Research note
How often do rejection reasons match the documented screening standard?
A local method for measuring how often rejection reasons point to a screening standard written before interviews, with separate checks for the reason and the evidence.
Published

Table of contents
Question and scope
The question covers candidates rejected after at least one screening step. It includes rejections at the first screen, after a work sample, and after a final interview. It excludes candidates who withdrew and searches that closed with no rejection record. The review covers one employer, one window, and the roles that use a written standard.
A case enters the sample when the candidate has a rejection record and the search has a dated standard. If the standard was written after the rejection, the case enters as a late standard. Record the source of the standard and the date it was approved. A standard reconstructed during the review should be marked as such.
The stop point is the reason recorded in the rejection summary, not a comment in an interview notes file. A reason must point to a named standard item and the evidence behind it. A summary that lists several impressions without a standard does not meet the test. Report the cases with no reason separately.
Definitions and evidence standard
Define a matching reason before reviewing any case. The reason must name one standard item and the evidence that failed it, such as a work sample that missed a required step. A reason that says fit or culture does not match unless the standard defines the behavior. Record the definition with the case so a later reader applies the same rule.
Use the standard, the rejection summary, and the candidate file as the primary evidence. A reviewer's memory can add context but cannot replace a dated record. If only a verbal reason exists, mark the case as unverified. Report the share of unverified cases with the result.
Separate a reason that matches the standard from a reason that is merely supported by evidence. A reason can cite a real problem that was never a requirement. Measure both so a well-written reason does not hide a standard that changed mid-search. The two numbers answer different questions about consistency.
Sampling and measurement
Freeze the list of rejected candidates before reviewing any of them. Include more than one recruiter, more than one role, and more than one rejection stage. Record the export date and the filter used. Do not remove a case because the record is thin, since the gaps are part of the finding.
If the list is large, take a random sample and record the selection method. Keep the full count beside the sample so a reader can see the coverage. A replaced case should be reported as a deviation. The sample should be large enough to compare at least two teams.
Measure the number of cases, the share with a reason that matches a standard item, the share with a reason but no standard, and the share with no reason. Add the share where the standard was approved after the first interview. Report the counts by recruiter and role. A single percentage hides which part of the record is weakest.
Findings and interpretation
Present the distribution before suggesting a cause. A cluster of unmatched reasons under one recruiter may mean the standard is hard to find. A cluster in one role may mean the standard was never written. Read the cases behind the largest cluster before naming a reason.
Compare the result with the search timeline. A standard approved before the first interview can anchor every later reason, while a standard written at the end tends to match the chosen person. Record the approval date and the first interview date. The comparison can show whether the problem is timing or wording.
Check whether unmatched reasons still described a real gap in the candidate's work. A reason can be honest and still fail the standard test. Record the evidence where the file supports it. The quality of the decision matters alongside the consistency of the record.
Limitations
The review measures the written record, not whether the hiring decision was fair in every respect. A matching reason can still come from a biased process. The result describes documentation consistency, not the full decision. Report it beside the standard review so a reader does not overread the number.
The standard may change during the window, and an older case may follow a rule that was appropriate at the time. Record the version rather than applying the current standard retroactively. A case that missed an old standard may not be a current defect. Keep the version note with the case file.
Candidates who withdrew are invisible to this method. The count may understate the problem if some people leave before a rejection is recorded. Pair the review with a count of withdrawals at each stage. The answer can reveal pressure that the rejection record cannot show.
Practical next step
Add the standard items to the rejection form and require one item and one evidence note for each decision. Send the standard to every interviewer before the first screen. A short form makes the match measurable. Keep the standard as a checklist beside the form.
Have a second reviewer read a sample of reasons before the search closes. A second reader can catch a reason that points to a preference rather than a requirement. Record the correction and the date. Tell the candidate only what the policy requires.
Test the process for one quarter and compare the match rate with the baseline. Keep the change if it improves consistency without slowing the search. If the form or the reviewer causes delays, fix that step. The goal is an explainable decision, not more paperwork.
Sources
References supplied with this research note.
| Source | Publisher | Context |
|---|---|---|
| U.S. Equal Employment Opportunity Commission, Prohibited Employment Policies and Practices | eeoc.gov | Primary reference supplied with this article. |
| U.S. Equal Employment Opportunity Commission, Recordkeeping Requirements | eeoc.gov | Primary reference supplied with this article. |
| National Institute of Standards and Technology, Privacy Framework | nist.gov | Primary reference supplied with this article. |