for burning food and 14% justified wife-beating for refusing sex with her husband.

Husband’s characteristics

The average age of husbands in the sample was 37.4 years. Slightly more than 23% (23.1%) had no education, almost 16% (15.6%) had primary education, 47.2% had secondary education, and 14.1% had higher education. Two major occupations were skilled and unskilled manual (37.6%) and agriculture, self-employed (25.4%). Slightly more than 19% (19.3) held sales, clerical positions; Almost 7% (6.6%) held domestic, service positions and 1.8% were not working. Only 9.4% of husbands had professional, technical or managerial positions. Slightly more than 37% (37.3%) of husbands used alcohol and 62.7% did not use alcohol.

Data in Table 2 show that 31% of ever married women have experienced physical violence in the 12 months before the survey. The corresponding figure for sexual violence was 8.3%. However, there are significant differences by household and individual characteristics. Among rural residents, 33.4% experienced physical violence and 9.7% experienced sexual violence. Among urban residents, 27.5% experienced physical violence and 6.6% experienced sexual violence. There is a negative association between wealth index and domestic violence. Slightly more than 15% (15.3%) of the richest category reported physical violence and 3.7% in the same wealth index category were subjected to sexual violence. These figures increased with level of poverty; the highest physical (45.5%) and sexual (13.6%) violence were reported among the poorest ever married women. The geographical distribution of the sample shows that there were more reports of physical violence of ever married women in the Central region (40.4%) and fewer in the North (23.6%). As for sexual violence, the East region had the highest rate, at 17%, whereas the West reported the smallest percent (3.7%). The Muslims had the highest proportion of physical violence (35.1%) and sexual violence (10.9%); Christians reported the lowest proportion—21.4% of physical and 4.4% of sexual violence. The ever married women who fell in the “other” religious category 29.1% experienced physical violence and 6.4% experienced sexual violence. Furthermore, women who had 5 children or more were more vulnerable to physical and sexual violence (43.5% and 11% respectively).

Table 2.

Percentage of Respondents who Experienced Physical and Sexual Abuse, India 2005- 2006 NFHS-III^a

	Physical Violence	Chi-Square/F	Sexual Violence	Chi-Square/F
Household variables
Place of residence
Urban	27.6	288.736	6.6	215.479
Rural	33.7		9.7
Household size^b
No violence experience	5.21	15.452	5.22	18.418
Experienced violence	5.28		5.36
Wealth index
Poorest	45.5	3,867.010	13.6	1,126.331
Poor	41.5		11.7
Middle	35.6		9.6
Richer	29.8		7.1
Richest	15.3		3.7
Region
South	30.2	1,225.747	4.1	1,623.721
North	23.6		7.5
Central	40.4		9.0
East	37.6		17.0
Northeast	28.7		9.2
West	25.0		3.7
Religion
Hindu	31.5	324.243	8.5	212.047
Muslim	35.1		10.9
Christian	21.4		4.4
Other	29.1		6.4
Number of living children
No children	21.7	1,331.175	7.6	146.740
1 to 2	26.8		7.2
3 to 4	36.4		9.3
5+	43.5		11.0
Wife’s characteristics
Age group
15 to 19	24.9	82.798	10.7	65.073
20 to 24	29.7		9.1
25 to 29	31.6		8.5
30 to 34	31.9		8.3
35 to 39	32.4		8.4
40 to 44	30.9		7.1
45 to 49	30.4		6.8
Age at first marriage
1<18	38.3	2,126.067	10.7	559.849
18 to 20	27.9		6.9
21+	18.3		4.8
Duration of marriage
0 to 4	19.3	966.244	6.8	51.909
5 to 9	30.0		8.3
10 to 14	33.9		9.1
15 to 19	34.5		8.9
20 to 24	33.9		8.7
25 to 29	35.0		7.8
30+	36.3		8.2
Education
No education	41.7	3,801.587	11.1	765.762
Primary	35.9		10.0
Secondary	22.6		6.0
Higher	8.9		2.5
Occupation
Not working	26.1	1,863.548	7.1	299.327
Sales, clerical	29.0		8.2
Professional, technical, managerial	15.1		4.5
Domestic, services	45.3		11.7
Agro-self employed	39.9		10.4
Skilled, unskilled manual	41.1		10.7
Wife-beating justified
Goes out without telling husband	37.8	613.779	10.1	114.539
Neglects children	36.3	524.872	9.4	55.609
Argues	38.2	676.380	10.3	138.872
Refuses sex	38.2	264.651	10.8	87.056
Burns food	38.6	424.329	10.2	75.840
Partner’s characteristics
Age
No violence experience	37.3	10.110	37.4	26.064
Experienced violence	37.5		36.8
Education
No education	42.5	2,898.461	11.4	603.324
Primary education	39.8		10.9
Secondary	27.5		7.2
Higher	13.7		3.8
Occupation
Not working	33.2	1,595.466	10.6	322.654
Sales, clerical	23.7		6.8
Professional, technical, managerial	15.8		4.1
Domestic, services	26.2		6.1
Agro-self employed	36.9		9.6
Skilled, unskilled manual	35.4		9.6
Drinks alcohol
No	23.1	3,449.209	6.0	793.871
Yes	44.4		12.2
All
Percent	31.0		8.3
Number of cases	69,416		69,432

All Chi Square and F values in this table are statistically significant at level p ≤ 0.001.

The values for household size are averages.

Wife’s characteristics

Table 2 data show that the highest level of physical violence (32.4%) was reported among respondents who were 35 to 39 years of age, whereas the highest level of sexual violence (10.7%) was reported among the 15 to 19 years old women. Data show that 38.3% of ever married women who first got married before the age of 18 years experienced physical violence and 10.7% experienced sexual violence. Of those respondents who got married at age 18 to 20 years, almost 28% (27.9%) and nearly 7% (6.9%) experienced physical and sexual violence respectively. Of those respondents who got married at age 21 years and older, 18.3% experienced physical violence and almost 5% (4.8%) experienced sexual violence. For duration of marriage, the highest proportion of physical violence (36.3%) was found among respondents who had been married for 30 years or more and the highest proportion of sexual violence (9.1%) was found among respondents who had been married for 10 to 14 years.

Table 2 also shows that ever married women who did not have an education reported the highest level of both physical (41.7%) and sexual (11.1%) violence. Of women who had higher education, almost 9% (8.9%) experienced physical violence and nearly 3% (2.5%) experienced sexual violence. Of those who had an occupation, respondents in domestic, services category reported the highest level of physical and sexual violence (45.3% & 11.7% respectively). The lowest physical violence (15.1%) and sexual violence (4.5%) was reported by respondents in professional, technical, and managerial category. Furthermore, amongst the categories of justifying wife-beating, the level of physical violence was the highest (38.6%) among those who justified wife-beating for burning food, whereas the highest report of sexual abuse (10.8%) was reported by those who justified wife-beating for refusing sex with partner.

Husband’s characteristics

The average age of husbands did not make a major difference for ever married women experiencing physical and sexual violence. The data show that husband’s education is negatively associated with both physical and sexual violence. The highest levels of physical (42.5%) and sexual (11.4%) were reported violence was higher among ever married women whose husbands had no education. Ever married women whose husbands held professional, technical, or managerial jobs reported less domestic violence than their counterparts whose husbands held other job categories. Of the women whose husbands used alcohol, 44.4% and 12.2% experienced physical and sexual violence respectively. The data show that alcohol drinking is associated with high rates of domestic violence.

Multivariate Logistic Regression Models—Physical and Sexual Violence Results

Physical violence

In Table 3 we present models showing the association between physical violence (one of our two dependent variables) and sociodemographic characteristics of ever married women. The results are presented in four models. The first model includes the household variables. The second model contains all the variables in Model I minus the wealth index to show how that variable changes the association between type of residence and both physical and sexual violence. The third model contains all variables from Model I, plus wife’s characteristics. Model IV includes all variables from Model III plus husband’s characteristics.

Table 3.

Odds Ratios of Logistic Regression of Wife’s Reporting of Physical Violence, India 2005-2006 NFHS-III

	Model I	Model II	Model III	Model IV
Household variables
Place of residence
Urban (ref.)	1.000	1.000	1.000	1.000
Rural	0.696****	1.251****	0.688****	0.733****
Household size	0.976****	0.959****	0.985****	0.988***
Wealth index
Poorest (ref.)	1.000		1.000	1.000
Poor	0.896****		0.961	0.997
Middle	0.708****		0.815****	0.864****
Richer	0.507****		0.661****	0.726****
Richest	0.206****		0.380****	0.448****
Region
South (ref.)	1.000	1.000	1.000	1.000
North	0.745****	0.638****	0.793****	0.858****
Central	1.259****	1.377****	1.441****	1.682****
East	1.091***	1.274****	1.248****	1.407****
Northeast	0.911***	0.977	1.056*	1.016
West	0.840****	0.774****	0.892****	1.031
Religion
Hindu (ref.)	1.000	1.000	1.000	1.000
Muslim	1.055**	1.105****	1.061**	1.385****
Christian	0.629****	0.565****	0.664****	0.646****
Other	1.227****	1.054	1.217****	1.074
Number of living children
No children (ref.)	1.000	1.000	1.000	1.000
1 to 2	1.554****	1.420****	1.261****	1.365****
3 to 4	2.111****	2.233****	1.385****	1.523****
5+	2.475****	3.029****	1.574****	1.756****
Wife’s characteristics
Age group
15 to 19 (ref.)			1.000	1.000
20 to 24			1.202****	1.237****
25 to 29			1.140**	1.158**
30 to 34			1.095	1.151*
35 to 39			1.109	1.202**
40 to 44			1.035	1.166
45 to 49			1.009	1.180
Age at first marriage
1<18 (ref.)			1.000	1.000
18 to 20			0.863****	0.875****
21+			0.770****	0.777****
Duration of marriage
0 to 4 (ref.)			1.000	1.000
5 to 9			1.435****	1.421****
10 to 14			1.566****	1.565****
15 to 19			1.529****	1.570****
20 to 24			1.437****	1.468****
25 to 29			1.502****	1.598****
30+			1.440****	1.565****
Education
No education (ref.)			1.000	1.000
Primary			0.975	0.997
Secondary			0.727****	0.789****
Higher			0.363****	0.450****
Occupation
Not working (ref.)			1.000	1.000
Sales, clerical			1.276****	1.251***
Professional, technical, managerial			1.425****	1.276****
Domestic, services			1.238****	1.181****
Agro-self employed			1.732****	1.467****
Skilled, unskilled manual			1.439****	1.271****
Wife-beating justified
Goes out without telling husband			0.915****	0.913****
Neglects children			0.844****	0.836****
Argues			0.858****	0.837****
Refuses sex			1.037	1.052
Burns food			0.976	0.964
Partner’s characteristics
Age				0.991****
Education
No education (ref.)				1.000
Primary education				1.088***
Secondary				0.959
Higher				0.850****
Occupation
Not working (ref.)				1.000
Sales, clerical				0.785***
Professional, technical, managerial				0.809***
Domestic, services				0.851**
Agro-self employed				0.725****
Skilled, unskilled manual				0.867**
Drinks alcohol				2.473****
Model chi square	5,796.649	2,955.077	7,443.793	9,419.781
Degree of freedom	17	13	44	54
Number of cases	69,325	69,325	66,761	62,300

Ref.: reference category

****

p ≤ 0.001. ***p ≤ 0.01.**p ≤ 0.05. *p ≤ 0.10.

As Table 3 suggests, most of the household variables are statistically significantly associated with physical violence in India. However, an interesting pattern emerged when place of residence and wealth index were both included in the logistic regression equation (Models I, III, and IV). Results show that, compared to urban residents, women living in rural areas were less likely to experience physical violence from their partners in all models. Such finding is contrary to what was observed in descriptive analysis (Table 2), which showed that rural residence were more likely to experience both physical and sexual violence than their urban counterparts. To determine the source of such a discrepancy, we ran various models (not reported here) with residence variable plus each of the other variables. The results showed that wealth index is the variable whose mediating effect changes the association between type of place of residence and physical and sexual violence (see Model II in both Tables 3 and 4). This result suggests that the positive association observed between rural residence and wife abuse is mostly due to the prevalence of poverty in rural areas.

Table 4.

Odds Ratios of Logistic Regression of Wife’s Reporting of Sexual Violence, India 2005-2006 NFHS-III

	Model I	Model II	Model III	Model IV
Household variables
Place of residence
Urban (ref.)	1.000	1.000	1.000	1.000
Rural	0.890****	1.374****	0.901***	0.983
Household size	1.000	0.987**	1.001	1.008
Wealth index
Poorest (ref.)	1.000		1.000	1.000
Poor	0.969		1.015	1.046
Middle	0.844****		0.919*	0.955
Richer	0.615****		0.744****	0.783****
Richest	0.313****		0.505****	0.576****
Region
South (ref.)	1.000	1.000	1.000	1.000
North	2.056****	1.822****	2.222****	2.564****
Central	1.978****	2.126****	2.233****	2.692****
East	4.018****	4.451****	4.521****	5.412****
Northeast	2.751****	2.876****	3.116****	3.220****
West	1.016	0.953	1.080	1.228**
Religion
Hindu (ref.)	1.000	1.000	1.000	1.000
Muslim	1.167****	1.235****	1.172****	1.527****
Christian	0.433****	0.406****	0.445****	0.449****
Other	0.781****	0.701****	0.755****	0.703****
Number of living children
No children (ref.)	1.000	1.000	1.000	1.000
1 to 2	1.085	1.027	1.010	1.038
3 to 4	1.194****	1.261****	1.010	1.051
5+	1.161**	1.353****	1.012	1.038
Wife’s characteristics
Age group
15 to 19 (ref.)			1.000	1.000
20 to 24			0.991	1.040
25 to 29			0.905	0.935
30 to 34			0.886	0.923
35 to 39			0.965	1.027
40 to 44			0.884	0.962
45 to 49			0.917	1.027
Age at first marriage
1<18 (ref.)			1.000	1.000
18 to 20			0.788****	0.792****
21+			0.760****	0.729****
Duration of marriage
0 to 4 (ref.)			1.000	1.000
5 to 9			1.140**	1.093
10 to 14			1.235***	1.161*
15 to 19			1.160	1.104
20 to 24			1.072	1.008
25 to 29			0.940	0.870
30+			0.910	0.844
Education
No education (ref.)			1.000	1.000
Primary			1.093**	1.102**
Secondary			0.884***	0.977
Higher			0.519****	0.686****
Occupation
Not working (ref.)			1.000	1.000
Sales, clerical			1.390***	1.233*
Professional, technical, managerial			1.565****	1.292***
Domestic, services			1.239****	1.180****
Agro-self employed			1.741****	1.456****
Skilled, unskilled manual			1.362****	1.124**
Wife-beating justified
Goes out without telling husband			0.907**	0.886***
Neglects children			0.916**	0.924*
Argues			0.932*	0.927*
Refuses sex			0.863***	0.867***
Burns food			0.991	0.969
Partner’s characteristics
Age				1.000
Education
No education (ref.)				1.000
Primary education				1.120**
Secondary				1.003
Higher				0.904
Occupation
Not working (ref.)				1.000
Sales, clerical				0.668***
Professional, technical, managerial				0.739***
Domestic, services				0.701***
Agro-self employed				0.581****
Skilled, unskilled manual				0.769**
Drinks alcohol				2.082****
Model chi square	2,500.272	1,962.319	2,855.584	3,281.163
Degree of freedom	17	13	44	54
Number of cases	69,341	69,341	66,776	62,310

Ref.: reference category

****

p ≤ 0.001. ***p ≤ 0.01.**p ≤ 0.05. *p ≤ 0.10.

Household size was negatively associated with physical violence. Ever married women in a household with more people were less likely to experience physical violence in Models I to III; however, when the husband’s characteristics were added in Model IV, there was a slight decline in the statistical significance (from p<0.001 to p<0.01).

As found in descriptive statistics, household wealth is negatively associated with physical violence in all models. There are also significant differences by region of residence. Data in all models show that ever married women living in North India were significantly less likely to experience physical violence than ever married women living in the South. Also, women living in the West were statistically significantly less likely to experience physical violence in Models I to III; but when the husband’s characteristics were added in Model IV the relationship became insignificant. Furthermore, Northeastern women were significantly less likely to experience physical violence only in Model I. In contrast, ever married women living in Central and East India were significantly more likely to experience physical violence in all models compared with ever married women in South India.

In terms of religion, compared with Hindus, ever married Christian women were significantly less likely to experience physical violence in all models. The Muslim category was statistically significantly more likely to experience physical violence in all models, although the association became even more significant when the husband’s characteristics are added in Model IV. Those ever married women in the “other” religious category were more likely to experience physical violence in Model I and III; however, when husband’s characteristics were added to the analysis in Model IV, the effect became insignificant. Ever married women with children were vulnerable to physical violence than women with no children. In all models the levels of violence were statistically significant.

In all models, among the wife’s characteristics, age groups 20 to 24 and 25 to 29 were statistically significantly more likely to experience physical violence than age group 15 to 19. Respondents in age groups 30 to 34 and 35 to 39 were statistically significantly more like to have experienced physical violence only in Model IV; the other age group categories were not significant. Age at first marriage was associated with physical violence for Models III and IV. Women who got married at age 18 to 20 years old or 21 and older were statistically significantly less likely to experience physical violence when compared to those who first got married at the age younger than 18 years old.

Marital duration was significantly correlated with physical violence in all models; although the relationship was not all linear. The data in Table 3 show that female education is significantly associated with a reduction in physical violence in India. Women with a secondary education or higher were statistically significantly less likely to experience physical violence. For occupation, all categories were significantly more likely to experience physical violence. Clearly, labor force participation increases women’s risk of physical violence in India. In contrast, ever married women whose husband’s had an occupation were significantly less likely to experience physical violence (Model IV). Ever married women who agree with statements that wife-beating is justified when a wife goes out without telling her husband, neglects children, or argues with her husband were more likely to be physically abused than those who disagree with such statements (see Models III and IV).

The remaining variables in Model III were all statistically significant. Ever married women whose husbands had a primary education were more likely to experience physical violence than those whose husbands had no education. Those whose husbands had higher education were significantly less likely to report physical violence experiences in the 12 months before survey. Also, husband’s age was a buffer against physical violence. Husband’s alcohol drinking was associated with wife’s report of physical abuse. Women living with partners who drink alcohol were significantly more likely to experience physical violence than those whose husbands did not drink alcohol.

Sexual violence

In Table 4 we present models showing the association between sexual violence (the second dependent variable) and sociodemographic characteristics of ever married women. Similar to physical violence, the analysis of sexual violence are presented in four models. Compared with their counterparts in urban areas, ever married women living in rural areas were less likely to experience sexual abuse in Models I and IV. However, like in the case of physical violence, such an effect is mediated through wealth index. When wealth index was omitted from the logistic regression equation (Model II), rural residents became more likely to have experienced sexual violence than their urban counterparts. This confirms our assumption of the mediating effect of poverty on wife abuse. In other words, the positive association observed in the bivariate relationship between rural residence and sexual violence is partly due to higher poverty in rural areas.

Among the geographical regions, all of the categories were strongly associated with increased likelihood of sexual violence when compared with the South with the exception of ever married women living in the West, who were significantly likely to experience sexual violence only in Model IV after adding the husband’s characteristics. As Models I to IV reveal, ever married women living in the North, Central, East, and Northeast were statistically significantly more likely to experience sexual violence than those in the South. However, when the husband’s characteristics were added to the models, the significant level disappeared for women in the Western region in Models I to III. Religion is also a strong correlate of sexual abuse. Compared with the Hindu religious category, only the Muslim religious category was significantly associated with an increase in the likelihood of sexual violence in all models; Christians and those in the “Other” category were significantly less likely to experience sexual violence across all models. Furthermore, all ever married women who had more than 2 (those in the 3-4 and 5+ categories) children were vulnerable to sexual violence only in Models I and II.

Among wife’s characteristics, significant levels were varied. Compared with those who first got married before age 18, women who got married at ages 18 to 20 and 21and older were statistically significantly less likely to experience sexual violence. For marital duration, compared with those who had been married for 0 to 4 years, only women who were married for 5 to 9 years and 10 to 14 years were more likely to be sexually abused. However, when the husband’s characteristics were added into the logistic regression equation in Model IV, only those who had been married for 10 to 14 years were significantly more likely to have experienced sexual violence.

For education, only ever married women with a primary education were significantly more likely to experience sexual violence. Those with higher educational attainment were less likely to experience sexual violence. Furthermore, ever married women in all occupation categories were vulnerable to sexual violence. Also, those who justified wife-beating were significantly less likely to experience sexual violence, except for burning food. It is surprising to note that only ever married women whose husbands had a primary education were more likely to experience sexual violence than those whose husbands had no education. Having an employed husband was a buffer to sexual violence for the ever married women. Wives whose husbands were working were less likely to experience sexual violence than those married to unemployed men. Finally, ever married women whose husbands used alcohol were significantly more likely to experience sexual violence than their counterparts whose husbands did not drink.

Discussion

Although domestic violence is a universal phenomenon that threatens the overall well-being of women, it is only recently that it has emerged as a global issue. In India, despite the tremendous impact it has on the woman, her family and society, it continues to be a “crime of silence” (Pillai, 2001). This study examined physical violence and sexual violence among the ever married women who were administered the domestic violence module as part of the 2005-2006 INFHS. We recognize that the reported statistics could be underestimated due to the sense of shame and embarrassment, coupled with the need to protect family honor that may keep many women silent. Also, cultural demands on women to be subservient and accepting of their husbands could be another reason why certain women may remain silent. Results are not consistent across physical and sexual violence; however, we discuss some of our most salient findings below.

Our findings show variation in domestic violence across geographic regions. That we should find variation in domestic violence reports across regions should not be surprising. This is especially true given the evidence regarding geographic variations in female autonomy (Dyson & Moore 1983) and social inequality (Bhengra, Bijoy, & Luithui, 1999) found in India. Their description of regional patterns of social norms is still relevant today. Ever married women in Central and East reported experiencing domestic violence, with women in the West, Northeastern, and North India reporting less physical violence than those in South India. For sexual violence, although all regions were highly correlated with sexual violence, the West was significantly associated with sexual violence only when the husband’s characteristics were added in the logistic regression equation. Nonetheless, these findings do not explain nuances within each region, which clearly needs further research. We also found that household size offers protection against physical but not sexual violence. Having more people living in a household may act as a buffer against domestic violence, particularly physical violence. It could be that household members share household work, thus taking away stress that might otherwise precipitate physical violence. As for sexual violence, we speculate that due to the private and intimate nature of the act, having more people in the household may not be as protective as with physical violence.

Wealth index is a household economic status indicator associated with domestic violence. Ever married women in the poorest category reported the highest rates of physical violence than their counterparts in the middle to richest categories for physical violence and in the richer to richest categories for sexual violence. However, we have to be cognizant of the fact that domestic violence is not limited to a particular social class. In relative terms, domestic violence may be less among the richest, when we consider that stress associated with poverty is unlikely to be present in this group. Studies from a range of settings show that although violence against partners cuts across all socioeconomic groups, women living in poverty are disproportionately affected (Djamba & Kimuna, 2008; Ellsberg et al., 1999; Kimuna & Djamba, 2008). Other studies have reported that economic security is protective against physical or sexual violence (Jejeebhoy, 1998b; Smith-Fawzi et al., 2005). But it may be, also, that ever married women in the richest households are more sensitive to public opinion and are more likely to provide socially desirable responses to survey researchers and to hide their experiencing violence than women living in poorer households. Also in recent years, domestic violence policies in India and significant changes in beliefs regarding gender differences have begun to change women’s attitudes toward violence and women from advantaged households seem to be those who are changing their attitudes most quickly indicating that they are decreasing their tolerance of intimate partner violence (Nayak, Byrne, Martin, & Abraham, 2003).

Education was associated with both physical and sexual violence with variation in experiencing physical or sexual violence. Ever married women who had primary education were less likely to report experiencing physical violence, but were vulnerable to sexual violence. First, the former finding echoes Jejeebhoy’s (1998b) finding that secondary education and higher were protective against women experiencing physical violence in Tamil Nadu. However, other studies from India report conflicting results. One study from rural Uttar Pradesh found a decreased risk of physical abuse with greater than 7 years of education among both husbands and wives. In summary, the relationship between education and domestic violence is complex and requires qualitative studies for better understanding these conflicting results.

We found that age at first marriage has a significant effect on both physical and sexual violence. For both physical and sexual violence, the relationship is negative, suggesting that women who marry at older ages may be more mature and may manage marital conflicts better than their younger counterparts. As for duration of marriage, we found that ever married women were vulnerable to physical violence; but only those women who had been married for 5 to 9 years and 10 to 14 years were more likely to experience both physical and sexual violence. We can speculate that it is because older women in an abusive relationship of long duration are less bothered about social stigma and are more forthright in admitting to domestic violence. This finding is similar to a study by Sahoo and Pradhan (2007), which noted that Indian women married for 5 years or longer were more likely to be beaten by their husbands.

Number of children was significantly associated with physical violence; but with sexual violence only in the model that did not contain wife’s and husband’s characteristics and only for women who had more than 3 children, suggesting that individual characteristics and socialization factors are perhaps more important in India. Furthermore, ever married women who had an occupation were at significant risk of experiencing both physical and sexual violence. These results could be explained by looking at traditional power structure in patriarchal Indian families and females earning money could be perceived as a threat to traditional male dominance. A man may be threatened by independence that a woman shows or by the monetary contribution that she makes to the household. Justification of wife-beating was associated with lower risk of both physical and sexual violence. This finding suggests that wives who believe that their husbands may use power to correct them may adopt submissive strategies just to avoid reprimand or confrontation.

In study after study, a man’s use of alcohol is a major risk factor for both physical and sexual violence. This finding is echoed in this study as well. The near universal response that alcohol drinking is a major precipitating factor for physical and sexual violence is consistent with many other studies on domestic violence (Djamba & Kimuna, 2008; Jejeebhoy, 1998a, 1998b; Kimuna & Djamba, 2008; Koenig et al., 2004; 2006; Rao, 1997). Although the literature does not show that alcohol use definitively causes domestic violence, it has consistently emerged as a risk factor for domestic violence albeit only suggestive.

Conclusion

Overall, this study is among the few that have used a national sample to examine the level and determinants of physical and sexual violence among ever married women in India. Our results show that gender role conditioning and cultural norms both contribute to domestic violence. Similar to other countries around the world, implicit cultural norms may be the culprit in domestic violence. As in the United States (Ferraro, 1993; Hart, 1993), battered women in India are hesitant to report and prosecute the perpetrators. It is perceived among Indian women that such violence is a problem of domestic violence and it is the woman’s fault for provoking it; and the tendency of laws and law enforcement to trivialize domestic violence.

Although policies have been created to curb such violence, women in India still conform to the adapted cultural expectation of society and the family. Prescribed gender roles place a heavy burden on these women. Social norms related to marriage dictate various circumstances in a married woman’s life. The patriarchal notions of male superiority and power and their socialization to accept the husband as head of the household seem to condition women to accept violence in their lives and relationships. Furthermore, keeping their marriages intact, despite experiencing violence, may not only be a traditional expectation, but it may also be for practical reasons such as financial support, care for their children, and having no other place to go. Moreover, in male dominated societies such as India, divorce and remarriage are often stigmatized and divorced women may face more hardships and may often be a target for further exploitation.

In summary, addressing domestic violence requires changes at the institutional and legal levels, following a human rights-based approach (Jewkes et al., 2002). Although the Indian government is obliged both by its Constitution and its affirmation of the Convention on the Elimination of All Forms of Discrimination Against Women to take effective steps to eliminate violence against women, the government has had little accountability for the massive gap between legal clauses and the reality of pervasive domestic violence (Bhattacharya, 2004; Jejeebhoy & Cook, 1997). Important steps that could be taken at the public policy level include reforming public institutions by making police officers more receptive to the needs of victimized women and enforcing existing legislation.

Footnotes

Declaration of Conflicting Interests

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

Funding

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

Bios

Sitawa R. Kimuna, PhD, is an associate professor of sociology at East Carolina University. She earned her MSc in mass communication, MA and PhD degrees in sociology from Kansas State University. Her research interests include gender based violence, aging, social demography/population dynamics, health in sub-Saharan Africa, race and ethnic relations: American and global and the social impact of mass media.

Yanyi K. Djamba, PhD, is the director of the Center for Demographic Research and professor of sociology at Auburn University at Montgomery (AUM). He earned MA and PhD degrees in sociology from Louisiana State University. Before joining AUM, He worked as a postdoctoral research fellow at the University of Texas at Austin, research associate at Brown University, assistant and then associate professor at Southeastern Louisiana University. He is the author of The Black Population in Alabama, State of Marriage in Alabama, Sexual Behavior of Adolescents in Contemporary Sub-Saharan Africa, and many articles in peer-reviewed journals, book chapters, and study reports.

Gabriele Ciciurkaite, MA is currently a PhD student in the department of sociology at the University of Kentucky. She earned her MA in sociology at East Carolina University in 2011.

Suvarna Cherukuri, PhD, is an associate professor of sociology at Siena College and the author of Women in Prison and Insight into Captivity and Crime (Cambridge University Press, India). Her research interests are in gendered carceral regimes and violence against women.

References

Ackerman

L. K.

Subramanian

S. V.

(2008). State gender inequality, socioeconomic status and intimate partner violence in India: A multilevel analysis. Australian Journal of Social Issues, 43(1), 81-102.

Agarwal

(1995). A field of one’s own: Gender and land rights in South Asia [Cambridge South Asian Studies Series No 58]. Cambridge: Cambridge University Press.

Agnes

(1992). Protecting women against violence: Review of a decade of legislation, 1980-89. Economic and Political Weekly, 27(17), 19-33.

Ahmed-Ghosh

(2004). Chattels of society: Domestic violence in India. Violence Against Women, 10, 94-118.

Ahmed

Koenig

M. A.

Stephenson

(2006). Effects of domestic violence on perinatal and early- childhood mortality: Evidence from north India. American Journal of Public Health, 96, 1423–1428.

Asling-Monemi

Pena

Ellsberg

M. C.

Persson

L. A.

(2003). Violence against women increases the risk of infant and child mortality: A case-referent study in Nicaragua. Bulletin of the World Health Organization, 81(1), 10-18.

Barzelatto

(1998). Understanding sexual and reproductive violence: An overview. International Federation of Gynecology and Obstetrics, 63, Supplement 1, S33-S42.

Bhattacharya

(Ed.). (2004). Behind closed doors: Domestic violence in India. New Delhi, India: Sage.

Bhengra

Bijoy

C. R.

Luithui

(1999). The adivasis of India, London, UK: Minority Rights Group International.

10.

Bruce

Clark

(2003). In technical consultation on married adolescents. Paper presented for the WHO/UNFPA/Population Council, Population Council, Geneva, Switzerland.

11.

Campbell

J. C.

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

12.

Cherukuri

(2008). Women in prison: An insight into captivity and crime. New Delhi: Cambridge University Press India.

13.

Coker

A. L.

Smith

P. H.

McKeown

R. E.

King

M. J.

(2000). Frequency and correlates of intimate partner violence by type: Physical, sexual, and psychological battering. American Journal of Public Health, 90, 553-559.

14.

Cokkinides

V. E.

Coker

A. L.

Sanderson

(1999). Physical violence during pregnancy: Maternal complications and birth outcomes. Obstetrics and Gynecology, 93, 661-666.

15.

Djamba

Y. K.

Kimuna

S. R.

(2008). Intimate partner violence among married women in Kenya. Journal of Asian and African Studies, 43, 457-469.

16.

Dyson

Moore

(1983). On kinship structure, female autonomy and demographic behavior in India. Population Development Review, 9, 35-60.

17.

Elizabeth

(1999). Patterns and Trends of Domestic Violence in India: An Examination of Court Records. Bangalore: Centre for Women and Law, National Law School of India.

18.

Ellsberg

Heise

Pena

Agurto

Winkvist

(2001). Researching domestic violence against women: Methodological and ethical considerations. Studies in Family Planning, 32(1), 1-16.

19.

Ellsberg

M. C.

Pena

Herrera

Liljestrand

Winkvist

(1999). Wife abuse among women of childbearing age in Nicaragua. American Journal of Public Health, 89(2),241-244.

20.

Ellsberg

Peña

Herrera

Liljestrand

Winkvist

(2000). Candies in hell: Women’s experience of violence in Nicaragua. Social Science and Medicine, 51, 1595-1610.

21.

Fernandez

(1997). Domestic violence by extended family members in India. Journal of Interpersonal Violence, 12, 433-455.

22.

Ferraro

(1993). Cops, courts, and woman battering. In Bart

A. P. B.

Moran

E. G.

(Eds.), Violence against women (pp. 151-176). Newbury Park, CA: Sage.

23.

García

(Ed.). (2000). Women, poverty and demographic change. Liege, Belgium: Oxford University Press for International Union for Scientific Study of Population.

24.

Hart

(1993). Battered women and the criminal justice system. American Behavioral Scientists, 36, 624-638.

25.

Heise

Pitanguy

Germain

(1994). Violence against women: The hidden health burden. [World Bank Discussion Paper, No 255]. Washington, DC: World Bank.

26.

Indian Express. (2005). Two-third married Indian women victims of domestic violence: UN. Retrieved from http://www.expressindia.com/news/fullstory.php?newsid=56501

27.

International Centre for Research on Women. (1999). A summary report of three studies, promoting women in development project. Washington, DC: Author.

28.

International Institute of Population Sciences & Macro International (IIPS). (2007). National Family Health Survey (NFHS-3), 2005–06. Mumbai, India: Author. Retrieved from www.measuredhs.com

29.

Jaising

(2009). Bringing rights home: Review of the campaign for a law on domestic violence. Economic and Political Weekly, 44(44), 50-57.

30.

Jaswal

(1999): Health records and domestic violence in Thane District,Maharashtra: A summary report. Mumbai, India: Department of Medical and Psychiatric Social Work, Tata Institute of Social Science.

31.

Jejeebhoy

S. J.

(1998a). Associations between wife-beating and fetal and infant death: Impressions from a survey in rural India. Studies in Family Planning, 29, 300-308.

32.

Jejeebhoy

S. J.

(1998b). Wife-beating in rural India: A husband’s right? Economic and Political Weekly, 33, 855-862.

33.

Jejeebhoy

S. J.

Cook

R. J.

(1997). State accountability for wife-beating: The Indian challenge. Lancet, 349, SI10-SI12.

34.

Jewkes

Levin

Penn-Kekana

(2002). Risk factors for domestic violence: Findings from a South African cross-sectional study. Social Science Medicine, 55, 1603-1617.

35.

Kapadia-Kundu

Dyalchand

Kulkarni

(2004). Physical abuse and its implications for the reproductive health of women in rural India [Unpublished].

36.

Karlekar

(1998). Domestic violence. Economic and Political Weekly, 33, 1741-1751.

37.

Kimuna

S. R.

Djamba

Y. K.

(2008). Gender based violence: Correlates of physical and sexual wife abuse in Kenya. Journal of Family Violence, 23, 333-342.

38.

Kishor

Johnson

(2004). Profiling domestic violence: A multi-country study. Calverton, MD: ORC Macro.

39.

Kiswar

(2005). Destined to fail: Inherent flaws in the anti-dowry legislation. Manushi: A Journal about Women and Society, 148, 1-6.

40.

Kishwar

(1988). Rethinking dowry boycott. Manushi: A Journal about Women and Society, 48, 10-13.

41.

Koenig

M. A.

Stephenson

Ahmed

Jejeebhoy

Campbell

(2006). Individual and contextual determinants of domestic violence in North India. American Journal of Public Health, 96(1), 131-139.

42.

Koenig

Zablotska

Lutalo

Nalugoda

Wagman

Gray

(2004). Coerced First Intercourse and Reproductive Health Among Adolescent Women in Rakai, Uganda. International Family Planning Perspectives, 30(4), 156-163.

43.

Krishnaraj

(2007). Understanding violence against women. Economic and Political Weekly, 42 (44), 90-91.

44.

Madhurima

(1996): Violence against women: Dynamics of conjugal relations. New Delhi, India: Gyan House.

45.

Malhotra

Mather

(1997). Do schooling and work empower women in developing countries? The case of Sri Lanka. Sociological Forum, 12, 599-630.

46.

Martin

L, Tsui

A. O.

Maitra

Marinshaw

(1999). Domestic violence in northern India. American Journal of Epidemiology, 150, 417-426.

47.

National Crime Record Bureau. (2010). Crimes against Women. New Delhi, India: Author.

48.

Nayak

M. B.

Byrne

Martin

M. K.

Abraham

A. G.

(2003). Attitudes toward violence against women: A cross-nation study. Sex Roles, 49, 333-342.

49.

Parker

McFarlane

Soeken

(1994). Abuse during pregnancy: Effects on maternal complications and birth weight in adult and teenage women. Obstetrics and Gynaecology, 84, 323-328.

50.

Pillai

(2001). Domestic violence in New Zealand: An Asian immigrant perspective. Economic and Political Weekly, 38, 965-974.

51.

Rao

(1997). Wife beating in rural South India: A qualitative and econometric analysis. Social Science and Medicine, 44, 1169-1180.

52.

Sahoo

Pradhan

M. R.

(2007, February 1-2). Domestic violence in India: An empirical analysis. Presented at the Seminar on Gender Issues and Empowerment of Women, Indian Statistical Institute, Kolkata, India.

53.

Singh

Mahapatra

Datta

(2008). Spouse abuse and infant and child mortality in India. Asia Pacific Population Journal, 23, 33-54.

54.

Smith-Fawzi

M. C.

Lambert

Singler

J. M.

Tanagho

Leandre

Nevil

(2005). Factors associated with forced sex among women accessing health services in rural Haiti: Implications for the prevention of HIV infection and other sexually transmitted diseases. Social Science & Medicine, 60, 679-689.

55.

Straus

M. A.

(1990). Measuring intrafamily conflict and violence: The conflict tactic (CT) scales. In Straus

M. A.

Gelles

R. J.

(Eds.), Physical violence in American families: Risk factors and adaptations to violence in 8,145 families (pp. 29-47). New Brunswick, NJ: Transaction.

56.

Suneetha

Nagaraj

(2006). A difficult match: Women’s actions and legal institutions in the face of domestic violence. Economic and Political Weekly, 41, 4355-4362.

57.

Verma

R. K.

Collumbien

(2003). Wife beating and the link with poor sexual health and risk behaviour among men in urban slum in India. Journal of Comparative Family Studies, 34, 61-73.

58.

Vindhya

(2002). Dowry deaths and domestic harassment of women in Andhra Pradesh: An analysis of socio-cultural dimensions and judicial outcomes of cases. Visakhapatnam, India: Institute of Development and Planning Studies.

59.

Visaria

(1999). Violence against women in India: Evidence from rural Gujarat. In Burton

Duvvury

Rajan

Varia

(Eds.), Domestic violence in India: A survey report of three studies (pp. 9-17). Washington DC: International Centre for Research on Women (ICRW).

60.

Visaria

(2000). Violence against women: A field study. Economic and Political Weekly, 43, 1742-1751.

61.

World Health Organization. (2002). World report on violence and health. Geneva, Switzerland: Author.

62.

World Health Organization. (2005). Multi-country study on women’s health and domestic violence against women initial results on prevalence, health outcomes and women’s responses. Geneva, Switzerland: Author.

63.

World Health Organization. (2006). Gender based violence in the Western Pacific region: A hidden epidemic? [Special issue on women’s health]. Geneva, Switzerland: Author.