Draft Honor Violence Study Report
Larger, more general studies frequently lack the rapport-building and cuing that lead to
accurate data on sensitive topics (Ellsburg et al., 2001). In a study assessing whether the
use of large-scale demographic surveys (such as the Demographic and Health Survey
[DHS]) yield lower estimates of the prevalence of violence than smaller, focused studies,
Ellsberg and colleagues reported that, compared with studies dedicated to the issue of
domestic violence, the prevalence of severe domestic violence was significantly lower in
the DHS. This outcome occurred despite the use of similar survey instruments that
included behavioral descriptors of violence.
Most researchers on violence against women agree that, at a minimum, a survey must
include measures to protect the safety of respondents and interviewers, crisis
intervention and referrals to specialized services for respondents, and special training
and emotional support for interviewers (Garcia-Moreno and Jansen, 2009: slide 10).
Ellsberg and Heise (2005), Watts et al. (2001), and Garcia-Moreno and Jansen (2009)
provide lists of considerations for studying women in violent situations. These
considerations include interviewing only one woman per household, not informing the
wider community that the survey includes questions on violence, not interviewing men
about violence in the same households or clusters where women have been asked about
violence, conducting interviews in complete privacy, using dummy questionnaires if
others enter the room during the interview, using candy and games to distract children
during interviews, using self-response questionnaires for some portions of the interview
in literate populations, training interviewers to recognize and deal with a respondent’s
distress during the interview, and referring women requesting assistance to local services
and sources of support or developing resources to provide such support.
These issues related to the measurement of violence toward women, in general, are also true for
measuring honor violence directed at women. Measurement of honor violence directed at women
involves unique factors also, including variation in cultural definitions of violence and the influence
of the community on reporting of these crimes. The issue of participant risk is particularly important
to the study of honor violence, which often involves lethal attacks. As one researcher reported,
“Collecting data on … violence against women that results in death (e.g., dowry-related deaths,
honor killings, and female infanticide) is particularly challenging” (Friends, 2013:13).
In addition, women who are victims of honor violence often do not recognize that they are victims
of violence, but rather “they interpret their treatment as an intrinsic part of their culture” (Gill,
2009:478). Weisz and colleagues (2000) note that “Islamic law supports husbands’ rights to use
physical force against their wives,” while “[i]n Arab cultures, even a stereotypical rape may be viewed
as an experience of promiscuity or infidelity, punishable by death” (Weisz et al., 2000:229). Any
attempt to accurately measure honor violence will require an understanding of the cultural meaning
of the violence to the victims. For this reason, incorporating questions about honor violence into a
victimization survey that frames the questions in context of crime may fail to obtain a report of the
This document is a research report submitted to the U.S. Department of Justice. This report has not
been published by the Department. Opinions or points of view expressed are those of the author(s)
and do not necessarily reflect the official position or policies of the U.S. Department of Justice.