The Week 1 discussion resonates with my experience, particularly the tension between compliance and learning. For me, power runs through both. Who defines the questions? Who controls access to evidence? Whose explanation carries weight? And who has the authority to act on the findings?
My strongest practical lesson is to invest heavily in understanding the actual situation at the outset. Inception is substantive evaluative work. If the underlying institutional story is still unclear, negotiate more time before fixing the questions and methods. Under funding constraints, we have even less room to spend scarce resources answering the wrong questions.
In one multi-donor fund evaluation, the arrangements involved a governing body, a technical secretariat, a UN-hosted fund management function, implementing organisations and national counterparts. Delivery operated through grants and intermediaries. Leadership, control over resources, delivery responsibilities and formal accountability sat in different places.
Reading the organisational chart was insufficient. The important questions concerned how those arrangements worked in practice: who could approve or delay a decision, who was answerable for delivery, and how much influence national counterparts could exercise through intermediary organisations.
From inception onwards, I made governance, partnerships and ownership explicit areas of inquiry. Documentary review and stakeholder accounts helped me examine the tensions between fiduciary assurance and operational flexibility, and between formal national ownership and the practical organisation of delivery. During analysis, I distinguished these structural issues from implementation weaknesses. Otherwise, the findings could easily have settled on “coordination problems” without explaining what needed to change.
That distinction carried through to the final report. I brought the findings into the assessment of governance, partnerships and sustainability, including future responsibilities. I also distinguished evidence of activities and outputs from evidence of longer-term benefits. Institutional strengthening and investment readiness mattered, but they did not automatically demonstrate resilience outcomes.
The lesson is to understand the underlying story well enough to test it. Inception should produce a grounded explanation and a credible inquiry, while leaving room for subsequent evidence to overturn our initial understanding. It should not become an indefinite search for certainty.
A different example comes from designing monitoring arrangements for a national environmental partnership involving government agencies, financing bodies and technical partners, with delivery envisaged through agency-led projects. Here, I distinguished programme-level results from agency and project-level monitoring and connected the framework to existing systems. This provides a practical basis for reducing duplication and clarifying who should collect what.
But “use existing evidence” needs qualification. Administrative systems capture what institutions are required and equipped to report. They may tell us much less about customary landowners’ priorities, women’s influence over decisions, or communities’ experience of change. Reusing evidence can save money while reproducing gaps in whose experience counts.
For Week 2, I would therefore suggest:
Protect the early diagnostic work. Study the institutional history, funding modalities, decision-making relationships and existing evidence before settling the scope.
Negotiate fewer, consequential questions. Agree which decisions the evaluation must inform and what evidence those decisions require.
Use existing evidence critically. Assess its quality, coverage and omissions before deciding where additional collection is essential.
Preserve opportunities to challenge interpretations. Community accountability requires more than collecting testimony; people need meaningful opportunities to question how their experience is represented.
Connect recommendations to authority. Distinguish what management can change from what requires governing-body or donor decisions.
What does not work is retaining an expansive scope while quietly reducing the evidence needed to support it. Nor should we assume that an evaluator can resolve institutional power imbalances simply by documenting them.
Our own power also deserves scrutiny. We choose sources, interpret disagreement and decide what reaches the final report. That responsibility begins at inception. A rushed understanding can shape the entire evaluation, however polished the final product. Spending enough time to understand the situation is one of the most practical ways I know to protect both the usefulness and the credibility of evaluation under pressure.
RE: Evaluation under pressure : Strengthening evidence use, accountability and learning amidst funding constraints (Joint EvalforEarth-ALNAP discussion)
United States of America
Stephanie Jill Hodge
Posted on 21/10/2026
The Week 1 discussion resonates with my experience, particularly the tension between compliance and learning. For me, power runs through both. Who defines the questions? Who controls access to evidence? Whose explanation carries weight? And who has the authority to act on the findings?
My strongest practical lesson is to invest heavily in understanding the actual situation at the outset. Inception is substantive evaluative work. If the underlying institutional story is still unclear, negotiate more time before fixing the questions and methods. Under funding constraints, we have even less room to spend scarce resources answering the wrong questions.
In one multi-donor fund evaluation, the arrangements involved a governing body, a technical secretariat, a UN-hosted fund management function, implementing organisations and national counterparts. Delivery operated through grants and intermediaries. Leadership, control over resources, delivery responsibilities and formal accountability sat in different places.
Reading the organisational chart was insufficient. The important questions concerned how those arrangements worked in practice: who could approve or delay a decision, who was answerable for delivery, and how much influence national counterparts could exercise through intermediary organisations.
From inception onwards, I made governance, partnerships and ownership explicit areas of inquiry. Documentary review and stakeholder accounts helped me examine the tensions between fiduciary assurance and operational flexibility, and between formal national ownership and the practical organisation of delivery. During analysis, I distinguished these structural issues from implementation weaknesses. Otherwise, the findings could easily have settled on “coordination problems” without explaining what needed to change.
That distinction carried through to the final report. I brought the findings into the assessment of governance, partnerships and sustainability, including future responsibilities. I also distinguished evidence of activities and outputs from evidence of longer-term benefits. Institutional strengthening and investment readiness mattered, but they did not automatically demonstrate resilience outcomes.
The lesson is to understand the underlying story well enough to test it. Inception should produce a grounded explanation and a credible inquiry, while leaving room for subsequent evidence to overturn our initial understanding. It should not become an indefinite search for certainty.
A different example comes from designing monitoring arrangements for a national environmental partnership involving government agencies, financing bodies and technical partners, with delivery envisaged through agency-led projects. Here, I distinguished programme-level results from agency and project-level monitoring and connected the framework to existing systems. This provides a practical basis for reducing duplication and clarifying who should collect what.
But “use existing evidence” needs qualification. Administrative systems capture what institutions are required and equipped to report. They may tell us much less about customary landowners’ priorities, women’s influence over decisions, or communities’ experience of change. Reusing evidence can save money while reproducing gaps in whose experience counts.
For Week 2, I would therefore suggest:
What does not work is retaining an expansive scope while quietly reducing the evidence needed to support it. Nor should we assume that an evaluator can resolve institutional power imbalances simply by documenting them.
Our own power also deserves scrutiny. We choose sources, interpret disagreement and decide what reaches the final report. That responsibility begins at inception. A rushed understanding can shape the entire evaluation, however polished the final product. Spending enough time to understand the situation is one of the most practical ways I know to protect both the usefulness and the credibility of evaluation under pressure.