November 2023 BREAKING THE CYCLE OF INTIMATE-PARTNER VIOLENCE: HARM REDUCTION BY ENGAGING ABOUT THE AFRICA GENDER INNOVATION MEN IN THE DEMOCRATIC REPUBLIC OF CONGO LAB Authors: Alev Gurbuz Cuneo, Kathryn Falb, Rocky Kabeya, Estelle Koussoubé, Rachael The Africa Gender Innovation S. Pierotti, Julia Vaillant1 Lab conducts impact evaluations of development KEY MESSAGES interventions in Sub-Saharan Africa, seeking to generate • Intimate partner violence (IPV) is a global problem and a widespread issue evidence on how to close in the Democratic Republic of Congo (DRC), where nationally more than gender gaps in earnings, one third of women have reported experiencing physical or sexual violence productivity, assets, and in the past year (2013-2014 DHS). The Africa Gender Innovation Lab (GIL) tested agency. The GIL team is the effectiveness of the Engaging Men through Accountable Practice (EMAP) currently working on over 80 program, a male-only discussion group intervention which aimed at preventing impact evaluations in more than 30 countries with the intimate partner violence (IPV) and promoting more egalitarian gender relations. aim of building an evidence EMAP was implemented by the International Rescue Committee (IRC). base with lessons for the • New analyses of the study data paint a nuanced picture of the impact of region. the intervention. While results from the impact evaluation show that on average The impact objective of GIL is the EMAP program did not lead to reductions in rates of IPV, further analysis shows increasing take-up of effective that there was positive impact of the intervention on some participants. policies by governments, development organizations, • We find that among men who were the most physically violent at baseline, and the private sector to the EMAP program was effective at reducing the probability and severity address the underlying of IPV perpetration. The results suggest that for men who are more violent than causes of gender inequality most of their peers, participatory discussions with less violent men may inspire in Africa, particularly in terms of women’s economic them to reduce their use of violence. In contexts of endemic violence, programs and social empowerment. such as EMAP can lead to a meaningful short-term reduction in harm to women, The Lab aims to do this by perhaps even without transforming prevailing norms about male superiority or the producing and delivering a acceptability of IPV. new body of evidence and developing a compelling • Although zero violence is a necessary and critical goal when promoting narrative, geared towards gender equality, these results reinforce the importance of measuring the policymakers, on what works contribution of policies and programs to reduce IPV’s occurrence, in addition to and what does not work in measuring effects on primary prevention and cessation of violence. promoting gender equality. 1 Nelsy Affoum contributed to the preparation of this policy brief. https://www.worldbank.org/en/programs/africa-gender-innovation-lab CONTEXT characteristics. Then, within each pair, one site was In the Democratic Republic of Congo, intimate partner randomly selected to receive the EMAP program and the violence is a pervasive problem. Women are frequently other was designated as part of the control group. In all exposed to physical insecurity, which affects their income sites, men who expressed interest in participating in EMAP generating activities and further contributes to increased were surveyed before the start of the program and again vulnerability to poverty and insecurity. In this context, approximately 9 months after the end of the program. violence prevention programs can lead to harm reduction Female partners of these men were also interviewed. The by preventing the onset of the use of violence (primary study’s data on violence come from women’s reports, prevention), by encouraging the cessation of the use of which were collected using a self-administered module violence, or by fostering a decrease in the frequency or on tablets to maximize confidentiality. severity of violence. To continuously improve program We used these data to evaluate whether there were design and inform targeting of future programs, it is differential impacts of the EMAP program depending important to understand these mechanisms of change on baseline types and levels of violence. To do so, we and to identify which participants experience positive grouped participants based on levels of violence reported impact. at baseline and estimated the program’s impact for each of those groups. This is an important feature of the study’s The Engaging Men through Accountable Practice (EMAP) design as it tests whether IPV prevention programs have program was intended to create a community of male the potential to achieve greater positive results on the allies who practice and promote gender equality and behaviors of relatively less or more violent men. non-violence. Trained male facilitators followed a scripted curriculum to lead 16 weekly discussions that encouraged participants to critically examine gender roles, power WHAT WE FOUND within relationships, and intimate partner violence. Adult Among men who volunteered to participate in men who were interested and willing to commit to non- EMAP, there was substantial variation in levels violence during the program period were encouraged to of pre-program violence perpetration in the past volunteer to participate. Although the program primarily year. Using a data-driven approach called latent class sought to engage non-violent men to elevate their roles analysis, we grouped men into four categories based on in the community as change agents, screening out all the levels of physical and sexual violence reported by violent men in the program recruitment process was not their female partners before the start of the program. The successful in communities with endemic violence. largest of the four categories, labeled “little to no IPV”, representing 70 percent of the sample, were men who Given this program design, we wanted to know if there was had not been violent or had perpetrated no more than variation in levels of pre-program violence perpetration one act of physical aggression in the past 12 months. among participants and whether program effectiveness The second largest group (15 percent of the sample) differed across relatively more or less violent men. Did reported moderate levels of physical violence and high the program encourage continued non-violence among levels of sexual violence. The third group (9 percent of those who were not using violence before the start of the the sample) reported high levels of physical violence and program? Did it encourage a reduction or cessation of moderate levels of sexual violence. Six percent of the violence among men who had been violent toward their sample reported high levels of both types of violence, intimate partner in the year before the program? labeled “systematic violence”. Additional information about the specific acts of violence HERE’S WHAT WE DID that were reported is presented in Figure 1. For example, Between 2016 and 2018, the Africa Gender Innovation in the “high physical and moderate sexual violence” group, Lab conducted a randomized control trial (RCT) to between 60 and 70 percent of women reported getting evaluate the effects of the EMAP program. Twenty-eight pushed and slapped, nearly 50 percent reported getting communities in the North and South Kivu provinces were kicked, and around 18 percent reported experiencing divided into pairs that were matched on sociodemographic forced sex. FIGURE 1: ACTS OF VIOLENCE AS REPORTED BY THE LATENT SUBGROUPS Systematic IPV High Physical & Moderate Sexual IPV Moderate Physical & High Sexual IPV Little to no IPV 0% 100% 0% 100% 0% 100% 0% 100% Push Slap Twist Physical violence Punch Kick Choke Weapon Sexual violence Forced sex Sex out of fear Degrading sex FIGURE 2: EFFECTS OF EMAP ON EXPERIENCE OF PAST YEAR IPV EMAP Control Comparison of endline rates of physical IPV Comparison of endline rates of sexual IPV between EMAP and Control, by subgroup between EMAP and Control, by subgroup 100 100 80 80 60 ** 60 40 40 20 20 0 0 Systematic High Physical Moderate Little to Systematic High Physical Moderate Little to IPV & Moderate Physical & High no IPV IPV & Moderate Physical & High no IPV Sexual IPV Sexual IPV Sexual IPV Sexual IPV Level of IPV at baseline ** = Stars indicate a difference that is statistically significant at the 5% level The EMAP program led to a statistically significant In contrast, EMAP was not effective at preventing reduction in physical IPV among women who the onset of the use of violence among men in the experienced high physical and moderate sexual “little to no violence” group. In both the treatment violence at baseline. We find a substantial behavioral and control groups, approximately 25 percent of the change among men whose partners reported high levels couples in this subgroup reported physical violence in the of physical violence prior to benefitting from the EMAP follow-up survey. In fact, in a separate analysis of violence program. This is an important and notable reduction in severity, there is suggestive evidence that the program harmful behavior. These findings suggest that men who may have had the unintended effect of slightly higher perpetrate violence against their female partners with levels of violence in this group. greater severity than average may be inspired to reduce their use of violence through participatory discussion with The EMAP program was not successful at less violent men. (significantly) preventing or reducing sexual IPV on average among any program participants. This suggests an avenue for additional program experimentation and research. There is a need for different types of interventions for reducing rates of forced sex and other forms of sexual violence within intimate relationships. CONCLUSIONS These analyses show that EMAP was effective at reducing the use of violence among some men who participated in the program. More specifically, the program was effective at reducing physical IPV among men who were most physically violent in the year before the start of the program. This has two main implications for policy and research. First, the results suggest that in some contexts, it may be possible to reduce harm to women, even before attaining more long-term goals of transforming prevailing norms about male superiority and the acceptability of IPV. The research showed that before the program began, most participants regarded only low levels of IPV as common and acceptable. These analyses demonstrate that men who were using higher than average levels of violence were encouraged by the program to reduce the severity of the violence that they perpetrate. Through the intervention, they may have learned from their peers that their actions were outside the bounds of acceptable behavior. This also indicates that the composition of discussion groups is an important feature of program design. Groups that are comprised of both more and less violent men may be important for encouraging progress among “outlier” men using violence levels above the group norm. Second, while zero violence is a necessary and critical goal, along the way, it is important to measure the contribution of policies and programs to reductions in IPV, in addition to measuring primary prevention and cessation of violence. While a fundamental shift in gender relations is an essential goal for the promotion of gender equality and women’s rights, it is important to measure shorter term improvements as well. Careful analysis of incremental changes can be useful for learning about how best to promote systemic transformative change in the future. For more information on the study, see the research paper here, the original impact evaluation paper here, and the in-depth qualitative study paper here. FOR MORE INFORMATION, PLEASE CONTACT Africa Gender Innovation Lab afrgenderlab@worldbank.org Photo credit: Vincent Tremeau / World Bank, John Hogg / World Bank, Dominic Chavez / World Bank 1818 H St NW Washington, DC 20433 USA This work has been supported through generous contributions from the World Bank’s Umbrella Facility for Gender www.worldbank.org/africa/gil Equality (UFGE), the World Bank’s Jobs Umbrella Multidonor Trust Fund, and the National Science Foundation (Kaur’s CAREER award SES 1848452). We thank Innovations for Policy Action (IPA), especially Nicolò Tomaselli, Henriette Hanicotte, Samuel Kembou Nzalé, Mireille Nuguhe Gbagbo and Augustin Kouadio for assistance with implementation. This project is a product of the World Bank’s Africa Gender Innovation Lab and Jobs Group.