Program Evaluation Readiness

Method, inputs and limits

Lists research-design documentation gaps and calculates explicitly comparable observed rates. Method version 0.2.0. Cannot establish a causal program effect, public-health outcome, political outcome, or validated research design.

Visible mathematics / decision rules

rate = numerator / positive denominator × explicit reporting scale.

descriptive difference = intervention rate − baseline rate, only for declared-comparable observations with matching units, definitions, and equal period lengths.

What this result cannot establish

• The six documentation items are a checklist, not a statistical design score; optional difference-in-differences is a descriptive arithmetic contrast.
• Documentation is self-declared. Even all six documented items do not establish independence, study validity, causation, or the absence of confounding.
• Comparison requires an explicit comparability declaration, identical unit and definition labels, equal elapsed period lengths, and positive known denominators. Before/after windows must be ordered and nonoverlapping; concurrent windows must match.
• No uncertainty interval, significance test, causal effect, prevented lives, or political/medical outcome estimate is produced.

What would falsify the assumptions?

• Changed definitions, non-comparable populations, measurement error, confounding, or a mismatched period can invalidate the descriptive comparison.
• Independent data collection and a qualified independent research-design review can contradict the supplied records.
• This describes rates and documentation, not resource creation or a causal estimate; the same division and checklist work without AI.
• MAR has no demonstrated benefit if manual review finds the same gaps; synchronized omissions or stronger unsupported causal claims would add risk.
• Assumptions fail if the supplied definitions, units, period, scope or source records are wrong or incomplete.
• An independent source audit or replication may contradict an input or the recorded dependency.
• This method creates no resources. Physical conservation is checked in the inventory and transfer ledgers; descriptive graphs and checklists are not resource-flow models.
• The same equations and review checklist can be reproduced with a spreadsheet or structured human-only process.
• This is deterministic arithmetic, a descriptive comparison or a structured governance check, not a fitted causal effect or forecast. Fictional inputs remain an exercise.
• Equal or worse measured errors, net review effort, disclosure burden or fallback under MAR would show no added benefit or added risk; no improvement is assumed.

{
  "observations": {
    "unit": "people",
    "origin": "synthetic",
    "sources": [
      {
        "id": "SYN-001",
        "kind": "synthetic",
        "label": "Fictional example",
        "url": null
      }
    ],
    "target": "Record the fraction of eligible people receiving the declared civilian service",
    "potentialConfounders": [
      "Population composition and unrelated service changes are not assessed"
    ],
    "parallelTrends": {
      "status": "not-assessed",
      "reference": null,
      "note": "No pre-trend evidence is supplied; no causal inference is supported"
    },
    "differenceInDifferences": [],
    "design": {
      "baseline": {
        "status": "documented",
        "reference": "SYN-001",
        "note": "Fictional aggregate baseline"
      },
      "comparator": {
        "status": "documented",
        "reference": "SYN-001",
        "note": "Fictional comparable group"
      },
      "period": {
        "status": "documented",
        "reference": "SYN-001",
        "note": "Equal 30-day periods"
      },
      "definitions": {
        "status": "documented",
        "reference": "SYN-001",
        "note": "Same observed service definition"
      },
      "confounders": {
        "status": "unknown",
        "reference": null,
        "note": "Not assessed in this fixture"
      },
      "independentReview": {
        "status": "missing",
        "reference": null,
        "note": "No independent reviewer"
      }
    },
    "comparisons": [
      {
        "id": "SERVICE",
        "label": "Observed civilian service access",
        "measureType": "share",
        "windowType": "before-after",
        "comparability": "declared-comparable",
        "comparabilityReason": "Same fictional definition and equal period lengths",
        "baseline": {
          "numerator": 40,
          "denominator": 100,
          "unit": "people",
          "definition": "Eligible people receiving the declared service",
          "periodStart": "2027-01-01",
          "periodEnd": "2027-01-31"
        },
        "intervention": {
          "numerator": 55,
          "denominator": 100,
          "unit": "people",
          "definition": "Eligible people receiving the declared service",
          "periodStart": "2027-02-01",
          "periodEnd": "2027-03-03"
        }
      }
    ]
  },
  "parameters": {
    "rateScale": 100
  }
}

{
  "documentedItems": 4,
  "designGaps": 2,
  "comparisonsAvailable": 1,
  "baselineRate": 40,
  "interventionRate": 55,
  "difference": 15
}

Authoritative background references

• NIST: AI Risk Management Framework — Voluntary governance background; not validation of MAR. Reviewed 2026-11-09. Reference, not endorsement or validation.

These sources inform definitions or review practices. They do not validate this implementation or endorse MAR.