A case study comparison of engaging community activists to prevent gender-based violence in Peru and Rwanda

ABSTRACT

Community mobilisation is recognised as an important strategy to shift inequitable gender norms and ensure an enabling environment to prevent gender-based violence (GBV). Yet there is a need to better understand the factors that facilitate effective community activism in particular contexts. Although fundamental to the success of mobilisation programmes, there is also limited appreciation of the experiences and agency of engaged community activists. This paper draws on qualitative evaluations from two community mobilisation GBV prevention programmes: the Gender Violence in the Amazon of Peru (GAP) Project and the Indashyikirwa programme in Rwanda. In Peru, participatory data was collected, in addition to baseline and endline interviews with 8 activists. In Rwanda, baseline and endline interviews and observations were conducted with 12 activists, and interviews were conducted with 8 staff members. The data was thematically analysed, and a comparative case study approach was applied to both data sets. The comparative study identified similar programmatic aspects that could hinder or enable activist’s engagement and development, and how these are embedded within contextual social and structural factors. We discuss these insights in reference to the current emphasis in public health on individualistic programming, with insufficient attention to how wider environments influence violence prevention programming.

KEYWORDS:

• Community activism
• intimate partner violence
• agency

Introduction

Community mobilisation is recognised as a promising approach to transform harmful attitudes, beliefs and norms underlying gender-based violence (GBV) and gender inequality (UN Women & WHO, 2020). These interventions typically engage volunteer ‘community activists’ who live or work locally and are trained and supported to engage with other residents in informal conversations and awareness-raising about violence (Maton, 2008). Community mobilisation approaches also work with key opinion leaders (Stern, 2021), and include efforts to hold government or institutions accountable, with communities coming together to pressure changes or ensure responsiveness to policies (Fox, 2015). Community mobilisation programmes have significantly reduced GBV within programmatic time frames of 3–5 years (Abramsky et al., 2014; Kerr-Wilson et al., 2020). Other evaluations of community mobilisation interventions have found no reductions in violence within similar time frames (Chatterji et al., 2020; Hatcher et al., 2019; Pettifor et al., 2018) suggesting that implementation and context are crucial for achieving such impacts. Indeed, community mobilisation interventions are highly shaped by their context, and appropriate mobilisation strategies vary across settings (Campbell & Cornish, 2010). Yet, there is limited documentation of the processes of change and the social and contextual factors that facilitate successful community mobilisation in various settings.

Evaluations of community mobilisation programmes have predominantly focused on the impact of activism on community members, with limited attention to the experiences of engaged activists. Yet the perspectives of activists are fundamental to understanding processes of social change as they witness first-hand the programmatic impacts for community members. Furthermore, meaningful participation and selection of activists can shape the long-term successes or failures of community mobilisation interventions (Minckas et al., 2020). Evaluations of mobilisation programmes have identified the importance of engaging staff and activists who embrace intervention principles and who are supported throughout their work to do so (Gevers et al., 2018; Jewkes et al., 2019). In addition, a lack of institutional support for staff and activist volunteers (Hatcher et al., 2019) and economic pressures on activists has been found to negatively impact community mobilisation interventions (Hargreaves et al., 2010). While promoting volunteerism can yield promising results for activism (Gevers et al., 2018; Minckas et al., 2020), there is limited understanding of how to best integrate volunteerism among activists (Michau, 2012). Some form of economic support for activists is arguably needed to sustain GBV prevention activities, especially in low-income settings (Mannell & Dadswell, 2017). Understanding which individuals are recruited to be activists, their motivations for initial and sustained engagement, and how they are trained and supported warrants more attention.

This paper aims to contribute to these gaps through comparing process evaluation data of programmes in Peru and Rwanda that recruited and supported activists to engage in community mobilisation activities to prevent GBV. The findings are structured according to Campbell and Cornish’s (2010) conceptual framework for community mobilisation, which argues that success or failure of community mobilisation interventions depends on how they interact with symbolic, material and relational dimensions of their social environment. Focusing on different aspects of the social environment is important given the current emphasis in public health on trials of individualistic programming, and limited awareness of how wider environments can hinder or enable violence prevention programming (Campbell & Cornish, 2010). We use the term agency throughout this paper to refer to the ‘socio-culturally mediated capacity to act’ (Ahearn, 2001) and approach agentic actions as context-specific (i.e. an action that challenges social structures in one setting may be different in another setting) (Campbell & Mannell, 2016; Markus & Kitayama, 2003). This is opposed to psychological approaches to agency that focus on rational choices of available options made by individuals from different social locations (Gillespie, 2012). As a reflection of this approach, this paper prioritises the perspectives of activists to document their development and engagement with the respective mobilisation programmes.

Interventions and settings

Gender violence in the Amazon of Peru Project

The Gender Violence in the Amazon of Peru (GAP) project was a community health worker-led mobilisation pilot for the prevention of GBV in Lower Napo River (LNR) communities, facilitated by the University College London and DB Peru, a local non-profit organisation in health and community development (Shannon et al., 2019). The project was implemented from November 2017 until December 2018 with funding from the Sexual Violence Research Initiative and the World Bank Group. The LNR region represents a group of 25 communities comprising approximately 5,000 inhabitants, in the Loreto Department of Peru. Many families lead a predominantly subsistence, agricultural life, reside in households with no access to running water or toilets, with most community members having only completed primary education (Murray, 2014). The LNR communities experience a high burden of GBV; a lifetime prevalence of 61% exists in rural Peru (WHO, 2005) and recent figures released by the organisation Promsex found that 79% of women between the ages of 18 and 29 in the Amazonian town of Mazan reported having ever experienced sexual violence (Chacaltana, 2006). Women living in the LNR have similar rates of violence to those living in Indigenous communities elsewhere, far exceeding those found in the rest of Peru (National Institute of Statistics and Informatics, 2020). This has been attributed to broader structural inequalities and systemic discrimination against Indigenous populations in Peru (Shannon et al., 2017). The Peruvian government has established a network of 73 women’s emergency centres (Centros Emergencia Mujer, CEM), which are intended to provide psychosocial, medical and legal support for women experiencing violence. However, CEMs are primarily located in urban or highly populated areas and are largely inaccessible to women living in Indigenous or remote communities, including the LNR region (Agüero, 2018).

The GAP Project used a Participatory Community-led Intervention Development (PCID) approach to work collaboratively with local promotores (community health workers) to identify risk factors for GBV and develop a set of targeted prevention activities for their communities (Mannell et al., 2021). The GAP project identified eight promotores, all with experience in community leadership and/or health. While women were encouraged to participate, the majority of engaged promotores were men because women felt they could not commit the required time to attend the required training due to other responsibilities. These promotores attended a series of workshops facilitated by the GAP Project team where they engaged in concept mapping, role play, and project management techniques to set priorities, design prevention activities, and identify resources required for implementation. Over a six-month period, the promotores implemented 57 prevention activities across 10 communities, including community meetings, household visits, hosting local government officials, local advocacy, developing posters and visual messaging, and creative engagement such as participatory theatre and film screenings. The promotores were unpaid volunteers but were reimbursed for all expenses incurred. A final series of workshops was held at the end of six months to conduct a participatory evaluation of the activities that had been implemented. All activities developed in partnership with the promotores were later published as a set of guidelines for violence prevention in the Amazon (Shannon et al., 2017). This was consistent with the participatory action research methodology that underpinned the PCID approach; an iterative process aimed at equitably involving community members and researchers whereby partners equally contribute expertise and share ownership of the process.

Indashyikirwa programme in Rwanda

Indashyikirwa ‘Agents of Change’ is an intimate partner violence (IPV) prevention programme that was implemented by CARE International Rwanda, Rwanda Women’s Network (RWN), and Rwanda Men’s Resource Centre (RWAMREC), funded by the UK Department for International Development (DFID) and evaluated as part of the What Works to Prevent Violence against Women and Girls Programme. The programme ran from August 2014 to August 2018 across 14 sectors in Eastern, Northern, and Western provinces of Rwanda, in predominantly rural, widely spread communities. Over the past few decades, there have been a significant growth of policies, laws, and programmes supporting women’s rights, and the Rwandan government has facilitated various initiatives to prevent GBV at a national and community level (Umubyeyi et al., 2016). Yet, GBV, including IPV, remains common, with an estimated one out of five women reporting having experienced physical or sexual violence by a partner in the past 12 months, and two out of five women having ever experienced emotional, physical, and/or sexual violence by a partner (National Institute of Statistics Rwanda [NISR], 2016). Indashyikirwa was comprised of four interlocking components: a couples curriculum; community activism with a sub-set of trained couples; opinion leader training and engagement, and women’s safe spaces. The couples curriculum was a five-month participatory training designed to help couples positively transform power imbalances in relationships, identify and manage triggers of IPV, and build skills for equitable, nonviolent relationships (Stern & Nyiratunga, 2017).

The community mobilisation component, inspired by the SASA! programme established by Raising Voices in Uganda, was designed to extend the relationship-level prevention of IPV fostered through the couples curriculum to the community level. Four hundred and twenty partners of couples that completed the couples curriculum carried out community-based activism for the remaining 19 months of the programme. RWAMREC staff offered activists a 10-day training in activism skills and coordinated monthly meetings and refresher trainings to provide ongoing support to activists. More trained couples than the programme had budgeted for expressed their interest to be activists, and the programme encouraged all activists to involve their spouses in mobilisation activities. Although the activist role was a voluntary position, activists were re-imbursed for transport costs. Criteria to be activists was basic literacy to be able to read the activism materials, commitment and availability to conduct at least three mobilisation activities per month. In 2017, the activist training was offered to an additional 80 partners of trained couples who had shown ongoing dedication to the programme, to widen the available pool of activists. Various mobilisation activities were employed including use of posters, quick chats, community conversations, theatre, women’s safe space discussions, household visits and advocacy with local authorities, reaching over 600,000 individuals (Stern et al., 2018).

These two case studies represent different ways to catalyse change to prevent GBV through community mobilisation. The programme in Peru relied on participatory development and community ownership, whereas the programme in Rwanda was more structured and instrumental to a wider programme model. The programme in Peru was shorter and implemented at a much smaller scale compared to the programme in Rwanda. This comparative case study provides an opportunity to assess variation in the engagement and agency of activists during distinct community mobilisation processes for GBV prevention.

Methods

Data collection in Peru

A participatory evaluation of the intervention activities designed by the promotores was completed as part of the PCID approach, which included the promotores’ individualised project planners (a 15-page activity book used to define, develop and monitor activities within each community), the promotores’ written reflections on project activities, drawings (e.g. community and risk factor mapping), and annual timelines of activities for each community. Additional data were collected to inform a process evaluation of the PCID pilot, including audio recordings of ten days of promotor leadership training, and 16 baseline and endline in-depth, semi-structured interviews with the eight promotores. The baseline and endline interviews were structured around project expectations, potential barriers and facilitators of GBV prevention, and how these expectations changed over the course of the project. The interviews were conducted by two qualitative female researchers, including the third author, in Spanish. Apart from one promotora (P1), all the interviewed promotores were men.

Data collection in Rwanda

The data collection for Rwanda consisted of two rounds of qualitative interviews and observations with a sub-set of Indashyikirwa activists. These interviews and observations took place in three intervention sectors (Rurembo Sector, Western Province; Gishari Sector, Eastern Province; and Gacaca Sector, Northern Province), which were purposefully selected to represent rural and peri-urban locations. After having completed the training and started the mobilisation activities, twelve partners of couples (two men and two women per sector), elected to be activists, were interviewed in November 2016 (baseline). They were recruited through RWAMREC staff and were asked about their impressions of the activism training, what motivated them to continue as activists, what they had been doing recently as activists, and whether they had faced any challenges. All activists were observed facilitating various mobilisation activities. In May 2018, the same activists were interviewed (endline) to document their progression as activists, towards the end of the programme. Three activists (one per sector) were observed facilitating mobilisation activities. For the observations, a researcher took notes on engagement and comprehension by participants, and facilitator skills.

To triangulate the activists’ perspectives with staff, interviews were conducted with 8 RWAMREC field officers and supervisors in May 2017 across all intervention sectors. Staff were asked to describe key successes and challenges of the community activism component. One female Rwandan qualitative researcher conducted both rounds of observations and interviews with activists, which were conducted in Kinyarwanda. The first author conducted the interviews with RWAMREC staff in English.

Ethical considerations

Ethical approval for the GAP evaluation was obtained from the Institutional Review Board of Universidad Peruana Cayetano Heredia in Peru (no. 102178), the DB Peru Board of Directors, and the Research Ethics Committee of University College London in the UK (no.5406.004). In November 2017, community leaders from all 25 LNR communities were invited to a three-day workshop to introduce the project and design local ethical guidelines for community engagement as part of a commitment to establishing a situated approach to ethics (Mannell & Guta, 2017). Promotores gave informed written consent to participate at the start of the project and this consent was discussed at key time intervals as the project progressed.

Ethical approval to undertake the Indashyikirwa evaluation was obtained from the Rwandan National Ethics Committee (RNEC) (REF: 340/RNEC/2015) and the National Institute of Statistics Rwanda (REF:0738/2015/10/ NISR). Secondary approval was also obtained from the South African Medical Research Council (REF: EC033-10/2015) and from the London School of Hygiene and Tropical Medicine. Before each interview, participants were given a chance to ask questions and to provide written or oral informed consent. All participants were given 2000 Rwandan Francs (approximately USD 2.50) as a token of appreciation for their participation in the study. All data has remained confidential and been anonymised for both studies, using codes alongside quotations. Demographic information of activists interviewed corresponding to the codes used can be found in Table 1.

Table 1. Coding information of Rwandan community activists interviewed.

ID	Gender	Province
MA01E	Male	Eastern Province
FA01E	Female	Eastern Province
MA02E	Male	Eastern Province
FA02E	Female	Eastern Province
MA01N	Male	Northern Province
FA01N	Female	Northern Province
MA02N	Male	Northern Province
FA02N	Female	Northern Province
MA01W	Male	Western Province
FA01W	Female	Western Province
MA02W	Male	Western Province
FA02W	Female	Western Province

Data analysis

All interviews were audio recorded, translated and transcribed into English. The transcripts were then thematically analysed by the first author (Rwanda) and third author (Peru) using a thematic coding network, with the support of qualitative analysis software. For the data from Rwanda, the first author debriefed with the qualitative researchers after data collection to capture their initial impressions, non-verbal and contextual insights. These research summaries were used to inform the analysis.

For both studies, thematic analysis was conducted to uncover predominant themes and provide a rich, detailed and holistic account of the data (Braun & Clarke, 2006). In writing this paper, the authors shared coded data for the respective datasets focused on activists’ engagement with the mobilisation activities, including enablers and barriers, and carried out comparative analysis of both datasets. The joint analysis was guided by a comparative case study approach, which justifies the importance of examining processes of sensemaking, shared purposes or identities as they develop over time, in distinct settings, regarding a similar phenomenon (Bartlett & Vavrus, 2017). We applied Campbell and Cornish’s (2010) framework that critically centres the social context to appreciate the environmental barriers and enablers for activists engaging with community mobilisation interventions.

Findings

To ease the comparative case analysis across both studies, we use the term ‘activist’ throughout the findings section. We note that in Peru, activists were identified as promotores, and in Rwanda, activists were identified as community activists or agents of change.

Material context

Two aspects of the material context are esepecially relevant to community mobilisation; the first is resource-based agency, and the second is experienced-based agency (Campbell & Cornish, 2012). Resource-based agency refers to the extent to which individuals have access to the resources needed to live healthy lives free from violence, including money, food and paid work (Mannell & Dadswell, 2017). This also includes funding for mobilisation interventions and compensation for tasks or responsibilities related to GBV prevention, which highlights the potential burden of voluntary positions for activists and constraints on the sustainability of activities, such as compensation for time and transport costs (Mannell & Dadswell, 2017). Experienced-based agency refers to concrete opportunities for engaged participants (including activists) to put learned skills or knowledge into practice (Campbell & Cornish, 2012). Small-scale successes can provide an experiential basis for more ambitious future action. For instance, in settings where women lack opportunities to exercise experienced-based agency, projects may begin to provide these, through facilitating their engagement in activities (Campbell & Cornish, 2012).

Activists in both settings shared a strong sense of commitment to their volunteer efforts as activists, in part, because of their own experiences of violence and personal transformation through the projects. As one activist in Peru emphasised: ‘I am doing this because I was exposed to family violence when I was young, and my parents were together. I lived this in my own flesh. That is why I started this project’ (P8, Endline). In Rwanda, a few female activists and staff members noted that women tended to be particularly committed as activists, given their own experiences of violence: ‘I think the women feel it’s really important for them since a lot of women are facing the same problems. They are more dedicated than men because it is a cause that helps the most’ (RWAMREC Field Supervisor, Western Province). The majority of activists in Rwanda emphasised that their dedication to activism was driven by having personally benefitted from the couples curriculum, including learning and applying skills to support quality, non-violent relationships. In Rwanda, more participants than budgeted for wanted to engage in activism, and many noted their pride at having been selected by their peers to continue in this work: ‘People perceive me as a woman who is an agent of change who wishes the community to be in good harmony, particularly couples’ (FA01N Endline). Yet, many activists in both settings identified competing financial and family responsibilities that could hinder their active engagement. One activist from Peru revealed: ‘A lot of times we have to leave our family, our kids at home and we come for a two- or three-day training. We leave a void. That is the effort that we put into it’ (P1, Baseline). Another activist from Rwanda explained how her financial circumstances limited her ability to regularly engage:

The challenge is the life condition. For instance, here at the cell office, most of the times there is an opportunity to find people to give a discussion to, but because of the poverty, we are always working, and we do not get time to catch that opportunity and then people don’t get a chance to benefit. (FA01N Baseline)

A few staff members in Rwanda reflected on the challenges that volunteering for activism can impose:

The activists are volunteers which imposes limits. It may become too much for them although they are doing a great job. They have other obligations; they need to look for money and look after their family. (RWAMREC Field Officer 1, Western Province)

Many activists in Rwanda were asked by community members or local leaders to facilitate mobilisation activities or offer support in areas they did not live. Although the Indashyikirwa programme provided activists with a monthly stipend to cover transport costs, some activists related that this amount was insufficient. One activist in Rwanda reported creating a mutual aid fund where each activist contributed a small amount of money that could be drawn on to meet individual financial needs, including to respond to activism opportunities.

Across both studies, activists noted that the community perception of being paid for their work, rather than operating as volunteers, risked undermining their legitimacy as activists. As one activist in Rwanda said: ‘There are some people who are like, come on, let us leave her! She is working for money and not us!’ (FA01E Baseline). Similarly, one activist in Peru remarked: ‘as an activist, we have no financial incentives, we don’t gain anything more than knowledge and books from the training you give us … this helps us to perform our role in our communities’ (P1, Baseline). In some communities in Peru, rumours circulated that activists gained money from their participation, which caused community members to question their authenticity:

Lately, I felt challenged by community members and the comments grew, saying that I am getting paid, that I have received a big mobile phone, that it belongs to the town and I should give it to the community, that I’m given medicine and I don’t give the medicine to the people. (P1, Endline)

Experience-based agency draws attention to how the programmes supported the skills and abilities of activists to engage in activism. In Rwanda, activists appreciated the programme-branded gear for increasing their confidence and pride, helping to see themselves and be seen as activists: ‘They gave us raincoats and vests which identifies us. Even when it is sunny, you wear it, and people recognise you; that is a change agent’ (FA02W Endline). In both settings, activists identified the importance of the programmes equipping them with knowledge and skills to effectively engage in activism. As an activist in Peru emphasized: ‘Thanks to this project I feel strengthened, but I do not receive money. But in this project, I got to know many of my values, so that many can come from behind in future’ (P5, Baseline). After having completed the intensive couples curriculum in Rwanda, activists felt more equipped to ‘practice what they preach’. Many activists in Rwanda initially lacked the confidence to facilitate mobilisation activities using participatory approaches and/or to speak publicly, which was often a novel experience. Yet, there was consensus on behalf of staff and activists that the skills and abilities of activists greatly improved over the course of the programme as a result of the ongoing trainings, support from staff, and with time and practice. One activist related how reflecting on her power within through the couples curriculum bolstered her confidence to speak publicly:

When we understood the power a woman has in herself, I liked it. As we were conducting discussions, I became more and more confident. Before, I could not speak in public, I could not conduct discussions. Now, I became confident, I conduct discussions and facilitate activism. (FA01W Endline)

Some couples noted that when they co-facilitated mobilisation activities or when one partner facilitated and the other sat in the audience, they could helpfully draw on each other for preparation and feedback.

An important way to foster the experience-based agency of activists identified across settings was supporting activists to deal with community resistance. One activist in Peru highlighted the importance of critical reflection to do so:

There are also a lot of people that don’t like what I do because they want to continue their [violent behaviour], but this to me is not a barrier because I am going to overcome this difficulty with suffering and discouragement, but in the discouragement, I will learn to find success, because that discouragement comes with a little reflection that I have. And that allows me to hold myself forward. (P2, Endline)

In Rwanda, the couples curriculum and activism training offered tips for activists to mitigate negative responses from community members, including opinion leaders. Programme supervisors met with activists on a monthly basis, and these meetings were identified as critical opportunities for activists to receive advice and support to mitigate challenges, including community backlash or resistance.

Symbolic context

The symbolic context includes social norms, cultural meanings and understandings associated with GBV and prevention activities (Campbell & Cornish, 2012), including how community members feel they should respond to GBV and perceived pathways for prevention (Mannell & Dadswell, 2017). The symbolic context also includes how different social groups are valued and respected. For instance, within a patriarchal society, ideologies about gender can limit how women and men understand the potential of women to make significant decisions or take leadership roles (Campbell & Cornish, 2012). Thus, promoting recognition of an individual or group’s worth, dignity, and legal rights can be critical for achieving change at the symbolic level.

Across both settings, activists identified the challenge of being perceived by community members as transgressing traditional gender norms as activists. Men, in particular, spoke about how they could be considered weak for engaging with work focused on preventing GBV. As a male activist in Rwanda noted: ‘Others think that we are making their women disrespect them; that we want them to be inganzwa (dominated men)’ (MA01W Baseline). Several female activists in Rwanda reported the negative perception of speaking publicly, including being called arrogant or bewitched:

Some men say, “Where did you get that knowledge from? Can our wives also be able to stand in front of people and conduct a discussion like that?” They look at me differently wondering where I got that knowledge from. They were astonished. (FA01N Baseline)

In Peru, a recurrent issue was the norm of household privacy, including of domestic violence, and the resistance to discuss these issues in public. As one activist said: ‘I am going to face those chauvinistic men, that are authoritarian in their homes. For example, some people do and undo at their house, and you can’t interfere in their life. That is a barrier that we will have to face’ (P3, Endline).

Activists identified strategies to mitigate such resistance from community members and challenge harmful gender norms. In Rwanda, the collective support from other activists bolstered activists’ capacities to resist negative social sanctions and model alternative, equitable norms. For instance, several male activists in Rwanda took the initiative of carrying babies on their backs, as a deliberate challenge to gender norms and as a way to spark community conversations. Several activists in Peru identified the importance of empathetic and respectful communication with community members, given the entrenched nature of gender norms:

We need to know how to talk to the person. For example, if I talk to this type of people, I

don’t reach out to them as an authority, but I talk to them as an elderly person … so that

they see me as a normal person. I empathise with him; I get in his shoes, that’s when the

other person is going to become aware and able to differentiate right from wrong. (P3, Endline)

The importance of activists showing humility, empathy and using participatory approaches was also identified in Rwanda, although this could be challenging. One field officer mentioned the need to continually emphasise the importance of activists respectfully engaging with community members and probing for critical dialogue:

Their probing skills are still low. That’s why it’s good we keep on visiting them and help them to probe. They also need to show more respect to community members. They tend to show that they know much, and it doesn’t help. Being humble would help more or valuing the community ideas would help. (RWAMREC Field Officer 1, Northern Province)

Activists in Peru and Rwanda identified the importance of safe spaces for community members to critically reflect on gender norms related to violence. Through the participatory process of identifying and addressing risk factors for GBV in Peru, activists and the wider community became increasingly aware of structural factors associated with GBV and were subsequently able to shape relevant primary prevention strategies. The education component drew attention to the gender inequalities that shape and define intimate relationships between men and women in LNR communities, and how violence may arise from these inequalities. Several activists and staff in Rwanda frequently identified the programme materials and visual aids as helpful tools for facilitating critical dialogue in more participatory, inspiring and engaging ways:

The tools we use are helpful because they show the consequence of IPV; like the image of this is what it feels like when you yell at your wife. Do you want her to feel this way; the wife you love? We have pictures how a happy family looks. It’s easy for them to realise how a family will end up if they behave positively. (RWAMREC Field Officer 1, Western Province)

Relational context

The relational context draws attention to the role that connections play in effective and sustainable GBV prevention activities (Mannell & Dadswell, 2017). Such connections include a project’s ability to mobilise and build bonding social capital (strong within-community support for project goals) and bridging social capital (supportive relationships between communities and agencies with the political or economic power to assist them in achieving their goals) (Campbell & Cornish, 2012).

Activists engaged with the Indashyikirwa programme had a rich source of bonding capital as the couples curriculum fostered rapport and led to a large network of activists: ‘Every village has at least four activists and we have time to meet and we exchange the challenges we have faced. If there is someone having a better idea about a certain challenge, we help each other’ (MA02E Baseline). Moreover, many activists were accompanied by their spouses, either as co-facilitators or attendees, which could help increase their legitimacy: ‘The people were more interested to follow the discussion because they could see that my husband also supports me and that his mindset is changing’ (FA02W Baseline). In Peru, each activist worked independently, whereby bonding social capital was more focused on building relationships between activists and community members. Some activists in Peru identified the challenge of building trust and credibility with community members: ‘Not everyone in the houses that I’m going to visit will see me with a good face. Because a lot of people are not aware, they say: “what does this promotor know?” I know more things than he does’ (P1, Baseline). Some male activists in Rwanda faced a similar challenge of distrust from community members, if they had been known to use violence or abuse alcohol prior to becoming activists. Activists demonstrating personal changes after completing the couples curriculum could gradually help overcome this challenge. As one staff member noted:

At the beginning, people didn’t believe in the activists because they knew them before to be in conflict. They were challenging them and accusing them of wanting them to change even though they haven’t. They would say they cannot trust someone who was in conflict and changed in 5 months. But now they saw a change in their families, so they trust the activists. (RWAMREC Field Supervisor, Eastern Province)

Another staff member in Rwanda commented on the importance of offering ongoing support to activists, to hold them accountable to changes fostered through the couples curriculum. Being held accountable to the programme values and messages could affect the participation of activists: ‘When we ask them why they stopped to work as an activist, they said people were challenging them for bad behaviours’ (RWAMREC Field Supervisor, Eastern Province).

In both settings, bridging social capital included fostering connections with local leadership, governments, and community organisations, which could help legitimise activists and secure access to activism opportunities. In Rwanda, programme staff met with all village leaders where Indashyikirwa was implemented before the initiation of activism, and some activists noted how critical this was: ‘After training those leaders, it made us be known and the leaders started to welcome us in meetings and groups. We would have had problems if the project trained only the change agents and they didn’t train the local government leaders’ (FA02W Endline). The programme also established quarterly meetings with opinion leaders, community activists and women’s safe space facilitators, which were highly valued by activists to support coordination. Some staff members related how important fostering links with local leaders was for activists’ sense of pride:

Some of them have told me when they are with local authorities dealing with other communities, they are happy. They feel they are a crucial man or woman in the community. Now they are requested by local leaders to support them for a family living with conflict. For them it is a privilege. (RWAMREC Field Officer 1, Eastern Province)

In Peru, the GAP team spent a substantial amount of time engaging with stakeholders beyond the community, including government actors in regional centres. One activist emphasised the importance of securing partnerships with directors of targeted organisations. Forming alliances with local leadership structures could help activists negotiate local resistance. The GAP project also harnessed the capital of external partners such as the NGO, DB Peru, which helped legitimise their efforts:

We went community by community presenting the project. We want to work. The community has been very interested in the issue because we have said that it is the first organisation of DB Peru that is interested in the issue of violence in the lower Napo. (P6, Endline)

Another important component of bridging social capital identified was ensuring referral links to survivor services, especially as mobilisation activities raise community awareness of violence. In Rwanda, activists were encouraged to refer community members to the project women’s safe spaces, although in some cases, community members preferred direct support from activists after having built trust and rapport. Several staff members expressed concerns that activists were not sufficiently trained in counselling to respond to such demands, including from local leaders:

People in the community are asking activists for counselling. In the community they are treated like mediators and counsellors, but they are just there to give information. Even local leaders ask them to resolve problems. It’s a challenge because what we are targeting for activism is to give information and not to resolve couples disputes. They need professional counselling training because we cannot avoid them helping other couples. (RWAMREC Field Supervisor, Eastern Province)

To mitigate this challenge, activists were offered a training in psycho-social support skills, and programme staff reminded opinion leaders of the voluntary and complementary role of activists. If they were given permission to intervene on survivors’ behalf, activists in Peru engaged directly with community leaders in response to cases of violence to develop strategies to support survivors.

Discussion

This study sheds light on the material, symbolic and relational contextual factors that enabled or inhibited the participation and agency of community mobilisation activists. Regarding resource-based agency at the material level, activists noted the importance of being committed as volunteers, and that being perceived as paid for their efforts could undermine their legitimacy as activists. In Rwanda, activists’ dedication as volunteers was fuelled by the intensive and benefits-based couples curriculum and being selected as activists through a competitive process. In Peru, activists’ motivation for participation often came through a desire to receive the leadership training offered by the project in combination with a commitment to address violence because of personal experiences. Yet, in both settings, it could be challenging for activists to commit as volunteers because of their financial needs, and given the potential costs to facilitate mobilisation activities, such as transport costs.

Findings related to the material context highlight important challenges for GBV prevention programming, particularly regarding the sustainability of such programmes in contexts with limited resources. The participatory mobilisation process underpinning both programmes suggests the possibilities of social change for women and/or couples experiencing violence, which may only instigate activities while programme resources are in place. This points to a persistent tension in GBV prevention programming (and development programming more broadly) around the potential harms of encouraging participation in time-limited interventions (Cooke & Kothari, 2001), and highlights the need for GBV prevention programming to involve communities in co-designing programmes that reflect local priorities and existing resources as a means of ensuring more sustainable interventions (Mannell et al., 2021).

Regarding experience-based agency, activists across both settings identified the importance of knowledge and skills training as a platform to confidently facilitate mobilisation activities. In Rwanda, it was helpful for partners to have their spouse present when facilitating activism, for support and feedback. Activists and staff in Rwanda identified the importance of having regular meetings with and observations by staff, to improve new skills such as public speaking and applying participatory methodologies. In both settings, involvement in the programme as activists could be a source of empowerment, which speaks to the importance of striving for inclusive recruitment of activists. In Peru, women had limited time to commit to the intensive training required, which hindered equitable inclusion of men and women. This likely reduced the potential for the GAP programme to achieve broader social changes because it limited the extent to which women’s perspectives on violence became part of participatory activities, while increasing the authority of men as ‘experts’ on GBV within their communities. As recognised by Flood (2015), exclusively involving men in GBV prevention interventions can reinforce men’s power and authority within communities as they benefit from programme activities intended to build local capacity. This finding also highlights the tension of participatory approaches whereby community-led programming can inadvertently re-produce existing power dynamics or hinder inclusivity. Across both settings, activists identified the importance of being supported to deal with resistance and backlash from community members, including to critically reflect on dominant norms and values, and through support from staff or other activists.

At the symbolic level, activists related difficulties in challenging gender norms that condone GBV or relegate it to the private sphere. To gain traction, activists identified the importance of being humble, respectful, and sensitive in their discussions around such entrenched norms. In both settings, activists were able to model alternative norms and behaviours through their own actions. For instance, in Rwanda, men carrying babies on their backs, and women speaking publicly, was identified to help challenge hegemonic gender norms. This supports other literature suggesting that the time spent by women engaging with a community mobilisation intervention can produce benefits, not only for themselves, but for other women in the community through social norm change (Abramsky et al., 2014; Gram et al., 2021). A few staff members and activists in Rwanda identified how focusing on the benefits of non-violence was helpful to counter resistance, as has been documented elsewhere (Michau, 2012). Across both settings, applying participatory approaches to allow for critical reflection on norms that drive violence was able to help mitigate resistance and foster alternative norms. This aligns with the evidence suggesting that because violence is normalised in many settings, communities’ reactions to prescriptive or top-down messages can backlash into defensiveness, confusion, or rejection (Michau, 2012; Minckas et al., 2020).

At the relational level, activists in Rwanda had increased bonding social capital after having completed the couples curriculum together and the programme recruited at least a few activists per village. Activists in both settings related the importance of taking time to build relationships and foster trust with community members, which supports literature emphasising the importance of meaningful engagement and trust among organisations, activists and target communities to create bonding social capital (Campbell, 2014). Trust of activists could be fostered if they were perceived to ‘practice what they preach’, which could be undermined if activists were known to have perpetrated or experienced violence. In Rwanda, activists related the value of the couples curriculum, and ongoing support from staff, to equip and hold them accountable to personal and relationship changes. To foster bridging social capital, activists in both settings identified the importance of regular and purposeful collaboration with leaders, government, and other influential stakeholders. This helped activists gain legitimacy, negotiate local resistance, provide access to activism opportunities, and as a source of empowerment. Another identified aspect of bridging social capital was to establish referral links to GBV response services.

Mannell and Dadswell (2017) built on the conceptual framework applied to this study by adding institutional dimensions of the social environment, such as government policy and law, education, and health. This additional level emphasises that community mobilisation should allow communities to challenge broader social and institutional structures that can undermine collective efforts to prevent IPV. Yet, many community mobilisation programmes do not provide sufficient space for community members to develop collective agency to challenge structural drivers of violence (Minckas et al., 2020; Stern, 2021). Interventions that address laws and policies are often separate from those that target the community level with different intervention models being used for different purposes (Mannell & Dadswell, 2017). A recent review of 10 programmes to prevent GBV using community mobilisation similarly found less evidence of work at the institutional level, partly explained by the challenges of working at this level given the small size of the organisations or focused scope of the projects (Stern, 2021). The findings in this study further support this challenge as there was limited data around activists’ engagement at the institutional level.

Limitations

As with all secondary analyses, the analysis presented here has some limitations. The comparison between the studies was conducted as a post hoc analysis of data from their respective impact evaluations and was not originally collected for comparative purposes. Yet the description of the programmes and contexts, analysis of detailed quotations, and triangulation of the case studies allowed for a robust secondary analysis of qualitative data. We attempted to mitigate social desirability bias on behalf of activists by using field researchers who identified themselves as external to the programme and emphasising the confidentiality of their answers, and in the case of Rwanda, triangulating the perspectives of activists with staff and observations of activists. In addition, the authors aimed to be reflexive of their positionality to the data and understanding of the contexts, through ongoing discussions with project team members in both contexts.

Implications

The framework for community mobilisation according to symbolic, material, and relational contexts (Campbell & Cornish, 2010) was pertinent for comparing the findings across two different community mobilisation programmes. This framework is helpful for planning and evaluating community mobilisation interventions, to consider how the engagement and agency of community activists may be affected by their social environment. The findings emphasise the critical role that activists play in the success of community mobilisation efforts, and the importance of careful and inclusive recruitment, offering sufficient training that responds to capacity gaps (i.e. facilitation skills, psycho-social counselling) and regular support. The intensive couples curriculum in Rwanda was especially appreciated for allowing couples to foster change in their relationships before encouraging others to do so. The findings from Rwanda also emphasise the value of couples activism, for fostering legitimacy, capacity and support of activists. There have been valid, historic concerns raised around the appropriateness of working with couples as part of violence prevention programming including that they condone reconciliation rather than empowering women to leave violent or abusive relationships or implying that both partners share responsibility for the violence (Armenti & Babcock, 2016). Yet, if managed responsibly, working with couples can reduce the frequency and severity of abuse in relationships (Dunkle et al., 2020). The findings from Rwanda further complements this evidence, although the strengths and challenges of working with couples as activists should be explored further.

The comparative findings speak to the importance of programmes supporting both resource and experience-based agency for activists. It is important to ensure that activists have their material needs met to carry out their work and are not financially burdened or hindered from engagement. Yet, there are various considerations and trade-offs around whether or not activists are compensated, and programmes should consider these in their contexts. Within their unique settings, programmes should identify potential barriers (including material barriers) to activists engaging with programmes. The findings also emphasise the importance of fostering regular opportunities for exchange and collaboration among activists and key stakeholders, including local leaders and civil society organisations. Supporting activists to compel change at the institutional level may require more attention to bridging social capital, including establishing strategic partnerships with other organisations, and/or capacity building in this domain.

Conclusion

Community activism is a promising and popular strategy to prevent GBV. More attention should be given to the environmental enablers and challenges underlying the agency and capacity of community activists, given their fundamental role to the success of such programmes. While community mobilisation plays out differently across specific contexts and communities, similar factors and best practices can be identified to support and enable activists’ engagement and development. This removes the onus from being entirely on the individual activist and importantly gives attention to the wider context where activists live and work. Finally, this study highlights the need for continued evaluation and learning within specific settings to deepen the understanding of the agency of activists engaged with community mobilisation efforts, and to explore gaps including activism with couples and how activists can support institutional level changes.

Acknowledgements

The evaluation of Indashyikirwa in Rwanda was conducted as part of the What Works to Prevent Violence? A Global Programme on Violence against Women and Girls (VAWG), which was funded by the Department for International Development (DFID) UK Government and managed by the South African Medical Research Council. The time spent by the first and fourth authors analysing the data and writing this paper was funded by Wellspring Philanthropic Fund. The authors are grateful for the input and support from CARE International Rwanda; RWARMEC; and RWN, and the participants who shared their experiences with us. The GAP Project in Peru was a collaborative partnership between University College London, DB Peru, and the local team of promotores. Funding was provided by the Sexual Violence Research Initiative and World Bank Group under the Innovations to Prevent Gender-based Violence scheme. The authors are deeply grateful for the participation of the LNR communities and leaders in the GAP Project, and for their commitment to addressing GBV.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The evaluation of Indashyikirwa in Rwanda was conducted as part of the What Works to Prevent Violence? A Global Programme on Violence against Women and Girls (VAWG), which was funded by the Department for International Development (DFID) UK Government and managed by the South African Medical Research Council. The time spent by the first and fourth authors analysing the data and writing this paper was funded by the Wellspring Philanthropic Fund. Funding was provided by the Sexual Violence Research Initiative (SVRI) and World Bank Group under the Innovations to Prevent Gender-based Violence scheme.