METHOD DOSSIERMSC-H-006Evidence foundations

Statistical decision methods: choose, quantify, validate

This territory links variable type, sampling design, estimand and error cost. The test or model comes after the question, never before it.

Direct answer

Choose a method consistent with the question and make decision uncertainty visible.

This territory links variable type, sampling design, estimand and error cost. The test or model comes after the question, never before it.

Model or deliverableQuestion → design → estimand → uncertainty → decision boundary

ESTIMAND

Unit of analysis: Decision × population × horizonControlled scientific terminologydecision × population × horizon
Exact estimand: A defensible chain from question to evidence
Model or deliverable: Question → design → estimand → uncertainty → decision boundary

01—07

01—08

01
Question
Statistical decision methods: choose, quantify, validate
02
Estimand
A defensible chain from question to evidence · decision × population × horizon
03
Data
population · unit_of_analysis · decision_horizon · outcome · assumptions
04
Model or deliverable
Question → design → estimand → uncertainty → decision boundary
05
Declared calculation
Question → design → estimand → uncertainty → decision boundary → A defensible chain from question to evidence
06
Uncertainty checks
Unique decision intent · Explicit identification assumptions · Reproducible evidence trail
07
Method-specific validation
Check data, estimate stability and interpretation limits. Unique decision intent · Explicit identification assumptions · Reproducible evidence trail
08
Limitations
Guarantee a correct decision from statistical significance alone.

Required variables

Checks · Controlled scientific terminology

Possible decision

Choose a method consistent with the question and make decision uncertainty visible.

Not supported

Guarantee a correct decision from statistical significance alone.

Scientific sources