preregistration instrument

Definition	Committing the hypothesis, the bar, and the analysis before the measurement is run, so that the record shows what was predicted. Chapter 8.
Example	The prediction, the bar, the null, and the seeds were committed before the run, and the commit hash is in the report.
Book	Data Mining as Observation, draft 0.2, commit `f3914f0`; entry id `preregistration`, kind instrument.
Status	no ledger row names this entry. Corrections: 1 item(s), see below.
Defining equation	none
Assumptions and scope	A preregistration commits the hypothesis, the bar, the analysis, and the missing-data rule before the measurement is run, so that the record shows what was predicted rather than what was found. Its bar must be anti-vacuous. A pass that any instrument would have produced, a threshold met by every reading, proves nothing, and the program’s own first PF5 pass is the case. Multiple looks at the data need a corrected threshold, and a preregistration that names one look needs none.
Prior art	none recorded
Evidence	`observation-theory-campaigns/experiments/PREREG-TEMPLATE.md:47-52`, `observation-theory-campaigns/ERRATA.md:90-110`, `lean/DataMiningAsObservation/Bonferroni.lean`
Reviewed	not yet reviewed; generated 2026-09-10 from records at the commits on the provenance page.

The claim, bar, null, and budget written before the measurement.

Equation

none

Conditions

A preregistration commits the hypothesis, the bar, the analysis, and the missing-data rule before the measurement is run, so that the record shows what was predicted rather than what was found.
Its bar must be anti-vacuous. A pass that any instrument would have produced, a threshold met by every reading, proves nothing, and the program’s own first PF5 pass is the case.
Multiple looks at the data need a corrected threshold, and a preregistration that names one look needs none.

Conditions are curated in entries.toml rather than read from a record.

Ledger

none

First stated

Chapter 8 section 8.8 of Data Mining as Observation, with the program’s template observation-theory-campaigns/experiments/PREREG-TEMPLATE.md:1-71 and Volume 14’s protocol geometric-observation/PROTOCOL.md:58-75.

Measurements

Where the book states it	Numbers, as the book’s sources table records them	Source
chapter 2 section 2.3	the missing-data rule as a required preregistration field	`observation-theory-campaigns/experiments/PREREG-TEMPLATE.md:47-52`
chapter 7 section 7.6	the preregistration fields	instructor working documents, not public [@bond2026course], `ECE_514-01_FA26_session-outlines.md:157-176`; chapter 8 section 8.8 of this book

Failures and corrections

observation-theory-campaigns/ERRATA.md:90-110 at f7b7c77. E3. Two wrong bar values in the campaign notes Found 2026-08-07, same pass. Not a published error. experiments/CAMPAIGN.md twice described the PF4-006 entry-point and timestep clause as having a bar of 1e-10. The sealed declaration and the committed record both set that bar at 1e-8. The 1e-10 figure is the separate translation-covariance bar. The narrative was wrong and the verdict was not, since the clause failed at 1.3e-2 against either number. The same document described PREREG-PF4-009 as passing all six bars. The sealed document groups its requirements into six bars and the governed runner reports eight items, and the mapping between them was never declared. Both statements are now made explicitly. Corrected in place, since these are working notes rather than sealed documents or published records. —

Invariance envelope

none declared

Machine checked

lean/DataMiningAsObservation/Bonferroni.lean, theorems family_error_le, bonferroni, one_look, at observation-data-mining f3914f0; what the check covers is stated in the book’s appendix C.

Used in

Data Mining as Observation chapters 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14.

sealed; ledger class; harness; certificate.

Status

Generated 2026-09-10 by encyclopedia/generate.py; book at observation-data-mining f3914f0; the commit of every record is listed in the encyclopedia’s provenance.

← precision, recallprincipal component analysis →