Research note
Can a payroll change be reproduced from its evidence?
A documentary test of whether a payroll change can be reconstructed from the request, approval, source, and confirmed result.
Published

Table of contents
Question and scope
This study asks whether a reviewer can reconstruct a completed payroll change from the request, the approval, the source, and the confirmed result. It examines the effective date, the old and new values, the approver, and the downstream confirmation. The scope is one pay period or one defined change type. The result is a traceability finding, not a payroll audit opinion.
The unit is one payroll change for one employee. If a change affects several fields, record each field separately when the approval or source can differ. Include changes that were reversed or corrected. Exclude test payrolls, and record the reason. Report the number of changes and the number of employees separately.
Method and evidence standard
Define the required evidence before sampling: the request, the source, the required approval, the effective date, the value change, and the confirmation that the change appeared in the payroll register. A request alone is not a completed change. A register entry without the approval is partial. Record the version of each document and the date it was used.
Sample across change types, departments, and preparers. Include changes that were processed late and changes that required a retroactive adjustment. Two reviewers should calibrate on a small set and discuss disagreements about whether a confirmation is specific enough. Preserve the initial rating and the adjudication. The study measures whether the change can be reconstructed, not whether the amount was correct.
Measures, findings, and interpretation
Report the share of changes with a complete request, source, approval, effective date, value record, and confirmation. Show the result by change type and by whether the change was late. Include the number of changes where the register value differed from the request and the number with no confirmation. Use counts and rates together so a small group is not overinterpreted.
Inspect the changes behind the largest gaps before assigning a cause. A cluster of missing confirmations may mean the pay run is closed before the check happens. A cluster of missing sources may mean the request form accepts free text without a document. The study cannot show whether the change complied with tax or employment law, and it should not be used to compare individual payroll specialists.
Limitations and practical next step
The review sees the documents and the register, not the conversations or the system state at the time. A change may be correct even when the evidence is incomplete, and a complete file may contain a value that was applied to the wrong employee. The study also depends on the sample and the version of the register. Note any system change or reprocessing during the period.
Choose one control to test, such as a required source link, an approval step before data entry, or a post-run confirmation check. Record the owner and the date, then recheck the same measure on the next pay period. Keep the sample and the decision log. A reproducible change record helps payroll, the employee, and a later reviewer answer the same questions.
Sources
References supplied with this research note.
| Source | Publisher | Context |
|---|---|---|
| U.S. Department of Labor, Fair Labor Standards Act | dol.gov | Primary reference supplied with this article. |
| U.S. Department of Labor, New Hire Reporting | dol.gov | Primary reference supplied with this article. |
| U.S. Equal Employment Opportunity Commission | eeoc.gov | Primary reference supplied with this article. |
| NIST Cybersecurity Framework 2.0 | nist.gov | Primary reference supplied with this article. |