Men’s Involvement in a Parenting Programme to Reduce Child Maltreatment and Gender-Based Violence: Formative Evaluation in Uganda

Parenting programmes involving fathers can reduce child maltreatment and gender-based violence. However, most parenting programmes find it difficult to recruit fathers. We piloted a 21 session parenting intervention, ‘Parenting for Respectability’, with fathers and mothers near Kampala, Uganda. Sixty-one fathers and 83 mothers were recruited initially and 52 fathers and 76 mothers retained to the end. We interviewed with 24 fathers and 16 mothers. Data were analysed thematically. Success in involving fathers was probably due to (a) the first 10 sessions being father-only, allowing them to share experiences before participating in mixed-sex sessions; (b) exploiting men’s pre-existing motivation to improve their children’s behaviour, thereby enhancing family respectability; and (c) the interactive, participatory delivery. Mixed sessions enabled couples to clarify conflicting perspectives regarding spousal relationships and gendered norms. However, men experienced social pressure to conform to conventional masculinity, suggesting the need to instil intervention values at community level.


Introduction
There is increasing evidence that parenting programmes which involve fathers have the potential to reduce child maltreatment (Cowan et al, 2009;McAllister et al, 2012;Panter-Brick et al, 2014a, b). However, most parenting programmes find it difficult to recruit fathers (Stahlschmidt et al, 2013). In this paper, we focus on how fathers engaged in a communitybased parenting programme 'Parenting for Respectability' in Uganda, designed to both reduce child maltreatment and, in the long-term, gender-based violence. It examines what did and did not work for men, the benefits they derived from their participation, and the challenges of such participation. It is part of a wider formative evaluation of the Parenting for Respectability Programme. The paper answers three main questions: -How do fathers perceive the relevance and acceptability of a parenting programme to address violence against children and improve spousal relationships? -How do fathers respond to a programme that is structured to ensure both mixed-and single-sex sessions? -What are the challenges of involving fathers in a parenting programme comprising both fathers and mothers?

Literature Review
Parenting, Child Maltreatment, and Gender-Based Violence Poor parenting, particularly in the child's early years, reflected in poor parent-infant attachment, and forms of child maltreatment such as harsh and abusive parenting, child neglect, failure to provide basic necessities, commercial or other exploitation of children, and exposing children to witnessing violence suffered by others in their families, is associated with a wide range of short-term and long-term physical, emotional, and health consequences for the child and society (Panel on Research on Child Abuse and Neglect and National Research Council 1993). In the short term, it affects a child's physical development and well-being, and leads to fear, anxiety and depression, conduct problems, cognitive delays, and schooling difficulties (Tremblay et al, 1991;McCoy, 2013;Altafim and Linhares, 2016). In the long term, child maltreatment is associated with the development of high-risk behaviours, such as smoking, illicit drug use, alcohol use, sexually risky behaviour, violent conduct, and persistent antisocial behaviour (Baldry and Farrington, 1998;Shader and Programs, 2001;Murray and Farrington, 2010;Knerr et al, 2013). In early adolescence, parents' expectations of their children's behaviour, expressed through their own sexual values, their advice and rules for their children, and their role-modelling (intentional or unintentional), all serve to influence their children's sexual behaviour (Wamoyi et al, 2015). In South Africa, males experiencing abuse or neglect in childhood is an important risk factor for committing rape as an adolescent or adult (Jewkes et al, 2011). Parents reproduce inequitable gender relationships through the intergenerational transmission of domestic violence: boys who witness domestic violence are far more likely to abuse their own partners (Roberts et al, 2010), and girls to tolerate intimate partner violence (Uthman et al, 2011). Gender-based violence, particularly intimate partner violence (IPV), is one of the key drivers of HIV infection in sub-Saharan Africa (Jewkes and Morrell, 2010;Wagman et al, 2015). Poor parenting is likely to occur in parents with a poor understanding of child development, who are less nurturing and have an authoritarian parenting style, and in parents who abuse drugs and alcohol, in parents who are depressed and have low self-esteem, when parents were abused themselves, and in parents who have a poor relationship with their spouses (Coore Desai et al, 2017).
In order to support the healthy development of children, and in the long term prevent genderbased violence and HIV transmission, interventions that support positive parenting are needed. Many parenting programmes are designed to support parents by developing knowledge, skills, and practices necessary to foster greater understanding of child development, build positive parent-child and parent-parent relationships, deal with challenging behaviour, and reduce child maltreatment (Ponzetti, 2016). Reviews of existing trials in low-and middle-income countries (LMICs) suggest that parenting interventions can be effective in improving parent-child interaction and parental knowledge on child development, and could address maltreatment of children and adolescents (Knerr et al, 2013;Chen and Chan 2015). High-quality father-child relationships are associated with many positive benefits for children and families (Stahlschmidt et al, 2013). Children of more involved fathers tend to have higher self-esteem (Amato, 1994), better mental health (Buckley and Schoppe-Sullivan, 2010), and greater cognitive competence and educational achievement (Flouri and Buchanan, 2004). In a Ugandan study, Warrington (2013) described different models of fathers' encouragement and support, and found that fathers with authority, education, money, and time stood apart from their communities in supporting their daughters. Such fathers were keenly protective of their daughters' welfare, gave attention and time to talk to their children, encouraged them at school, and were described by their daughters as key figures and role models who built their self-confidence.
In spite of the known benefits of fathers' involvement in co-parenting, most parenting programmes find it difficult to recruit fathers because of various barriers including cultural, institutional, professional, operational, content, resource, and policy constraints (Panter-Brick et al, 2014a, b). For instance, if a programme is presented as being for 'parents', few fathers participate since it is widely interpreted as intended for mothers (Bayley et al, 2009). It is also often assumed by policy makers and programme implementers that fathers are difficult to work with and that they are deficient or unconcerned about parenting (Panter-Brick et al, 2014a, b). In addition, men are often presented as the perpetrators of violence against women and children, with little examination of how the socialisation of boys and men might encourage violence. While the evidence on best practices in engaging fathers in parenting programmes is increasing, very little comes from Low-Income Countries (LICs).

Parenting Interventions Addressing Gender-Based Violence and Child Maltreatment in Sub-Saharan Africa
A range of social interventions addressing parent-child relationships, child maltreatment, and attachment have been evaluated, mostly in high-income countries (HICs) (Knerr et al, 2013). Some promising interventions from HICs have recently been adapted and rigorously evaluated in countries in sub-Saharan Africa (SSA). Families Matter! (FMP), which was adapted in Kenya from the US programme Parents Matter! (2004)(2005)(2006) and is now being replicated in Tanzania, uses parent-child dyads and encourages parent-child communication about sexual risk reduction and positive parenting skills (Vandenhoudt et al, 2010). A contrasting example is a programme of lay home visiting in South Africa, adapted from health visitor programmes in Britain, which promoted sensitive and responsive parenting and secured infant attachment by targeting women in economically disadvantaged areas from late pregnancy up to 6 months postpartum (Cooper et al, 2009).
There are now an increasing number of programmes developed within Africa, many of them focusing on sexual health and gender transformation. For instance, Mema Kwa Jamii (MkJ) (''Good Things For Communities'') was developed in Mwanza, Tanzania, to promote parentchild communication and positive socio-cultural norms and behaviours around reproductive and sexual health (Hayes et al, 2005). A 10 session Sisters for Life programme was developed in South Africa to increase women's HIV knowledge and awareness and help them challenge cultural norms and taboos around sexuality (Phetla et al, 2008), and this is now being replicated in Mwanza, Tanzania. Another promising intervention in South Africa is Men as Partners (MAP) which works closely with male peer groups to explore values related to sexuality and gender relations as well as sexually transmitted infections (including HIV) and reproductive health problems (EngenderHealth and PPASA, 2001). The Sinovuyo Caring Families Programme for Young Children was developed in Cape Town, South Africa in 2012, and focused on the systematic development of an evidence-informed, locally relevant parenting programme for socioeconomically disadvantaged parents of children aged 3-8 years. Process evaluation of this programme concluded that evidence-based parenting programmes are feasible when situated within relevant cultural contexts (Lachman et al, 2016). In Rwanda, the MenCare+ programme, adapted from Programme P, was delivered by Promundo through the health sector and involved men in maternal and reproductive health, neonatal and child health, couple communication, and violence prevention. An evaluation of the pilot programme in 2014 reported positive shifts in men's family roles and their increased interaction with children (Doyle et al, 2014).
In Uganda a range of programmes exist to address gender-based violence. They tend to emphasise children's and/or women's concerns more than men's, and few address both spousal relationships and child maltreatment. The SASA! intervention by Raising Voices works at a community level to prevent domestic violence and advocates a supportive environment for women's rights and equality in Uganda through legislative and policy reform (Abramsky et al, 2016). However, SASA! does not address child maltreatment as a major goal (Abramsky et al, 2012). The SHARE Project (Wagman et al, 2015) adapted prevention strategies from Raising Voices and Stepping Stones to address intimate partner violence in central Uganda. In contrast, the Plan Uganda manualised 12 session maternal well-being and child development intervention in Lira, which was shown to be effective in an RCT, addressed both parental and child outcomes. One of the most innovative features of that study was the family focus, with sessions for fathers, sessions about mother-father communication, and men as community volunteers, with seven of the 13 volunteers being men (Singla et al, 2015).
Interventions with men and boys have recognised the value of explicitly addressing masculine norms which perpetrate gender-based violence and child maltreatment, through activities that encourage self-reflection and rehearse new behaviours and a sense of responsibility and positive engagement as fathers (Cowan et al, 2009;Dolan, 2013;Doyle et al, 2014). Father-centred programmes that deliberately seek to address outcomes associated with both intimate partner violence and child maltreatment in Uganda are therefore increasing, although few such interventions have been rigorously evaluated. The REAL FATHERS Initiative, a 12 session father mentoring programme, is one of the few men-specific interventions being delivered in Northern Uganda, to promote positive images of men as fathers, reduce child exposure to violence at home, and break the cycle of intergenerational violence. Formative evaluation of this programme found a positive impact of the intervention on intimate partner violence and a significant reduction in physical child punishment (Ashburn et al, 2016). Finally, the Gender Roles, Equality and Transformation (GREAT) Project was implemented mainly with male youths in northern Uganda in 2012 to support the formation of life-stage-specific strategies which motivate youths to transform gender norms and reduce gender-based violence. Evaluation of the pilot phase reported some changes in behaviour and attitudes related to gender roles, with men increasingly helping with household chores and encouraging equitable decision making (IRH George Town University, 2016).
The Parenting for Respectability Programme: Goals, Structure, and Delivery Parenting for Respectability (PfR) was designed to modify four main familial processes that predict child maltreatment and sexual and gender-based violence in Uganda: poor attachment and parental bonding, harsh parenting, inequitable gendered socialisation, and spousal conflict and disrespect. It is structured as a 21 session community-based parenting programme with a core of 10 single-sex group sessions and 11 mixed-sex group sessions led in a semiparticipatory way by a trained facilitator. The programme was first delivered to parents in Wakiso District near Kampala, Uganda.

Programme Theory and Principles
A central principle of this programme is to harness the target group's existing motivations for parenting. With respect to fathers, in Uganda little attention has been paid to one of the most important dimensions of fatherhood: the need to maintain the family's respectability, in large part achieved through the appropriate behaviour of the children and their parents; (Siu et al, 2013;Wamoyi and Wight, 2014). Meanwhile there is a growing recognition that interventions with men and boys should explicitly address masculine norms which perpetuate violence, through activities that encourage self-reflection and provide the scope to rehearse new behaviours and a sense of responsibility and positive engagement as fathers (Dolan, 2013). Parenting for Respectability attempts to do this, as well as harness the core motivation to bring up well-behaved, respectable children.
The programme is intended to operate primarily at the interpersonal level, between the facilitators delivering the programme, parental care-givers, and their children. It is also intended to work at the community level, through engaging formal and informal opinion leaders, involving participants' neighbours and families in homework exercises, and through community events, thus changing community norms. Ideally it is complemented by other interventions in Uganda at a macro level, which address the structural context and determinants of gender-based violence, such as poverty and socio-economic inequalities. Operating at this level is beyond the scope of this intervention. The intended intermediate and long-term outcomes of the programme are to (i) improve parent-child connectedness; (ii) encourage more equitable socialisation of girls and boys; (iii) reduce harsh parenting; and (iv) reduce spousal conflict and inter-partner violence.

Structure of the Programme and its Formative Evaluation
There were two stages in the development and formative evaluation of the Parenting for Respectability programme. Stage 1 involved the development of the draft content and facilitators' manual, structured under five broad themes: introducing the programme, including the benefits of fathers' involvement; parental bonding and attachment; gender socialisation; Men's Involvement in a Parenting Programme achieving children's good behaviour and respect through positive discipline and relationships; and spousal/couple relations. Other sessions were recommended by participants: marital relationships and alcohol, and the closing ceremony and award of certificates, as summarised in Fig. 1.
Stage 2 involved piloting the programme to determine the appropriateness and relevance of the content and overall delivery strategy, including negotiating access to communities. This was done in two parishes in Wakiso District near Kampala: Kigungu, a rural, predominantly fishing village, and Bweya, a peri-urban neighbourhood with predominantly formal employment. Six groups of 15-20 parents, each of whom had children aged 0-17 years (total 128 parents, 52 men), were recruited to take part in the programme, one fathers' and one mothers' group from pre-existing savings groups in Kigungu, and two fathers' and two mothers' groups in Bweya. The programme was delivered in three phases, first with two groups in Kigungu, then midway through it was started with two groups in Bweya, and when that was completed, with a further two groups in the same community. In each phase the two groups started as single sex for the first 10 sessions. In Session 11, the groups were combined and subdivided into two mixed-sex groups, to whom the second half of the programme was delivered. Across all groups there were 16 parental couples (see Table 1).
Weekly 2-3 h sessions were conducted through a participatory approach using brief lectures, discussion, reflection groups, posters and pictures to improve knowledge, raise awareness, and modify values. Practical skills were developed through role play and exercises to practice at home with ones' spouse or children. Problematic sessions or activities within sessions were revised as necessary and tested again or dropped altogether. For example, the original activity 'Persons and Things' under the session 'Gender Socialisation', was dropped following initial testing. It required some participants to act like persons and others like 'things', that is, as if they cannot think, feel, or make decisions, but instead must do only what the 'persons' tell them. The goal was to increase awareness of gendered power and unequal relationships. However, participants, both male and female, found it offensive to act as 'things' and likely to reinforce gendered stereotypes. This activity was replaced with another adapted from the SASA! manual titled 'The New Planet: The Sun, the Moon and the Stars', which was more respectful and acceptable because it related closely to their aspirations for new ways to relate in their families. In Sessions 11-20 the mixed-sex groups were encouraged to address conflicting gendered perceptions of parenting and spousal relationships.
The sessions were led by CN, an experienced community development trainer, assisted by 12 lay facilitators (6 males, 6 females) recruited from the parental groups. They had no formal professional education but received training from the Parenting for Respectability Programme Team to lead sessions on a voluntary basis. Single-sex groups were facilitated by pairs of same sex facilitators, and the mixed groups facilitated by a pair of mixed-sex facilitators. Facilitators were mentored and supervised closely by CN.

Data Collection
This study collected data using semi-structured interviews with a sample of 24 men. They were purposively selected to ensure representation of men who had: consistently attended the sessions (N = 8); missed some sessions (N = 4); disclosed having difficult relations with their spouse early in the programme (N = 2); a spouse who had also attended the programme (N = 6); and repeatedly reported positive changes in their families during programme sessions (N = 4). These data were complemented with four interviews with women interviewed with their spouse (parental couples). Interviewing couples together had several potential benefits. First, it offered an opportunity to the couples to clarify issues and come to a consensus on their responses. Second, it allowed us to compare their responses and those obtained from husbands and wives separately, and third, it helped us to assess their relationship from how the couple discussed and behaved towards each other.
Each of the sessions with the six parental groups was observed by an experienced qualitative social science researcher (RS or FZ), who noted how individual activities were implemented, participant engagement, and key issues of concern. CN also made notes immediately after each session to record what went well and what required revision. These notes were combined in debriefing reports and discussed by the entire team, and where necessary adjustments were made to the next sessions in light of the lessons from the previous one.

Data Analysis
Data were analysed thematically (Braun and Clarke, 2006). GES and RS read through the transcripts. Initially all authors discussed their impressions of the data and generated preliminary themes. Following a consensus, additional interviews were recommended with eight men to answer further questions about men's involvement in the programme. GS then read the transcripts and observation notes to enrich and consolidate the existing themes. GS initially drafted the manuscript which was successively revised with DW and JS.

Ethics Declaration
The study was reviewed by the Research and Ethics Committee of the Uganda Virus Research Institute (UVRI) and approved by the Uganda National Council for Science and Technology (UNCST). It was then introduced to the technical and political leadership of the district. At the community level, parents of children aged 0-17, as well as formal and informal community leaders, helped develop the programme, in particular selecting facilitators and determining how groups should be constituted and when the programme should be delivered. The Entebbe Municipal Community Development Office assisted in contacting the study sites and mobilising local leaders to create awareness and support for the programme. The local leaders then approached existing groups and asked them to select interested volunteers to join the programme. All interviewees provided consent. Identification numbers, rather than names, are used to reference the participants Findings Involving Fathers in the Programme: What Worked

Recruitment and Retention
A core aim of the programme was to recruit fathers as well as mothers. We therefore designed the programme to start with single-sex groups, with men being recruited into fathers' groups and the benefits of their involvement discussed. Fathers valued starting in single-sex sessions: The fact that we first attended as single sex it was good before joining together. We could also bring up a challenge in the home and could be discussed [P 7].
Discussions in single-sex sessions were more open than in the later mixed-sex groups and fathers discussed issues at much greater length than mothers, requiring far more time for the same exercises. This was mainly because they had more varied views than women about parenting and family relationships, but they were also more argumentative and assertive.
The successful recruitment and retention of parents, especially men, is one of the main achievements of the programme. In Kigungu, 20 men and 23 women were recruited, of whom 17 men and 19 women were retained to the end of the programme. A further two men and four women joined the programme when some participants requested to bring their partners at the start of the mixed sessions. In Bweya, initially 11 men and 26 women were recruited. Subsequently male numbers rose to 17 and dropped to 14, while female numbers rose to 28. In the second round of delivery in Bweya 19 men and 25 women participated. Across all groups the men were aged between 19 and 65, with most between 30 and 40 years. Participation who received the programme is summarised in Table 1.
We have little information on how typical these men were of others in their communities. Even in Kigungu where participants were initially recruited from pre-existing savings groups, more men were recruited from outside those groups since they had few male members. Generally, the men seemed to have fairly varied occupations and family relationships. Kigungu men were predominantly fishermen, while in peri-urban Bweya men had more varied forms of employment, including working on a construction site, as a cleaner, a roadside market trader, and as a motor cycle taxi rider commonly known as bodaboda. Since Kigungu fishermen went fishing late in the evening and returned in the early morning they had time to be involved in the intervention during the day, in contrast to Bweya men. Consequently it was more difficult to recruit and retain men in Bweya, revealing that an important contextual factor affecting recruitment of fathers was their predominant occupations in the local economy.

Motivation
There seemed to be two main initial motivations for men to join the programme. First, it addressed their pre-existing concern that their children should grow up well behaved and respectful of their elders, thus not undermining the respectability of their family. One father commented in the Kigungu men's group: It's terrible, you know we are fishermen and keep moving around the lake… as a result, we produce so many children and sometimes take 5 years without seeing them. What can we do for the children and ourselves to have those benefits, to achieve respect? We want good children in future''.
Another father [P 7] reported that he joined and completed the programme because he ''had challenges with family'' while another [P 1] said ''I was already interested in acquiring parenting skills so as to improve my family life''.
Second it was clear from men's comments, both after the initial sessions and after completing the programme, that they welcomed an intervention which seemed to recognise them as equal partners in raising their children and credited them with being concerned about their children's well-being and behaviour.
The truth is that fathers need this training more than women because they too are concerned about children. Whenever you train women only and they learn these things and when the man lacks information on the parenting of children, it is very easy for what a woman has learnt to fade quietly. If a man is trained and understands it well it is easy for the woman to move with him instead of the woman pushing them into a man. Some men do not want to be told or advised on anything, so it is very, very important for the father to attend, [P 11].
Men especially appreciated the programme because the content explicitly addressed men's concerns to raise well-behaved children. A father stated: For the men to join, it was to help them get the same package because if you only teach the woman and the man has not understood then you will not have helped, and also because it [the programme] had other challenges and things that were concerning the men [P 19].
To these fathers, the programme emphasised co-parenting in a way that made sense to them as it specifically framed their concern to shape children's discipline within their traditional status and aspirations as family heads, as P 19 went on to explain: The men to join the programme shows that discipline of a child has to come from the man and learn from him as the head of family. Most times men leave the children with the women and so they learn women manners alone and lack the behaviors from the man. But they have to behave like the mother and the father and grow up with two gifts and that is the future homes which will please people because people love good manners.
Exploring Non-conventional Norms: Significance of Programme Environment In all three men's groups, participants explored the expectations and constraints of conventional masculinity, and they were more willing to question these norms in the single sex than mixedsex sessions. Commenting on the learning about new roles and child skills in single-sex sessions, P 7 stated: Men's Involvement in a Parenting Programme The clearest example of discussions questioning conventional norms and roles was debates in Kigungu between a woman's partner and his friends and her previous partner, who she had betrayed. However, there was some evidence that even within the sessions men were not ready to practise conventional maternal behaviours beyond specific exercises. For instance, during the formation of mixed groups in Kigungu, although a man was ready to hold his baby in a role play he handed it back to the mother as soon as the role play ended. More generally, it was clear that some men intended to continue restricting their spouses' financial and domestic power.

Engaging Couples and Integrating Fathers' and Mothers' Groups
We promoted the programme as suitable for parental couples to participate in together. Initially parents were recruited irrespective of whether they joined as a couple, but during the formation of mixed-sex groups, midway through the programme, we tried to allocate couples to the same group to complete the programme together. We involved the groups themselves to suggest the best way to involve couples, assessed the practicality and acceptability of this approach, and evaluated their overall participation during the mixed sessions. During the formation of mixedsex groups for the second half of the programme, the participants agreed on the following criteria for involving couples. Existing couples were allocated to one or other group randomly, and other participants were encouraged to recruit non-participating partners. To ensure cultural sensitivity, affinal relatives, particularly fathers or mothers-in-law, were allocated to the alternative group since they should not mix and openly discuss sensitive topics. Another important consideration was the presence of participants' former partners, or the new spouses of their former partners. These were separated to minimise sensitive discussions. A role play was acted out by the participants to highlight some potential benefits of working together in mixed groups, and to identify possible issues for future discussion in such groups, with participants especially identifying the need to discuss marital relationships and communication between couples, and between parents and their children.
From the observations and the interviews conducted after the mixed sessions, it was evident that couples found it extremely helpful to attend the mixed sessions together, and that other participants benefitted if couples shared their experiences openly in the group. Attending mixed-sex sessions only occasionally resulted in couples openly discussing sensitive issues, but it enabled them to clarify conflicting perspectives regarding spousal relationships, negativegendered norms, and parenting practices. It helped them to discuss sensitive issues openly while at home, and made it easier for partners, particularly women, to encourage or remind their spouse what was learned and the need to adhere to it. A couple that was interviewed together about participating in mixed-sex sessions highlighted their value, with the wife saying: It was better that our partners be around and we could learn together. We used to share what we learnt together. If I attend alone -we cannot kukwatagana bulungi (understand each other well), but if we are together you can tell the partner and we could change. If I study alone, the other partner cannot change [P 23].
Her husband [P 3] confirmed this, suggesting that it made it easier for couples to remind each other of appropriate behaviour, saying, ''It is good to attend together [as husband and wife] because while at home we find it easy to remind ourselves what was discussed in the session, what is right and what is not''. Another father's account clarified this: It is like I have said, that if both attend this program it helps in that one does not pull ropes with the other because both learnt the same thing, and for that reason parenting children becomes easier because both of them know what to do and no one is forcing the other to do something, [P 1].
Other men suggested that unlike single-sex sessions, attending sessions as a couple encouraged transparency between the couple to confront the issues affecting them without attempting to hide: It is like studying in the same class then you all leave when you have understood the idea at once and 'no pulling ropes' but if it's different groups there is a possibility of first hiding it from your partner [P 17] Perhaps the most important value of mixed sessions was that it legitimised further discussions of topical issues at home, as spouses usually referred to the lessons they both learnt from the programme. A mother [P 24] had been unhappy with her husband's drinking habits but had for a long time felt powerless to discuss them with him. She appreciated the value of mixed sessions as it gave her confidence to demand change at home: I feel the training was relevant to couples -like the alcohol session -you [husband] drink and the wife suffers from the smell of beer without okwetonderamunno (apologizing to your partner Attending the parenting sessions as a couple also offered an opportunity to learn new perspectives on parenting from opposite sex participants. Among the several examples, a wife [P 22] who attended the mixed sessions with her husband observed that, 'When we were in the mixed group it helped us because men brought new ideas which we did not know'.
In the mixed-sex groups, participants shared their experiences and views more readily if exercises started in single-sex sub-groups with views fed back to the plenary group as a collective view without it being explicitly linked to individuals. It seems that because of their partners' presence, or more generally the opposite sex, participants were reluctant to state their views openly, and they were also readier to discuss issues if they did not sit next to their partners. Suggesting that they valued the opportunity to exchange views with members of the opposite sex in a less threatening context, another mother [P 27] stated: There was no problem [with the mixed sex sessions] and when we mixed it was okay. It helped us because we were able to give in our views and men gave in their views.
We noted, however, that for couples participating in sessions together, discussing issues initially in single-sex sub-groups allowed spouses, particularly wives, to communicate difficult messages to their partners by having them expressed as the collective view of the sub-group. Women who feared or felt oppressed by their partners saw this as an extremely valuable opportunity, providing a safe forum to raise concerns about the negative habits of their partners. The facilitator actively assisted this. To maximise the benefits of having parental couples in the sessions, some activities, especially home practice exercises given at the end of each session, were designed to involve both husband and wife or father and child. Men appreciated the idea and reported positive benefits pertaining to improvement in relationships. However, the requirement to do home practice activities did not always work with great success, particularly with a spouse. There were several reasons for this. For some couples, it was lack of time. P 9 reported: ''Sometimes we failed to get time because my wife was away or she is busy''. But for many, it was due to limited skills and practice in starting a conversation with the spouse. Given the perceived advantages of mixed sessions, many participants recommended that to improve recruitment of parental couples and maximise the benefits of the programme, married persons should only be enrolled to the programme as couples. For example, one couple [P 3 and P 23] recommended: Make it a rule that people attend as couples, because if the programme is for the couple changes could come, but those attending alone can't teach others [P 3].
A father whose wife did not participate in the programme felt that she had missed a great opportunity to take part in an important programme, and further confirmed the challenge spouses found while helping their non-participant spouse: As for me I got the content except that some issues did not come out clearly because my wife did not get firsthand information from the trainings because she was busy by that time and could not attend. That is where there was a gap, but for me I got them [P 12].
Although there was no objection to the mixed-sex sessions, there were some challenges to couples participating together, since some found it difficult to discuss sensitive family issues. One father described how delicate it was to discuss sensitive family issues during the sessions and in the presence of a spouse: There were people who pretended that they are working together but I think that the fear to be together in one group was also there. 'How can I say something when my partner is seated there and when we go home these other people will not be there but yet it was much safer to learn something individually' [P 12].

Perceived Impact of the Programme
Generally participants evaluated the early influences of the programme on their parenting skills and spousal relationships very positively. Many confessions and testimonies during the sessions suggested that parents absorbed the key messages from the programme and it was striking that almost all parents reported dramatic changes to their relationships. Nevertheless, it is important to note that those involved in parenting programmes often claim they are having profound effects, especially to those they think associated with the programme, when, in fact, the impact is rather limited and/or short lived. Self-reported impacts of parenting education are especially subject to social desirability response bias (Ponzetti, 2016). Parenting for Respectability Programme has not yet undergone an outcome evaluation -ideally through a randomised controlled trial which minimises potential bias (Eccles et al, 2003;White, 2013) -so evidence of effectiveness is partial and we do not know what the longer-term impacts might be.

Impact on Parenting
Several aspects of parent-child relationships were, however, reported to have improved due to this pilot parenting programme. Men reported that the programme made them more aware of their limited understanding of child development and upbringing. The two programme components that they particularly valued were learning about positive parenting and bonding and attachment. With regard to bonding and improved child-parent relationships, fathers reported that they had learnt about a wide range of valuable ways to interact with their children, which were never taught at school or learnt in any other forum. They referred to the importance of emotional engagement, play, physical care, and consulting with children, as one father [P 10] revealed: Several things which I cannot forget: I learnt things that I did not learn from school -relationships between spouses, relationships with children, playing with children, asking children what they want, and I tell them what I also want-all these are important.
The home practice exercises given at the end of each session greatly helped fathers improve their relationships with their children: The homework was good if you could give it time and you do it. It could build up a relationship you had not expected with children because up to now there are times when you see that the children are coming closer to you. Because whenever I came back from the sessions I could play with this young boy and you could observe that the gap in between you and him reduces. Currently this young child cooks some things and tells me that daddy I have prepared a cake and he serves me and I also receive it; I think it was mainly because of that homework [P 7].
With regard to positive discipline, fathers reported that as a result of participating in the programme, they were open to learning alternative disciplinary methods to corporal punishment.
I only knew corporal punishment as a means of discipline but now we know that we can relate well and explain things clearly to children, and they will do the right thing [P 13].
Indeed, from their reports of attempting to implement programme lessons, fathers appeared more consistent than mothers in applying alternative methods of discipline, such as 'grounding'. However, there were a few fathers who acknowledged that they found it difficult and rather strange to rely only on positive methods, without using corporal punishment, while others believed that alternative methods only worked in the short term. For instance, P 9 said: Positive punishment like denying a child an outing works, say for the next two days they (children) change, but after sometime they forget.

Impact on Couple Relationships and Roles
Initially we wondered how men could be persuaded of the value of gender equitable relationships and roles. We developed interactive sessions with activities that portrayed positive images of men involved with their families and focused on building respectful supportive spousal relationships. Although the exercise on gendered power did not ask them to adopt this as an ideal, and none expressed such an aspiration, many men greatly valued acquiring skills to improve their relationships with their spouses. One of the most commonly reported benefits of the programme was that it recognised men's role in raising children, and helped them realise that increased involvement in children's upbringing is positive for both the child and the father. This motivated fathers to seek more positive relationships and resulted in greater satisfaction with their identities as men: Men also reported a reduction of spousal conflicts and increase in mutual respect following the programme. Among many examples a husband reported [P 7]: We managed to learn what annoys the wife and each of us could know our mistakes. We used to air them out during the session. Both men and women reported that they had come to appreciate that good communication between partners is key for a peaceful relationship at home, and several men resolved to communicate better with their wives.

Men's Involvement in a Parenting Programme
Women also emphasised that they had become increasingly conscious of negative impact their own poor communication skills usually had on their relationships and they were ready to change. This was expressed both during the sessions and the interviews. For example, during the session on 'Healthy Marital Relationships', a mother revealed: ''We have a problem in the way we talk to our husbands''. Across the interviews, there was a striking consistency in the accounts of the couples, whether they were interviewed separately or together as a couple, regarding the impact of the programme on marital harmony due to improved communication between partners. For example, a father [P 8] described how his poor relationship with his spouse had improved, although he still attributed the fault fairly exclusively to her: I have had problems with her in the family until the project came here to teach us. This project has helped her to change her behaviour. At one time I left my home and went and hired another house and stayed away for 3 months out of touch. But because of our children we have now reunited. She has now learnt how to talk with people, and to interact with the partner and how to treat the children. When she heats up I cool down and I think when she remembers the sessions, her tempers cool down.
Meanwhile his wife [P 21], interviewed separately, gave a fairly consistent account: I have learned good behaviour like talking with people, and handling my husband. Each could behave his own way. But now we sit together and plan. He can even give me money for savings when I don't have.
The programme also helped men to recognise that their wives were overwhelmed by child care responsibilities, leading them to greater appreciation of co-parenting. In particular, men found it useful that the programme addressed the traditional roles of women and men and helped them realise the value of sharing some domestic tasks, without them losing their status as husbands: Traditionally, the man is supposed to be above the woman. This was addressed so that you balance, whereby you could do activities done by the woman and you would lose nothing, and the woman could do likewise [P 7]. I would expect my wife to cook the food even if she was sick. And if she went away to attend to something else outside the home, I would wait for her irrespective of what time she would be back. But now I can organise and cook [P 3].
The opportunities to learn some practical skills during the sessions were not only considered useful but also surprising, as men learnt that they too could do some roles which are considered the reserve of females: There was when we dressed the baby. This session surprised me in that we can do all the things although we have that thinking that those are not mine. But if you are serious you can do them [P 2].

Potential to Further Disempower Mothers
We established that husbands tended to police their wives following their participation in the programme, potentially reinforcing male power in relationships. During the sessions it became apparent that most male participants thought they were more receptive to the programme messages than their partners, especially with regard to positive parenting. For example, some fathers complained that their partners tried to avoid harsh parenting for a day or two after attending sessions but then reverted to hitting their children. One father, after a session stated: The mothers change for only 3 days after the session, however, after the 3 days, they return to their normal behaviour of shouting and beating the children as if they are not attending the training.
Unlike mothers, fathers seemed to find it easier to practice alternatives to corporal punishment, which might relate to them having far less domestic work to complete. For many mothers, a quick slap, for example, was considered more practical than lengthy admonishments such as imposing 'time out' for the child.
Entrenched Norms of Masculinity Despite fathers' readiness to get more involved in their children's upbringing and to improve their relationships with their spouses, beyond the programme they still experienced social pressure to adhere to the norms of conventional masculinity. This came from their relatives and the wider village membership and undermined their efforts to enact new dimensions of being a father. In feeding back about a home practice exercise at the start of a session, a father reported how his sister had found him getting involved in his children's games and thought it very unusual. She teasingly suggested he may be losing his mind, saying: My brother, don't you think that you need to make a will because you are about to die? Are you yourself anymore? What are you trying to do?
In referring to the potential value of linking programme participants across groups and communities, this man might have wished to empower participants to challenge dominant norms: May be what was missing was that we heard you people were also teaching in another place. The facilitator who came from that side came wearing a t-shirt that was resembling mine and I think what was missing was to link us with the people of that side. So what was missing was to unite us so that we are a bigger group because whenever the group is big then the program becomes powerful. Actually that was it to just unite us with those you had already taught and maybe we have a dinner with them or even exchange ideas [P 17].
Other men also underscored the value of ongoing support in their ability to sustain the new behaviour. One participant, [P 3] suggested: Whatever we learnt could help to bring good behaviour and respectability, but people need to be reminded all the time and not to stop at the end of the sessions; they need to be reminded because sometimes people backtrack. People need to remain in the group as they refresh themselves.
Booster sessions for men may be one such strategy to sustain behaviour change, found elsewhere to be effective in enhancing the long-term effectiveness of preventive interventions (Braukhaus et al, 2003).

Discussion and Conclusion
This paper reports the lessons learned about how to engage fathers in a community-based parenting programme being piloted in two neighbourhoods in Wakiso District, Uganda. There is a strong rationale to involve fathers in parenting or caregiving programmes but men are not traditional targets of such programmes, particularly in the global south. Most parenting programmes find it difficult to recruit and retain fathers, and it is often difficult to constitute a group with sufficient numbers (Scourfield et al, 2016).
The Parenting for Respectability programme had few exclusion criteria and recruitment and retention of fathers was one of the most important achievements of our initial piloting. There Men's Involvement in a Parenting Programme Ó 2017 The Author(s) 0957-8811 The European Journal of Development Research Vol. 29, 5, 1017-1037 are many possible reasons for this success. At the start we deliberately presented the Parenting for Respectability programme to men as being for fathers, with a parallel programme available for mothers. Had it been presented as meant for 'parents' in general it was likely to have been interpreted as being intended primarily for mothers, which would discourage fathers (Bayley et al, 2009). Previous research suggests that a clear recruitment process and an understanding of what is involved in the parenting programme were some of the most important factors that motivated fathers' participation and retention (Axford et al, 2012;Tully et al, 2017). We found that a particular attraction to fathers was harnessing one of their most important pre-existing motivations regarding parenting: the concern to raise well-behaved and respectful children, which is the hallmark of family respectability and father's reputation (Siu et al, 2013;van den Berg et al, 2013). Fathers seemed more enthusiastic and keen to achieve their children's good behaviour and respect than mothers, right from an early stage. Discussing with fathers, the benefits their involvement can have on various aspects of child development can greatly motivate them in their parenting efforts (Best Start Resource Centre, 2012). Improving spousal relationships also motivated men to attend the programme. In a recent study on parenting practices, Ugandan parents have identified troubled intimate partner relationships as detrimental to positive parenting (Boothby et al, 2017).
The structure of the Parenting for Respectability programme was an important factor for its success. Men did not object to starting with single-sex sessions and midway through the programme combining their groups with women's groups. Working in single-sex groups allowed men to discuss sensitive issues with fellow men in an open, less threatening manner. It seems that in single-sex groups, men were more able to freely explore non-conventional masculinity because the programme created an environment in which they did not fear ridicule, that is a setting in which they regarded each other as peers rather than as competitors. Similar benefits of an all-male setting were found in the 'Dads Only' Parenting Programme in the UK (Dolan, 2013), in the One Man Can father's workshops in South Africa (van den Berg et al, 2013), and in the gender-based physical activity intervention among Football Fans in Training in Scotland, UK (Hunt et al, 2014). Men tend to be highly motivated to attend parenting programmes when they are presented with an opportunity to meet fathers in circumstances similar to their own (Scourfield et al, 2016).
Conversely, attending sessions as a couple enabled the participants to clarify conflicting perspectives regarding spousal relationships and negative-gendered norms, as we had hoped, having drawn on the widely implemented Stepping Stones programme (Welbourn and Bollinger, 1995), and as has more recently been found in Kampala (Starmann et al, 2017). Mixed-sex groups also offered an opportunity to learn new perspectives on parenting from the opposite sex. Research with low-income families in the USA has found that in parenting programmes involving both single-sex and mixed-sex groups, fathers in the couples' groups had more consistent, longer-term positive improvement in their relationships with spouses and engagement with children than those in fathers-only groups (Cowan et al, 2009).
While the findings from this formative evaluation are promising, we observed the potential for the programme to unintentionally reinforce male power in parenting relationships. This is a key issue to resolve in the next stage of developing the programme. Fathers felt that they were more receptive to the programme than their wives and criticised the latter's parenting styles, particularly their use of physical punishment. Therefore, by encouraging men to take an active part in what women might regard as the 'female' sphere of parenting, and introducing new principles by which parenting should be conducted, the programme might inadvertently encourage men to monitor, criticise, and undermine their partners' capacity and behaviour in an area of domestic life in which their partner previously had autonomy. This may exacerbate a gendered power imbalance, rather than improve it by promoting new expressions of masculinity.
This study has also highlighted the barriers men face in changing their conventional masculine practices within their wider social networks. In South Africa challenging social contexts, including the dominant masculinities of peer networks, have been found to limit change among male participants in the Stepping Stones and Creating Futures interventions aimed at gender transformation (Gibbs et al, 2015). In a fathers' parenting programme in the UK, demonstrations of involved fathering clashed with certain masculine ideals shared by peers, such as demonstrating that one is knowledgeable, leading fathers to fear to disclose their attendance and to conform to traditional fathering templates (Dolan, 2013). A fundamental question thus remains: how can the programme's goals be shared beyond those fathers involved in the programme, to help them overcome the social pressure to conform to normative masculinity? There is no single answer to this, but at the community level, an initiative suggested by one of the respondents [P 17] might help: to link-up programme participants from different communities. This may create a new community of practice and a critical mass that may support further learning and sustain the new interpersonal and parenting behaviours advocated by the programme. At a broader level, modifying deeply entrenched hierarchies of gender and generation, and the norms that both reproduce and are reproduced by them, requires long-term interventions at macro as well as meso and micro levels. This has been demonstrated by projects in south west Uganda (Wagman et al, 2015) and in northern Uganda (Ashburn et al, 2016). Parenting for Respectability therefore has to be complemented by other initiatives operating at a national level. We conclude that the success of the Parenting for Respectability programme in involving fathers and parental couples was probably due primarily to three factors: its exploitation of parents' pre-existing motivation to improve their children's behaviour, thereby enhancing family respectability; the first half of the programme being single sex; and its interactive or participatory delivery. Our experiences so far suggest that individual women and men derive benefit from a relatively short-term intervention. This formative evaluation has demonstrated that the sensitivity to men, and perceptions of male behaviour, in the way the training was conceived and delivered, was welcomed by fathers providing them space to share ideas and learn new skills. Local, individual initiatives like this demonstrate that awareness can be raised and behaviour modified, suggesting that investing in such interventions can support both parents and children.
There were some limitations to the study. One is the bias in relying on participants' selfreports to assess the impact of the programme. Second is that the small qualitative sample limits generalisability of results. A large scale 'Before-and-After' outcome evaluation is currently underway to assess whether the Parenting for Respectability Programme is effective in modifying familial factors underlying child maltreatment and intimate partner violence. A randomised controlled trail will be conducted if the results are promising.