Qualification: Replication of null results: Absence of evidence or evidence of absence?

Record: kaal:position:2026-08-08-331 · 2026-08-08

The inference remains bounded. Pawel and coauthors show that a non-significant estimate does not by itself establish the absence of an effect. The reported interval crosses zero and remains compatible with effects from -0.0502 to +0.0985. The result therefore leaves direction unresolved under the reported design. It does not show that the effect equals zero. Evidence for absence requires a method designed for that question, such as equivalence testing or a Bayes factor. This qualification does not recalculate Kaal's estimate, identify an equivalence margin, or test the registered endpoint, the multi-model cohort, or the Agentic Reputation Substrate. It limits the interpretation to what the estimate and confidence interval can support.

Affirmed commentary position. This record extends a source-bound scholarly claim but is not a verbatim paper claim.

Holds when

The response is limited to the exact eLife passages and the one mapped Kaal claim.
External evidence level: complete public eLife article in Europe PMC JATS XML with a concordant Crossref record, an official eLife article page, and exact proposition-bearing passages.
Mapping review tier: independent substantive scholarly-growth qualification.
The paper analyzes null results and replication methodology across scientific studies. It does not evaluate Kaal's registered endpoint, multi-model cohort, or Agentic Reputation Substrate.
It does not recalculate Kaal's estimate or specify an equivalence margin for the registered endpoint.
The qualification addresses interpretation only. It does not independently validate the numerical estimate or its confidence interval.
Crossref Spanish returned HTTP 429, Semantic Scholar search and three inherited identity refreshes returned HTTP 429, and arXiv timed out. No blocked response was promoted.

Current debate

Replication of null results: Absence of evidence or evidence of absence?

Scholarly basis

kaal:claim:7261018-035
Wulf A. Kaal, Empirical Evaluation of the Agentic Reputation Substrate: Deliberation, the Composition of Error, and the Registered Measurement of Agency Costs in a Controlled Multi-Model Cohort (2026). SSRN: https://ssrn.com/abstract=7261018
Source PDF sha256: 1d6cbe544bd0055133f7cf8ff308be4fde955867bd8dc764992b8d516af15fa8

Evidence and mapping

Evidence: complete public eLife article in Europe PMC JATS XML with a concordant Crossref record, an official eLife article page, and exact proposition-bearing passages
Review tier: independent substantive scholarly-growth qualification
Mapping confidence: 0.98
Mapping ambiguous: false

Topics

research-methods consensus-and-security scholarly-growth-coverage scholarly-literature statistical-inference confidence-intervals null-results evidence-provenance

Provenance

Affirmed in kaal-review:2026-08-12:scholarly-growth-7261018-035-reviewed-v1 on 2026-08-08. Review record.

Verify

Canonical markdown sha256: 4f06d39a4fb9432cd1f9c6717bf9003f9eb10b43f48341a87704d1cbcc65aa22
curl -s https://wulfkaal.github.io/positions/2026-08-08-331.md | sha256sum