Onboarding And Retention Research

How complete are remote I-9 files at the review gate?

A bounded method for checking whether a remote employment eligibility file shows a complete employee section, an authorized review, and a retained record.

| 7 sources

How complete are remote I-9 files at the review gate?
Onehire, one file
Tworeview paths
Missingevidence kept visible
On this page
  1. Key takeaways
  2. Question and scope
  3. Population and unit
  4. Evidence standard
  5. Sampling
  6. Reviewer calibration
  7. Measures and reporting
  8. Limitations
  9. Privacy and role boundaries
  10. From finding to controlled change
  11. Worked example: a file returned for a missing review
  12. What the reviewer should record
  13. Correction and recheck
  14. Reporting the sample
  15. Review cadence
  16. Sources and methodology
  17. Related research
  18. Questions about the findings

Key takeaways

  • Define the review gate before sampling files.
  • Separate a missing section from an unauthorized review.
  • Record the correction and the recheck date.

Question and scope

This study asks whether a remote employment eligibility file shows the employee section, the authorized review, and the retained record required by the employer's own written procedure. It treats the file as an administrative record and does not decide whether the employee is authorized to work. The result is intended to improve intake, review, and retention controls rather than to produce a compliance opinion. The scope is one employer or one defined hiring group during a fixed period. The study may compare locations or review methods, but it should not generalize to an industry rate. A small sample can reveal a process gap without proving how common that gap is across the employer's entire workforce.

StageEvidenceFailure modeUse
Employee sectionDated entries and signaturesBlank or inconsistent fieldReturn to owner
ReviewReviewer identity and methodUnsupported completionRecheck
RetentionStored copy and access recordMissing or unreadable fileRestore or escalate

Population and unit

The population should be drawn from a frozen export of remote hires who completed the review process during the window. Record the systems used, the query time, the locations included, and the status filter. Include hires whose files were later corrected, because a correction is part of the record. The unit is one hire's eligibility file at the review gate. If a file contains multiple documents, treat the file as one unit and record the document count separately. A reopened or corrected file remains linked to the same hire, with the correction event recorded rather than replacing the original observation.

Evidence standard

Define what counts as a complete employee section, an authorized review, and a retained record before opening any file. For the employee section, specify which fields, signatures, dates, and document types are required. For the review, specify who may conduct it, what method is acceptable, and what evidence must be recorded. For retention, specify where the file must live, how access is controlled, and how long it must remain available. A file that exists only in an email attachment is not the same as a file stored in the approved system. The standard should distinguish missing, unreadable, and not applicable values so the reviewer does not collapse different problems into one code.

Sampling

Draw the sample from the frozen frame using a documented method. A simple random sample may be enough for a small group, while a stratified sample can ensure coverage of locations and review methods. Record the seed, any replacement, and the reason for exclusion. Include files that were completed on the first pass, files that were returned, and files that were corrected after the start date. A sample of only clean files will overstate completeness. A sample of only problem files will describe failure modes but not the base rate.

Reviewer calibration

Two reviewers should classify a small calibration set before the main review. Compare their answers field by field and discuss the cases where they disagree. Revise the standard and repeat the calibration until the common definitions are clear. During the main review, preserve the initial ratings and the adjudication. If one reviewer marks a file complete and the other marks it partial, the difference may reveal an unclear rule rather than an error. Report the disagreement instead of hiding it in a final consensus.

Measures and reporting

Report the number of files in the frame, the number sampled, and the number complete, partial, or unverifiable. Show the result by element rather than only as an overall rate, because a file can have a complete employee section and a weak review record. Include the age of open corrections at the observation date. Use counts and percentages together, and state the denominator for every rate. If the sample is too small for a stable subgroup estimate, say so. A table with short definitions is more useful than a chart that implies a precision the design does not support.

Limitations

The study observes files that reached the review frame and may miss hires whose files were never created or never exported. It cannot determine whether the underlying document is authentic or whether the employee is authorized to work. Those questions belong to the authorized reviewer and, where applicable, the relevant government process. Reviewer judgment and system changes can affect the result. A new form version, a portal update, or a change in the review method may alter what the team can record. Note the relevant dates and avoid comparing periods as if the instrument stayed constant.

Privacy and role boundaries

Use the minimum personal information needed for the study. Replace names with study keys, restrict access to the crosswalk, and follow the employer's retention rule. Do not copy identity documents into the analysis file when a reference to the approved record is enough. The onboarding coordinator may prepare the sample and classify administrative fields. The authorized employer reviewer decides whether a file is acceptable and what correction is required. The study should not be used to make a legal conclusion about an individual's eligibility.

From finding to controlled change

Choose one change based on the most common or most consequential gap. It might be a required field, a second-review step, a different storage location, or a clearer correction reason. Name the owner, the due date, and the measure that will show whether the change helped. Recheck the same measure after two cycles. If the result does not move, examine whether the change addressed the cause or only the visible symptom. Keep a short decision log with the finding, the change, the owner, and the result so the next reviewer can see why the process looks the way it does.

Worked example: a file returned for a missing review

A remote hire submits the employee section and the supporting document before the start date. The coordinator marks the employee section received and routes the file to the authorized reviewer. The reviewer finds that the document was viewed but no review method or reviewer identity was recorded. The file is returned with a reason code rather than marked complete, and the next action names the reviewer who must complete the record. The correction adds the reviewer identity, the review date, and the method used. The coordinator rechecks the file against the standard and records the new observation without deleting the original return. This example shows why a missing review is different from a missing document, because the document may be present while the evidence of review is not.

What the reviewer should record

For each sampled file, record the hire identifier, the documents present, the fields completed, the review method, the reviewer, and the storage location. Note whether the file was first-pass complete, returned, or corrected. A short rationale should explain every partial or unverifiable rating so a second reviewer can follow the decision. The record should not include unnecessary identity details in the analysis file. A study key and a reference to the approved source are enough. If the reviewer needs to inspect a document, the inspection should happen in the approved system and the analysis should record only the reference and the finding.

Correction and recheck

A correction should name the missing element, the person who supplied it, the date, and the evidence that the file now meets the standard. The recheck should be performed by someone who did not make the correction when the risk is high. Record the recheck result and keep the earlier return visible. If the same correction appears in several files, check the form and the review instructions before coaching the same coordinator. A recurring missing reviewer field may mean the system does not require it, not that the coordinator forgot. The study should point to the design issue when the pattern supports it.

Reporting the sample

Report the frame size, the sample size, and the number of files that were complete, partial, or unverifiable. Show the result by element and by location or review method when the subgroup is large enough. Include the number of corrections and the time between the return and the recheck. Use a short table with the definitions beside the counts. Avoid a single headline percentage that hides whether the gap is in the employee section, the review, or retention. End the report with one recommended change, one owner, and the next sample date.

Review cadence

A quarterly sample may be enough for a stable process, while a monthly sample is more useful during a system change or a hiring surge. Choose the cadence based on risk and volume rather than on the calendar alone. Record the cadence in the procedure so the team knows when the next review will happen. Between samples, keep an exception log for files returned or corrected. The log helps the team see whether the process is drifting before the next formal review. If the log and the sample disagree, investigate the difference rather than assuming one source is wrong.

Sources and methodology

A bounded documentary review of remote employment eligibility files with a frozen population, a written evidence standard, and a two-reviewer calibration step. Public guidance frames the questions but does not validate a local file or supply a legal conclusion.

  1. USCIS — Form I-9 current.
  2. U.S. Department of Justice — I-9 and E-Verify current.
  3. U.S. Department of Labor — New Hire Reporting current.
  4. IRS — Form W-4 current.
  5. NIST Privacy Framework current.
  6. U.S. Small Business Administration — Hire and manage employees current.
  7. U.S. Equal Employment Opportunity Commission current.

Source count: 7. Last verification date: November 3, 2026.

Questions about the findings

What is the unit of analysis?

One hire's eligibility file at the review gate, with corrections linked to the original observation.

Does this study determine work authorization?

No. It observes whether the administrative file meets the employer's written evidence standard.

Put the finding to work

Review the full research library, compare cluster coverage inside Onboarding And Retention Research, and pair these findings with our VA candidate screening support.

remote I-9onboarding evidenceHR operations