Thrive in Conversation: Creating an evidence-based theory of change for fatherhood engagement

Interview with presenters Dr Anjalee Kohli and Kate Doyle

A recent Thrive in Conversation webinar on fatherhood programmes– which are recognised as a promising approach to prevent violence against women and children – prompted many questions and comments from the audience. Here, we ask two of the webinar speakers, Dr Anjalee Kohli and Kate Doyle, some of the questions.

Anjalee and Kate were part of a team that performed an in-depth analysis of two existing interventions that have demonstrated reductions in intimate partner violence and violent discipline and are currently being scaled in rural settings in Rwanda and Uganda: the REAL (Responsible, Engaged, and Loving) Fathers programme in Uganda and the Bandebereho (‘role model’) programme in Rwanda.

Q: Can Bandebereho and REAL Fathers be adapted and replicated by researchers in other places/contexts?
A: Yes they can. Both Bandebereho and REAL Fathers are open source and curricula are available on the internet. However, part of what we hoped to contribute through the theory of change is the idea that people do not have to adapt only these two programmes to work with fathers to prevent violence against women and violence against children. The theory of change can be a guide for folks who are interested in developing new programmes as well. If organisations are interested in adapting existing curricula, we suggest they seek out these resources, which can support thoughtful adaptation to fit context and population:

Prevention Collaborative:
Adapting Curriculum-based Violence Prevention Programmes
Prevention Triad

SVRI and Equality Insights Lab:
ADAPT+ Framework
Online course on adapting IPV prevention programs

The Government of Tanzania, as an example, is planning to adapt the Bandebereho programme for delivery in Tanzania under the PAMOJA project.

Q: Why are only three out of 30 districts implementing Bandebereho, even though the programme started in 2019 in Rwanda? What were the bottlenecks for not scaling up?
A: Great question! The slow or gradual approach to scaling was a concerted choice that was made by Rwanda Men’s Resource Centre (RWAMREC), Equimundo and the Rwanda Biomedical Center. We knew that when scaling, there can be a big push to scale fast and reach many people, and at the same time to reduce the cost of an intervention. Research has shown how these pressures can lead to unsuccessful attempts to scale programmes and even to real harms, like increasing violence. Thus, we decided to approach scaling gradually, to be able to learn what it takes to adapt the programme from an NGO-delivered approach to one that is delivered by Government.

We have been doing that one district at a time – modifying certain aspects of the approach (e.g. training of community health workers), testing them, and then adapting them further based on research and practice-based knowledge. Each of the three districts has served as an important testing ground for refining and optimising the approach, while also building up Government capacity to take on the programme responsibilities – from training to implementation, supervision, monitoring and reporting (see for example Equipping community health workers in Rwanda to deliver a gender transformative parenting program to prevent violence against women and children at scale). The findings have informed the development of a national scale up strategy that can now guide future scaling in Rwanda.

Q: Can you say a bit more about the public testimony piece; what that was, how it played out, and why it was an important component?
A: For REAL Fathers, public testimony was part of the original design of the programme as we understood that fathers were influenced by what others in their community expected of them. Public testimony was designed for young fathers to express publicly, to others in the community, how they changed, what they were proud of, and the behaviours they planned to continue enacting. It was a way of building social accountability. The public testimony took place at the end of the project, when men who participated in REAL Fathers gathered in a large group with their wives. Other members of the community who were invited included local leaders and influential people. At times, the wives would testify on their own behaviour or their partner’s behaviour and how it changed their relationship. Community members and mentors would witness this, offering them an avenue to hold men accountable if they deviated, to come to them for advice, and to celebrate their achievements. By having their community publicly celebrate their achievements, we think it increased men’s investment in their commitments.

For the Bandebereho programme, public testimony came about in an organic way. During the piloting of the programme, men and women who participated sought out opportunities to speak publicly about the important changes they were making in their relationships. Community leaders also sometimes called upon Bandebereho participants to share what they had learned and how it had benefited them at community meetings. Over time, this grew into a more structured part of the programme, with the official integration of a closing ceremony, with space for public testimony, as the programme has scaled. The closing ceremony brings together local leaders, Bandebereho participants and their families, and other community members they wish to invite. Men (and women) appreciated the opportunity to reflect on where they have come from and where they are now, and the benefits that the changes they have made have had for their relationships and their family development and wellbeing.

Q: The community aspects of your programmes seem to be important determinants of the theory of change. What might happen in big cities in Latin America, for example, where the community links between participants is less strong?
A: Indeed, these programmes include community support and public testimony as change catalysts. The question raised is an important one. When adapting these programmes to new settings or designing/refining a new or existing intervention using the theory of change, it is important to understand who are the people and communities that influence participants and their behaviours. Neither REAL Fathers nor Bandebereho have been adapted to a big city in Latin America. Some formative work in the communities, including consultations with the participants and their influential groups, could help reveal who these communities are and how they can be engaged in the programme.

With REAL Fathers, we were able to do some adaptation work for a rural setting in West Bengal, India, in partnership with the Children in Need Institute. We used a participatory process, conducting conversations and assessments with fathers, their partners, and community members. We learned about the importance of parents and in-laws to fathers and mothers’ behaviours and had to consider how to engage them in the adaptation of the project.

We are familiar with a different project – Transforming Masculinities – that was designed to prevent gender-based violence, improve family planning conversations and uptake, and improve gender equality. The approach engaged faith leaders and congregations. The project was later adapted for use in the large urban capital city – Kinshasa – in the Democratic Republic of Congo. While it is not a fatherhood approach (and not part of the theory of change), we learned a lot about adapting the intervention to an urban setting. Much of the learning came post-project when we analysed what made the programme successful and what we could improve. While people in rural settings are often tightly networked across different faith and community groups, in the urban setting, faith communities are just one group that people belong to. Their close friends, peers, colleagues, family members, and other networks may not overlap with the congregation. Therefore, efforts to shift norms and behaviours in the urban setting may need to consider multiple social networks and design community support to resonate across these settings. This brief shares more information on what we learned.

There is a lot to learn about how to do this and whether these programmes are well-suited to an urban environment.

Q: Are there cultural differences in how girls and boys are treated during upbringing?
A: Across the globe, prevailing gender norms – the societal expectations of how girls and boys (and women and men) should behave – deeply influence parenting practices, including how parents and caregivers (and others around them) interact with and treat children depending on their sex or gender identity.

Research has shown that gender norms influence children’s opportunities and behaviour from an early age. Children learn how to view themselves and the world around them through their interactions with their parents and siblings, shaping their attitudes and behaviour. Parents may consciously or unconsciously transfer harmful gender norms to their children by raising boys and girls differently (this is called ‘gender socialisation’). These norms and gender stereotypes influence what parents think their children can or should do based on their sex – including how or where they play, their toys, the stories they are told, if they do household chores, whether or for how long they go to school, and at what age they marry. This can perpetuate gender inequality and power imbalances that limit children’s opportunities (e.g. girls’ access to education, boys’ opportunities for emotional connection) and also create risks for future violence. These norms can interact with other cultural practices and beliefs to shape the attitudes, behaviour, expectations, and opportunities for girls and boys.

See more in Parenting Programmes to Reduce Violence Against Children and Women: Why it is important.

Q: Do you think that cultural practices influence fatherhood and parenting styles in different regions? Do different cultures impact differently on ideas about how to bring up children, and about violence against children and women?
A: Certainly there are different cultural practices across the globe which influence fatherhood and parenting. However, by and large, unequal gender norms and power dynamics – which are a key risk factor for both violence against children and violence against women – are similar in many places.

Prevailing gender norms promote the idea that caregiving is women’s responsibility, while men’s role as a father is to be the provider or breadwinner. They promote men’s dominance in the family (often as the head of the household) and perpetuate the idea that violence is an acceptable way to ‘discipline’ women and children’s behaviour. Naming, discussing, and changing these deeply ingrained norms is a core aspect of gender transformative programming with parents, like Bandebereho and REAL Fathers. These approaches are rooted in the culture and community, where fathers and mothers consider their current context and behaviour, share their experiences, and propose new ways forward. When developing or adapting a programme to work with parents, you should also consider how these norms play out in your context (e.g. specific cultural practices) and how to integrate discussion of these within your programme.

Q: How were men/fathers selected for these programmes and were they men who expressed using intimate partner violence and harsh discipline?
A: Both programmes were designed as primary prevention interventions – meaning men were not chosen because they were known to use violence. However, in both Uganda and Rwanda, violence against women and violence against children are prevalent. So while men were not recruited because they used violence, many men were in fact using violence.

When we first piloted and evaluated REAL Fathers in Uganda (2013-2015 and 2016-2018), we included men who were fathers aged 16-25 parenting a child between 1-3 years old. Fathers were expected to be living with a partner, even if they were not formally married, living in the selected district, and express an interest and commitment to participating in the programme. In Senegal and Rwanda, we revised the recruitment criteria to include fathers who were parenting children 0-5 years, as we believed the programme would work well for fathers with children in this age group. We also adjusted the age criteria to reflect early fatherhood – 18-35 years in Senegal and 18-30 years in Rwanda.

For Bandebereho, men were chosen for the programme because they were fathers of young children (under five years) or were expecting a child, were partnered, and aged between 21-35 (in the pilot – the age range has expanded during scale up).

Both programmes take a positive and strengths-based approach to working with men. They encourage men to participate in a programme that will help them to achieve the goals they have for their families and to have more peaceful homes. This framing is important for reaching men and encouraging them to participate in the programmes.

Q: What was the age range of the fathers? Is acceptance potentially greater among younger demographics?
A: For REAL Fathers, we had different age ranges of fathers. In Uganda, it was 16-25 years. Later adaptations of REAL Fathers expanded this to 18-35 years to reflect when men more commonly had young children aged 0-5. The programme was designed to reach men who were younger and relatively newer to being a partner and a parent, as we expected that they may be more open to learning new information, skills, and behaviours. We have not used the programme with older men or unpartnered men. We expect that if it was to be adapted for these groups, some of the elements would need adaptation to address what motivates and influences their behaviours.

For the Bandebereho programme, men had to be between the ages of 21 and 35 years. The pilot programme also included separate interventions for younger, unpartnered men and for young women as well. We have found that younger men (and women) are often more open to change than older men and women. Reaching couples before they are married or have a child can also be an important opportunity to work with them to develop healthy relationship skills, co-parenting skills, and shared caregiving.

Q: Were there any incentives for fathers to attend mentoring sessions?
A: REAL Fathers did not provide an incentive for fathers to participate. For Bandebereho, participants in the pilot did receive a transportation stipend for participating in each session (2000 Rwandan francs or about US$2.50 at the time). While the stipend helped support strong attendance and retention in the pilot (men attended on average 14.1/15 sessions), we have found that attendance is still quite high even when incentives are not given. Now that the programme is being delivered through Government, no incentives are provided to participants, and attendance remains strong, with men attending on average 13.2/17 sessions (the curriculum was expanded).

Q: Are these programmes focusing exclusively on the nuclear family? It would be interesting to know if there’s potential to engage all caregivers (including relatives, paid support etc) or if that’s not really relevant to child outcomes.
A: These two programmes – given their focus on preventing men’s use of intimate partner violence against women and parents’ use of violence against children – do focus predominantly on the nuclear (heterosexual) family, where these particular dynamics play out. However, there is tremendous opportunity for engaging a much broader range of stakeholders and caregivers for improving child outcomes, which may depend on the setting or context. For example, in some settings, programmes engage with mothers-in-law and other relatives, because of their influential role in shaping parenting practices.

Q: Do the programmes displace and/or undermine authority and responsibility of ‘traditional’ bearers of wisdom and knowledge about domestic/married life and child rearing e.g. aunts and uncles, grandparents co-habiting with couples, church/faith actors such as mothers’ and men’s guilds providing day-to-day teachings?
A: REAL Fathers is designed to work within cultural systems, bearers of wisdom and knowledge, and not to displace people. What we mean is, the programme was designed in Uganda by involving fathers, mothers, and other community members to understand what they wanted fathers to do as partners and parents, what they were concerned about, how they teach each other, and how a programme like REAL Fathers could address it. The programme builds on these expectations of what a community wants to see as a good father and good partner, and good community member. However, we have not included many of these other groups named in the question. To the best of our knowledge, the programme has not displaced these other groups; we are not negating their role nor challenging what they have to offer.

Q: How do these programmes emphasise the children of special needs groups like children with disability and intersex children?
A: The Bandebereho programme was not initially designed with a particular focus on disability. However, over time, efforts have been made to make the programme more disability-inclusive through partnership with local disability rights organisations. While the programme does not address the specific needs of children with disabilities, we are working to better engage parents with disabilities and parents of children with disabilities, through outreach and through training facilitators on disability inclusion.

This is an important question and one that has not been addressed in REAL Fathers. Future adaptations or iterations of the programme would do well to consider this.

Q: How do you define gender equality in parenting?
A: Each person might have a different way of defining what gender equality in parenting means to them. In our work, we describe gender transformative parenting programmes as working with both female and male parents and caregivers to promote caring, equitable relationships and nonviolent interactions for the whole family.

Alongside improved parenting practices, programmes often seek multiple changes that can benefit children’s physical and mental health, development, and wellbeing, such as:

Caring, supportive, and nonviolent parent-child and partner relations
Equitable relationships where partners share responsibility for caregiving and power in making decisions about their relationship, household, and children’s lives
Parent/caregiver capacity to raise children with equal care and opportunities for play, learning, and education, free from gender stereotypes

See more in Parenting Programmes to Reduce Violence Against Children and Women: What gender transformative programmes look like.

Q: The Bandebereho programme led to a change in child behaviour (as reported by parents). Did you also measure impact on other child development outcomes, more from an education perspective?
A: The six-year follow-up to the Bandebereho randomized controlled trial did show important differences in children’s behaviours (internalising and externalising behaviours) between the children of Bandebereho parents and those of the control group. We also measured other child development outcomes through direct assessments with children using the International Development and Early Learning Assessment (IDELA), but did not find any differences. However, the intervention was also not designed to address these specific outcomes.

REAL Fathers has not assessed this. Given that our programme focuses on fathers of young children, it would be interesting to understand better how it affects a young child’s development in the short and long term. The programme is not designed to address educational outcomes.

Q: In terms of the evaluation of both programmes, have you looked at the impacts on time use for caregiving and domestic work? Do you see any impacts on reducing women’s unpaid caregiving/domestic work time through redistribution among women and men?
A: Another great question! We have assessed impacts on time use for the Bandebereho programme. We collected data from both men and women about who generally is responsible for a variety of household and childcare tasks (e.g. woman usually does, partners share the task equally), as well as whether they performed that task in the last week and if so, how much time did they spend on it. We found at both 21 months and six-year follow-ups that both women and men in the Bandebereho intervention reported greater sharing of these tasks between partners, compared to couples in the control group. We also found that men in the Bandebereho group reported spending more time on these tasks than men in the control group (about 52 more minutes at 21 months and 30 more minutes at six-year follow-up), but women continued to spend much more time than men on these tasks and a similar amount of time as women in the control group. However, this analysis only focused on six key tasks, not the full list of tasks that we asked about – we are currently doing a deeper analysis of the full time use data to understand more about how these dynamics shifted.

After those initial findings at 21 months, we conducted additional qualitative research with couples who had participated in Bandebereho to find out more about why we weren’t seeing a decrease in women’s time spent on caregiving tasks when men were spending more time on them. Women reported that although men were doing more, they sometimes had to redo the work men did, or that tasks often required regular maintenance (i.e. they couldn’t be done just once in a week), that men sometimes were not doing. RWAMREC then made some changes to the Bandebereho curriculum to integrate additional focus on unpaid care work, including to promote more discussion of expectations around household tasks between partners. Subsequent data collected during the scale up found that there was a slight decrease in the amount of time women spent on these tasks, as men’s time went up.

The evaluation of REAL Fathers included questions about whether fathers reported being engaged in caregiving and household activities. We did not ask this as a time-use question, which would have given us better information on the extent to which fathers are doing these tasks.

Q: Which instruments or assessment tools do you use to measure internalising and externalising behaviours?
A: For the Bandebereho randomized controlled trial we used the Strengths and Difficulties Questionnaire (Parent report) and collected data from both mothers and fathers about the same index child.

Q: During the trial study, how did you control the contamination of intervention to the control group especially regarding education provided and knowledge shared?
A: For the randomized controlled trial of Bandebereho, randomization took place at the individual level, given that the programme was already operating in the communities where the evaluation took place. Facilitators recruited 25 men who met the eligibility criteria, who were then randomized to the intervention or to the control group. As the men lived in the same communities, and the programme encourages men to share what they have learned with others, we were initially worried about potential spillover/contamination. We collected a lot of additional data on men’s social networks that might be able to help us understand if/how spillover occurred. However, we ultimately found that this was not needed. While men did talk to their friends and neighbours, this did not have the same impact as the intensive intervention.

For REAL Fathers, we randomized by village/community so that some communities received REAL Fathers as part of the intervention group and others were in the control group. We did this to reduce the possibility of spillover. We think this strategy was sufficient. In Rwanda, we were able to do some qualitative work on what fathers and mothers were sharing with others, as part of an organic diffusion of information and skills. They named people in their community as those with whom they shared the most information and skills with. Fathers and mothers described having other couples approach them asking for guidance, information, and to learn what they were doing differently. While some of these community members said they were trying out these new skills in their relationship, they still wanted to participate in the programme to fully learn the skills.

If you would like to speak to Dr Anjalee Kohli and Kate Doyle about the research, you can contact them by emailing anjalee.k232@gmail.com/ k.doyle@equimundo.org.

Useful resources