Working for the Well-Being of Children: The Value and Efficacy of Adopting a Cooperative, Inter-agency Approach

This chapter presents a case study illustrating the positive value of cooperation between a research team and policy leaders in strengthening policies that support children’s well-being in Sierra Leone, one of the poorest countries in the world. Having reviewed briefly the science of cooperation and its implications for policy development and practice, it outlines action research with a robust design in which communities themselves took action to support children’s protection and well-being. Its focus, however, is on the cooperative processes surrounding the action research that enabled policy influencing. It highlights the importance of taking a cooperative, inter-agency approach that actively engages researchers and policy leaders in the development and use of the research. It shows the value of enabling government cooperation and ownership at multiple levels, working with a broker to help manage power dynamics, and drawing on the complementary strengths of academics, UN agencies, NGOs, and government collaborators. It illustrates how a cooperative strategy can be useful if there is careful attention to task interdependence, power asymmetries, and ethics. It also examines the usefulness and limits of this approach in other contexts.


The Value of Cooperation
In work globally on humanitarian action and long-term development, inter-agency coordination, which entails cooperation, is essential for achieving effective outcomes and accountability to affected populations. Global guidelines such as the IASC Guidelines on Mental Health and Psychosocial Support in Emergency Settings (2007) prioritize inter-agency coordination, as do nearly all the UN NGO standards and guidelines. Calls for coordination are well founded since no agency can alone meet the totality of the needs that exist in a particular setting. It is essential to divide up the work according to a collective plan, share information, and work together for the wider good. Without coordination, different agencies may take divergent, even clashing approaches. They may also cause unintended harm by, for example, duplicating efforts, causing assessment fatigue, or concentrating efforts on the capital city, with little attention given to remote areas where the needs are greater.
Despite these considerations, coordination remains the Achilles heel of the humanitarian enterprise (Slim, 2015). Poor coordination often stems from the competition that is built into the humanitarian architecture. Since agencies compete for funding, access to information and affected populations, and status, they often keep assessment information to themselves, raise funding for themselves, and emphasize their own achievements and needs rather than collective achievements and needs ). Thus, ongoing attention is needed to enable inter-agency cooperation. In this respect, it is useful to review briefly key themes and findings from research on cooperation.

The Science of Cooperation
Extensive research, including studies that use quasi-experimental designs and include comparison groups, attests to the positive effects of cooperation. In both laboratory and field experiments, cooperation by members of a group frequently leads to better performance in solving diverse tasks than does competition in which group members are pitted against each other (e.g., Deutsch, 1949;Johnson, Maruyama, Johnson, Nelson, & Skon, 1981;Stanne, Johnson, & Johnson, 1999). In schools in the USA and other countries, educators have used cooperative learning strategies in which small groups work together on a learning task. Reviews of cooperative learning have reported that cooperative learning increases learning and academic achievement (Curry, Damicis, & Gilligan, 2011;Johnson & Johnson, 1989;Slavin, 1996). Also, cooperative learning strategies can improve inter-group relations (Johnson & Johnson, 1989), including relations between majority and ethnic minority groups (Slavin & Cooper, 1999). Extensive evidence indicates that intergroup cooperation is an effective means of improving group relations and reducing hostility and negative attitudes across groups that had been in competition (Deutsch, 1973;Kelman, 2002;Sherif, Harvey, White, Hood, & Sherif, 1961).
However, cooperation strategies do not work better than competitive strategies in all contexts. Even early in the study of cooperation, some findings indicated that both individual and group competition increased levels of both motivation and performance (Julian & Perry, 1967). Also, meta-analyses have found that the effects of cooperation and competition depend on the nature of the task, in particular the means interdependence of the task (Stanne et al., 1999;Tauer & Harackiewicz, 2004). A task is means interdependent when it calls for or requires individuals or groups to coordinate their efforts in completing the task. For example, when two children get to play when they have shoveled snow from the family driveway, they can complete their work by cooperating, that is, by both shoveling and not duplicating or interfering with each other's efforts. This cooperation is mutually beneficial because it enables both children to finish shoveling more rapidly, allowing more time for play. In contrast, an independent task (such as subtracting 1 from 3 or riding a bicycle) does not require individuals or groups to coordinate their efforts. In general, cooperation yields better performance on tasks that are means interdependent, whereas competitive strategies are favored on independent tasks (Stanne et al., 1999). Thus, it is very difficult to say whether cooperation or competition in general yields superior results.
Cooperative strategies can be difficult to use and may have unintended, negative effects. Cooperative strategies are difficult to use in situations such as protracted political conflict (Bar-Tal, 2000;Maoz, 2011) or strong racism. Also, the use of cooperative strategies can produce harmony across groups that differ in ethnicity and power. However, if they overemphasize harmony, cooperative strategies can serve to maintain the status quo, decrease the salience of minority identities, and reduce motivation to work for social change in patterns of inequality (Dovidio & Banfield, 2015).

Implications for Cooperative Efforts to Strengthen Policies Related to Children
The science regarding cooperation has valuable implications for developing more cooperative approaches to using research findings to influence policy. First, the cooperative strategy should emphasize the means interdependence of the task at hand-enabling children's well-being. This could be accomplished by pointing out that policies relating to children should be based on policy-relevant evidence. If the researchers and policy leaders cooperate in the development of the research, the research will more likely be seen as policy relevant by the leaders who are positioned to change policies related to children. At the same time, the researchers and policy leaders have different yet complementary roles, with the former conducting the research and the latter working to ensure that quality children's policies exist and reflect the best available evidence. These complementary roles are interdependent and reflect a common, shared commitment to children's well-being. If the roles are implemented with keen respect for interdependence and the value of joint decision-making, a spirit of collective ownership can develop that enriches and deepens the collaboration. Second, careful attention should be given to power inequities and the humanitarian imperative Do No Harm. The voices of large, international agencies should not drown out the voices of smaller, national agencies that have less power and visibility but have a grounded understanding of children's lives and situation. Also, the emphasis should be less on harmony between the different actors and finding areas of agreement than on learning deeply about children's situation and doing what is in children's best interests. These themes, which likely apply to wider issues of humanitarian and development coordination, will be evident in the case study presented below.

The Case Study
The case study features the cooperative, inter-agency processes around research to improve children's protection and well-being. Before discussing the cooperative processes, it is useful first to provide an overview of the research itself.

The Research
The research was conceptualized by the Inter-Agency Learning Initiative (ILI) on Community-Based Child Protection Mechanisms and Child Protection Systems. Save the Children (via Sarah Lilley) was selected to coordinate a global Reference Group, with one of its members-the Columbia Group for Children in Adversity (via M. Wessells)-serving as its technical arm. The ILI drew on the findings of a global review of community-level processes and mechanisms for protecting children . The review reported that the dominant approach was nongovernmental organization (NGO) led or facilitated and used a top-down approach. Frequently, NGOs led child protection assessments and then helped to establish in communities Child Welfare Committees (CWCs). The CWCs consisted of groups of local people who monitored violations against children, reported criminal violations to authorities, and took steps to prevent violations at community level. The review, however, found that such approaches tended not to be highly effective or sustainable in large part because local people did not take ownership of them (Wessells, , 2015. Seeing them as NGO projects, communities tended neither to take responsibility for the CWCs nor to use their own creativity and resources to support them.
The research described below was designed to develop and test a bottom-up, community-led approach, which complements more top-down approaches (Wessells, 2015(Wessells, , 2018. This approach placed power in the hands of communities, enabling them to identify the key harms to children, select which one they wanted to address, plan how to address it, and then implement the community-led action while monitoring and periodically taking stock of the action. The research took place in Sierra Leone, one of the world's poorest countries. In Sierra Leone, policies regarding children's protection had placed relatively little stock in community-led action on behalf of children. Via its 2007 Child Rights Act, the emphasis had been on establishing and supporting CWCs at village level.

Design
The study consisted of action research (McIntyre, 2008) in which communities defined a problem, developed and took steps to address it, and evaluated and refined their approach. The design entailed a two-arm randomized cluster trial (see Wessells, 2015Wessells, , 2018 for detailed information on the trial and the intervention). In each of two districts, there were two clusters of three villages which were approximately similar in regard to size, child protection issues, socioeconomic status, and services available. On a random basis, one cluster in each district was assigned to the intervention condition, with the other clusters serving as comparison groups. Quantitative, survey measures were taken from all young people 13-19 years of age in the communities at the baseline (pre-intervention) and midline (1.3 years after the intervention had begun). Also, narrative, qualitative data were collected before the intervention and at the midline. A planned end-line data collection was scuttled by the Ebola crisis.

Stages
The work was conducted in multiple stages, the first of which involved ethnographic learning about the communities. A key finding (Wessells et al., 2012) was that local people did not report harms to children to CWCs but overwhelmingly preferred to handle them by means of traditional mechanisms. For example, if a girl became pregnant, the girl's family developed a "compromise" with the boy's family, resulting in the boy's family paying for the girl's lost education and the boy marrying the girl. Asked whether they would report a criminal offence such as rape of a child to authorities, nearly all the participants said "No," indicating low confidence in the police or the magistrates and concerns about corruption and likely inaction. Overall, people viewed the main harms to children as teenage pregnancy out of wedlock, out of school children, heavy work, and maltreatment of children not living with their biological parents. The ethnographic findings were fed back to local collectives in a participatory manner, with space created for reflection about "What are we [community members] going to do about these harms to children?" In Stage 2, the planning phase, with the help of a facilitator who enabled a slow process of dialogue and also inclusivity, the communities selected which harm they wanted to address in a manner that linked them with formal stakeholders. The communities chose to address teenage pregnancy, a third of which was owed to sexual abuse (Coinco, 2010) and which typically forced the young mothers into survival sex. With significant leadership from young girls and boys, the communities designed a community action to prevent teenage pregnancy. Broadly, the community action consisted of family planning, sexual and reproductive health, and life skills, implemented in a highly participatory manner (Wessells, 2015).
In Stage 3, the intervention communities implemented their community-led action, with ongoing monitoring and periodic reflective evaluations in which they stepped back, identified strengths and weaknesses of the community action, and made any needed adjustments. The action included activities such as collective dialogue and decision-making, child leadership and messaging via street drama and radio, peer education, support from local authorities, and linkage of communities with health services.

Results
The results at this stage were promising and featured high levels of community ownership and diverse signs of the intervention effects in addressing teenage pregnancy (Stark et al., 2014;Wessells, 2015). High levels of community ownership were evident in how many people volunteered their time and work, without material compensation, and regularly referred to the intervention as "ours," stating that NGOs and the government support them but do not lead the intervention. Many teenagers said they used contraceptives regularly, and teenage girls reported that because of the intervention, they said "No" more frequently to unwanted sex. Both girls and boys said that they had learned how to discuss and negotiate with their partners in regard to sex and also how to plan their sexual activities in light of wider, life goals.
In contrast to previous low use of health posts, many teenagers and parents in the intervention condition visited the health posts regularly for contraceptives or advice, and villages frequently invited nurses and other health staff to visit in order to educate villagers about puberty, sex, and preventing teenage pregnancy. A reduction in teenage pregnancies occurred only in the intervention condition, as confirmed by participant observations and interviews with health post staff, monitors, teenagers, and adults (Wessells, 2015). Also, the action process significantly improved communities' collaboration and linkage with the local health posts.
Unfortunately, the Ebola crisis erupted in Sierra Leone in August 2014, radically disrupting the intervention and data collection and life in general (Kostelny et al., 2016). Nevertheless, the promising results had a significant policy impact, due in no small part to the cooperative process employed.

The Cooperative Process and Its Policy Impact
The cooperative process was deliberately cultivated in all phases of the research as a means of influencing policy and practice. This section examines the nature of the cooperation between the research team and policy leaders at different stages and also considers how various challenges were handled through cooperative processes of dialogue, negotiation, and collective problem-solving.

Country Selection
Although the ILI had decided to conduct the action research in one country in each of West and East Africa, it had not initially chosen which countries to work in. Since the ILI recognized that achieving a national policy impact would be more likely if the United Nations Children's Fund (UNICEF) encouraged and supported it, an invitation to consider participation was sent to the UNICEF's country offices in each region. The Sierra Leone and Kenya offices of UNICEF indicated their interest in supporting the action research. Sierra Leone had emerged from a decade-long, horrific war only in 2002, whereas Kenya offered a somewhat more stable, developmental setting. The Sierra Leone case study is featured below. The references to UNICEF from here on in the paper pertain to the UNICEF country office in Sierra Leone.

The Design Phase
The broad research design expressed the common vision of the ILI for strengthening the evidence base and simultaneously developing a more community-owned and community-led approach. This vision resonated well with UNICEF, which had, as a result of a previous child protection system mapping, learned about the need for greater community involvement and deeper learning about community processes. Intentionally, however, the design was not complete and something that was imposed by outsiders. There was a need to adapt it to the particulars of the Sierra Leone context and in particular to select a limited number of sites where the work could be most illuminating. Wanting to avoid imposing outsider ideas, the ILI set out to select sites collaboratively, through open dialogue with national partners.
Appropriately, UNICEF recommended that the ILI talk initially with the National Child Protection Committee (CP Com), which was chaired by the Ministry of Social Welfare, Gender, and Children's Affairs (MSWGCA), and the members of which were international and national NGOs whose representatives were mostly Sierra Leonean. Late in 2010, M. Wessells and S. Lilley met with the CP Com, explained that they came in a spirit of co-learning, outlined the action research and its purpose, and invited discussion about two regions where the action research should be conducted. This discussion was pivotal in helping to cement a common vision that efforts to strengthen child protection practice and policy should be guided by quality evidence. Common ground for cooperation arose also from the mutual appreciation that Sierra Leonean communities have rich cultural traditions that include many practices that support children's well-being.
The discussion of priority sites for the conduct of the action research generated an initial list of many options. However, it was explained that in order to learn in depth, the action research could not be done in a large number of areas. Only two geographic regions could be included, and that one reasonable way to approach selection would be to select sites that reflect the regional, ethnic, and linguistic diversity of Sierra Leone, which remains a mostly rural, agrarian society. This explanation sparked conflict, as members of national NGOs pointed out that intensive needs existed in all districts of Sierra Leone and suggested that it would be unfair to favor some over others. Seeking to handle this conflict in a constructive, respectful manner, the action researchers affirmed the widespread needs in Sierra Leone and noted the need for evidence regarding the effectiveness of a communityled approach. The team was not suggesting that a proven approach be extended to some communities but denied to others. They suggested that if the approach worked, the approach could be taken to scale and used in many different districts. As the dialogue continued, the CP Com members came to see the benefits of learning deeply in two sites, and the Minister himself indicated that they could subsequently extend the needed support to other areas. This conversation was important substantively, and it also helped to build a norm of constructive conflict management, which itself is key in taking a cooperative approach.
The CP Com then suggested that it would be useful to conduct the action research in both the southern, Mende-speaking region and in the northern, Temne-speaking area of Sierra Leone. The agreement of the research team with this approach cemented the grounds for cooperation, and the CP Com asked to be kept updated on the site selection and to periodically meet to discuss findings as they came out. Since additional information and site visits were needed to identify the specific sites within each area, various agencies offered to help make connections with particular district offices and enable the local visits. This multilevel dialogue process led the research to select particular Chiefdoms within Moyamba and Bombali districts, respectively. A positive outcome of this site selection process was that the CP Com members began to take ownership of the research, seeing it as truly cooperative and "theirs."

The Ethnographic Phase
To strengthen a cooperative approach, an in-country Reference Group was also formed that, like the global Reference Group, was chaired by Save the Children and included UNICEF and NGOs such as Plan, World Vision, ChildFund, ActionAid, and Concern, among others. When the ethnographic research was conducted, two of the researchers came from the former agencies. David Lamin of UNICEF, who in many respects was the driving force behind the research in Sierra Leone, served as the Moyamba team leader, and a child protection worker for Save the Children became one of the researchers who lived in the villages. This arrangement deepened the cooperation between UNICEF and Save and the research team, and it marked a turning point for the researcher from Save, who said that this experience had transformed his understanding of communities and of the problems inherent in dominant NGO approaches.
A significant moment in the cooperative approach occurred when, in 2011, the ethnographic phase findings were presented to the CP Com, with the Minister of the MSWGCA as chair. In presenting the findings, the author felt apprehensive since the findings showed the limits of the approach taken over the previous 4 years by the MSWGCA, UNICEF, and CP Com. That approach was based on the 2007 Child Rights Act, which had mandated the establishment of a CWC in each village and envisioned an effective referral system for reporting violations against children to authorities. In contrast, the ethnographic findings showed that the vast majority of local people did not use the CWCs and other formal aspects of the child protection system, which they saw as embodying outsider values and approaches. Showing that people viewed child rights as harmful and preferred instead to use their traditional processes and mechanisms, the research indicated a significant disconnect between the formal and the nonformal aspects of the child protection system.
Fortunately, the Minister did not respond defensively and thanked the research team for its findings, saying that he had heard some about these kinds of problems and that the research had provided a more systematic picture of communities and the CWCs. This welcoming remark opened the door to honest group dialogue and reflection about the limits of a top-down approach, the grip of traditions on local communities, and how to engage with communities in more meaningful, effective ways. A key part of the discussion was the recognition that communities already do much to protect their children and that the agencies and workers around the table should support appropriate local efforts. In this manner, the group demonstrated its openness and commitment to learn from research findings and its willingness to reflect together about what was not working and how to improve their inter-agency approach. As people took ownership for the findings and discussed what more effective approaches might look like, the spirit of cooperation and co-learning grew richer, inviting additional action. Subsequently, UNICEF convened a workshop aimed at developing more participatory approaches to introducing child rights and ending the backlash against child rights that the ethnographic research had reported.

The Community-Led Planning and Action
Continuing the upward spiral of cooperation, the research team provided regular updates to the CP Com at regular intervals (every 6 months) as the intervention communities selected which harm to children to address, developed a plan for addressing it, and took action themselves to reduce the harm. These updates included collaborative learning about how to stimulate high levels of inclusivity and ownership via power sharing with communities and community efforts to reach out to hear the voices of the most vulnerable and marginalized children and families. These updates enabled significant problem-solving dialogue about how to achieve meaningful child participation, a priority that the global child protection sector has found difficult to achieve. This period also saw increased cooperation with UNICEF in influencing multiple partners. UNICEF organized periodic workshops with different NGOs to help share the findings, approaches, and learnings from the community-led work. Conducted in a reflective mode, these workshops not only shared what the research team was learning and how it worked but also invited each partner agency to share about how it engaged with communities and evaluated its work. Collective reflection on the agency sharing increased awareness of the need for greater power sharing with communities and for increased levels of community ownership and sustainability. These workshops boosted the desire of different agencies to learn how to support a more community-led approach.
UNICEF also played a key brokering role in influencing the Government of Sierra Leone. Because UNICEF worked closely with the MSWGCA and its Minister, it knew the main power dynamics within the Ministry, understood the key leverage points, and identified when the timing was ripe for action. On a regular basis, D. Lamin organized meetings between himself, the Minister, and M. Wessells. These meetings were mostly for purposes of updating about the action research, its approach, and its key findings to date. However, they also provided a steady message of encouragement to listen more deeply to communities, build on existing community resources, and enable and learn from community action on behalf of vulnerable children. These meetings were critical in increasing the support of the Ministry, which looked to UNICEF for vision and leadership on behalf of children. With UNICEF's encouragement of a more community-led approach, the Ministry could move with confidence and without image loss in shifting direction away from the top-down approach inherent in the previous Child Rights Act.
Cooperation with UNICEF was crucial also in managing the rapid turnover (every 6 months or so) in the Minister post, which risked a loss of continuity and the shift of the MSWGCA agenda in directions away from the community action approach featured in the research. UNICEF managed this challenge effectively by cultivating strong working relationships with mid-level managers who tended to stay in posts for several years. Regular meetings between mid-level managers, D. Lamin, and M. Wessells helped to engender support for the action research at multiple levels within the MSWGCA and to avoid gaps in support that might have developed otherwise.
A similar process occurred within UNICEF, which also underwent changes in its Representative and the Chief of Child Protection posts. Regular presentations and dialogues were organized within UNICEF to update key staff, including the Representative and the Deputy Representative, on the approach, findings, and value added of the community-led approach, and also to enrich the understanding and interest of its child protection workers. The latter was vital for insuring that bottom-up approaches got included in the national efforts to strengthen the child protection system. The meetings also enabled dialogue that helped to develop a holistic vision in which top-down and bottom-up approaches were seen as complementary elements of strengthening national child protection systems.

Toward Policy Change
The inter-agency research approach and findings converged with the findings of other studies (e.g., Child Frontiers, 2010; Krueger, Thompstone, & Crispin, 2013) that highlighted the importance of community action on behalf of children. By late 2015, the MSWGCA had developed and gained Cabinet approval for a new Child Welfare Policy. Consistent with this research, the policy emphasizes the importance of the government supporting community efforts to protect children, and it cautions against the tendency to add particular structures as instruments for protecting children.
It is inherently difficult to measure whether the action research in particular significantly influenced the new national Child Welfare Policy. However, three different Ministers of the MSWGCA told the author and UNICEF that they saw the research findings as guiding the formation of a new policy that placed greater emphasis on supporting community action to protect children. Also, two mid-level managers within the MSWGCA said that this research had been pivotal because it was cooperative, systematic, and able to document using a rigorous design the effects of community-led action on behalf of children. Of interest to the author was that when MSWGCA stakeholders discussed the action research, they spoke of it as if it were their own and a rich source of co-learning. Key UNICEF stakeholders confirmed independently that the action research and its cooperative process had contributed significantly to the development of the new policy.

Limitations
The cooperative approach described here for influencing national policy is likely useful in diverse contexts, yet it is not universally applicable. If, for example, a national government was engaged in or supportive of corruption or violations against children, a cooperative approach would be inappropriate. Further, the community-led approach used in the action research would not be appropriate in zones of armed conflict that are suffused with spies and tensions. In such environments, group discussions could be misperceived as political organizing and could lead to participants being killed.

Conclusion
In concluding, it is useful to revisit the points raised earlier regarding means interdependence, attention to inequalities, and the importance of adhering to the Do No Harm principle. The cooperative approach in this case study recognized the high levels of interdependence among different agencies and stakeholders involved in policy and practice related to children's well-being. Indeed, it illustrates that the human component-relationships and processes of cooperationcan be as important as are the technical aspects of the research. Begun early on and applied in a continuous manner, a cooperative approach builds the shared vision, the spirit of interdependence, and the sense of collective ownership that enables policy leaders to take research findings on board and use them to develop more effective child protection policies. Because the process worked through the CP Com, small, national agencies had a significant voice alongside of large, international agencies. Throughout the discussions, attention was given to avoiding unintended harm and respecting children's dignity and rights.
It is important to recognize the limitations of a case study approach (Yin, 2018), which does not enable wide generalizations. It would be a mistake to assume that the kind of cooperative approach presented in this case study would apply readily to many different contexts. Still, the strategy of early and ongoing cooperation warrants further research. This case study is offered in a spirit of inviting additional work aimed at documenting and evaluating approaches to enabling research on children to actually influence policy changes.
An important challenge for the future is to enable wider use of this kind of cooperative approach for purposes of achieving a policy impact. This wider use will require a more intentional effort on the part of researchers and policy leaders to build strong relationships, explicitly recognize the interdependence of research teams and policy leaders, and do the challenging, time-consuming work of cooperation. It will also require strong "soft" skills such as listening, building trust, negotiation, and collective dialogue, reflection, and problem-solving. More thought should be devoted to developing these skills as part of the effort to prepare the next generation of researchers who can strengthen policy and practice in regard to children's protection and well-being.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.