Exploring the Low Rates of Reporting Domestic Violence in Bihar, India

Exploring the Low Rates of Reporting Domestic Violence in Bihar, India

Portia Bajwa, Kelsey Foreman, and Charlotte Sall

Abstract

This needs assessment study explores the underreporting of domestic violence against women survivors. It focuses on the Indian state of Bihar, which has one of the highest rates of domestic violence in the nation and yet a low rate of reporting. Social norm theory is used as a lens to explore this discrepancy. The study draws on secondary data from the India 2015-2016 National Family Health Survey (NFHS-4) to examine: (1) the extent to which there is a discrepancy between the prevalence of domestic violence and the rate of reporting DV for married women in the state of Bihar, and (2) how barriers to reporting domestic violence relate to social norms in Bihar. Results from a multivariate logistic regression (N = 1053) indicate that social norms are not predictive of the rate of reporting DV in Bihar. Possible explanations for these findings are discussed.


Over the past 40 years, domestic violence (DV) has emerged as a global concern and is now recognized as a human rights issue (Kishor & Johnson, 2004; Heise, 2011). The United Nations (1993) defines violence against women as “any act of gender-based violence” related to “physical, sexual, or mental harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or in private life.” Despite the fact that DV has gained worldwide recognition as a social problem, its prevalence is underestimated, and cases of DV are notoriously underreported (Felson & Pare, 2005; Straus, Gelles, & Steinmetz, 1980). Understanding why cases go unreported is key to developing appropriate and effective interventions aimed at reducing DV (Ahmed-Ghosh, 2004; Koenig, Stephenson, Ahmed, Jejeebhoy, & Campbell, 2006). In India, domestic violence is one of the most common crimes against women. An estimated 21 percent of women over the age of 15 have experienced abuse from their husbands (Chaudhary, 2013). The state of Bihar, one of the least developed in India, with comparatively low levels of female literacy and autonomy (Jejeebhoy & Santhya, 2018), has the country’s highest rate of DV: 59 percent of ever-married women areestimated to have experienced domestic abuse (Chaudhary, 2013; Chachra, 2017; “Bihar,” 2008). Much of this information comes from a national health survey as well as statistics on dowry-related deaths, as many women do not utilize official reporting systems (such as contacting the local police) and do not feel comfortable disclosing their DV experiences to members of their community (Krishnan, 2017). 

While some research suggests that the lack of reporting to official systems relates to a lack of awareness of women’s rights and protections (Krishnan, 2017; Jhamb, 2011), other findings suggest that India’s social norms—including patriarchy, religious and cultural beliefs about marriage, and asymmetrical gender expectations (Koenig et al., 2006; Chaudhary, 2013; Chachra, 2017)—help sustain DV (Koenig et al., 2006; Ahmed-Ghosh, 2004). These norms can perpetuate notions that DV is justified (Chaudhary, 2013; Sahoo & Raju, 2007). Furthermore, those who wish to report abuse fear the consequences, which may include divorce, family disillusionment, lack of financial resources, or spousal retaliation (Chachra, 2017; Kalokhe et al., 2017). In addition, there is a lack of trust in the institutional criminal justice system when it comes to reporting DV (Kishor & Johnson, 2004; Abrams, Belknap, & Melton, 2001). 

The current study is a preliminary needs assessment that seeks to examine: (1) the extent to which there is a discrepancy between the prevalence of domestic violence and the rate of reporting DV for married women in the state of Bihar; and (2) how barriers to reporting DV relate to social norms in Bihar.


REPORTING IN BIHAR

The system of reporting DV in Bihar is shaped by the 2005 Protection of Women Against Domestic Violence Act (PWDVA) (Dubochet, 2012; Jhamb, 2011). Under PWDVA, Bihar is expected to protect women from domestic violence by providing immediate shelter services and orders of protection (“An Analysis,” 2016). Thirty-five out of the thirtyeight districts in Bihar have a helpline where individuals are able to report domestic violence (Krishnan 2017). Despite these resources for reporting, 86 percent of women claimed that they were not aware of these institutions, and of the women who were aware, 80 percent said that they would not know how to go about the process of reporting DV to the helplines (Krishnan, 2017). These findings indicate a gap between the systems in place for reporting DV and the likelihood of victims to report. Although the PWDVA was created as a system to increase rates of reporting DV, the outcomes have not improved much due to inconsistent and biased implementation (Govinderajan, 2016). Even when women make use of the system, court visits can be delayed for years (Govinderajan, 2016). Due to these limitations of official systems of reporting in Bihar, it is also important to consider reporting to unofficial systems, such as social supports. These modes of reporting can have a positive impact for the women in terms of increasing their coping strategies and enhancing their overall safety. Additionally, increasing the amount of people who know about the violence increases society’s overall awareness of DV prevalence and subsequent perceptions that DV is problematic. Allowing for more open discourse on the topic of DV may positively affect a woman’s willingness to report (Paluck & Ball, 2010).


SOCIAL NORM THEORY AND DOMESTIC VIOLENCE REPORTING

Social norm theory provides a framework for understanding the problem of domestic violence reporting (Berkowitz & Perkins, 1986). The theory posits that individual behavior is motivated by perceptions of how other members of social groups think and behave, regardless of the accuracy of those perceptions (Berkowitz, 2005; Paluck & Ball, 2010). Termed “pluralistic ignorance” (Miller & McFarland, 1991; Toch & Klofas, 1984), these misperceptions have strong power over individual behavior (Berkowitz, 2005). Norm theory operates under the premise that actions are grounded in attitudes, so behaviors only change once underlying beliefs are restructured (Paluck & Ball, 2010). Although norm theory was created in a Western context, interventions based on the framework have been implemented and evaluated across the globe (Heise, 2011). Thus, norm theory can be utilized as a lens for attempting to understand India’s cultural values around domestic violence and reporting.

Evidence from studies in low- and middle-income countries documents that both the wife’s level of acceptance toward beating and the husband’s level of control over female behavior are predictive of a country’s DV rate (Uthman, Lawoko, & Moradi, 2009; Rani, Bonu, & Diop-Sidibe, 2004; Guoping et al., 2010). In terms of reporting, perceptions of being alone in their experience can exacerbate the feeling of isolation and prevent women from sharing, even to a friend (Felson & Paré, 2005). Given the high prevalence of domestic violence in Bihar, norm theory suggests that many women may not realize the true rates of DV. Thus there may be a confluence of (1) societal norms regarding attitudes toward domestic violence and (2) pluralistic ignorance whereby women feel alone in their experiences of violence, which could account for low rates of reporting.


NORM THEORY AND EVIDENCE-BASED PRACTICES

Norm theory calls for interventions that are preventative, targeting root causes of social problems (Berkowitz, 2005; Heise, 2011). As a result, interventions often do not show immediate outcomes, so few evidence-based practices document their success (Heise, 2011). One strategy aimed at changing social norms is awareness-raising campaigns. For example, Oxfam’s “We Can” campaign was created to address social norms that contributed to violence against women in Bangladesh, India, Sri Lanka and Pakistan (Raljan & Chakraborty, 2010). A mixed methods impact evaluation of “We Can” suggests the campaign has succeeded at increasing societal acceptance of women who speak out about their experiences with domestic violence (Raljan & Chakraborty, 2010; William & Aldred, 2011). Another norm-targeting strategy is edutainment programs, which are interventions combining media and dialogue to effect social change. One such example is the 2008 “Bell Bajao” campaign, begun by the Indian organization, Breakthrough, with the goal of using multimedia and grassroots organizing to change norms surrounding DV (CMS Communication, 2011). In a pre- and post-test evaluation (pre N=1204; post N=1590), there was a significant decline in the belief that an abused wife should remain silent (15.8% baseline; 5.7% endline) (Heise, 2011). While interventions targeting social norms are increasingly popular, their effectiveness has not yet been thoroughly evaluated.


JUSTIFICATION FOR FURTHER RESEARCH

Social norms affecting domestic violence and reporting are deeply entrenched, yet norm theory offers a framework for designing future interventions to address these norms. While evaluative research on normbased DV interventions is still fairly underdeveloped, preliminary evidence shows that norms can be changed with well-designed interventions (Raljan & Chakraborty, 2010; William & Aldred, 2011; Bradley et al., 2011; CMS Communication, 2011).


METHODS

Sample
The sample was obtained from the Demographic and Health Surveys (DHS) Program from the results of the India 2015-2016 National Family Health Survey (NFHS-4), which is a nationally-representative household survey that provides data for a wide range of indicators in the areas of population, health, and nutrition (IIPS & ICF, 2017). A two-stage sample design was used. First, cluster sampling was employed to select either villages (rural) or Census Enumeration Blocks (urban). Second, randomselection was used to select individual households within those (IIPS & ICF, 2017). Of the 572,000 respondents included in the total sample, approximately 46,000 women and 6,000 men lived in Bihar (IIPS & ICF, 2017). This study’s sample included only married women who received the DV module and who then declared to have experienced any form of violence (emotional, less severe physical, severe physical, and/or sexual violence) (N = 1053). The sample was constricted based on the dataset, which did not include DV experiences outside of marriage (most likely due to traditional cultural norms surrounding intimate relationships in India).

Study Design and Data Collection
The present study utilizes secondary data obtained from the original DHS survey, which was administered through questionnaires from March 16 to August 8, 2015 via individual interviews. Information was collected in 19 languages (including Hindi, Bengali, and Punjabi) using Computer Assisted Personal Interviewing (CAPI) (IIPS & ICF, 2017). To ensure privacy and anonymity, an informed consent statement was read aloud before each interview, and any information collected was deidentified upon completion of data processing (IIPS & ICF, 2017).

Measures
All variables were measured through questions from the NFHS-4 questionnaire. The official questionnaire consisted of approximately 400 possible questions for women (430 if the DV module was included) (DHS, 2017b), and 55 were included in the present study.

Experience of domestic violence. A woman’s experience of DV was measured from responses to 25 dichotomous questions based off of the Revised Conflict Tactics Scale (Straus & Douglas, 2004; Begum, Donta, Nair, & Prakasam, 2015; Rowan, Mumford, & Clark, 2018). Examples of these questions include, “Did your husband/partner ever insult you or make you feel bad about yourself?” and “Did your husband/partner ever push you, shake you, or throw something at you?” (DHS, 2017a). If the woman was previously married, these questions were also asked of her previous husband(s). If the answer was yes to any of these questions, the respondent was then asked about the frequency (“Often,” “Sometimes,” or “Not in the last 12 months”) (DHS, 2017a). Based on the answers to those questions, the type of DV that a woman experienced was then classified into four broad categories: emotional violence, less severe violence (physical), severe violence (physical), and sexual violence (DHS, 2017a).     

Reporting domestic violence. Whether or not a woman told anyone about her experience of DV was measured through one dichotomous question. If the woman answered yes to any of the questions regarding type of violence experienced, the woman was then asked “Have you ever told anyone about this?” (DHS, 2017a).

Risk factors. Risk factors of DV were measured using a dichotomous question of whether or not the respondent’s husband drinks alcohol (DHS, 2017a).

Social norms. Social norms related to justification of DV were measured through five dichotomous questions, including “Is beating justified if the wife doesn’t cook food properly?” and “Is beating justified if the wife argues with the husband?” (DHS, 2017b). The answers to these questions were compiled into a composite score measuring attitudes toward DV (α = .83). “Yes” responses were coded as 1 and “No” responses were coded as 0. The maximum composite score is 5, signifying a belief that inter-spousal physical violence is generally justified, and the minimum score is 0, signifying that beating is generally unjustified.

The respondents’ opinions on social norms related to women’s empowerment and gender roles were measured through five questions, including “Who should have the greater say when deciding what to do with the money the wife earns from her work?” and “Who should have the greater say when making major household purchases?” (DHS, 2017b). For this study, these questions were recoded into a binary variable of the wife having either no power or some degree of power. “Husband” was coded as 0, representing the wife having no power, “Wife” and “Wife and husband jointly” were coded as 1, representing the wife having some degree of power, and “Don’t know/depends” was coded as missing due to the small number of respondents (n = 15). These five questions were compiled into a composite score (α = .83) where the minimum score was 0, signifying that the wife should not have any power over household decisions, and the maximum score was 5, signifying that the wife should have at least some degree of power over household decisions.

Control variables. Based on prior peer-reviewed studies on domestic violence in India, the following sociodemographic characteristics were included as control variables: wife and husband’s age, wife and husband’s education level (no education or some education), wife and husband’s employment status (currently employed or not currently employed), type of residence (rural or urban), religion (Hindu or Muslim), total number of children, and wealth index (poor or not poor) (Begum et al., 2015; Rowan et al., 2018).


RESULTS

A multivariate logistic regression was run to assess which factors are associated with whether or not married women in Bihar reported their experience of domestic violence to anyone. The factors in the regression analysis included sociodemographic characteristics of both the husband and the wife, variables relating to the experience of violence, and variables relating to social norms (See Appendix). Controlling for all other variables, the wife’s level of education is the only sociodemographic variable significantly associated with whether or not she reported her DV experience to anyone (β = .75, p = .01). Women who had some level of education were twice as likely to report their DV experiences when compared with women with no education (ExpB = 2.12). Respondent’s wealth index is marginally significant (β = -.63, p = .06). Of the experience of violence variables, and keeping all other variables constant, whether or not women experienced emotional violence is significantly associated with reporting (β = -.84, p = .001). Women who experienced emotional violence were 58 percent less likely to report DV when compared with women who did not experience emotional violence (ExpB = .43). It is important to note that this variable does not represent women who exclusively experienced emotional violence. This finding indicates that the women who experienced emotional violence alone or in combination with physical violence were less likely to report DV than women who experienced physical or sexual violence without accompanying emotional violence.

Additionally, keeping all other variables constant, whether or not the husband drinks alcohol is significantly associated with reporting (β = -.53, p = .03). Women were 41 percent less likely to report their DV experiences if their husband drank alcohol (ExpB = .59).

Finally, controlling for all other variables, lasting effects of violence is significantly associated with reporting (β = .30, p = .03). Women who exhibited lasting physical effects of violence (bruises, burns, scars, etc.) were 34 percent more likely to report than women who had no lasting physical effects of violence (ExpB = 1.34).

When controlling for all other variables, neither attitudes toward DV being justified (β = .002, p = .97) nor the women’s empowerment scale (β = .03, p = .65) are significantly associated with whether or not women reported DV.


DISCUSSION

The data confirm a substantial discrepancy between women experiencing DV and women reporting it: out of the 1,053 women who experienced some form of domestic violence, only 94 women (8.9%) stated that they reported it to someone.

According to the logistic regression model used for this study, there are several predictors that influence whether or not a woman reports, and these explanations are important when considering women’s possible motivations for remaining silent and subsequent policy recommendations aimed at addressing this discrepancy.

Experience of emotional violence. Women who experienced emotional violence were 58 percent less likely to report than those who did not experience emotional violence. These findings align with previous research, as women who experience less severe forms of violence are less likely to report that violence (Naved, Azim, Bhuiya, & Persson, 2006; Rowan et al., 2018). It is important to note that these women may have experienced other forms of DV as well, suggesting that emotional violence significantly decreases the likelihood of reporting other forms of DV. The literature suggests that emotional abuse is often the first type of DV that a woman experiences (Karakurt & Silver, 2013), and women who experience emotional violence before other types of violence may be less inclined to report because they may grow accustomed to violence that gradually increases in severity (Naved et al., 2006; Rowan et al., 2018). 

Lasting physical effects of violence. Women who have shown lasting physical effects of violence are 34 percent more likely to report DV. Lasting effects of violence may be a sign of more severe abuse, which could explain why this variable predicts reporting. Additionally, women who experience severe abuse might consider their lives to be in danger, the desire for survival outweighing the social stigmas and risks of reporting (Rowan et al., 2018; Fanslow & Robinson, 2010). Another hypothesis is that someone may have seen cuts or bruises on a woman and directly asked her about violence, thus prompting her to report.

Husband drinks alcohol. Women were 41 percent less likely to report DV if their husband drank alcohol. This finding may reflect the perceived risk that accompanies reporting DV. It is likely that women whose husbands drink believe that there is more risk in reporting the violence they have experienced, as reporting could lead to violent retaliation if the husband were to find out (Berg et al., 2010).

Women’s education. The only other significant predictor of reporting was the woman’s level of education. Women who had some level of education were more than two times as likely to report domestic violence as women with no education. This finding reflects how more educated women may be more aware of their rights as well as of the official reporting systems that exist in Bihar (Rowan et al., 2018; Andersson et al., 2010).

Experience of emotional violence. Women who experienced emotional violence were 58 percent less likely to report than those who did not experience emotional violence. These findings align with previous research, as women who experience less severe forms of violence are less likely to report that violence (Naved, Azim, Bhuiya, & Persson, 2006; Rowan et al., 2018). It is important to note that these women may have experienced other forms of DV as well, suggesting that emotional violence significantly decreases the likelihood of reporting other forms of DV. The literature suggests that emotional abuse is often the first type of DV that a woman experiences (Karakurt & Silver, 2013), and women who experience emotional violence before other types of violence may be less inclined to report because they may grow accustomed to violence that gradually increases in severity (Naved et al., 2006; Rowan et al., 2018).

Lasting physical effects of violence. Women who have shown lasting physical effects of violence are 34 percent more likely to report DV. Lasting effects of violence may be a sign of more severe abuse, which could explain why this variable predicts reporting. Additionally, women who experience severe abuse might consider their lives to be in danger, the desire for survival outweighing the social stigmas and risks of reporting (Rowan et al., 2018; Fanslow & Robinson, 2010). Another hypothesis is that someone may have seen cuts or bruises on a woman and directly asked her about violence, thus prompting her to report.

Husband drinks alcohol. Women were 41 percent less likely to report DV if their husband drank alcohol. This finding may reflect the perceived risk that accompanies reporting DV. It is likely that women whose husbands drink believe that there is more risk in reporting the violence they have experienced, as reporting could lead to violent retaliation if the husband were to find out (Berg et al., 2010).

Women’s education. The only other significant predictor of reporting was the woman’s level of education. Women who had some level of education were more than two times as likely to report domestic violence as women with no education. This finding reflects how more educated women may be more aware of their rights as well as of the official reporting systems that exist in Bihar (Rowan et al., 2018; Andersson et al., 2010).

Social Norms and Reporting

The present study initially sought to utilize norm theory as a framework for understanding the low rate of reporting in Bihar. If India’s social norms portrayed domestic violence as a natural and acceptable part of marriage, then norm theory would suggest that these perceptions influence individual behavior, thus hindering women’s likelihood to report (Berkowitz, 2005; Paluck & Ball, 2010). None of the social norm variables, however, predicted whether or not a woman reported her experience with DV. Neither women’s empowerment variables nor overall social attitudes toward DV significantly impacted the rate of reporting.

Instead, the data painted a different picture. The mean for the women’s empowerment composite score was 3.55 out of 5, signifying that most respondents believe that women should hold at least some power in household decisions. Additionally, the mean for the DV composite score was 1.78 out of 5, meaning that there were, on average, between one and two situations where respondents thought that spousal beating was justified. This indicates that the social norms (that women should have some power and that DV is generally not justified) contradict the study’s initial review of the literature about gender norms in India (Uthman, Lawoko & Moradi, 2009; Rani, Bonu, & Diop-Sidibe, 2004; Gouping, 2010). The discrepancy between the social norms in the literature and the social norms in the sample illuminates a conceptual challenge with using norm theory to explain low rates of reporting in Bihar.

One hypothesis for this study’s social norms results is social desirability bias, meaning that women may have selected social norms based on what they thought the right answer should be instead of what they actually believed to be true. Another explanation for the unexpected social norm data relates to the study’s sample. By definition, social norms are the perceived norms of an entire community (Berkowitz, 2005; Paluck & Ball, 2010). In this study, however, data on social norm variables was collected only from women who have experienced some form of domestic violence. Utilizing a more representative sample (including men and women who did not experience DV) to examine Bihar’s social norms may lead to results that are more reflective of the literature on social norms in India. Lastly, these social norm results are limited in that they only measure attitudes toward domestic violence being justified and beliefs that women should have some power when making household decisions. These social norms are not inclusive of all norms that can potentially influence a woman’s likelihood to report DV in Bihar.


CHALLENGES AND RECOMMENDATIONS

The use of secondary data presented a significant challenge for the present study. Due to fear of survey fatigue, DHS restricts the number of questions in each module, therefore limiting the DV measures for this study. A wider array of questions would be necessary to gain a fuller picture of the factors contributing to the low rate of reporting in Bihar. Data pertaining to how and to whom domestic violence was reported would further contribute to knowledge of reporting and stigma. Notably, the variable of reporting DV only provided information on whether or not a woman reported her abuse, not to whom it was reported. The lack of information on whom the DV was reported to limits our knowledge on the use of formal versus informal systems of reporting. Future research should utilize a refined definition of reporting to better understand women’s motivations when choosing whether or not to report to both informal and formal systems. 

Another notable challenge is that investigation of the discrepancy between prevalence-of-DV and reporting-of-DV relies on self-reported data. This poses a few conceptual challenges. All women who indicated that they experienced some form of domestic violence reported that violence to the surveyor, which then led to the question of whether they told someone else about the DV they experienced (the question about reporting). The fact that the data collection inherently relies on self-reports while seeking to measure reporting adds a layer to response bias that must be noted. To address this challenge, future studies should consider working with community-based groups to conduct surveys. These local groups may be able to collect more reliable data on sensitive topics like DV, since they are “insiders” that community members would be more likely to trust compared to external surveyors.

Finally, there are likely more women who experienced violence than who admitted it in the survey, so there are many challenges in interpreting the accuracy of the data. This challenge emphasizes the need for data on how women report their DV experience (what they share and do not share, how honestly they describe the frequency and severity of their experiences, and what happens when they do report). Considerable attention should be given to designing a culturally sensitive survey instrument in order to more effectively measure domestic violence and reporting in Bihar. Only with a more robust understanding of the factors contributing to low reporting can interventions be developed to address this problem.


REFERENCES

Abrams, M. L., Belknap, J., & Melton, H. C. (2001). When domestic violence kills: The formation and findings of the Denver metro domestic violence fatality review committee. Project Safeguard, Denver.

Ahmed-Ghosh, H. (2004). Chattels of society: Domestic violence in India. Violence Against Women, 10(1), 94-118.

An analysis of perceptions of perceptions of domestic violence and efficacy of the implementation of the PWDVA (2005) act in Bihar. (2016). CARE India. Retrieved from https://www.careindia.org/wp-content/uploads/2017/05/study-report-R2.pdf

Andersson, N., Cockcroft, A., Ansari, U., Omer, K., Ansara, N., Khan, A., & Chaudhry, U. U. (2010). Barriers to disclosing and reporting violence among women in Pakistan: Findings from a national household survey and focus group discussions. Journal of Interpersonal Violence, 25(11), 1965-1985.

Begum, S., Donta, B., Nair, S., & Prakasam, C. P. (2015). Socio-demographic factors associated with domestic violence in urban slums, Mumbai, Maharashtra, India. India Journal of Medical Research, 141, 783-788.

Berg, M. J., Kremelberg, D., Dwivedi, P., Verma, S., Schensul., Gupta, K., ... Singh, S. K. (2010). The effects of husband’s alcohol consumption on married women in three low-income areas of greater Mumbai. AIDS & Behavior, 14, 126-135.

Berkowitz, A. D. and Perkins, H. W. (1986). Problem drinking among college students: A review of recent research. Journal of American College Health, 35, 21-28.

Berkowitz, A. D. (2005). “An overview of the social norms approach.” In L. Lederman and L. Stewart (Eds.). Changing the culture of college drinking: A socially situated prevention campaign (pp. 193-214). New York: Hampton Press.

Bihar tops states in domestic violence. (2008). The Hindu. Retrieved from http://www.thehindu.com/todays-paper/tp-national/tp-otherstates/Bihar-tops-States-indomestic-violence/article15279971.ece

Bradley, J. E., Bhattacharjee, P., Ramesh, B., Girish, M. & Das, A. (2011). Evaluation of stepping stones as a tool for changing knowledge, attitudes and behaviors associated with gender, relationships and HIV risk in Karnataka, India. BMC Public Health, 11, 496.

Chachra, M. (2017). 10 years of domestic violence act: Dearth of data, inadequate implementation, delayed justice. India Spend. Retrieved From http://www.indiaspend.com/cover-story/10-years-of-domestic-violence-act-dearth-of-datainadequate-implementation-delayed-justice-85613

Chaudhary, S. (2013). Domestic violence in India. Journal of Indian Research, 1(2), 146-152.

CMS Communication. (2011). End line survey on domestic violence and HIV/AIDS, 2010. Breakthrough: New Delhi, India.

Demographic and Health Surveys. (2017a). Domestic violence module. Retrieved from https://dhsprogram.com/pubs/pdf/DHSQM/DHS7-Module-DomViol-Qnnaire-EN-27Jan2017-DHSQM.pdf

Demographic and Health Surveys. (2017b). Model woman’s questionnaire. Retrieved from https://www.dhsprogram.com/pubs/pdf/DHSQ7/DHS7-Womans-QRE-EN-07Jun2017-DHSQ7.pdf

Dubochet, L. (2012). Protecting Women from Domestic Violence. Oxfam. Retrieved from https://policy-practice.oxfam.org.uk/publications/protecting-women-from-domestic-violence-288437

Fanslow, J. L, & Robinson, E. M. (2010). Help-seeking behaviors and reasons for help seeking reported by a representative sample of women victims of intimate partner violence in New Zealand. Journal of Interpersonal Violence, 25(5), 929-951.

Felson, R. B. and Paré, P. P. (2005). The reporting of domestic violence and sexual assault by nonstrangers to the police. Journal of Marriage and Family, 67(3), 597-610.

Guoping, H., Yalin, Z., Yuping, C., Momartin, S., & Ming, W. (2010). Relationship between recent life events, social supports, and attitudes to domestic violence predictive roles in behaviors. Journal of Interpersonal Violence, 25(5), 863-876.

Govinderajan, P. (2016). A curious change to India’s domestic violence act. The Diplomat. Retrieved from https://thediplomat.com/2016/11/a-curious-change-to-indiasdomestic-violence-law/

Heise, L. L. (2011, December). What works to prevent partner violence? An evidence overview. Working paper. Retrieved from http://www.dfid.gov.uk/R4D/PDF/Outputs/Gender/60887-Preventing_partner_violence_Jan_2012.pdf

International Institute for Population Sciences (IIPS) and ICF. (2017). National Family Health Survey (NFHS-4), India, 2015-16: Bihar. Mumbai: IIPS.

Jejeebhoy, S. J., & Santhya, K. G. (2018). Preventing violence against women and girls in Bihar: challenges for implementation and evaluation. Reproductive health matters, 26(52), 92-108.

Jhamb, B. (2011). The missing link in the Domestic Violence Act. Economic and Political Weekly, 45-50.

Kalokhe, A., et al. (2017). Domestic violence against women in India: A systematic review of a decade of quantitative studies. Global Public Health, 12(4), 498–513.

Karakurt, G., & Silver, K. E. (2013). Emotional abuse in intimate relationships: The role of gender and age. Violence and Victims, 28(5), 804–821.

Kishor, S. and Johnson, K. (2004). Profiling domestic violence: a multi-country study. Measure DHS: ORC Marco.

Koenig, M. A., Stephenson, R., Ahmed, S., Jejeebhoy, S. J. & Campbell, J. (2006). Individual and contextual determinants of domestic violence in North India. American Journal of Public Health, 96(1), 132-138.

Krishnan, L. (2017). After a decade of implementation of PWDV Act in Bihar, study on domestic violence shows some mixed results. CARE India. Retrieved from https://www.careindia.org/blog/after-decade-implementation-pwdv-act-bihar-studydomestic-violence-shows-some-mixed-results/

Miller, D.T. & McFarland, C. (1991). When social comparison goes awry: The case of pluralistic ignorance. In J. Suls & T. Wills (Eds.) Social comparison: Contemporary theory and research (pp. 287-313). Hillsdale, NJ: Erlbaum.

Naved, R. T., Azim, S., Bhuiya, A., & Persson, L. A. (2006). Physical violence by husbands: Magnitude, disclosure and help-seeking behavior of women in
Bangladesh. Social Science & Medicine, 62, 2917-2929.

Paluck, E. L. and Ball, L. (2010). Social norms marketing aimed at gender based violence: A literature review and critical assessment. New York: International Rescue Committee.

Raljan, A. and Chakraborty, S. (2010). Regional report of the Assessment of “We Can” Phase II, Draft II. London: Oxfam.

Rani, M., Bonu, S., and Diop-Sidibe, N. (2004). An empirical investigation of attitudes towards wife-beating among men and women in seven sub-Saharan African countries. African Journal of Reproductive Health, 8(3), 116-136.

Rowan, K., Mumford, E., & Clark, C.J. (2018). Is women’s empowerment associated with help-seeking for spousal violence in India? Journal of Interpersonal Violence, 33(9), 1519-1548.

Sahoo, H. and Raju, S. (2007). Domestic violence in India: Evidences and implications for working women. 

Straus, M. A. & Douglas, E. M. (2004). A short form of the Revised Conflict Tactics Scales, and typologies for severity and mutuality. Violence and Victims, 19, 507-520.

Straus, M., Gelles, R., & Steinmetz, S. K. (1980). Behind closed doors: A survey of family violence in America. New York.

Toch, H. & Klofas, J. (1984). Pluralistic Ignorance, Revisited. In G. M. Stephenson and J. H. Davis (Eds.), Progress in Applied Social Psychology, Vol. 2 (pp. 129-159). New York: Wiley.

United Nations, General Assembly. (1993). “Declaration on the Elimination of Violence against Women.” Retrieved from http://www.un.org/documents/ga/res/48/a48r104.htm

Uthman, O.A., Lawoko, S., & Moradi, T. (2009). Factors associated with attitudes towards intimate partner violence against women: a comparative analysis of 17 sub-Saharan countries. BMC International Health and Human Rights, 9, 14.

William, S. & Aldred, A. (2011). Change Making: How we adopt new attitudes, beliefs and practices. Insights from the We Can Campaign. London: Oxfam GB.


APPENDIX

Significant factors of whether or not women reported their experience of domestic violence.

Independent VariableBS.E.Exp(B)
Experienced any emotional violence-.838.250.433**
Experienced any severe violence0.1170.2781.124
Experienced any sexual violence0.0680.2521.071
Husband/partner drinks alcohol-0.5260.244.591*
Lasting effects on violence composite score0.2950.1371.344*
Attitudes towards DV composite score0.0020.0641.002
Women’s empowerment composite score0.0310.0681.031
Age of women-0.0090.0170.991
Age of household head-0.0070.0100.993
Number of children ever born0.0260.0671.026
Type of residence: Rural vs. urban-0.3450.3290.708
Wealth index: Poor vs. not poor-0.6330.3370.531
Women's education: No education vs. some education0.7490.2962.115**
Husband’s education: No education vs. some education-0.710.2590.932
Wife’s employment status: Employed vs. not employed0.5170.3171.677
Husband’s employment status: Employed vs. not employed0.4980.4141.646
Constant-2.6481.1400.071

*p<.05, **p<.01


PORTIA BAJWA is a second-year master’s student at the School of Social Service Administration with a focus on clinical social work and global social development. She is currently placed at Heartland’s Marjorie Kovler Center, where she provides psychotherapy to torture survivors who are seeking asylum in the United States. Her academic interests center on mental health and trauma recovery across an array of cultural contexts. Prior to enrollment at the University of Chicago, Portia worked as a teaching assistant in Spain. She earned her bachelor’s degree in Psychology at the the University of Washington in Seattle.

KELSEY FOREMAN is a second-year master’s student at the School of Social Service Administration with a concentration in social administration. She is currently working in the field of early childhood and is interested in understanding how trauma impacts children’s development, particularly in international contexts. Prior to enrolling at the University of Chicago, she earned her bachelor’s degree in intercultural communication from Pepperdine University. In the future, she hopes to continue working in the field of early childhood with a focus on implementing trauma informed practices.

CHARLOTTE F. SALL is a second-year social administration master’s student pursuing a certificate in Global Social Development Practice. Her academic interests include refugees and migratory populations, organizational theory, program evaluation, and comparative perspectives to social work. Prior to SSA, Charlotte worked with Roma youth in Serbia and taught English in rural South Korea. For her second-year field placement, Charlotte spent six months at an education non-profit in New Delhi, India, where she focused on research and evaluation of early childhood interventions. Charlotte graduated from Princeton University with a degree in sociology.

Return to Advocates' Forum 2019