How to avoid unintended consequences in intimate-partner violence prevention programs

The views expressed in the Let’s Talk Development blog are solely those of the author(s).

In the quest to promote women's rights and opportunities and reduce intimate-partner violence (IPV), well-intentioned programs can sometimes yield results opposed to their goal. A recent paper sheds light on the surprising and concerning outcome of an IPV prevention program in Rwanda. The program, designed to improve communication and promote gender equality among couples, unexpectedly led to a significant increase in IPV. This blog explores the key findings, implications, and broader lessons for policymakers and designers of similar interventions.

The study, a randomized control trial, documented the negative impact of a 22-week couples training program aimed at preventing IPV in rural Rwanda. The program, funded by the World Bank, led by the Rwandan Ministry of Gender and Family Promotion and implemented by local and international NGOs, sought to improve communication, shift gender attitudes, promote gender equality among couples, and ultimately shift behaviors such as intimate-partner violence perpetration. Contrary to the program's objectives, the intervention resulted in significant increases in IPV in treated communities. Women who participated in the program as well as non-treated women who lived in treated communities reported higher rates of emotional, physical, and sexual violence compared to those in the control group.

The data used were collected through a study involving more than 2,000 couples in about 100 communities in Rwanda's Eastern Province. The couples were randomly assigned to either the treatment group, which received the 22-week training program, or the control group. The study employed a two-stage randomization process: first, communities were randomly assigned to either the treatment or control group, and then within treated communities, couples who expressed interest in the training were randomly assigned to participate or not in the program. This produced three groups: a treated group, a spillover group, and a control group. Data collection included baseline and endline surveys, which measured various indicators such as emotional, physical, and sexual IPV, gender attitudes, well-being, and economic empowerment. The researchers also used innovative measurement techniques and extensive robustness checks to ensure the accuracy and reliability of the findings.

Couples who did not directly participate in the training but lived in the same communities as treated couples experienced the largest increases in IPV. These unintended negative ‘spillover’ effects may stem from male backlash against perceived identity threats. As women in the program became more progressive in their attitudes and aspirations, men felt pressured to reassert their dominance, leading to increased violence. The negative outcomes were less severe among couples who directly participated in the training sessions, though still present. Even in these directly trained couples, women still experienced significant increases in IPV compared to the control communities, just not as extreme as the spillover group.

These findings contribute to ongoing debates about the effectiveness of IPV prevention programs and highlight the complexities involved in strengthening women's roles in the household and communities through behavioral change interventions. The importance of examining how interventions may impact individuals' sense of identity and the potential for backlash cannot be overstated. These findings contrast with earlier studies in Rwanda and beyond that found significant reductions in IPV, raising questions about the external validity of the results and the effectiveness of such interventions once indirect effects at the community levels are considered.

Developing strategies to mitigate the risks of backlash is crucial for the success and safety of such programs. The new research provides a cautionary message about the potential unintended consequences of interventions aimed at shifting deeply held social norms. Policymakers, program designers, and researchers working on IPV prevention should consider incorporating specific measures to address backlash risks. Alternative forms of programming may be more appropriate depending on the context. For example, framing such programs as improving broad outcomes for families and communities rather than emphasizing gender-equality objectives explicitly has been found to mitigate backlash in some circumstances. Other programs that may provoke less resistance include those affecting change through storytelling, such as edutainment programs, which may avoid direct confrontation between couples. Additionally, adolescent-focused interventions may be effective by shaping attitudes when young men have more malleable perspectives and less to "lose" in terms of power.

Changing deeply held social norms is a complex and challenging process. This paper ultimately highlights how considering norms, identity and backlash risk can inform better program design. By understanding how interventions affect individuals' sense of self and status within their communities, policymakers and practitioners can develop more effective strategies for promoting gender equality. The research calls for approaches that acknowledge and address the complex interplay between gender norms, identity, and social change.