MAKING THE INVISIBLE VISIBLE: Measuring Domestic Violence and Its Consequences

1
MAKING THE INVISIBLE VISIBLE Measuring
Domestic Violence and Its Consequences
Sunita Kishor (Ph.D.) Senior Gender
Advisor Demographic and Health Research Macro
International, Inc. Maryland, USA
2
Overview

• Gender and gender-based violence (GBV)
• The role of the Demographic and Health Surveys in
  making the invisible VISIBLE
• The nitty-gritty of measurement substantive
  and ethical challenges
• What we now know prevalence, correlates, and
  consequences
• Continuing challenges

3
A digression What is
gender?
...the different roles, rights, and obligations
that culture and society attach to individuals
according to whether they are born with male or
female sex characteristics
Not just different, but UNEQUAL Not just
UNEQUAL, but SUBORDINATE
GBV an inherent part, an expression, an
outgrowth, of this gender inequality
4
What is gender-based violence?

• Gender-based violence is violence involving men
  and women, in which the female is usually the
  victim and which is derived from unequal power
  relationships between men and women. Violence is
  directed specifically against a woman because she
  is a woman, or affects women disproportionately. 
  
• United Nations Population Fund (UNFPA)
• Gender Theme Group

5
Gender-Based Violence Over the Life
Cycle
Violence by states (e.g. rape in
war) Trafficking Acid throwing Non-partner
rape/harassment/violence Dowry deaths/honor
killings Differential access to food/medical care
Psychological abuse Coerced sex/rape/harassment
Physical violence Violence during pregnancy
Female infanticide FGC/M Sex-selective
abortion Differential access to food/medical
care Violence in pregnancy Physical and
sexual violence Psychological abuse
Others
Family members
Intimate partner
Prebirth/ Infancy
Adolescence Reproductive Older
ages
(adapted from Watts and Zimmerman, 2002)
6
Sources of Domestic Violence Data

• Institutional sources
• Police, hospital, court records data from
  shelters
• Incomplete coverage provide data on only very
  severe cases
• Qualitative, ethnographic studies
• Cannot be generalized good for in-depth study
• Survey sources
• Surveys on GBV Rare and costly
• e.g., WHO Multi-Country Study of Domestic
  Violence
• Surveys on other topics with questions on GBV
  e.g., Demographic and Health Surveys (DHS)

7
What is the DHS?

• A demographic, nutrition, and health survey
  program for the developing world since1984,
  primarily funded by USAID
• Nationally-representative household surveys with
  large samples (typically 6000 India-100,000)
• Usually repeated every 5-7 years
• Provide indicators for monitoring at the national
  and sometimes, sub-national level
• Comparable across countries and over time
• More recently include biomarkers, e.g., syphilis
  and HIV, and domestic violence

DHS surveys considered the gold standard by
countries and development partners
8
(No Transcript)
9
DHS Countries with Domestic Violence Data

• Latin America/Caribbean
• Bolivia 2003
• Colombia 1990, 1995, 2000, 2005
• Dominican Republic 2002 2007
• Haiti 2000 2005
• Honduras 2005
• Nicaragua1997/98
• Peru 2000 2004
• Eurasia
• Azerbaijan 2006
• Moldova 2005
• Turkmenistan 2001
• Ukraine 2007

• Africa
• Cameroon 2004
• DRC 2007
• Egypt 1995 2005
• Ghana 2008
• Kenya 2003
• Liberia 2006/07
• Malawi 2004
• Mali 2006
• Rwanda 2005
• South Africa 1998
• Uganda 2006, 2009
• Zambia 2001/02 2007
• Zimbabwe 2005/06
• Asia/South Asia
• Bangladesh 2004, 2007
• Cambodia 2000 2005
• India 1998/99 2005/06

Data not yet available.
10
Measurement Challenges

• Defining what violence to measure
• Ensuring validity of the measures the how
• Ensuring safety and ethical standards
• Determining what else to measure other than
  prevalence

11
Validity of Data

• No catch-all term to capture violence
• Recommendation Ask about different types of
  violent acts separately, eg have you been
  slapped NOT have you experienced any violence or
  have you been beaten?
• Risk of underreporting of violence
• Recommendations Build rapport, ensure privacy,
  provide multiple opportunities to reveal abuse

12
Measures of Spousal Violence

• Physical violence Any of the following acts
  perpetrated by her husband
• Pushed her, shook her, or threw something at her
• Slapped her
• Twisted her arm or pulled her hair
• Punched her
• Kicked her, dragged her, or beat her up
• Tried to choke her or burn her on purpose
• Threatened her or attacked her with a weapon

• Sexual violence Any of the following any of the
  following acts
• Forced her to have sexual intercourse when she
  did not want it
• Forced her to perform sexual acts she did not
  want to

• Emotional violence

13
Selected Domestic Violence Indicators in the DHS

• Prevalence of spousal violence ever and 12
  months
• Timing of initiation of spousal violence
• Injuries related to spousal violence
• Wives initiating spousal violence
• Violence by others (ex-spouse, boyfriend, family,
  others)
• Forced first sexual intercourse
• Violence during pregnancy
• Lifetime prevalence of sexual violence by anyone
• Help seeking by abused women

14
Safety and Ethical Concerns

• Confidentiality and informed consent
• Only one woman per household receives the DV
  questions
• Men not asked DV questions in the same household
• Informed consent obtained
• Translators not used for this section
• Interview discontinued if privacy not assured
• Respondent and interviewer safety
• DHS guidelines are modeled on WHOs Guidelines
  for the Ethical Conduct of Domestic Violence
  research (WHO, 2002)
• DHS has sought to strike the right balance
  between ethical concerns and information needs

15
Prevalence of Spousal Physical or Sexual
Violence Selected Countries
Percent of ever-married women age 15-49
Only 1-7 of married women have ever initiated
violence against their husbands.
16
Correlates of Spousal Violence India 2005-06
Percent of ever-married women age 15-49
17
Correlates Husbands Drinking of
AlcoholAdjusted odds of experiencing violence
among ever-married women age 15-49
Odds of experiencing violence 1 if husband does
not drink then odds of experiencing violence if
husband
are

(ns)
Kishor and Johnson, 2004
18
Intergenerational Aspects of ViolenceAdjusted
odds of women (ever-married, 15-49) reporting
violence ever
Odds of experiencing violence if father did not
beat mother 1 Adjusted odds of experiencing
violence if father beat mother
Kishor and Johnson, 2004
19
Immediate Health Consequences Among women who
have ever experienced spousal violence
Percent who have had
20
Selected Adverse Health Outcomes
Never experienced spousal violence, odds 1 If
experienced violence, adjusted odds
Had an STI
Discontinued modern contraceptive
Base line risk 1.0
Base line risk 1.0
21
MORE Adverse Health Outcomes
Never experienced spousal violence, odds 1 If
experienced violence adjusted odds
Pregnancy termination
Unwanted birth
Base line risk 1.0
Base line risk 1.0
Kishor and Johnson, 2006
22
What we do know

• Spousal violence varies from about 14 in
  Cambodia to 59 in Uganda
• Typically, half or more of the women who have
  ever experienced spousal violence have
  experienced it in the past 12 months
• Spousal sexual violence is reported by 5-20
  percent of women
• Physical violence is the most common type of
  violence
• Violence begins early in the marriage
• Violence results in injuries
• Women rarely seek help
• Further analysis shows that there are strong
  links to health outcomes for mothers and their
  children

23
What Perpetuates GBV?

• Gender norms that
• Rigidly define gender roles and what a good
  woman is
• Tolerate and accept male aggression and risky
  sexual behavior as part of being a man
• Give men the right to discipline and control
  wives
• Cultural norms that treat domestic violence as a
  private matter
• Womens lower status in society that results in
• Limited access to and control over resources
• Less education and skill development
• Dependency on men, on sons

24
DHS Question on Norm Socialization

• Is a husband justified in hitting or beating his
  wife if she
• Neglects the children
• Answers back/argues with him
• Goes out without telling him
• Refuses sex with him
• Burns the food

25
Evidence of Norm Socialization
One of the most common reasons Neglects the
children
26
Solution Multifaceted Interventions A
Health-Sector Example
Communication for social behavior change
programs
HEALTH SERVICE DELIVERY PROGRAMS

• Correct diagnosis
• Appropriate care
• No harm to patient
• Care attuned to physical psychological needs

Effectiveness of health provision
Community mobilization programs
Health policy programs
Womens RH
Womens experience of GBV
Norms Laws Institutions Funding Policy discourse
27
Gaps and Continuing Challenges

• Measurement of
• mens experience of violence from women
• other types of intimate partner violence
• Doing more to minimize under-reporting
• Recognize the limitations of large scale surveys
• Doing more in-depth and longitudinal studies to
  fill in gaps
• Going beyond the measurement of prevalence
• Analysis, dissemination, action

28
www. measuredhs.com
29
For Specific Questions

• sunita.kishor_at_macrointernational.com

Photo credits Photos courtesy of M/MC Photoshare
at jhuccp.org/mmc References Kishor S. and K.
Johnson. 2006. Reproductive Health and Domestic
Violence Are the Poorest Women Uniquely
Disadvantaged? Demography Vol. 43(2 )
293-307 --. 2004. Profiling domestic violence A
multi-country study. ORC Macro, Calverton,
Maryland Watts. C. and C. Zimmerman. 2002.
Violence against women global scope and
magnitude Lancet Vol. 359( 9313)1232
1237 World Health Organization. 2001. Putting
women first Ethical and safety recommendations
for research on domestic violence against
women. Department of Gender and Women's Health,
Geneva, Switzerland