HC 934 Violence against Women and Girls

Written evidence submitted by the Gender Violence and Health Centre (GVHC)

London School of Hygiene and Tropical Medicine


Members of the Gender Violence and Health Centre (GVHC) at the London School of Hygiene and Tropical Medicine welcome the opportunity to share our views on the DFID’s evolving programme on violence against women and girls. In this submission, we concentrate on elements of DFID’s programme that relate to research and evaluation, our primary area of expertise, but touch briefly on larger issues affecting the effectiveness of the UK government’s commitment to addressing violence against women and girls (VAWG).

Specifically we discuss:

· The strength of the UK government’s leadership on VAWG internationally

· Benefits and weaknesses of DFID’s current approach to research and generating evidence to inform policy and programmes

· Underexploited opportunities to create synergies between DFID programme areas and strategies of work

· Broadening DFID’s humanitarian programming on VAWG

· Increasing integration of VAW programming across DFID’s programmes

· Improving policy coherence across Whitehall and the wider UK government

To help tackle violence against women and girls (VAWG) more effectively, and improve the processing of generating evidence to inform future work, we propose that the UK Government:

1. Lengthen the average timeline for research grants designed to generate evidence for informing policy and programmes on VAWG, recognising that change may not occur over short times

2. Ensure that interventions are optimized before investing funds to evaluate their impact

3. Adapt Dfid’s model of funding research and programme consortia (RPCs) to its work on VAWG, including a one year inception period for programmes designed to evaluate complex interventions related to VAWG

4. Commit to piloting new interventions before seeking to evaluate them for impact

5. Consider adopting a "programme science" approach to expanding and evaluating programmes to address VAWG in low income countries

6. Re-orient DFID’s portfolio of projects on VAWG to better align them with its theory of change

7. Broaden DFID’s humanitarian work on VAWG from conflict-related sexual violence to include the full range of gender-based abuses that displaced women and women in conflict experience

8. Strengthen and institutionalise technical expertise on VAWG within DFID, including the provision of basic orientation information on VAWG to all DFID staff, and institutionalise senior positions within the organisation, to serve as technical experts and support the achievement of programme and policy commitments

9. Seek synergies between DFID programming on early childhood development and child health, and its programming on VAWG, and recognise the value of programmes to address VAWG for other areas of health and development, including HIV, maternal health, and mental health.

10. Ensure that the Ministerial Champion on Violence Against Women and Girls Overseas has the access, authority and resources to drive the international section of the cross-government strategy forward across all government departments.

11. When competing contracts, ring fence funding for women’s rights organisations and prioritise applicants that have a demonstrated track record of working effectively on women’s issues, and be sensitive to the risks that small but established civil society groups may be less well placed to bid for large contracts, but have greater expertise in this area.

1.0 Introduction

1.1 In response to the request for input into the UK’s International Development Committee’s inquiry about programming to prevent violence against women and girls, we would like to offer our insights as a group of experts who have worked for the past two decades researching and campaigning to stop violence against women and girls. Together we work at the Gender Violence and Health Centre (GVHC) of the London School of Hygiene and Tropical Medicine, with the group having expertise in impact evaluation, qualitative research, quantitative data analysis, systematic reviews and economic evaluation.

1.2 Researchers at GVHC have been involved in implementing several large-scale and small projects funded by DFID. This includes interventions research in Mpumalanga Province, South Africa, that demonstrated a 55% reduction in intimate partner violence; an on-going assessment of efforts to reduce trafficking and its associated harms among women in the Mekong Region of South East Asia; research in Tanzania on the relationship between women’s economic development and violence; and a research programme consortium (RPC) known as STRIVE, that is dedicated to generating evidence on how to address the structural drivers of HIV, including gender inequalities and violence. This research programme includes prevention intervention research in Uganda and Tanzania. In addition, GVHC has are now involved with trials of partner violence interventions in Bihar, India (funded by DFID) and Cote d’Ivoire. Overall, our group’s expertise focuses primarily on prevention of violence against women and girls (VAWG) and its health-related consequences, but we are increasingly focusing also on the links with child abuse. As well as primary data collection, our group is known for its cross-national research and guidance, including Prof Watts chairng the Global Expert Review on the extent of intimate partner violence and its health impacts; Dr Zimmerman’s leadership in the development of guidelines on support to survivors of trafficking; design and Dr Heise’s synthesis of evidence on effective prevention.

1.3 We draw on this body of work and our long-standing collaborations with Southern and International partners to offer the following observations and suggestions regarding DFID’s programme of related to VAWG.

2.0 Strength of UK leadership role internationally on VAWG

2.1 Global leadership DFID is to be congratulated for taking a global lead in policy dialogue, interventions and generation of evidence on VAWG. In our assessment, DFID is one of the four governmental entities (including AusAID, Irish Aid and the recently announced US global programme on VAWG) that have elevated violence to the level of global attention that it deserves. DFID has expanded funding available to support work to reduce violence and support victims, and has made a visible commitment to pursue evidence-informed programming in this area.

2.2 Emphasis on evidence DFID has proven its dedication to ‘evidence-informed’ programming through commissioning a large-scale literature review conducted by Dr. Lori Heise of our group, entitled: What works to prevent partner violence? This review summarises the available evidence on factors causally linked to partner violence and on what is known from research to reduce either abuse or the proximate determinants of abuse [1]. In addition, they commissioned the Gender and Development Network (GADNet) to develop a theory of change to guide their programming [2] and a series of guidance notes for DFID programme officers on evaluating violence interventions [3] and on implementing community-level prevention programmes [4].

2.3 In addition, DFID should be recognized for seeking to strengthen the global evidence base on strategies to effectively address VAWG, including female genital cutting, partner violence and violence in conflict situations. DFID has demonstrated this commitment by sponsoring several large impact assessments of programmes to reduce violence, including its current programming to design and evaluate interventions to reduce partner violence in Bihar India and to evaluate efforts to reduce human trafficking in the Mekong region. Soon they will be announcing major initiatives to combat FGC and support further programming and evaluation research through its planned "innovations fund on VAWG." The details of these initiatives are not yet available so it is impossible at this juncture to comment on their design or implementation. How these two initiatives are structured, however, is likely to influence greatly the future effectiveness of DFID’s programming.

3.0 Strengths of DFID programming on VAWG

3.3 Mapping and theory of change. DFID recently developed a theory of change to guide its programming and completed a mapping of its work globally on VAWG. This is an important first step in rationalising its investments and reconciling its grant-making with its new theory of change. However, in our view, DFID’s current portfolio suggests an over-commitment to reforming formal justice, social service and health systems as a core strategy for responding to the needs of victims and for deterring future violence. Since less than 10% percent of survivors in most settings ever reach out to formal institutions [5], strategies focused exclusively on formal services miss the lion’s share of the problem. Moreover, while services can assist individual women to escape violence, they have a limited preventive effect. Our research suggests that globally, almost 3 in 10 partnered women will experience violence, with higher values in many developing country settings. Such high levels of violence, alongside generally low levels of formal service use, highlight the critical need to invest in prevention, as well as broader legal and support services for survivors of violence.

3.4 Nonetheless, we see positive shifts toward greater emphasis on prevention and cultivating community-led strategies for responding to the needs of victims in DFID’s new theory of change. The ToC outlines the four specific sets of interventions DFID should pursue to eliminate all forms of VAWG:

1. Empower women and girls

2. Change the social norms that condone and perpetuate VAWG

3. Build political will and legal and institutional capacity to prevent and respond to VAWG

4. Increase the provision of comprehensive services for women and girls that are victims and survivors of violence

3.5 To reconcile its programming with its revised TOC, DFID will need to greatly increase its emphasis on empowering women and girls and changing social norms that condone and maintain VAWG. This will require shifts in country-level priorities and the ability to monitor whether the required shifts in investment are occurring.

3.6. Adapting DFID’s RPC model to VAWG. DFID has several existing models of programming and evaluation that could productively be expanded in its work on VAWG. One especially relevant model is DFID’s strategy of supporting Consortia of researchers and implementing agencies over 6 to 8 years to generate a programme of structured learning about a topic such as VAWG. Typically, DFID sets out the desired outcomes and terms of engagement, and groups of researchers, NGOs, and governmental bodies come together as a team to compete for the available funding. The first round of competition requires a broad brush vision of the proposed programme of work and a firm commitment to collaborate. DFID then picks the top 3 or 4 contenders and they are invited to submit full proposals for consideration. Importantly, DFID provides £10,000 up front to these teams to meet and work on the design of their proposal. This opportunity to meet is critical to the long-term success of these endeavours because it allows all partners to have input into the design of the work plan from the beginning. Establishing the terms of engagement and building trust early on, helps ensure that the partners will be able to negotiate the differences in power and perspective inherent in such collaborations (North South divides; differences in disciplinary perspective; and orientations toward research versus implementation, etc.).

3.5 Sustained Funding. Also critical to the success of this model of knowledge generation is the 6 year time frame and the incorporation of a 1 year "inception" period for planning. Sustained funding over at least 6 years is an almost essential feature of evaluating interventions that are attempting to address deeply embedded patterns of behaviour like violence. One of the weaknesses of the current funding regime more broadly is the requirement to demonstrate "results" in 2 to 3 year time frames. As described further in paragraph 4.3, such time frames are generally unrealistic when attempting to evaluate complex, social interventions. DFID’s programming on VAWG is moving toward more sustained funding, as evidenced by its support for violence work in Bihar. But the majority of its support is still relatively short term, given the nature of the task at hand.

4.0 Weaknesses in current approach to VAWG programming

4.1 Unrealistic Expectations. We are concerned that DFID has unrealistic expectations about the state of knowledge in the VAWG field and in the costs and challenges of generating new knowledge to inform policy and programmes. While the field has gained many well-founded insights into factors that contribute to gender-based violence in different settings, and there is starting to be evidence on the impacts of different models of intervention, the field of violence prevention is less than a decade old and the evidence available to inform interventions varies greatly by setting and type of violence. Thus far, we have more evidence to inform programming in the areas of FGC and partner violence than we have in sexual exploitation, violence in conflict situations, child sexual abuse, acid throwing, or honour crimes. Even with FGC and partner violence, however, experience trying to prevent these forms of violence rather than dealing with their consequences, is relatively new. Even newer, are efforts to rigorously evaluate those approaches that are being tried. For this reason, it is very important that new interventions are evaluated, but new initiatives need to be sensitive to the state of the field and knowledge.

4.2 Importantly, the state of knowledge into how to prevent and respond to VAWG effectively is no less developed than many other areas of social policy, which also share a dearth of clear evidence to guide investment and policy making. Thus absence of evidence should not be used as an excuse for inaction; rather it argues for moving forward in a thoughtful and informed way. It is important that DFID explicitly seeks to maximise opportunities for thoughtful learning and sharing in its investment in this area.

4.3 Evaluation strategies for complex interventions are costly, because they must be implemented at a community rather than an individual level. This means that to implement a randomized controlled trial (RCT)-the gold standard of evaluation research-often requires randomising whole communities or neighbourhoods, rather than individuals, to receive an intervention. Since the intervention must be implemented in a large number of communities in order to achieve statistical power to detect an effect, community randomised trials can easily cost £1+ million or more. The number of communities or clusters that must receive the intervention likewise, frequently taxes the ability of the implementing agency to deliver and monitor the programme. As a result, there is tension between the need to move quickly, the quality control that can be maintained on the intervention, and the cost and size of a trial. It is our experience that DFID, like many donors, frequently underestimate the complexity and cost of delivering the type of "evidence" they seek. This leads to corners being cut that undermine the validity of the entire exercise.

4.4 Need to optimize programming before evaluation. The short time frames of project support and DFID’s commitment to generate evidence has resulted in a rush for evaluation, with the risk that substantial sums are invested in evaluating programmes that are sub-par. The pressure to generate evidence quickly means that programmes are being evaluated before their implementation and design have been fully optimized. We recommend that in all complex evaluations DFID adopt a 1 year inception period, where researchers and programme staff work together to optimize the design and delivery of the programme before formal impact evaluation begins (i.e. baseline data are collected). We further recommend that prior to investing large sums to evaluate the potential impact of newly designed interventions, that they be piloted to work out problems, inform the ultimate intervention design, and generate the site-specific data (such as inter-cluster variation estimates) necessary to ensure that the evaluation will have the power to capture an impact, should it be there. Presently, the desire to move quickly is jeopardising the validity of programme findings and potentially wasting substantial sums of money.

4.5 Adopting a programme science model. DFID’s violence portfolio should include other models of evaluation that can help generate vital learning in areas where formal impact assessment is impossible or not yet warranted. Increasingly the field of HIV prevention is adopting a model of assessment known as "programme science." Programme science embeds researchers into large scale programmes as they are developed and rolled out, encouraging active learning and the immediate feedback of lessons learned into the refinement of the programmes [6]. Programme science not only seeks to answer whether something "works", but also uses research to help "build the helicopter as you are flying it." It is geared toward learning by doing, constant programme modification, and capturing insights along the way. Given the need to move forward with programming while generating evidence, DFID could be well-served by this type of research model.

5.0 Broaden DFID’s humanitarian programming on VAWG

5.1 In conflict-affected settings, women and girls appear to be equally, if not more likely to experience a range of non-conflict-related violence, such as domestic violence, acquaintance rape, child sexual abuse [7]. For example, findings from the GVHC-IRC survey in Cote d’Ivoire demonstrates that even in conflict affected settings, the burden of partner violence is larger than that related to the conflict [8]. Given this reality, DFID’s humanitarian programming should broaden its focus from conflict-related sexual violence to include the full range of gender-based abuses that displaced women and women in conflict experience.

6.0 Increase integration of VAW programming across DFID programmes

6.1 While the portfolio of work on VAWG within DFID is ambitious, it is currently siloed, rather than being integrated systematically across the work of the organisation. There is a need for a stronger focus on VAWG as a core issue across all of DFID’s programming, in particular, in the areas of water and sanitation, education, health and emergency response. Although programming in various areas has begun to identify the potential ‘gendered’ effects of proposed activities (e.g., water and sanitation assessment of safe distances for women to walk for water), to date, these programmes are not incorporating the input of local and international experts in violence against women. VAWG is a technical area, and it is important that appropriate input is received when designing programmes if there is going to be a more meaningful investment in addressing the potential causes and consequences of violence.

7.0 Strengthen and institutionalise technical expertise on VAWG within DFID

7.1 Institutionalise expertise within DFID Like other areas of development work, VAWG is an area of technical expertise. There are real dangers that ill-conceived programmes can actually put women at greater risk of violence, or reinforcing gender inequalities in the community. Interventions need to be carefully designed and implemented, with considerations of women’s and girl’s safety and agency at their core, and it is important that all staff have basic understanding of the issues at hand. In addition, those developing country programmes and international policy requires access to individuals with deep knowledge and experience with VAWG. Recently DFID issued an RFA to contract individuals with technical expertise to assist their staff in designing country-level programmes. This is a step in the right direction. However, DFID requires at least some senior, full time staff in charge of this area of expertise. Presently, DFID headquarters seems to be relying on seconded staff from women’s organisations to fulfil this role. The post is neither full-time nor permanent and has been filled with a host of different individuals who have rotated through, making continuity difficult, both within DFID and in relation to its work with groups and agencies in this area.

8.0 Pursue synergies between VAWG programming and work with children

8.1 As detailed in the DFID and ESRC funded document, What Works to Prevent Partner Violence, there are many strategic reasons to better integrate violence with DFID"s work with children and social protection programmes. One of the strongest causal factors in perpetration of partner violence is exposure to violence during childhood. Witnessing or experiencing violence affects a child’s developing brain, making them more likely to exhibit anti-social behaviour in adolescence and violence in adulthood; it likewise distorts their understanding of acceptable ways to exert power and get what you want. There are many unexploited opportunities to reduce the long term impacts of violence on children’s development and to help prevent future violence by working through parenting and early childhood development programmes. Research has shown that early intervention can overcome the negative impacts of early exposure to violence on future behaviour.

8.2 Increasingly, the impacts of VAWG on health and development are being recognised. Effective prevention to address VAWG, including programmes to empower women, and shift gender norms about the acceptability of violence, and men’s and women’s roles and expectations in relationships, have the potential to achieve multiple development and health impacts, including in HIV, reproductive health and mental health. For this reason, it is important that progamme funds seek to achieve these synergies, as they provide important opportunities for DFID investments to achieve large returns on investment.

9.0 Improve policy coherence across Whitehall and the wider UK government

9.1 As noted in the testimony of ActionAID, there are a number of governmental processes that address the UK’s work on VAWG, including:

1. The Cross Government Violence against Women and Girls Strategy (Home Office)

2. The DFID Gender Strategy

3. The National Action Plan on UNSCR 1325 Women, Peace and Security (FCO)

4. The Building Stability Overseas Strategy (FC0)

5. The Preventing Sexual Violence in Conflict Initiative (FC0)

9.2 Policy coherence among these various initiatives should be strengthened, in order to achieve greater impact and ensure that opportunities to advance women’s rights and address VAWG more broadly are not missed. A case in point is the FCO’s recent project on sexual violence in conflict. While attention to violence by the FCO is welcome, this initiative is needlessly narrow and threatens to squander important political capital at the G-8 by focusing exclusively on investigating and prosecuting perpetrators of sexual violence in war. Various organisations, including GVHC and the members of GADNET, have pushed for a broader agenda focused on comprehensive efforts to support survivors of violence and to prevent violence from happening in the first place.

9.3 To better align agendas, the UK government should ensure that the Ministerial Champion on Violence Against Women and Girls Overseas has the access, authority and resources to drive the international section of the cross-government strategy forward across all government departments, especially among the FCO, DFID and MoD.

10. Recognize and maintain the historical leadership of women’s rights organisations in efforts to combat VAWG

10.1 Until recently, small women’s organisations have been virtually the only actors willing to champion the cause of VAWG. They have deep experience with the issue that should be valued and further cultivated. New opportunities for funding have begun to attract more mainstream development actors, some with little understanding of gender-based violence. In crafting its request for proposals and funding schemes, DFID needs to recognise the danger that small civil society groups are least able to compete for large grants, but have the largest expertise. For this reason, DFID should ring fence some funding for women’s rights organisations, and prioritise applicants that have a demonstrated track record working effectively on women’s issues.

February 2013


1. Heise, L., What works to prevent partner violence: An evidence overview, 2012: London, UK.

2. Department of International Development, A theory of change for tackling violence against women and girls, in CHASE Guidance Note Series 2012, Department of International Development: London, UK.

3. Department of International Development, Guidance on monitoring and evaluation for programming on violence against women and girls, in CHASE Guidance Note Series 2012, Department of International Development: London, UK.

4. Department of International Development, A practical guide on community programming on violence against women and girls, in CHASE Guidance Note Series 2012, Department of International Development: London, UK.

5. Garcia-Moreno, C., et al., Prevalence of intimate partner violence: findings from the WHO multi-country study on women's health and domestic violence. Lancet, 2006. 368(9543): p. 1260-9.

6. Blanchard J.   & Aral S. Program science Program Science: an initiative to improve the planning, implementation and evaluation of HIV/sexually transmitted infection prevention programmes Sex Transm Infect 2011; 87 :1 2-3

7. Watts C, Zimmerman C. Violence against women: global scope and magnitude. Lancet 2002;359(9313):1232-7

7. Hossain, Mazeda, personal communication, London School of Hygiene and Tropical Medicine, February 3, 2013

Prepared 18th March 2013