Guest Post : Do we need numbers?

By the Centre for Poverty Analysis (

You can follow CEPA on facebook at and on twitter at

Gender-based violence is a violation of human rights, and yet is a well known and frequently occurring problem in Sri Lanka. While GBV is considered almost synonymous with domestic violence, community, workplace, school, places of worship and public transport, among others, are all places in which violence and abuse befall women, and to a lesser extent, men. Under the larger umbrella of GBV, these acts of violence can be further categorized as domestic violence (DV), intimate partner violence (IPV), genital mutilation, rape, assault and sexual assault, among others. While the issue has been studied at length over the last three decades, there is a lack of representative data on the prevalence of GBV, as well as DV, IPV and other forms of violence, which may be affecting the design and implementation of effective policies to eliminate these practices.

Numerous studies have been conducted on the incidence of GBV and related phenomena by state, non-state and academic institutions, which has resulted in an array of findings. For example, the average prevalence of DV in Sri Lanka is estimated to range from 30-60% (WHO); IPV is estimated at 40% (Jayatilleke 2010). The majority of the studies on GBV are conducted using small, non-representative samples. As such, there are no national figures on the prevalence of these issues. Considering that GBV causes serious impacts on the health and wellbeing of individuals, there should be strong state and non state mechanisms in place for documenting incidents of GBV, tools for data collection, and avenues of rehabilitation and justice accessible to the victims. In order for these systems to be implemented, there needs to be valid and representative national data to support currently established national policies and laws, and for further support structures and resource allocation to be put in place. As long as the data remains ambiguous about the extent of the problem in Sri Lanka, it is difficult to establish the cause and implications of the phenomenon.

Perceptions of Sri Lankans are also important to understanding the causes of GBV. The Demographic and Health Survey 2006/2007 contains one question which probes the attitudes of women towards wife beating, which shows that women feel it is acceptable to beat a wife if she argues with her husband (41%), goes out without telling him (35%) and neglects the children (41.8%). Such attitudes are similar over age, location, and district, but differ according to the level of education received. On a larger scale, this survey could be used as a tool to measure perceptions of men and women on issues such as patriarchal attitudes, women’s role in society, stigma regarding abuse, appropriate sentences for abusers and other issues that could help us better understand the extent of the problem in Sri Lanka.

While it is established that more national level data on GBV is needed, we can speculate about the many reasons for why it is not collected. The numbers that are currently available come through various channels, including the Women in Need counseling center, the Women and Children’s Police desks, hospitals and other research and community organizations. The constraints for data collection on a larger scale could be due to the sensitivity of the subject matter and the need for specially trained enumerators. An example which illustrates this difficulty is the Oxford Poverty and Development Initiative (OPHI) study, which developed an international survey module on ‘missing ‘dimensions of poverty’ – that is, dimensions of poverty which are largely missing from nationally representative survey data. One such dimension is physical safety and the draft module contained a section on domestic abuse. However, it was considered too sensitive and difficult an area for country level enumerators to probe without extensive training and was dropped from the pilot surveys in Sri Lanka, Nigeria and the Congo. Another constraint to data collection is the respondent’s reluctance to discuss issues of GBV due to issues of confidentiality, social stigma and fear of repercussions.

In the face of these challenges, what suggestions can researchers make? One option is to include GBV related questions in household surveys such as the DHS, to be administered by staff specially trained for the task. However, it could be argued that this method wouldn’t provide comprehensive data due to prioritizing information and time constraints in administering the questionnaire. As such, if a household survey is seen as too big an instrument to tackle the issue, considering cost of designing the survey, staff training, sampling, administering the survey, data entry, cleaning and analysis, then special surveys could be constructed for this specific purpose.

In New Survey Methodologies in Researching Violence against Women, Walby and Myhill look at three possible survey options: generic national crime surveys, dedicated domestic violence surveys and violence against women surveys (2001). The two latter surveys appear more effective, drawing on a range of quantitative and qualitative methods to provide context to the issue. Many countries, including USA, UK, Canada and Germany have surveys dedicated to gathering information on domestic violence and violence against women. “These large-scale national surveys are very important in providing the robust evidential basis of the extent and nature of gendered violence” (Walby, 2004)

In conclusion, it is important for researchers and implementers to have access to national figures on the severity of GBV in the country, in order to study in-depth the causes and consequences of the issue, as well as for implementers, to put in place preventive and rehabilitation structures. It may also aid government policy and regulations in alleviating GBV as a phenomenon. However, how to obtain such data, while safeguarding the rights of the respondents, is a challenge that still needs to be resolved.


Jayatilleke, A. C., Poudel, K., Yasuoka, J., Jayatilleke, A. U. and Jimba, M. (2010) Intimate partner violence in Sri Lanka. BioScience Trends 2010; 4(3):90-95.


Oxford Poverty and Human Development Initiative, The Missing Dimensions of Poverty Data Research Theme ( and notes from training workshop and study visit attended by CEPA February/March 2009


Walby, S., and Myhill, A. (2001) Survey Methodologies in Researching Violence against Women British Journal of Criminology


Walby, S., (2004) Domestic Violence: Developments in survey methodology. Presented to European conference on ‘Everyday violence and human rights’.

Women’s Empowerment and Demographic and Health Outcomes. The Demographic and Health Survey 2006/2007, Chapter 14.

World Health Organization (WHO) gender-based violence (gbv) basic information sheet, Sri Lanka (


One thought on “Guest Post : Do we need numbers?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s