Evidence use in community-led learning ecosystems: An OTT Talks live

This OTT Talks Live episode, hosted by Enrique Mendizabal, discussed the findings and implications of a recent study on evidence use in community-led learning ecosystems, produced by OTT with funding from the Jacobs Foundation. The conversation featured Marcela Morales (formerly OTT, now Pouzas Philanthropies), Ross Hall (formerly Jacobs Foundation, now Fondation Botnar), and Yasmin Bekkouche (Jacobs Foundation).

Motivation behind the study

Ross Hall, who commissioned the report, explained that the rationale for the study stemmed from an urgent need to shift learning outcomes for young people. The central idea was to help every young person learn to thrive together, which requires developing not only literacy and numeracy but also a wider range of skills to create a better world, necessitating new evidence for holistic development.

Read the report here.

Read the learning brief here. 

He outlined three key reasons for commissioning the report:

• Wider learning environments: To support holistic development, young people need more diverse learning environments beyond traditional schooling (e.g., learning through projects, in nature, in the community, through play, making, moving, meditating). This also requires new types of evidence.
• Shift to learning ecosystems: Education systems need to become more organic and adaptable learning ecosystems, moving away from mechanical, standardised models. This involves greater ownership at the child, school, and community levels, fostered by trust and collaboration. This systemic shift demands better informed decision-making and, consequently, better evidence.
• Understanding evidence: The report was commissioned to gain a deeper understanding of both the evidence for learning ecosystems and the use of evidence within them.

Watch the full OTT Talks Live here:

Report findings: Community-led learning ecosystems

Marcela Morales, a co-author of the report (with Rachel Makokha and Emma Broadbent), presented the study’s findings. She emphasised that the review offers a closer look at how community-led learning ecosystems are being built by community actors working in partnerships with schools, civil society, and local institutions.

Definition and importance:

• What they are: Evolving networks of actors working in and around education, centring schools but extending beyond them to involve families, local governments, NGOs, museums, sport clubs, and local businesses. They are shaped by local priorities, relationships, and lived experience, not just formal institutions or national curricula.
• Where learning happens: Learning occurs everywhere – in classrooms, but also in public spaces, mentoring groups, and neighbourhood projects.
• Why they matter: They challenge three fundamental assumptions of traditional education models:
  • What learning is: Broadening focus beyond academic basics to include social, emotional, and practical skills like collaboration, agency, and identity.
  • Where learning happens: Breaking open school gates and activating community spaces as learning environments.
  • Who is involved: Shifting from a model where schools deliver and communities receive, to one where everyone (parents, teachers, local leaders, young people) becomes a co-creator in education, fostering true ownership.

Evidence in community-led learning ecosystems:

• Evidence is generated, interpreted, and acted upon within communities to inform decisions, adapt practices, and foster a shared understanding. It is seen as a process, not a product.
• Unlike linear, technocratic approaches, where researchers produce evidence for practitioners to implement, evidence in these ecosystems is collective, fluid, and not fixed, but deeply embedded in relationships and practice.
• The report identified four overlapping types of evidence:
  1. Research-based evidence: RCTs, academic studies.
  2. Pedagogical evidence: Generated by teachers through lesson planning, classroom feedback, and peer exchange.
  3. Sector-generated data: Attendance figures, standardised test scores.
  4. Practice-informed evidence: Local knowledge, lived experience, contextual reflection.
• Crucially, these initiatives continually blend and repurpose all types of evidence, challenging conventional assumptions about who counts as an expert, what constitutes valid knowledge, and where change begins. This calls for an expanded understanding of evidence to make visible the real learning that occurs outside of classrooms (e.g., internships, youth clubs, volunteering, family caregiving).

Real-life examples: Marcela shared examples showcasing the breadth of these ecosystems:

• Brookings Family Engagement Initiative: Combines research with active engagement, co-creating evidence with families and schools across 60+ countries and documenting the learning process itself.
• Education Endowment Foundation (UK) Research Schools Network: Supports peer-to-peer learning among educators, with schools acting as evidence brokers and adapting findings.
• Educació360 (Spain): A regional movement framing learning as 360 degrees, expanding access to play, arts, and enrichment activities, and positioning public spaces as learning actors. For more on Fundació Bofill’s approach, visit here.
• Remake Learning (Pittsburgh, US): A peer network of 500+ organisations linking schools with libraries, museums, and community spaces.
• Lively Minds (Ghana and Uganda): Focuses on early childhood development by training mothers as volunteer play leaders, blending qualitative and quantitative data to track child development and community engagement. Some ecosystems form around formal structures, while others emerge from community action, civil society partnerships, or creative coalitions.

Lessons for funders: Marcela emphasised the following lessons for funders:

• Supporting community-led learning ecosystems requires investing in relationships, capacities, and reflexive spaces that allow ecosystems to take root and evolve, rather than just scaling up a blueprint.
• Funders need to shift the way they think about funding and prioritise funding the interpretation of evidence, not just its generation.
• Making sense of evidence together is where transformation begins, requiring spaces, time, and roles for communities to reflect, interpret, and act collaboratively. This means embracing complexity, co-creation, and different forms of knowing.
• Funders must become partners in the learning process, not just in results.

Discussion on ecosystem diversity and knowledge brokers

Enrique Mendizabal inquired about the characteristics that distinguish more structured from less structured systems.

Marcela noted that these ecosystems come in diverse forms, some resembling traditional academic institutions, while others emerge as local, practical solutions. The diversity offers essential lessons.

She emphasised that every actor can be a knowledge broker – teachers, parents, local NGOs, even learners.

The “glue” bringing actors and structures together is lived experience and proximity to the community. Insights from parents or volunteers are just as valuable as those from formal research institutions. The key is having structures that allow these diverse sources of knowledge to come together and inform thinking.

The Jacobs Foundation’s approach and examples

Yasmin Bekkouche shared how the findings align with the Jacobs Foundation’s approach, emphasising that sustainable change cannot be externally imposed; interventions must be locally and contextually grounded.

Systemic Change Strategy: The Foundation’s strategy is built on three interconnected pillars:

1. National level: Strengthening evidence use in decision-making.
2. Intervention level: Supporting evidence-based teaching and learning.
3. Interconnection (subnational/local): Using pooled funding for locally driven, evidence-based initiatives at district or subnational levels, centred on adaptive learning ecosystems.

Community of change approach:

• This approach, core to the third pillar, is a three-stage process:
  1. Gathering People: Creating multistakeholder coalitions or “change communities” to form learning ecosystems.
  2. Discussing Evidence: Discussing a broad range of evidence and local insights.
  3. Co-creating Projects: Developing projects to transform local systems.
• A key insight is that “evidence needs relationships to drive change”. Evidence is not just provided but co-constructed, leading to greater use by communities and practitioners. The Foundation focuses on creating space for these relationships to foster co-creation.

Example: Community of excellence programme in Ghana:

• This programme, facilitated by UNICEF and T-Tel with support from the Ministry of Education and Ghana Education Service, applies the “community of change” approach. Yasmin stressed the importance of acknowledging multiple models and avoiding a normative approach.
• One supported model is “Managing for Learning”, a participatory methodology enabling local stakeholders (head teachers, district officials, students, community leaders, parents) to define priority issues and actions to improve learning.
• This leads to a “learning and transformation agenda” for each district, grounded in local evidence and insights. Diverse “change leaders” champion this agenda.
• Promising results include periodic accountability sessions where teachers share learner progress with the community and parents, fostering evidence creation, communication, and discussion.
• Since 2022, there has been a strengthening of local decision-making, with previously inactive District Education Oversight Committees now playing a crucial role through regular meetings, school visits, and data-informed decision-making.
• This approach, alongside others, will be scaled to 140 districts (2,000 communities) between 2025 and 2028 by the government, demonstrating a virtuous cycle between local initiatives and national reform.

Implications for funders

Ross Hall reflected on the implications for funders, emphasising the importance of cohering the field of learning ecosystems in three ways:

1. Funding learning ecosystems: Directly funding the creation of learning ecosystems from which everyone can learn.
2. Funding evidence generation about these ecosystems: Supporting the generation of evidence about how these model ecosystems come about, telling and sharing their stories.
3. Funding “weaving”: This is often the most difficult but critical aspect. Weaving involves holding reflective spaces where communities make sense of evidence and apply it in practice. This relational work requires complex coordination and collaboration within and between communities, as well as across different levels (local, district, national, and international).

Why “weaving” is hard to fund: Ross offered a deeper reflection on the challenges involved in financing the weaving functions.

• It is under-recognised as a type of work, often done voluntarily or mislabelled.
• It is undervalued because it’s hard to attach tangible outputs.
• It’s often seen as mere “facilitation,” despite requiring a sophisticated type of facilitation to navigate competing perspectives, diverse evidence types (scientific vs. intuitive, quantitative vs. qualitative), and maintaining a whole-system view.

Remaining questions and future learning

Yasmin acknowledged that many questions remain. While encouraged by the Ghana experience, there’s much to learn:

• Building evidence capacities: How to do this at local levels without overburdening communities.
• Meaningful evidence types: Understanding what types of evidence are most meaningful in different contexts and expanding the restrictive view of evidence.
• Long-term effects and model comparison: Needing more evidence on the long-term impacts of these approaches and comparing different models across contexts.
• Community support for evidence use: Learning what support communities truly need.
• Scaling while remaining flexible: A key challenge is how to scale interventions while maintaining flexibility and adaptability to context.
• Continuous learning: The work succeeds through continuous listening, discussion, and sharing experiences among practitioners and stakeholders.
• It’s important to embrace complexity in funding and measuring impact, and funders must be transparent about these questions.

Audience questions and panel reflections

1. Are caregiver/parent perceptions of evidence incorporated into the study and how?
   • Marcela noted that caregiver perspectives are central but rarely captured in conventional M&E. Parents are often involved in shaping learning environments and offering feedback, as seen in Argentina’s initiatives.
   • However, parents’ views on credible evidence often don’t align with standard education metrics; they prioritise children’s happiness, engagement, and confidence, while schools report on foundational skills.
   • Efforts to surface these insights through storytelling and parent groups exist but are underdeveloped, requiring better methods and investment. Ross added that this differing valuation of holistic vs. narrow academic outcomes is a persistent problem.
2. What is the role of arts-based approaches in generating qualitative evidence?
   • Marcela mentioned examples like Fundació Bofill in Barcelona and the use of storytelling, theatre, and participatory photography to surface young people’s perspectives.
   • These are powerful tools for generating insight, especially where conventional indicators fall short, but they are often undervalued in formal evaluation frameworks.
   • Validation requires shifting how rigour is defined, recognising that understanding lived experience needs different methods. Funders can support this by investing in mixed-method approaches, creating space for community-generated data, and valuing outcomes like belonging, confidence, and connection. This ties back to changing the understanding of education’s role.
3. What is the balance between aiming for scaling vs. local relevance?
   • Marcela stated that community-led learning ecosystems challenge the assumption that evidence must scale in the conventional sense. Local, context-specific evidence, drawing from practice and trust, can drive meaningful change.
   • Funders and policymakers should prioritise supporting the infrastructure that enables local actors to generate and adapt evidence, rather than just producing evidence for centralised use.
   • This means rethinking scale as “scaling capabilities” (the ability of communities to learn, adapt, and collaborate using evidence) rather than scaling programs. This approach is slower but more sustainable.
   • Ross added that part of this work involves creating spaces for authentic conversations about what is and isn’t working, having the courage to admit failures, and committing to change. Yasmin emphasised funders’ responsibility to support learning from ineffective approaches, shifting to more efficient ones, and even encouraging “failure” in transformative work, as long as learning occurs.

The conversation concluded with an emphasis on the ongoing nature of learning and the need for continued collaboration and discussion around these complex issues.