Comparing records
How to compare two conflicting records
Short answer
- Step 2
- Compare coverage, counting unit and reference period 2
Step 1: Pin down each record's vintage
Write down the publisher, the table or filing, the release date and the stage label of each figure. BEA publishes advance, second and third GDP estimates, then annual and benchmark updates 1. BLS says the latest 2 months of payroll estimates are based on incomplete returns, and a monthly figure is final only after two revisions 2.
Step 2: Match scope, unit and period
Check who or what each record counts. BLS says the household survey counts self-employed and agricultural workers, which the establishment survey excludes 2. It counts multiple jobholders once, and the establishment survey counts each job 2.
Check the reference period. The household survey uses the calendar week containing the 12th, and the establishment survey uses the pay period including the 12th 2. If scope or period differ, the records may both be right.
Step 3: Look for revisions and corrections
Check whether either figure has been revised or corrected since you copied it. ONS notes revisions in the next release, and shows error corrections in a banner titled Correction 4. Eurostat places revision information under Data revision policy and Data revision practice in dataset metadata 3. Eurostat treats corrections of errors as unscheduled revisions 3.
Step 4: For company figures, check for restatements
If the records are financial statements, check whether the company filed a Form 8-K under Item 4.02. That item covers a conclusion that previously issued statements should no longer be relied upon, and a notice from the auditor to the same effect 5. Under PCAOB AS 2820, an auditor's report on restated statements includes an explanatory paragraph stating that the previously issued statements have been restated, with a reference to the note that describes the correction 6.
Step 5: Read the quality documentation
Producers publish notes on methods and limits. US federal rules require recognized statistical agencies to publish descriptions of methods, data quality indicators and known limitations or sources of error, and to identify errors discovered after release 7. The IMF Data Quality Assessment Framework lists revision studies and revision policy and practice as elements of quality 8.
Step 6: Record the result
- If one record is a later vintage of the other, use the later one and cite both dates 1.
- If scope or period differ, report both and state the difference 2.
- If one record was corrected, cite the correction notice 4.
- If accounts were restated, cite the filing that marks the earlier figures as not to be relied upon 5.
Common questions
Which figure should I use if two official records conflict?
If they are vintages of the same series, the later release reflects more complete data 1,2. If they measure different things, report both and say how they differ 2.
Where do I find why a figure changed?
In the producer's revision notes or metadata. Eurostat keeps them under Data revision practice 3. ONS notes revisions in the next release 4.
How can I tell if company figures were restated?
Check for a Form 8-K Item 4.02 filing 5 and an auditor's explanatory paragraph on restated statements under AS 2820 6.
Do statistical agencies have to tell users about errors?
In the US, 5 CFR 1321.6 requires recognized statistical agencies to identify errors discovered after release 7.
Rules and charges change. Check the official source linked on this page before you pay or travel. Spotted an error? Email hello@recordsdisagree.com.
Sources
- Bureau of Economic Analysis: Comparisons of Revisions to Real GDP (October 2026)
- U.S. Bureau of Labor Statistics: Employment Situation Technical Note (2 October 2026)
- Eurostat: Data revision policy
- Office for National Statistics: The Office for National Statistics' Revisions Policy and Correction of Errors Policy (9 August 2024)
- U.S. Securities and Exchange Commission: Form 8-K Current Report (OMB Number 3235-0060)
- PCAOB: AS 2820 Evaluating Consistency of Financial Statements
- Legal Information Institute: 5 CFR 1321.6 Credibility and accuracy
- International Monetary Fund: Data Quality Assessment Framework (DQAF) for National Accounts Statistics (May 2012)
Related
- Why official records disagree
The main reasons two official records show different figures for the same thing: timing, revisions, corrections, scope, definitions and restatements.
- Scope and period mismatches between official records
How differences in coverage, counting units, reference periods, fiscal years and geography make two correct official records show different numbers.
- Data quality frameworks compared
How the IMF DQAF, the European and UK codes of practice, UN principles, US rules and ISO 8000 define data quality, revisions and errors.
- Financial restatements explained
How companies correct errors in published accounts under US rules and IFRS, what a Form 8-K Item 4.02 notice means, and how EU enforcers act.
- Statistical revisions explained
Why GDP, jobs and other official figures change after release, how agencies schedule revisions, and how a revision differs from a correction of an error.