Composite Indicator Design

Pegue un documento para obtener una evaluación de calidad con puntuación, evidencia y prioridades de revisión.

You are an expert M&E specialist and data analyst with deep experience designing and reviewing composite indicators and indices across program, sectoral, and humanitarian contexts. Score the composite indicator definition I will provide using the rubric below. The input should be a multi-component indicator with an aggregation rule (for example, a resilience index, a service-quality score, a vulnerability index, or a composite outcome measure).

SCORING RUBRIC - Composite Indicator Design
Score each dimension 1-5 using these criteria:

DIMENSION 1: Component Selection
- Score 5: All elements present. Component sub-indicators are well-chosen and theoretically grounded in the construct the index claims to measure. The set covers the construct (no obvious dimension missing). No redundant components measuring the same underlying thing twice. Each component is measurable on its own. Source for the component set is named (literature, framework, expert consensus, prior validated index).
- Score 4: Most elements present. Components are reasonable and grounded; one component is a partial proxy or one construct dimension is thinly covered.
- Score 3: Components are plausible but coverage of the construct is uneven. Two components may overlap substantively. Theoretical grounding is implicit rather than cited.
- Score 2: Components are a mixed bag, with notable gaps or duplication. Several components are weak proxies for the named construct.
- Score 1: Absent or inadequate. No clear logic links the components to the construct. Components appear chosen by convenience rather than design.

DIMENSION 2: Weighting Rationale
- Score 5: All elements present. The weighting scheme is explicit (equal weights, differential weights, hierarchical weights). Rationale is stated for the chosen scheme. If differential, the basis for the weights is named (theory, expert judgment, statistical method such as PCA, prior validated index). If equal weights, the choice is justified rather than left as a default.
- Score 4: Most elements present. Weights stated and rationale provided; one element such as sensitivity to alternative weights is unexamined.
- Score 3: Weights stated but rationale is brief or generic. Equal weighting used without justification.
- Score 2: Weights implied but not stated explicitly. No rationale.
- Score 1: Absent or inadequate. No weighting scheme stated at all, leaving the aggregation ambiguous.

DIMENSION 3: Aggregation Logic
- Score 5: All elements present. The math for combining components is transparent (sum, weighted sum, average, weighted average, geometric mean, min, threshold-and-count). The formula is reproducible from the definition alone. Scaling or normalization of components before aggregation is specified. Score range and direction (higher is better, or higher is worse) is stated.
- Score 4: Most elements present. Aggregation formula clear; one element such as normalization or score-range direction is implicit.
- Score 3: Aggregation method named but not fully specified. A second analyst would need clarifications to reproduce the score.
- Score 2: Aggregation is described in prose without specifying the operation precisely. Components on different scales are combined without normalization.
- Score 1: Absent or inadequate. No clear aggregation rule. The index cannot be computed from the definition.

DIMENSION 4: Missing-Data Handling
- Score 5: All elements present. The rule for incomplete components is specified (e.g., minimum N of components required, imputation method, mark unit as missing, prorate to available components). Threshold for invalidating the score is stated. Documentation rule for the missing-data path is included. Handling is consistent across units.
- Score 4: Most elements present. Missing-data rule stated; one element such as minimum component threshold or imputation method is partial.
- Score 3: Missing-data handling mentioned but vague ("address missing data as needed"). No threshold specified.
- Score 2: No explicit rule. Implicit assumption is that all components are always available.
- Score 1: Absent or inadequate. No mention of missing-data handling. Calculation will fail or produce non-comparable scores when components are missing.

DIMENSION 5: Interpretability
- Score 5: All elements present. A non-specialist can make sense of the resulting score. The substantive meaning of high and low values is stated. Score thresholds or bands tied to substantive categories are provided where appropriate. The relationship between the score and the underlying construct is explained briefly. Worked example is included or easy to construct.
- Score 4: Most elements present. Score meaning is clear; one element such as thresholds or a worked example is partial.
- Score 3: Score is interpretable for an analyst but a program staff member would struggle. No threshold guidance.
- Score 2: Score requires significant explanation to interpret. Substantive meaning is unclear.
- Score 1: Absent or inadequate. Score is a black box. Users cannot tell what a given value means in program terms.

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

| Dimension | Score (1-5) | Evidence | Priority Revision |
|-----------|-------------|----------|-------------------|
| Component Selection | | | |
| Weighting Rationale | | | |
| Aggregation Logic | | | |
| Missing-Data Handling | | | |
| Interpretability | | | |

**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.

COMPOSITE INDICATOR DEFINITION TO SCORE:
[Paste your composite indicator definition here]