Public evidence

engineering.uncertainty.pooled_sd

Machine-checked verification for calculation version 1.0.0 · tier engine_tested

Open calculator API Capability Download JSON MCP Review policy

Verification

8/8 calculation tests · Production surface contract 4/4

Golden tests 0 / 0
Boundary tests 5 / 5
Property tests 3 / 3
Production surface contract 4 / 4
Engine tested Yes
Source checked Yes
Artifact integrity PASS
Production attestation Embedded snapshot: unpublished · Build schema 1.0.0 ready · Semantic contract ✓ · Attestation report not published on origin · Live production status STALE (1 capability; 156 remain CURRENT) @ 2026-09-14T03:54:53.650Z
Semantic contract PASS
Evidence attested at 2026-09-14T00:43:05.498Z
Last production check 2026-09-14T00:43:05.498Z
Public / cache —
Origin —
Supplemental domain review Not performed
Named expert review Not performed

release artifacts aligned PASS
schema valid input executes PASS
engine success output validates schema PASS
schema rejects incomplete input PASS

Fingerprint sha256:9ce4e6f368b473483310c22bdcf9392af1368150be6a13bececbd7f2dcfa96d7

Input schema sha256:9df2c14e2a0bd90df92a6b43f0ace6bbdf7a0291d6514b2febc035bdc4024748 · Output schema sha256:c655b77686e92e8138f442f7f6f36a4ead7cdd219d3a6d2b5f9f6e69859b2c5b · Build-time SHA-256 digests prove published artifacts agree with each other (release_integrity.scope=workspace). Production attestation GETs canonical URLs twice (public/cached view and origin), including Capability HTML, via npm run attest:production --write (published at /.well-known/calculatorx-production-attestation.json) and fails if the public Capability page advertises a different calculation_version. This is artifact consistency, not an independent immutable release log.

Build & scope

Engine identity and declared limitations.

Tool: pooled-sd
Engine: pooled-sd
Engine build: 2026-09-15
Evidence core: 2026-09-15T23:34:31.244Z
Rendered: 2026-09-15T23:38:12.910Z
Last reviewed: —

Assumptions

Pooled standard deviation: s_p = √(Σ ν_i s_i² / Σ ν_i) from k≥2 stated groups. ν_i > 0 and finite.
s_i ≥ 0. Both zero is valid (s_p = 0).
Not Type A from one sample, not Welch–Satterthwaite ν_eff, and not ISO 21748 u of a mean.

Limitations

Takes stated s_i and ν_i — not a single sample of observations
k=1 is rejected so this cannot wrap Type A / identity s
Does not compute Welch–Satterthwaite ν_eff
Does not compute ISO 21748 u of a mean

Sources

Standards and references supporting this tool specification. Methods are listed separately — they are algorithms, not bibliographic sources.

JCGM 100:2008 — Evaluation of measurement data (GUM) — 4.2.4 Type A evaluation from several series
Supports: Independent series of observations of the same measurand may be pooled: s_p² is the ν-weighted mean of the sample variances.
ISO 5725-2 — Repeatability and reproducibility of standard measurement methods — Pooled repeatability standard deviation
Supports: s_p is the usual pooled-within estimate from several groups. This page takes stated s_i and ν_i rather than a collaborative-study table.

Published checks

Golden, boundary, and property cases included in this evidence build.

ID	Kind	Status	Detail
`boundary-pool`	boundary	pass	Expected s_p=5 · nu=2 · k=2 · Actual s_p=5 · nu=2 · k=2
`boundary-pool-equal`	boundary	pass	Expected s_p=5 · Actual s_p=5 · nu=11 · k=2
`boundary-pool-n`	boundary	pass	Expected s_p=5 · Actual s_p=5 · nu=2 · k=2
`boundary-pool-one`	boundary	pass	Expected VALUE_OUT_OF_RANGE · Actual VALUE_OUT_OF_RANGE
`boundary-pool-missing`	boundary	pass	Expected MISSING_REQUIRED_INPUT · Actual MISSING_REQUIRED_INPUT
`property-pool`	property	pass	5
`property-pool-equal`	property	pass	5
`property-pool-alias`	property	pass	5