Socioeconomic Determinants of Domestic Violence in Northeast India: Evidence From the National Family Health Survey (NFHS-4)

Abstract

Violence against women is now widely recognized as an important public health problem, owing to its health consequences. Domestic violence generally refers to violence against women that generally takes place in the matrimonial home, which includes physical, emotional, sexual as well as economical violence occurring in the adult relationship. Domestic violence has negative impact on the demographic outcome of a country or region, which includes the health and health seeking behavior of women. The present study is an attempt to investigate the socioeconomic factors that would significantly influence domestic violence in the northeastern region of India. The northeastern region of India comprises of eight states, including Manipur and Sikkim, which are reported having the highest and the least prevalence of domestic violence in the whole country according to NFHS-4. Considering these reports and having a different socioeconomic setup as compared to other states in the mainland India, we selected the region for the current study. Data from NFHS-4 conducted during 2015–16 is being employed for the present study. Findings show that education of women, wealth, women’s working status, religion, parity of women, and husband’s alcohol use are some of the significant covariates that might have influenced domestic violence. The overall prevalence of physical, emotional, sexual, and severe physical violence among women of northeastern India are 24.7%, 11.8%, 6.4%, and 5.3%, respectively. Logistic regression analysis computes the relative risks of categorical variables via the odds ratios. Those women who got higher education and are in upper wealth quintiles have less risk of experiencing domestic violence as compared to women in lower education and lower wealth quintile in northeastern India. There is no rural urban difference in experiencing domestic violence in the region.

Keywords

violence against GLBT domestic violence alcohol and drugs sexual assault

Introduction

Domestic violence (DV) generally refers to violence against women that generally takes place in the matrimonial homes. The type of violence includes not only physical but also psychological in which the violent behavior attempts to dominate or gain a complete control over the victim’s right to live. Though the main perpetrator in a DV happens to be the husband, all persons including the in-laws and persons intimately or otherwise connected to the women through marriage are often involved in many cases of DV.

United Nation’s framework for nodal legislation on DV states that all acts of gender-based-violence: physical, psychological, and abuse by a family member against women in the family, ranging from single assault to aggravated physical assaults, kidnapping, threats, intimation, coercion, stalking, humiliating verbal use, forcible or unlawful entry, arson, destruction of property, sexual violence, marital rape, dowry related violence, female genital mutilation, violence related to exploitation through prostitution, violence against household workers, and attempts to commit such acts shall be termed DV.

Global Scenario

Internationally, one in three women have been beaten, coerced into sex or abused in their lifetime by a member of her own family (Heise et al., 1999). The most common type of abuse to women worldwide is DV or the physical, emotional, and sexual abuse of women by their intimate partners. There are other forms of abuse, such as dowry harassments, which are linked to traditional or customary practices and are limited to specific regions and communities. In the developing countries, the problem of violence is gradually increasing with different dimensions. Abuse by spouses or partner abuse is not only physically or psychologically devastating to women, but it can also be similarly devastating for children (Rahman & Kabir, 2006). The international community has recognized that DV is a serious violation of human rights, and direct demage to general health and well-being and autonomy of women. It adversely affects women’s self-esteem and quality of life. DV occurs in all countries of the world but its prevalence varies greatly across the world and even within the sub-Saharan Africa. Irrespective of social, economic, religious, and cultural groupings, men have been identified as the main perpetrators of DV against women (Owusu Adjah & Agbemafle, 2016).

About 63% of the Pakistan population lives in rural areas (World Bank, 2011) and women living in rural areas are at increased risk of DV than their urban counterparts. Various factors include rural women’s relatively weaker social position, low literacy rate, and lower level of awareness about their social and legal rights, and due to the structural disadvantages, rural women lack capacity building opportunities and access to economic resources and are usually dependent on men for their day-to-day subsistence (Zakar et al., 2015). The socio-economic status of women in Pakistan is low, particularly in rural areas. Overall, women in Pakistan face various types of discrimination. The Global Gender Gap Report (World Economic Forum, 2013) showed that Pakistan is ranked at 135 out of 136 countries experiencing worst gender disparities.

DV is one of the most costly and severe health problems facing in today’s society affecting all cultural backgrounds, sexual orientations, ages, ethnicities, education levels, and social economic classes. According to the National Intimate Partner Violence and Sexual Violence Survey (Black et al., 2011), a nationally representative study of over 9,000 women, approximately one in three women (36%) in the United States have experienced rape, physical violence, and/or stalking by a domestic partner in their lifetime. Generally, women who belong to marginalized or oppressed groups are at greater risk for experiencing DV, as they are less likely to have the resources available to flee abusive situations. In fact, studies suggest that as a woman’s financial stability increases, the likelihood that she will experience DV decreases.

Indian Scenario

India is not an exception to the menace of DV, and DV of different degrees in varying dimensions has been continuously reported from across the states in India. In the Indian context, DV includes causing any harm or injury to the safety, life, health, or well-being of the aggrieved woman by committing any physical, sexual, verbal, or economic abuse. Moreover, it also includes any injury or harm done to the aggrieved woman or her relative with a view to coerce her. Threats to commit such violence are also covered under this definition.

India is a vast country having several communities, which are distinct in their geography, language, and culture. In several places of India, violence faced by women on a regularly basis goes unreported. As per National Crime Records Bureau (NCRB, 2017), a crime has been recorded against women in every 3 minutes in India. United Nation Population Fund Report (UNPFR), indicated that around two-third of married Indian women are victims of DV attacks and as many as 70% of married women in India between the age 14 and 49 are victims of beating, raped or forced sex.

The third national family health survey revealed that there is considerable variation across the states in the prevalence of DV (NFHS-3, 2005–06). As reported by the National Family Health Survey-4 (International Institute of Population Sciences [IIPS] and ICF, 2017, www.rchiips.org/nfhs) 27% of women have experienced physical violence since age 15 and 6% have ever experienced sexual violence in their lifetime. Further the same report indicates that 31% of ever married women have experienced physical, sexual, or emotional spousal violence. The most common form of spousal violence is physical violence (27%) followed by emotional violence (13%) and sexual violence (6%). It is also reported that there is a decline in DV ever experienced by ever-married women from NFHS-3 (2005–06) 37% to NFHS-4 (2015–16) 29%. Among all the states of India, Manipur is reported to have the highest percentage of DV (55%) and Sikkim is reported to have the least DV (3.5%) as ever experienced by ever-married women in the age group 15–49 years. Both these two states are in the northeastern region of India.

Review of Literature

A substantial amount literature has been found which attempt to investigate the quantum of DV of varied dimensions all over the world. They have also tried to measure the socioeconomic and cultural backgrounds pertaining to cause or influence DV. Age of husband and wife, social status, and religion are considered important explanatory covariates for DV in India (INCLEN, 2000). Murthy et al. (2004) see that family sizes, type of marriage, and husband’s occupation are significant covariates to influence DV. Women who have been separated or divorced from a prior union tend to face more violence due to the stigma attached to the concept of divorced (Chowdhury et al., 2018).

Many researchers also found that the lifestyle characteristics of husband, such as smoking, alcoholism, drugs, etc., can significantly influence DV (Leonard & Blane, 1992; McKenry et al., 1995; Rao, 1997). Gendered socialization process is also indicated responsible for DV (Sahu, 2003). Another study among Uttar Pradesh men by Gerstein (2000) is of the view that low educational level and poverty are important reasons for DV. Further, marriage at a younger age makes women vulnerable to DV (Hindin, 2002; Mishra, 2000; Rao, 1997).

Studies shows that among Indian communities identified some individual-, family-, and community-level factors that determine the risk of DV. Demographic factors, such as age, number of children, family type, and marriage type, also influence the risk of violence (Rao, V). Stress-related factors within the family, including husband having a low-educational level, family living in poverty, and having multiple children, have a positive association with DV.

The Present Study

In view of the reported prevalence of DV in different dimensions, researchers have attempted to assess the situation besides exploring the cause and consequences, which pertains to the victim women, in particular, and the society, in general. The present article attempts to assess and investigate the situation in the northeastern region of India. The northeastern region of India comprises of eight states viz. Assam, Arunachal, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim, and Tripura. These states have socioeconomic and demographic characteristics, which are more or less different from the mainland population of India. Specifically, the economic activity in the region is quite different from the mainland as having little to no industrialization and mainly depends on agricultural activities. All states are dominated by tribal population except Assam where tribal population accounts for 12% only. The main religious groups in the region are Hindu, Muslim, Christians, Buddhists, and some unrecognized local faiths still exist though fewer in number. The population of the region is sparse as compared to other parts of the country and shares only 3.57% of the whole population of India while the geographical area covers 7.57% of the total area of the country. All states except Meghalaya follow patrilineal norms while in Meghalaya there are ethnic groups who follow matrilineal norms. While studies on DV are carried out for other parts of mainland India, no such studies are extensively done exclusively for this region. However, as cited earlier, NFHS-4 (2015–16) reported the highest and the least percentage of DV ever-experienced by ever-married women in two states of the region. The authors hope that the findings of the present study, which mainly focus on the region, will be a contribution to the overall discussion on the issue of DV around the world.

We hypothesize that DV is a widespread phenomenon and variation in its prevalence occur across the northeast Indian states, as these states differ from each other in overall development as well as demographic characteristics. Also, it is hypothesized that differences occur within the population of these states based on their socioeconomic characteristics such as habitation (rural or urban residence), age, religion/caste affiliation, education, occupation, and income/wealth. The purpose of the present study is to report the prevalence of various forms of DV against women and to explore the socioeconomic factors which would significantly influence DV in the region states. This article deals with the violence faced by women, perpetrated by their husbands and other family members within their conjugal homes.

Data and Methodology

The National Family Health Surveys (NFHS-4) (2015–2016) is the fourth in the NFHS series that provides information on population, health, and nutrition and violence against women in India. The first round of the NFHS was conducted in 1997–98 all over the country and the fourth round was conducted in 2015–16. Because of the growing research on DV, which concluded a negative effect of DV against women’s health, it is important to include information on DV in the NFHS. In NFHS-4 information was obtained from ever-married women on their experiences of violence committed by their current or former husbands or by others. All the four NFHS surveys have been conducted under the stewardship of the Ministry of Health and Family Welfare (MoHFW), Government of India. MoHFW designated the International Institute for Population Sciences (IIPS), Mumbai, as the nodal agency for all the surveys. Funding for NFHS-4 was provided by the United States Agency for International Development (USAID), the United Kingdom Department for International Development (DFID), the Bill and Melinda Gates Foundation (BMGF), UNICEF (United Nations International Children’s Emergency Fund), UNFPA (United Nations Population Fund), the MacArthur Foundation, and the Government of India. ICF provided technical assistance through the DHS (Demographic and Health Surveys) Program, which is funded by USAID.

Data for this study were obtained from the NFHS-4 (2015–2016) coordinated by the IIPS. This is a nationally representative probability sample of all women age 15–49 (N = 699,686), from which 79,729 women were administered the gender-based violence module. In northeast India, 98,702 women in the age group 15–49 are interviewed out of which only 11,056 women are selected and interviewed for DV module. Out of 11,056 women, only 8,766 women are ever married on which the present analysis is carried out. The present study excluded never married women in the age group 15–49. The northeastern region of India comprised of eight states, namely Arunachal Pradesh, Assam, Manipur, Meghalaya, Mizoram, Nagaland, Sikkim, and Tripura. In Arunachal Pradesh, 1,299 women’s in the age group 15–49 are selected and interviewed for DV module, 2,630 women for Assam, 1,137 women for Manipur, 721 women for Meghalaya, 976 women for Mizoram, 860 women for Nagaland, 524 women for Sikkim, and 619 women for Tripura.

Dependent Variable

The present study considered DV as dependent variable (D.V.). In NFHS-4, information was obtained from never-married women on their experience of violence committed by anyone and from ever-married women on their experience of violence committed by their current and former husbands and by others, more specifically, violence committed by the current husband for currently married women and by the most recent husband for formerly married women. For the present study, we have considered only ever-married women. It also provides information on four types of violence against women: physical, sexual, emotional, and severe. The violence was measured by asking all ever-married women if their husbands ever committed the following to them:

Physical violence. Pushing, shaking, throwing something to the woman, slapping, punching or hitting by a harmful object, kicking or dragging, strangling or burning, threatening with a knife/gun/any other weapon.

Emotional violence. Ever been humiliated by husband/partner, ever been threatened with harm by husband/partner, ever been insulted or made to feel bad by husband/partner.

Sexual violence. Ever been physically forced into unwanted sex by husband/partner, ever been forced into other unwanted sexual acts by husband/partner, ever had arm twisted or hair pulled by husband/partner, ever been physically forced to perform sexual acts respondent did not want to.

Severe violence. Bruises, eye injuries, dislocations, severe burns, wounds, broken bones, broken teeth, and others.

Independent Variables

The present analysis focuses on three sets of independent variables: (a) household-level factors (wealth index, marital status, husband drinks, parity of women, and husband’s education); (b) individual-level factors (age of women, women’s education, working status of women); and (c) community-level factors (types of residence and religion).

Statistical Analysis

To examine the association of socioeconomic variables and DV, a dichotomous variable for DV has been generated. DV is considered “yes” (coded-1) if ever experienced any form of violence and “no” (coded-0) if never experienced any form of violence. The association between DV and all socioeconomic variables which are thought to be associated with DV is examined using the bivariate analysis which produces chi-square p values for the significance of the measure of association. All covariates that are significant at 5% (p < .05) are considered potential covariates in the logistic regression analysis. Binary logistic regression models were fitted to estimate the relationship between sociodemographic variables and violence against married women. Furthermore, the impact of various background characteristics on women’s experience of DV has been assessed through a multivariate logistic regression analysis (Hosmer & Lemeshow, 2000). Two logistic regression models are fitted wherein the first model has three forms of violence combined viz. physical, emotional, and sexual violence as one D.V. and in the second model severe violence only is taken as the D.V. The probabilities are interpreted in terms of odds ratios (OR) which gives the likelihood of experiencing the event in the different categories of a covariate as compared to a reference category. SPSS version 21 was used to conduct all the statistical analyses.

Variables and Categories in the Model

A conceptual framework for predicting the socioeconomic determinants of DV in northeast India is proposed in Figure 1. This framework suggests that DV is a function of individual-, household-, and community-level variables.

Figure 1.

Framework for the determinants of domestic violence.

Individual-level variables. The individual-level variables used in the analysis are age of women (subsequently categorized into age groups), education of women (illiterate, primary, secondary, higher), and working status.

Household-level variables. The household-level variables included in the analysis are wealth of the family (poorest, poorer, middle, richer, richest), marital status (married, widowed, divorced, separated), husband’s alcohol use (yes/no), and parity of women (no. of children born).

Community-level variables. The community-level variables included in the analysis are type of place of residence (rural or urban), religion (Hindu, Muslim, Christian, no religion, and others).

Results

We have computed the percentage of women who have ever experienced a form physical, emotional, sexual, and severe physical violence during their past life committed by husband or partner and presented in Table 1. In the whole country and in the northeastern states, Manipur has topped the list in experiencing physical violence with 51.8% sexual violence 13.8%. For emotional violence, Arunachal reported the highest with 16.3%, and in severe violence, Tripura reported highest with 12.3%. Sikkim reported the least percentage in all forms of DV.

Table 1.

Percentage of Women Who Ever-Experienced Domestic Violence (DV).

State	No. of Women (%)	No. of Women Who Experienced (At Least Once) DV
State	No. of Women (%)	PV (%)	EV (%)	SV (%)	Severe V (%)
Arunachal	1,299 (14.8)	350 (26.9)	212 (16.3)	91 (7.0)	66 (5.1)
Assam	2,630 (30)	633 (24.1)	296 (11.4)	140 (5.3)	121 (4.6)
Manipur	1,137 (13)	589 (51.8)	160 (14.1)	157 (13.8)	86 (7.6)
Meghalaya	721 (8.4)	170 (23.6)	84 (11.8)	31 (4.3)	47 (6.5)
Mizoram	976 (11.1)	154 (15.8)	97 (9.9)	28 (2.9)	44 (4.5)
Nagaland	860 (9.8)	87 (10.1)	88 (10.2)	51 (5.9)	19 (2.2)
Sikkim	524 (6)	10 (1.9)	14 (2.7)	5 (1.0)	2 (0.4)
Tripura	619 (7.1)	169 (27.3)	82 (13.2)	58 (9.4)	76 (12.3)
Overall states	N = 8,766	2,162 (24.7)	1,033 (11.8)	561 (6.4)	461 (5.3)

Note. P = physical; E = emotional; S = sexual; V = violence.

Table 2 presents the prevalence of various forms of DV reported by ever-married women belonging to different socioeconomic characteristics. Percentage of women who ever experienced DV is presented for each category of socioeconomic variables.

Table 2 .

Prevalence of Various Forms of Domestic Violence Ever-Experienced by Ever-Married Women in Different Socioeconomic Characteristics.

Category of Socioeconomic Characteristics

Percentage of Domestic Violence

Physical

Emotional

Sexual

Severe

Age group

15–29

30–49

23.7

25.2

12.1

11.6

6.5

6.3

5.1

6.5

Women’s working status

Yes

22.1

31.2

16.3

5.6

8.5

5.1

8.1

Women’s education

Illiterate

Primary

Secondary

Higher

31.6

21.6

19.6

15.8

13.7

10.1

6.3

8.1

7.5

5.6

3.6

9.1

7.2

4.8

1.4

Wealth index

Poorest

Poorer

Middle

Rich

Richest

35.9

28.8

21.2

18.5

13.2

16.5

10.4

8.8

5.8

7.7

5.7

4.4

3.4

9.1

7.3

5.4

3.7

2.3

Parity of women

Up to 2

More than 2

28.2

10.4

13.7

5.9

4.7

7.7

Marital status

Married

Widowed

Divorced

Separated

42.4

54.4

8.5

32.8

49.6

6.1

5.5

21.5

5.4

7.9

21.6

25.6

Husband’s education

No education

Primary

Secondary

Higher

Do not know

31.5

27.1

22.4

20.5

23.1

15.5

14.6

10.4

7.5

15.4

7.6

7.8

4.2

2.6

8.7

7.0

5.8

2.8

5.1

Husband drinks

Yes

15.6

36.4

7.1

17.9

3.7

2.7

10.3

Residence

Urban

Rural

21.3

25.8

10.6

12.2

5.5

6.7

5.2

6.3

Religion

Hindu

Muslim

Christian

No religion & other

26.3

28.6

21.7

12.9

7.8

6.6

6.7

5.8

4.1

6.1

5.6

7.4

The percent distribution of sampled women according to background characteristics is presented in Table 3. The association between each socioeconomic variables with the variable representing at least one form of DV is tested using chi-square test of association and the p values are shown in the last column of Table 3. Significant p values (<.05) indicates that all socioeconomic variables except “age” (which is not significant at 5% level) are highly associated with DV experienced by ever-married women in the region.

Table 3.

Percent Distribution of Sampled Women According to Background Characteristics.

Socioeconomic Characteristic

No. of Women (%)

Chi-square

p Value

Individual-level variables

Age group

15–29

30–49

3,063 (34.9)

5,703 (65.1)

.43

Women’s education

Illiterate

Primary

Secondary

Higher

1,955 (22.3)

1,587 (18.1)

4,648 (53)

576 (6.6)

<.001

Women’s working status

Yes

6,287 (71.7)

2,476 (28.3)

<.001

Household-level variables

Wealth index

Poorest

Poorer

Middle

Richer

Richest

1,372 (15.7)

2,697 (30.8)

2,296 (26.2)

1,696 (19.3)

705 (8)

<.001

Marital status

Currently married

Widowed

Divorced

Separated

8,177 (93.3)

343 (3.9)

125 (1.4)

121 (1.4)

<.001

Husband’s education

Illiterate

Primary

Secondary

Higher

Do not know

1,532 (17.4)

1,621 (18.5)

4,658 (53.1)

925 (10.6)

39 (0.4)

<.001

Husband drinks

Yes

4,953 (56.5)

3,813 (43.5)

<.001

Parity of women

Up to 2

More than 2

5,038 (57.2)

3,728 (42.5)

<.001

Community-level variables

Type of place of residence

Urban

Rural

2,249 (25.7)

6,517 (74.3)

<.001

Religion

Hindu

Muslim

Christian

No religion & others

3,479 (39.7)

930 (10.6)

3,452 (39.4)

905 (10.3)

<.001

The Logistic Regression Models

First, bivariate logistic regression models are fitted by considering the socioeconomic variables, which are significant at the chi-square test separately, and the results are presented in Table 4. The OR are estimated along with the p values given within the braces. The aim of these beginning models is to see whether there is any variable which may not be important to consider as a potential covariate for DV in the multivariable models. All p values are significant at 5% level in the column for physical violence except the one variable (religion) which in turn indicates that all other variables are potential to influence DV to be included in the multivariate logistic regression model. However, religion will also be included in the multivariable model as it is important social characteristic. It is not appropriate to interpret at this stage, the OR until a final adjusted model is fitted by adjusting for different socioeconomic variables. Two models are attempted in the final multivariable adjusted model. The first model (Model 1) is fitted by considering DV as a combined effect of all three forms of violence viz. physical, emotional, and sexual violence. The explanatory covariates are those socioeconomic variables which are significant in the bivariate analysis (Table 4). The second model (Model 2) is fitted by considering experience of severe violence as D.V. and with all other socioeconomic variables as in the first model. The result of fitting these two models is presented in Table 5 which consists of estimated OR and corresponding p values.

Table 4.

Results of Bivariate Logistic Regression Analysis Between Socioeconomic Characteristics and Prevalence of Various Forms of Domestic Violence.

Socioeconomic Characteristics	PV(Odds Ratio/p Values)	EV(Odds Ratio/p Values)	SV(Odds Ratio/p Values)	Severe V(Odds Ratio/p Values)
Women’s working status
No (ref)Yes	1.63 (.00)	1.74 (.00)	1.57 (.00)	2.0 (.00)
Type of place of residence
Urban (ref)Rural	1.28 (.00)	1.16 (.049)	1.23 (.047)	1.38 (.00)
Women’s education
Illiterate (ref)PrimarySecondaryHigher	0.79 (.00)0.59 (.00)0.52 (.00)	0.84 (.08)0.59 (.00)0.35 (.00)	1.91 (.97)0.86 (.16)1.12 (.41)	0.78 (.04)0.50 (.00)0.14 (.00)
ReligionHindu (ref)MuslimChristianNo religion & others	1.12 (.15) 0.77 (.00)0.96 (.68)	1.09 (.43)0.90 (.18)1.11 (.32)	1.00 (.97)0.86 (.16)1.12 (.41)	0.91 (.58)0.92 (.46)1.23 (.14)
Wealth indexPoorest (ref)PoorerMiddleRichRichest	0.72 (.00)0.48 (.00)0.40 (.00)0.27 (.00)	0.82 (.037)0.58 (.00)0.48(.00)0.31 (.00)	0.84 (.16)0.61 (.00)0.47 (.00)0.35 (.00)	0.76 (.036)0.50 (.00)0.40 (.00)0.24 (.00)
Marital statusMarried (ref)WidowedDivorcedSeparated	0.9 (.67)2.3 (.00)3.7 (.00)	0.7 (.16)3.9 (.00)7.9 (.00)	0.8 (.65)2.0 (.00)4.1 (.00)	1.5 (.04)4.8 (.00)6.0 (.00)
Parity of womenUp to 2 (ref)More than 2	1.39 (.00)	1.39 (.00)	1.19 (.039)	1.57 (.00)
Husband’s educationNo education (ref)
PrimarySecondaryHigherDo not know	0.80 (.00)0.63 (.00)0.56 (.00)0.64 (.26)	0.92 (.46)0.63 (.00)0.44 (.00)0.99 (.98)	1.03 (.81)0.77 (.023)0.53 (.001)0.31 (.26)	0.79 (.08)0.60 (.00)0.30 (.00)0.57 (.44)
Husband drinksNo (ref)Yes	3.0 (.00)	2.8 (.00)	3.0 (.00)	4.0 (.00)

Note. Ref = reference category; P = physical; E = emotional; S = sexual; V = violence.

Table 5.

Multivariate Logistic Regression Analysis Between Socioeconomic Characteristics and Prevalence of Various Forms of Domestic Violence.

Socioeconomic Variables	Model 1: DV ⁾ …			Model 2: Severe V⁾ …
Socioeconomic Variables	p Values	Odds Ratio	p Values	Odds Ratio
Women working status
No (ref)Yes	0.00	1.63	0.00	1.33
Women educationNo education (ref)
PrimarySecondaryHigher	0.150.010.19	0.890.830.83	0.120.010.00	0.800.710.25
Wealth indexPoorest (ref)PoorerMiddleRicherRichest	0.000.000.000.00	0.810.600.480.29	0.350.020.000.00	0.880.710.490.34
Husband educationNo education (ref)
PrimarySecondaryHigherDo not know	0.900.770.020.22	0.9881.0231.3230.603	0.910.680.890.35	1.021.060.960.49
ReligionHindu (ref)MuslimChristianNo religion & others	0.000.000.16	1.570.750.88	0.050.060.541	1.430.811.09
Marital statusMarried (ref)WidowedDivorcedSeparated	0.000.000.00	0.682.473.31	0.310.000.00	1.245.224.82
Parity of women
0–2 (ref)3–above	0.00	1.29	0.00	1.51
Place of Residence
Urban (ref)Rural	0.10	0.89	0.05	0.78
Husband drinks alcohol
No (ref)Yes	0.00	3.23	0.00	4.03

Note. Ref = reference category; P = physical; E = emotional; S = sexual; V = violence.

Model 1: DV has three forms of violence combined, viz. physical, emotional, and sexual violence.

Discussion

The present study is based on a sample 8,766 ever-married women from the northeastern region of India who provide complete information on DV ever-experienced by them. The prevalence of spousal violence is low in Sikkim. In Sikkim, most of the husbands are educated (91%) and percentages of alcohol by men are as low as 39% as compared to other states. About 82% of married women are literate in Sikkim, which is a good reason for low prevalence of DV. Women are mostly involved in making household decisions, such as purchase on major household items, visits to their own family and relatives, etc. This indicates a higher status of women in Sikkim, which is a potential reason for lower rate of prevalence in DV.

The prevalence of spousal violence is highest in Manipur. Though the literacy rate is high (94%) among men in Manipur, the percentage of alcohol use by men is very high (60%) as compared to other states, and this may be the reason for high prevalence of DV experienced by women in Manipur.

Further the present study selects 10 potential socioeconomic variables, which are thought to be associated with DV as ever-experienced by ever-married women in the eight northeastern region of India. The rationales for selecting these socioeconomic variables are mainly based on past experiences, which are supported by literatures of similar studies on DV around the world. While examining the statistical association of these ten variables the age of women is not found to be significantly associated with experience of DV reported by women in the region. The remaining nine variables are highly significant (see p values in Table 3) to DV so that they are potential to further investigate with other statistical analysis such as regression models. Because of ease of interpretability of results logistic regression model is a natural choice for a dichotomous D.V. to be regressed by these socioeconomic variables we have fitted two such models (Table 5). The interpretations of the OR in Table 5 are as follows. The OR for a particular category is interpreted as the likelihood of experiencing DV reported by women as compared to the reference category indicated by (ref). If OR is greater than unity then the likelihood is more in that category of women, whereas if it is less than unity then the likelihood is less as compared to the reference category.

At the individual-level variable working status of women significantly influence DV in northeastern region of India. Those women who work outside of homes are 63% more likely to ever experience DV and 33% more likely to ever experience severe violence as compared to women who do not work outside home. Two categories out of three in women’s education are not significant. However, we can conclude from the OR that higher educated women are less likely to ever-experience DV and severe violence as well.

At the household-level variable, wealth of family is a significant characteristic to influence DV. Women in higher wealth quintiles are less likely to ever experience DV and severe violence. For example, as compared to women in the poorest wealth quintiles, women in the poorer and middle wealth quintiles are 19% and 40% less likely to ever experience DV. Further women in the richer and richest wealth quintiles are very less likely (52% and 71%) to face DV. Similarly, experience of severe violence is also less in higher wealth quintiles as compared to the poorest wealth quintile.

Husband’s education as a household-level variable cannot significantly influence DV as most categories are found to be insignificant. Marital status of women is also an important household-level variable to influence DV. Women who are not currently married have more chance of reporting DV than widowed women. However, divorced and separated are more likely to ever experience DV as well as severe violence as compared to currently married women. Parity of women which is defined by the number of births born to a woman is highly significant to influence DV and severe violence. Women at higher parity is 29% and 51% more likely to ever-experience DV and severe violence, respectively, as compared to women in lower parity. Another very important household-level covariate is husband’s characteristic of alcohol use. Those women whose husband’s drink alcohol are 3.23 and 4.03 times more likely to report DV and severe violence, respectively, as compared to women whose husband’s do not drink alcohol.

At the community-level variables place of residence is not a significant covariate to influence DV. However, religion which was not significant in the bivariate analysis is now a significant covariate in the multivariate model. As such Muslim women are 57% more likely to ever-experience DV but Christian women are 25% less likely to report experience of DV as compared to Hindu women in the region. Similarly, Muslim women are 43% more likely to ever-experience DV, but Christian women are 19% less likely to report experience of DV as compared to Hindu women.

Conclusion

DV is a universal phenomenon. Current or former husbands are mostly responsible for violence among majority of women though other members of the conjugal home are also responsible. We have attempted in the present study to explore socioeconomic characteristics of women that have significant association with the occurrence of DV. This article is primarily aimed at finding the prevalence of DV in northeastern states of India and thereby investigating the socioeconomic covariates which significantly influence the violence. While age of women is considered an important factor for occurrence of DV in other states of India (INCLEN 2000) analysis shows that age of women is not a significant covariate to influence DV in the region. Education of women is a protector covariate for DV which is in line with other studies cited in the literature. However, education of husband cannot be considered as a significant covariate for the same. Working status of women is a significant explanatory covariate and those women who work outside of home are more vulnerable to DV. The influence of husband’s habit of drinking alcohol highly significant as found in other studies also. Interestingly, there is no significant difference between rural and urban women to experience DV in the region.

DV has been recognized in India since 1983 as a criminal offence under the Indian Penal Code 498-A. However, it was not until the enactment of the Protection of Women from Domestic Violence Act 2005 (PWDVA), which came into effect in 2006 that civil protections were afforded to victims of DV. PWDVA provides a comprehensive definition of DV, which included all forms of physical, emotional, verbal, sexual, and economic violence, and the threats to commit such violence. The DV Act is applicable to all over India except the state of Jammu and Kashmir. It is a civil law, which focuses on the reliefs given to the aggrieved woman such as compensation, protection, right to residence in the shared household, etc. It covers all kinds of violence faced by a woman at her shared household. Despite the enactment of PWDVA, violence against women and girls continues to be a major challenge and a threat to women’s empowerment in India. The findings in this study could well be recognized by policy and program planners in order to tackle the issue of DV faced by married women in the region.

Footnotes

Acknowledgments

We would like to thank the anonymous reviewers whose comments were extremely useful in improving the article.

Declaration of Conflicting Interests

Funding

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

ORCID iD

Sharatchandra Haobijam

Author Biographies

Sharatchandra Haobijam is a research scholar (statistics), Department of Statistics, Manipur University, Manipur, India. His research interests include statistical modelling and analysis of demography data specially mortality, fertility, and domestic violence with its reason and consequences.

Kshetrimayum Anand Singh, PhD (statistics), is an assistant professor of statistics in Department of Statistics, Manipur University, Manipur, India. His research interests include statistical modelling, survival analysis, and analysis of demographic data.

References

Black

M. C.

, Basile

K. C.

, Breiding

M. J.

, Smith

S. G.

, Walters

M. L.

, Merrick

M. T.

, Chen

, & Stevens

M. R.

(2011). The national intimate partner and sexual violence survey (NISVS): 2010 summary report . National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Chowdhury

S. R.

, Bohara

A. K.

, & Horn

B. P.

(2018). Balance of power, domestic violence, and health injuries: Evidence from demographic and health survey of Nepal. World Development , 102, 18–29. https://doi.org/10.1016/j.worlddev.2017.09.009

Gerstein

(2000). In India, poverty and lack of education are associated with men's physical and sexual abuse of their wives. International Family Planning Perspectives , 26(1), 44–45.

Hindin

J. M.

(2002). Who is at risk? Factors associated with intimate partner violence in the Philippines. Social Science and Medicine , 55(8), 1385–1399.

Heise

, Ellsberg

, & Gottemoeller

(1999). Ending violence against women (Population Reports, Series L, No. 11). GTZ.

Hosmer

D. W.

, & Lemeshow

(2000). Applied logistic regression (2nd ed.). John Wiley & Sons, Inc.

(IIPS)

International Institute of Population Sciences

ICF. (2017). National Family Health Survey (NFHS-4), 2015–16 . www.rchiips.org/nfhs

International Clinical Epidemiologists Network (INCLEN). (2000). Domestic violence in India 3: A summary report of a multi-site household survey . International Centre for Development and Population Activities.

Leonard

K. E.

, & Blane

H. T.

(1992). Alcohol and marital aggression in a national sample of young men. Journal of Interpersonal Violence , 7(1), 19–30.

10.

McKenry

P. C.

, Julian

T. W.

, & Gavazzi

S. M.

(1995). Toward a biopsychosocial model of domestic violence. Journal of Marriage and the Family , 57(2), 307–320.

11.

Mishra

(2000). Women and human rights (Chapter 5). Kalpaz Publications.

12.

Murthy

M. S. R.

, Ganesh

, Srivirajarani

, & Madhusudan

(2004). Proximate determinants of domestic violence: An exploratory study on role of menstrual problems and life style of men. Demography India , 33(1), 85–105.

13.

National Crime Records Bureau (NCRB). (2017). Crime in India 2017 - Volume 1_0.pdf

14.

Owusu Adjah

E. S.

, & Agbemafle

(2016). Determinants of domestic violence against women in Ghana. BMC Public Health , 16, 368. https://doi.org/10.1186/s12889-016-3041-x

15.

Rao

(1997). Wife-beating in rural South India: A qualitative and econometric analysis. Social Science and Medicine , 44(8), 1169–1180.

16.

Rahman

M. M.

, & Kabir

(2006). Gender based violence against female adolescents and factors affecting it in Bangladesh. Genus , 62(1), 9–20.

17.

Sahu

(2003). Contextualizing domestic violence from women’s perspective: A study in a slum community in Orissa [Unpublished M.Phil. dissertation]. IIPS.

18.

World Bank. (2011). World bank annual report, 2011 . https://openknowledge.worldbank.org/handle/10986/2378

19.

World Economic Forum. (2013). WEF_GenderGap_Report_2013.pdf

20.

Zakar

, Zakar

M. Z.

, & Abbas

(2015). Domestic violence against rural women in Pakistan: An issue of health and human rights. Journal of Family Violence , 31(1), 15–25. https://doi.org/10.1007/s10896-015-9742-6