Content area
Social norms, the informal rules that influence behavior, play essential roles in shaping people’s behavior. Community-based norms-shifting interventions (NSIs) identify gender and other social norms linked to unhealthy behaviors and implement activities to promote collective change by encouraging communities to reflect on and question these norms. Though NSIs are gaining international traction in social and behavior change programming for health promotion, how change occurs needs to be clearly understood in African and other contexts. To build understanding and guidance for future NSI design, the applied-research Passages Project and collaborating non-governmental organizations in West and Central Africa conducted realist evaluations of four NSIs focused on adolescent/youth sexual and reproductive health, operating in Democratic Republic of Congo, Niger, and Senegal. The evidence base for the realist synthesis came from four quasi-experimental outcome evaluations and 19 rapid implementation studies, which confirmed the four program Theories of Change. The synthesis findings identified eight norms-shifting mechanisms common across NSIs: information provision; dialogical, experiential approaches; role modeling; safe spaces; within-community meetings; planned diffusion; cross-community meetings of change agents; and community-service linkages. NSIs directly, at times indirectly, engaged reference groups that uphold norms, explaining their theoretical roles operationally. These findings led to middle-range theory showing how NSI activities, mechanisms, and reference group engagement should, over time, lead to norms-shifting outcomes. Design implications include developing a fuller understanding of how program components, as norms-change mechanisms, lead to effects; being deliberate about when and how to engage reference groups; and recognizing systems complexity and the subsequent need for NSI implementation elasticity.
Introduction
Social norms, or the informal rules that influence people’s behavior, play an essential role in shaping behavior. Community-based norms-shifting interventions (NSIs) include gender and other social norms as a focus for social and behavior change (SBC) (1). They complement other health promotion and SBC strategies, such as those focused on knowledge, attitudes, agency, or structural supports, by encouraging communities to reflect on and question norms related to behaviors and promote collective change (1,2). Multiple theories exist about how norms, alongside individual and structural determinants, lead to behavior (1). NSIs are gaining traction in international development as a new programming area in SBC (1,3). However, a limited understanding remains of how programs influence norm shifting, including which SBC strategies foster norm shifting, how activities lead to normative change, and whether norm shifting leads to behavior change (2–4).
The Passages Project (2015–2022) aimed to advance the science, practice, and scale-up of NSIs serving adolescent and youth sexual and reproductive health (SRH). Passages hosted the Learning Collaborative to Advance Normative Change, a global community of practice to share and learn about social norms. These joint efforts synthesized much of the adolescent/youth SRH and norms literature, convened consultations, and developed papers on the measurement, practice, and scaling of community-based NSIs.
Passages used a realist evaluation approach tailored for community-based interventions and reflected a program-practical approach to the evaluation of NSIs to explore and build an understanding of how these types of interventions work to achieve change. The realist evaluations used a conceptually grounded collaborative evaluation approach, meaning they defined their approaches using SBC frameworks, theories of how norms influence behavior and shift, and evidence from previous pilots. By collaborative, we mean that researchers and practitioners came together to refine program theory, make sense of evidence gaps, apply methods, interpret findings and make sense of them for program adaptation, inform program scale, and build evidence for a global audience. Immediate practical use of findings was of priority (5).
Passages partnered with four non-governmental organizations (NGOs) in the Democratic Republic of Congo (DRC), Niger, and Senegal that had developed adolescent and youth-focused NSIs, with some outcome evidence of behavior change. The four realist evaluations were founded on collaboratively developed program Theories of Change (ToCs) to include norms-shifting and then systematically built understanding and evidence of how the NSI works by examining the relationship of context, mechanisms, and outcomes (6). Program ToCs were depicted visually as cause–effect chains with project activities grouped into components of similar activities, instead of examining pathways of each activity separately within a ToC. This provided a way to engage program stakeholders in thinking about how an intervention works without getting into a level of specificity that could obscure the discussion. Each realist evaluation designed monitoring, learning, and evaluation studies to gather evidence to address evidence gaps along pathways and assess outcome normative effects (5).
At the start of the synthesis of realist evaluations, the team noted the relative lack of published literature on NSIs for adolescent and youth SRH; therefore, this synthesis focused on the four Passages projects (7,8). We developed an innovative approach grounded in the principles of realism and realist synthesis to use program theory, consultations, implementation monitoring of the community response to programming, and evidence from each NSI realist evaluation to answer core questions about how the NSIs shift norms and behaviors (5), in the process developing explanatory meaning grounded in understanding how NSIs work (8,9).
Specifically, we sought to understand in a programming sense how NSI activities lead to norms shifting by answering the following three questions:
-
What are the common mechanisms across NSIs leading to norm shifting?
-
What are the commonalities among change agents who foster norms-shifting community reflections?
-
How do the NSIs engage reference groups? Furthermore, when do reference groups become norms-shifting agents supportive of new behaviors?
Methods and analytical approach
Description of NSIs
Table 1 describes the four NSIs included in this synthesis: Girls’ Holistic Development; Growing Up GREAT!; Masculinité, Famille, et Foi; and Husbands’ Schools. They operate in three country contexts, reach different adolescents and youth as primary change groups, and aimed for different results depending on life stage. Collectively, though, all used a combination of activities that (i) fostered a more enabling normative environment in peers, adults, and providers to improve young people’s options and choices, (ii) built adolescent, youth, and adults’ awareness, capacities, and agency to act on SRH, and (iii) improved adolescent and youth linkages to health and education services.
Table 1.
Description of the four norms-shifting interventions included in the synthesis.
| Senegal | DRC | DRC | Niger | |
|---|---|---|---|---|
| Primary change group | Very young adolescent girls | Very young adolescent boys and girls, in- and out-of-school | Young couples, first-time parents in urban church congregations | Married men (by extension, spouses) |
| Influencers (reference groups, power holders) | Girls: peers, youth, and adult leaders | Girls: parents, friends, big sisters, teachers | Women: friends, mother, husband, pastor’s wife, health worker | Men: male peers, model husbands |
| Setting | Rural communities in Southern Senegal engaging girls, families, reference group members, community-influential adults | Urban-poor neighborhoods in Kinshasa engaging girls and boys, families, teachers, health providers, reference group members | Urban/peri-urban Protestant congregations in Kinshasa engaging young couples, pastors, other influential congregation members | Rural communities around the country engaging pro-RH leaning men, reference groups and other influential community members |
| Theory of Change | Key components leading to change | Key components leading to change | Key components leading to change | Key components leading to change |
| Outcomes-behavior and norms shifts | • Young girls remain in school, marry later, have children later | • Young girls and boys respect and communicate with peers and adults, stay safe from pregnancy, STIs | • More couples use modern FP services | • More women use antenatal, delivery, and FP services |
DRC: Democratic Republic of Congo; Adol: adolescent; FP: family planning; RH: reproductive health; STI: sexually-transmitted infections
Evidence sources
Participatory workshops with staff and other stakeholders allowed defining/refining change theory to include norms and individual outcomes. The collaborative approach reflected realist values that fostered the co-creation of learning agendas and identified evidence gaps (5). Table 2 summarizes how evidence gaps were addressed with rigorous evaluation and smaller program implementation studies.
Table 2.
Building evidence and understanding of how norms-shifting interventions work: evaluation, implementation research, experiential mapping: 2016–2021.
| Girls’ Holistic Development | Growing Up GREAT! | Masculinité, Famille, et Foi | Husbands’ Schools |
|---|---|---|---|
| Overall program and evaluation research focus guided by program Theory of Change | |||
| Assess how poorly understood implementation effects lead to individual and community socio-cultural support for girls | Assess multi-level effects in gendered behavior and norms in home, school, and services environments | Assess implementation fidelity, diffusion of social influence, and behavior change in couples, community members, and congregation leaders | Assess the implementation effects of gender norms shifting, diffusion of social influence, and acceptability of RH services use in couples and communities |
| Quasi-experimental evaluation research studies assessing social norms and behavioral changes (n = 5) | |||
| • Cross-sectional endline survey in intervention and comparison villages to understand norms, behavior change, community cohesion, and collective action. In-depth interviews and focus group discussions with adolescent girls, grandmothers, and caregivers in select intervention villages to explore norms and behavior change pathways | • Longitudinal quantitative outcome evaluation with intervention and comparison of adolescent girls and boys to assess gender socialization, norms shift, and behavior change | • Cross-sectional baseline and endline surveys in intervention and comparison congregations with eligible, participating couples. Assess MFF intervention effects (norms change, partner violence, FP) for couples. Assess the diffusion of norms and behavior change with congregation members | • Qualitative research comparing model husbands, their wives, and non-model husbands and wives and community leaders in pilot and scale-up communities to understand whether and how Husbands’ School activities lead to shifts in individual and community gender dynamics and male engagement in RH |
| Targeted, primarily qualitative rapid studies and reviews of monitoring data (n = 19) | |||
| Norms assessment | Norms assessment | Norms assessment | Implementation fidelity |
| For details, see: https://reproductive-health-journal.biomedcentral.com/articles/10.1186/s12978-021-01295-5 | For details, see: https://www.irh.org/interventions/growing-up-great/ | For details, see: https://www.irh.org/interventions/transforming-masculinities/ | For details, see: https://www.irh.org/resource-library/husbands-school-niger/ |
RH: reproductive health; MFF: Masculinité, Famille, et Foi; FP: family planning; GHD: Girls’ Holistic Development; GUG!: Growing Up GREAT!; HS: Husbands’ Schools
Analytic approach
Concept mapping
As the four realist evaluation studies closed, the cross-project analysis of NSI theory and realist evaluation findings began during a 2019 workshop with NGO project staff implementing the NSIs. Examining the change pathways in the program ToCs, we discussed, analyzed, and compared their Context–Mechanism–Outcome patterns (5). We were particularly interested in the set of preliminary findings of common change mechanisms that emerged to build understanding of what constitutes good NSI program practice of community-based NSIs. Outcomes, defined in early program theory development, focused on SRH, with behavioral and norm outcomes reflecting the life stage of primary participants, whether very young adolescents or married youth. All NSIs were community-based and explicitly linked to services.
These four interventions were grouped together in the synthesis of realist evaluations because of common contextual factors, while recognizing that there are also contextual differences. The common contexts we considered important in bringing together these four programs are: being implemented in communities where: the opinions, knowledge and values of elders, religious leaders, family members influence individual actions in SRH; the relative power inequity between youth and parents/elders might influence youth choices and intended behaviors; intergenerational and across-gender discussions of SRH are taboo; men hold greater decision-making power in SRH and over family life; and where young people are in significant life stage transitions (e.g. unmarried, newly married, first time parent).
Synthesizing change mechanisms and refining a common program theory
A four-person team worked to further specify the primary mechanisms identified in the workshop and explore the role of reference groups. The team used an operational definition of mechanisms established in realist and implementation science to bring comparative rigor to the analysis (10,11). Change mechanisms are project elements that lead to changes in attitudes, behaviors, and intentions of those who engage. Operationally, mechanisms are (a) activities implemented by project social change agents, and (b) the resulting new ideas and changes in reasoning seen among those who participated in the NSIs (10). We aimed to understand commonalities and differences in mechanisms related to program outcomes with findings in the different studies (Table 2), which had been designed to develop an understanding of the relationship between component implementation and their observed effects.
Project activities were grouped into components (Table 1) to facilitate program implementation analysis. We examined the roles and engagement of social change agents as the human driver of change mechanisms. The team used commonly accepted characteristics of social change agents to bring comparative rigor to the analysis: their actions are broader and deeper than knowledge transfer. They conduct activities with community members that promote knowledge building and critical reflection of social norms via open dialogue that expands social network linkages and increases social cohesion (12). We aimed to understand commonalities across NSIs in discussions with NSI implementers, specifically the types of change agents they engaged, and what distinguishes a social change agent, that is, the characteristics of agents who support social change processes.
Synthesizing the role and engagement of reference groups as social change agents
The team addressed another key question: how to consider reference groups (sometimes called influencers in SBC programs) in the relational mix (Table 2). Reference groups uphold normative expectations of primary change groups; however, how they interact with NSI activities needs to be better understood. Reference groups for specific behaviors were identified as the Passages–NSI collaborations began (Table 1) through social norms formative assessments (Growing Up GREAT! and Masculinité, Famille, et Foi), past project action-research studies (Girls’ Holistic Development), and NGO field workers’ social–cultural understandings of community influencers (Husbands’ Schools). Reference group types were identified in norms assessments via a series of questions asked of primary group members on social influence, for example, from whom do you seek counsel?, who advises you on issues relating to X behavior? (13). Those most frequently cited were considered more influential. With this information, we could analyze ToC change pathways to identify where and when reference groups show up during implementation. This complemented observational information gathered from NSI field staff on when and how reference groups were exposed or became involved with the NSI activities and their effects.
We included several time-related assumptions which added nuance to the synthesis and a more realistic reflection of the contextual complexity of change processes. We assumed that social change mechanisms operate on a continuum over time rather than an on/off switch (10). Such assumptions are supported by social change theory, particularly EM Rogers’ Diffusion of Innovation Theory, which reflects how new ideas enter and diffuse into a community.
Confirming results of the synthesis with NSI implementers
The team held discussions with the management and field staff of the four NSIs to validate and provide additional inputs to the findings, proposed middle theory, and scale-up design recommendations.
Results
We share the main findings, beginning with a brief discussion of social change agents and how reference group members assumed social change agent roles. We then discuss activities and underlying norms-change mechanisms common across the four NSIs, sharing NSI examples.
Social change agents
Two types of change agents linked to projects are common across the four interventions. The first type, Staff Change Agents, work within the project team. The second type, Community Member Change Agents, are project volunteers who lead community activities. Both types reflect characteristics of social change agents per our operational definition above. They are trusted actors who accompany a community in reflection that leads to new ideas, attitudes, consideration of norms, and, over time, changed behaviors.
Reference groups as social change agents
NSIs aim to shift the ideation of reference groups so that they can foster change within primary change groups. Reference group types common across NSIs include peers and family members. However, social norms assessments also identified influential people less visible to NSI staff. For example, younger adolescent girls in Growing Up GREAT! sought advice and counsel from teachers and ‘big sisters’ in their neighborhoods, and boys sought advice from teachers and grandmothers.
Projects engage in different ways with reference groups and at different times in project implementation: directly as NSI social change agents as the project began, as participants in NSI activities, and indirectly via planned diffusion. Table 3 offers examples of each of these strategies.
Table 3.
Examples of how norms-shifting intervention (NSIs) engaged reference groups.
| Form of engagement | Example | Study and monitoring findings |
|---|---|---|
| Directly, as social change agents | Girls’ Holistic Development sought forward-thinking grandmothers for Grandmother Leader training. In turn, the Grandmother Leaders, acting as change agents, reached out to other older women in the community to increase solidarity and capacity to advocate for girls’ rights | Over time, the project-supported social change agents (Grandmother Leaders) evolved into reference groups for others in the community. |
| Directly, as participants in NSI activities | Growing Up GREAT! directly reached parents/caregivers because they represent an important reference group for younger adolescents. Parents/caregivers participate in video sessions modeling gender-equal parenting by local families, followed by discussion. These project participants/reference group members undergo personal and collective transformation | Parents/caregivers began to support norms shifting publicly, reaching out to other parents/caregivers to share their learnings and advocate for other parents/caregivers to practice more gender-equitable behaviors towards younger adolescents |
| Indirectly via diffusion | All NSIs employed planned diffusion of new ideas and behaviors to reach others in the community. Planned diffusion could reach new reference group members via participants who attended NSI activities | Men living in communities where Husbands’ Schools operate became interested in reproductive health and male engagement after listening to comments by peers who heard about or attended school outreach activities. Parishioners and pastors participating in Masculinité, Famille, et Foi shared their experiences with peers in other congregations, who became interested in furthering gender discussions in their congregational communities |
Norms-change mechanisms shared across NSIs
Change mechanisms are within project activities and initially implemented by project-supported social-change agents. The eight change mechanisms in Table 4, identified by NSI staff in the 2019 workshop and then further analyzed, reflect common mechanisms used by the four NSIs to foster norms-shifting. Following the operational definition of a mechanism, the table shows project components (groupings of similar activities) implemented by social change agents (column 1) explicitly designed to shift participants’ ways of thinking or reasoning about an issue (column 2). The term ‘reasoning’ is used broadly and contextually to include cognitive, emotional, and other ways a person’s reasoning shifts. Mechanisms are explained through illustrative examples drawn from studies and staff observations (column 3).
Table 4.
Common norms-shifting intervention (NSI) mechanisms: Component actions and related changes in participant reasoning.
| Mechanisms | Examples | |
|---|---|---|
| Component actions by change agents | How they change the reasoning of participants | |
| Information provision | Improves the quality of critical reflection on norms and behaviors | Masculinité, Famille, et Foi: gender champions and pastors provided FP information via facilitated reflective dialogs on masculinities and men’s roles in FP and intimate partner violence with couples and during sermons |
| Dialogical, experiential approaches | Allows critical reflections grounded in real life | Growing Up GREAT!: younger adolescent clubs and parent video-session discussion and reflection on one’s life (experiential) allowed internalization and incorporation of new gender-equitable ideas by younger adolescent girls, boys, and their parents/caregivers |
| Role modeling or testimonials | Demonstrates publicly new attitudes and behaviors and breaks norms on appropriate behavior | Husbands’ Schools: model husbands (school members) were trained and shifted their attitudes and community roles. They became public advocates for RH services use and discussed with other husbands in their homes. Some even accompanied their wives to services |
| Using safe spaces | Permits a frank examination of sensitive issues | Growing Up GREAT!: VYA club settings and parent video sessions created safe spaces with peers to discuss sensitive topics of puberty, reproductive health, family planning, adolescent wellbeing, and gender equality in teen and adult actions |
| Within-community meetings (intergenerational, same-sex, mixed-sex gatherings) | Equalizes information access; fosters shared understanding across community groups and different age/sex/social groups | Girls’ Holistic Development: staff organized meetings – Days of Solidarity, Intergenerational Forums, Grandmother Appreciation – which engaged leaders, parents/caregivers, girls and boys, and health and education providers in discussions, leading to a shared understanding of issues and support for girl health and wellbeing actions |
| Change amplification mechanisms | Examples | |
| Component actions by change agents | How they amplify change in reasoning of participants | |
| Planned diffusion | Maximizes the reach of social change mechanisms at a community level | Husbands’ Schools: model husbands shared RH information and new ideas with their wives, who then diffused RH information and ideas of men’s support for RH services to their peers. This approach was critical since men were not socially allowed to interact with women |
| Cross-community meetings of change agents | Creates conditions for solidarity and reciprocity among social change actors | Girls’ Holistic Development: annual meetings of Grandmother Leaders and stakeholders across villages allowed sharing ideas about school retention, delayed marriage, and childbearing, fostering regional solidarity for pro-girl actions |
| Community-service linkages (HS and MFF for health services; GHD for schools) | Allows achievement of service behavior aims, an endpoint of social change | Masculinité, Famille, et Foi: gender champions (change agents) included local health workers in reflective dialogue sessions during congregation meetings for couples to create interpersonal relationships between congregations and providers, encouraging visits to seek FP services |
FP: family planning; RH: reproductive health; VYA: very young adolescents; HS: Husbands’ Schools; MFF: Masculinité, Famille, et Foi; GHD: Girls’ Holistic Development
The mechanisms acted throughout the project (not as one-time-only events), with some people engaging with them earlier and others later. The definition of change mechanisms in Table 4 helps clarify how NSIs differ from projects seeking to increase factual knowledge to improve services uptake or other behavior change but not aiming to shift community norms.
The social change mechanisms at the top of Table 4 use a variety of program-initiated sparks to foster social change.
-
They catalyze critical reflection processes to elicit individual reflection on emotions, values, experiences, and behaviors and to generate discussion, understanding, and consensus-building.
-
They use some form of behavioral modeling where change agents and community members model the behaviors they seek from the community.
-
The location where interventions occur helps create spaces for frank discussion, thus catalyzing change mechanisms. Locations included after-school clubs, congregations, and other community gathering sites.
The three mechanisms at the bottom of Table 4 amplify the effects of the above change mechanisms.
-
All ToC reflected planned diffusion processes to spread new normative possibilities and model behaviors promoted by the project to other community members not directly reached by project activities. Qualitative evidence indicated that diffusion of new ideas was occurring, which reflects social diffusion theory as a pathway to health behavior changes (14). NSIs might structure organized diffusion by asking NSI participants to speak to a spouse, family, or congregation. It could also be less structured, such as local media reports or public discussions by local leaders.
-
Cross-community meetings of social change agents. These regularly-occurring meetings create interpersonal and inter-community linkages that foster solidarity and increase motivation and support among change agents.
-
Linking services to NSI activities. While NSI may foster new ideas and help concretize intentions to act, people may need help figuring out what to do next. A specific linkage to services can increase the odds of more people acting on their choices.
Discussion
Our synthesis of realist evaluation findings builds a common understanding of the relationships between activities, contexts, and expected norms and behavior change outcomes. We present a summary of results, followed by a proposed middle theory and design implications for future NSIs.
Summary of findings
NSIs play essential roles in health promotion, especially when norms influence health behavior. NSIs aim to influence community-level norms about what is expected and acceptable behavior of its members, which, when aligned, creates social environments supportive of behavior change. Community-based NSIs focus less on personal knowledge and self-efficacy improvements and more on group dialogues, including intergenerational conversations, around normative expectations and their link to health outcomes. Early in the Passages Project, research led to nine characteristics of successful NSIs: seeking to affect community-level change, engaging people at multiple levels, creating safe spaces for critical reflection, using organized diffusion, being grounded in culture and community value systems, addressing power imbalances, accurately assessing norms that influence behaviors, addressing misconceptions, and introducing new normative possibilities (2). The synthesis of realist evaluations deepened these initial findings. Such distinctions imply different approaches to program design, strategies, and activities.
The different NSIs shared eight norms-shifting mechanisms. The change mechanisms’ modus operandi included interpersonal critical reflection processes, modeling new attitudes/behaviors, and planned diffusion to spread new ideas and actions to the larger community. Outcome studies (Table 2) indicate that all NSIs deepened the reach and community or population-level exposure to new thoughts and actions, facilitating changing reasoning and support for normative and SBC shifts.
Social change agents were the driving force of change mechanisms. While initially project-affiliated and supported, new community change agents emerged later in implementation, including reference group members that we assume are motivated by information and ideas emanating from project activities.
The synthesis helped clarify the roles of reference groups in NSI programming. The rapid studies show that even if reference group members identified by the NSIs are not initially aligned with new ideation aiming for improved health outcomes, NSIs can shift the reasoning and actions of reference group members, who then help change the logic, behaviors, and norms of the primary change group, other reference groups members, and other community groups.
Proposed middle theory
Based on this synthesis of realist evaluations, Figure 1 summarizes our middle theory of how NSIs should lead to norms shifting. The NSI activities implemented over the project’s life (top arrow) aim for an outcome (Outcomes square) that reflects a community’s adoption of new norms and behaviors. The community context changes in response to NSI activities and other factors (bottom arrow). NSIs operate at this project implementation/community reaction intersection: the helix in the middle grows larger over time to reflect this dynamic and deepening intersection of NSI activities and changing community context.
Figure 1.
How norms-shifting interventions (NSIs) foster social norms shifts at the community level.
The NSI-focused effects above the helix reflect the growing number of new change agents, individuals who from their participation in project-initiated reflection begin to speak and act in ways supportive of the NSI aims, and, over time, the involvement of reference group members. The community-focused effects below the helix show the deepening changes in community reasoning over time. To offer a more concrete explanation, NSI activities and skilled change agents directly reach a subset of community members, triggering reflections on new ideas. Community members react to NSI activities and interpret them in ways that lead to new reasoning (new beliefs, attitudes, behaviors, and normative expectations). The ideas spread to others in the community, who also interact and interpret new ideas in ways that resonate with them. Expanding interactions and feedback loops grow between members. In response to these changes, reference groups’ influence grows to support changes. This expanding web of support for new ideas and normative expectations helps create tipping points of solidarity for new behaviors and normative change.
Implications for designing community-based NSIs
The findings on mechanisms, social change agents, reference groups, and time lags needed to support change processes have several implications for designing NSIs. (1) Ensure a complete hypothesis of the norms-change mechanisms. How an NSI conceives norms-change mechanisms can strengthen the design, implementation, monitoring, and learning of NSIs. Not all activities affect participant reasoning on norms and behaviors, but they may create an enabling environment for this. (2) In addition to building NSI staff and community change agents’ technical and interpersonal skills, it is essential to take time during base training and later on-the-job moments for personal reflection and group discussion by change agents regarding their personal values, norms, and biases. The resulting clarity of intention of NSI aims by change agents fosters coordinated norms-shifting during NSI implementation. (3) Understand from the beginning who are key reference groups, and then decide how and when to engage them. Reference groups influence behaviors of primary change groups; understanding how a particular reference group can influence behaviors and which norms they uphold is vital for program design. Being purposeful in which reference groups a project works with and how they are engaged can be more meaningful and facilitate feedback loops and the development of new community change agents. 4) Allow for design elasticity that supports community change over time. NSI effects deepen in a primary change group, reference groups, and the community. Over time, as more community members agree with values and norms, others explore and uptake them. Systems shifts that occur imply that projects may need to reduce one strategy and embark on new strategies or reflections to support further changing understandings and normative expectations.
Strengths and limitations
The findings are relevant to NSI-type programs working in similar contexts and with similar population groups, in our case those NSIs working to improve SRH of adolescents and youth. Further research will help understand how consistent NSI mechanisms are across other-sector outcomes and population groups. The Passages realist evaluation and synthesis approach involved an innovative adaptation of realist evaluation to be program-practical, conceptually and realist-values grounded, and collaborative to ensure context was included in understanding evidence and program directions. Because of this focus, some of the complex and detailed approaches used in many realist evaluations were not part of the approach we used.
Conclusion
Our findings contribute to middle-level theory that integrates high-level social norms, behavior change theory, and empirical evaluation research and is valuable across social development sectors, not only health. It clarifies how projects can systematically engage reference groups in social change efforts. Interrogating the four program theories of change was crucial in synthesizing realist evaluations. These important learnings offer a new way of understanding norms shifting and related program design while considering the community context that can allow for, limit, or interact with planned norm-change strategies. Each change affects another, possibly magnifying this or other effects or leading to feedback loops for collective change, contributing to sustained community-level shifts.
We are grateful for and acknowledge our close collaborations in Niger with Issa Sadou (United Nations Fund for Population Activities (UNFPA)– Niger) and Mohamed Haidara (SongES), the developers of the Husbands’ Schools concept and its male engagement approach, who continue to support school refinement and expansion in Niger and West Africa. In Senegal, our close collaborators include Mamadou Coulibaly and Judi Aubel (Grandmother Project: Change Through Culture), the developers of the Girls’ Holistic Development (GHD) concept and its grandmother-inclusive approach, who continue to support GHD refinement and expansion. In DRC, our close collaborators include Frankie Quirke (Tearfund) and Ospey Nzagire (Église de Christe au Congo), who led the adaptation of Transforming Masculinities from Eastern Congo to urban Kinshasa. Our close collaborators of Growing Up GREAT! include Pierrot Mbela, Jennifer Gayles, and Brad Kerner (Save the Children), who led the adaptation and scale-up of Growing Up GREAT! in urban Kinshasa, and Eric Mafuta (Kinshasa School of Public Health), who led the series of rapid studies and participatory adolescent-led evaluations of Growing Up GREAT! and DRC’s Global Early Adolescent Study. We also thank the blind reviewers of different versions of the manuscript; their thoughtful questions and comments improved the final manuscript.
Declaration of conflicting interests
The authors have no conflicts of interest to declare.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the USAID-supported Passages Project, Cooperative Agreement Number AID-OAA-A-15-00042.
ORCID iD
Susan Igras https://orcid.org/0000-0001-9836-9479
© The Author(s) 2025