Smart Library on Globalization
Topics |  Bibliography |  Authors |  Expert Panel |  Help Print PDF Add
Local women are an invaluable source of data and knowledge about sexual assault. Local women in Liberia designed and administered a survey and gathered high quality data on sexual assault.

Collecting high quality statistical data on human rights abuses is usually the domain of specialists. But, it need not be so.

In fact:

  • Rural women in the midst of on-going conflict can design and administer a survey and collect high quality data within their own communities,
  • Survey instruments can be tailored to specific conflicts in specific countries,
  • The results of statistical research need not be limited to specialists, but can be tailored in a way that is useful to people in local communities.

Women's Rights International, in collaboration with the Women's Health and Development Program, carried out a study in Monrovia, Liberia on the rates and nature of violence against women during the Liberian civil war. What was notable about this study was the fact that the survey was designed by local women, administered by local women and packaged in a way that it would be useful to the local community. Academic specialists were involved in the research, but they worked with women in local communities to carry out the research. The result showed that research designed and carried out by local people in local contexts is not “second rate” research.

There are definite benefits to using this method, which the authors call “action research.” Research carried out in this way has special features.

Methodology Is Tailored to Its Purpose

Should you gather research during the conflict or wait until it is over?

Gathering data during an ongoing conflict has definite risks.

  • Safety is an issue, not only for the people gathering the data, but also for the respondents. Accusing a neighbor, a militia member or a member of the military of a crime puts the respondent at risk for retribution.
  • Data collection is an issue. Security problems, curfews and war related complications can hamper data collection.

However, these risks may be outweighed by the benefits, which include:

  • Data gathered immediately may be more accurate and more useful,
  • Victims may be more willing to report recent violations since they represent current concerns. A woman who was raped in the past may be less willing to talk about the incident and more interested in putting her past behind her.
  • Conflicts can be protracted—sometimes lasting years—and it is irresponsible to wait until it is “safe” to gather crucial information that can be used to create much needed intervention programs.

So, while there are definite risks, gathering data during conflicts can pay off.

What Is the Purpose?

There are different reasons for gathering data, and these reasons shape the way that the data is gathered and who gathers it.

If the purpose is to gather data in order to prosecute in international tribunals, then international interviewers may be the best way to go.

However, if the goal of gathering data is to understand the impact of the violence on women's lives and tailor programs to address their needs, then local women may be best suited to this task.

The purpose also shapes just what data is gathered. For instance, if the interest is ultimately to bring criminals to justice, then names, dates, places, etc. are all critical. On the other hand, if the goal is to better the local situation, then gathering detailed personal information on assailants may not be needed—and in fact may put the women at risk. Members of fighting factions may be living in the same villages or towns. It could be years before local people feel safe enough to give out names.

Additionally, if the data is to be used to better the local situation, then local people need to be in control of that data. International organizations and donors may want certain kinds of information immediately and for their own purposes. However, these purposes may be at odds with the purposes of the locals (as with the case of prosecuting criminals versus living in safety). These organizations and donors need to be sensitive to the needs and situation of the local residents in determining what to do with the information gathered.

Using Respondents to Design the Survey

Why use local women to design a survey instrument? Surely, academics trained to create surveys would be better at it.

Not necessarily. But, the situation is not either/or. When experts trained in research and survey construction and administration work with local people to design and administer a survey, the outcomes may actually be more valid. Researchers from outside the community have their own cultural filters. They may enter with an understanding of how, say, violence against women plays out in other conflicts of countries. The problem is, these filters may miss information vitally important to the particular context.

For example, research on other conflicts shows that there are certain places or situations in which women are at a greater risk of sexual violence. However, when researchers met with small groups of Liberian women to talk about their range of experiences, they discovered that sexual assault was linked to being forced to cook for a soldier or fighter. Women are often detained at checkpoints or when their village is attacked and pressed into domestic service (cooking) for one of the fighters. In the context of the women studied, the rates of sexual assault are much higher for women who are forced to cook for fighters than for those who are not. So, as it turns out, asking whether a woman was forced to cook for a fighter is an important piece of information to include on the survey that would likely not have appeared had an outside researcher simply asked where they were sexually assaulted. A woman may be willing to share that she was conscripted into cooking for a fighter, but less forthcoming about a sexual assault. Knowing the local context of high risk for sexual assault can help the researchers and the locals get a better sense of what the real rates of assault are.

A Word on Wording

The wording of a question is critically important for getting accurate data. One of the challenges is that what counts as rape in one context may not be the same as in another context. Indeed, the study found that the understanding of rape varies from woman to woman.

In the small group discussions among the Liberian women, it became clear that there was a continuum along which sex and war related. Some defined rape as a sexual assault only when the woman sustained a visible injury. According to other woman, only an unmarried woman or girl could be raped. If a married woman was assaulted, it would not be considered rape.

Even the meaning of “force” varied. Not all women understood being “forced to have sex when they didn't want to” in the same way. A woman who entered into a relationship with a soldier or fighter in order to survive or gain protection might not understand that as a “forced” relationship—that she was forced to trade sex for food, protection, etc. From the woman's perspective, the relationship may have been voluntary. Certainly, brutal rape was forced sex. But, was sex within a relationship with a fighter forced? Allowing the women to design the questions and tailor the wording to the particular situation opened up a broader understanding of the use of force within gender relationships in this context.

Disseminating Research

Research findings can be disseminated at many levels. This means that different formats are called for. Highly technical reports may be appropriate in an international setting, where complex charts and tables are the norm. At the local level, not only is the level of training likely to be different, but the uses to which the information will be put is different as well.

In the Liberian setting, the researchers developed a story methodology. By drawing on the typical experiences in common settings, researchers could communicate findings such as central tendency in a way that made sense to the local women. Liberian women used example characters, settings and experiences drawn from the research to tell stories about women in the typical situations. What happened to them and why? Even women who are not literate can use graphs or figures tailored in such a way that they communicate the basic statistical ideas.

The goal of action research is to draw on the local knowledge to increase the local and general understanding. The aim is to use this new information—valid and locally relevant—to better the lives of the women in the local setting.

Data and Methods:

Data Sources:

A survey of 205 Liberian women in Monrovia, Liberia. Local women met in small discussion groups to help design the survey. Surveys were administered in neighborhoods, markets, internally displaced persons' camps and high schools.

Funding Sources:

Not reported.

Full Text Availability:
Full text available at

Jennings, PJ., and Swiss, S. 2000. "Statistical Information on Violence Against Women During the Liberian Civil War." In Statistics, Development, and Human Rights: Proceedings of the International Association of Official Statistics. Montreux, Switzerland.

© Copyright 2018 CLG Portal. All Rights Reserved. Powered by