Build an Operational Learning Loop

You are an expert M&E advisor specializing in learning and adaptive management. Score the learning agenda I will provide using the rubric below.

SCORING RUBRIC - MEL Learning Agenda Quality Score each dimension 1-5 using these criteria:

DIMENSION 1: Strategic Relevance

• Score 5: Every question addresses a genuine uncertainty in the program theory (an untested assumption, a contested mechanism, or a context-specific unknown). Questions reflect input from program staff, partners, and intended users. None of the questions are already answered by existing evidence or are peripheral to program decisions.
• Score 4: No fewer than 80 percent of questions address genuine uncertainties tied to program theory. One or two questions may be lower-priority but remain plausibly useful. Stakeholder input is visible in question framing.
• Score 3: Half or more questions address genuine uncertainties. The remainder are interesting but not tied to a specific program assumption or decision. Stakeholder input partial or implied.
• Score 2: Fewer than half of questions address genuine uncertainties. Several questions are already answered by existing evidence, OR several are peripheral (descriptive curiosities rather than learning questions).
• Score 1: Questions are generic ("what works?") or unrelated to the program theory. No connection to documented assumptions or program decisions.

DIMENSION 2: Answerability

• Score 5: Every question is phrased so it can be answered with data the program will collect or can reasonably gather within budget and timeline. The data type, source, and method that will answer the question are implied by the question's wording (e.g., "how does X vary across rural and peri-urban sites" implies disaggregated quantitative data).
• Score 4: No fewer than 80 percent of questions are answerable with available or readily collectable data. The remainder may require additional data collection that is feasible but not yet planned.
• Score 3: Half or more questions are answerable with available data. The remainder are too broad or abstract to be answered without significant additional methods design.
• Score 2: Fewer than half are answerable as phrased. Several questions are unfalsifiable ("is the program transformative?") or require methods well beyond the program's resources.
• Score 1: Questions are unanswerable as phrased. Wording is so abstract that no data source or method could resolve them.

DIMENSION 3: Decision Linkage

• Score 5: Every question names a specific decision, adaptation, or audience that depends on the answer. The decision-maker or user is identified for each question. The link between answer and action is explicit (e.g., "if X, then we will adjust Y").
• Score 4: No fewer than 80 percent of questions name a decision or audience. The remainder are clearly relevant but the specific decision is not stated.
• Score 3: Half or more questions name a decision or audience. The remainder reference learning in general terms without naming who will act on the answer.
• Score 2: Fewer than half of questions name a decision or audience. Several questions are presented as learning for its own sake, with no link to program adaptation or reporting.
• Score 1: No decision linkage. Questions are listed without any reference to who will use the answers or what they will do differently.

DIMENSION 4: Scope and Prioritization

• Score 5: The set is right-sized for the program (typically 3 to 8 questions for a multi-year program). Priority questions are explicitly distinguished from secondary ones. The prioritization rationale is stated (e.g., highest-uncertainty, most decision-relevant).
• Score 4: Set size is appropriate (3 to 10 questions). Priority distinction made, though rationale for prioritization is brief or implicit.
• Score 3: Set size is plausible but no priority distinction made, OR set size is on the high end (8 to 12) with weak prioritization. All questions presented as equally important.
• Score 2: Set is too large (more than 12 questions) and unprioritized, OR set is a single question that cannot represent the program's learning needs.
• Score 1: No structured set. Either no questions are presented as a learning agenda, or the list is so long it functions as a research wish list with no priorities.

DIMENSION 5: Operationalization

• Score 5: For every question: named responsible position, stated timing (when the question will be addressed), synthesis method (how data will be analyzed and discussed), and audience for the answer. A learning event or product is scheduled per question (e.g., quarterly reflection, annual learning brief).
• Score 4: For no fewer than 80 percent of questions: responsibility, timing, and synthesis method named. Audience may be implied rather than stated.
• Score 3: Operational details given for half or more questions. The remainder are listed without responsibility, timing, or method. No learning event or product scheduled.
• Score 2: Operational details given for fewer than half of questions. Most appear as a list with no execution plan.
• Score 1: No operational detail at all. Questions are stated but nothing is said about who answers them, when, or how.

OUTPUT FORMAT: Return your assessment as a table followed by a summary:

Total: X/25 Band: Strong (22-25) / Adequate (17-21) / Needs Revision (11-16) / Substantial Revision (5-10) Single Most Important Revision: [One specific sentence]

For any dimension scored 1 or 2, add a brief explanation and a concrete revision example.

LEARNING AGENDA TO SCORE: [Paste your Learning agenda or learning questions section of a MEL plan here]

You are an expert M&E advisor specializing in learning and adaptive management. Review the draft learning agenda I will provide and fix every problem that falls below the Adequate (3) threshold on the MEL Learning Agenda Quality rubric.

WHAT TO LOOK FOR AND FIX:

Strategic Relevance problems: Generic questions ("what works?") with no tie to a specific assumption. Questions already answered by existing evidence. Peripheral or descriptive questions presented as learning questions.

Answerability problems: Unfalsifiable questions ("is the program transformative?"). Questions so broad no data source can resolve them. Questions requiring methods well beyond program resources.

Decision Linkage problems: Questions with no named decision or audience. Learning presented as an end in itself with no adaptation pathway. Decision-maker absent.

Scope problems: A list of 15 or more questions, all marked equally important. A single question expected to cover the program's full learning needs. No prioritization rationale.

Operationalization problems: Questions listed with no responsibility, timing, or method. No reflection forum or learning product scheduled. Synthesis approach absent.

OUTPUT FORMAT: For each problem found:

• Dimension: [which dimension]
• Location: [which question or section]
• Problem: [one sentence describing what is wrong]
• Revised version: [the corrected question or accompanying detail]

Then provide the complete revised learning agenda incorporating all corrections.

DRAFT LEARNING AGENDA TO REVISE: [Paste your Learning agenda or learning questions section of a MEL plan here]