Insight · Public Safety Selection & Promotion

The Day the List Posts: Promotional Processes Fail at the Hearing, Not the Exam

By Tabitha Weinstein · June 5, 2026 · 5 minute read

A promotional list is among the most scrutinized documents a public safety agency publishes. The day it posts, every candidate who missed the cut reads it like an auditor. Some of them appeal. The exam, the assessment center, the cut-off score, and the documentation behind them have to hold up at that moment, months after test day, in front of a hearing officer who was not in the room when any of the decisions were made.

The scrutiny arrives after the celebration

Agencies experience a promotional process in two phases. Phase one is operational: announce, test, score, rank, post. It ends with congratulations in the briefing room. Phase two starts about a week later, when the first grievance lands. Phase two is where promotional processes actually succeed or fail, and almost nothing in phase one prepares an agency for it unless defensibility was engineered in from the start.

I have sat in those hearings. I have testified in them as an expert witness, defending promotional processes covering thousands of sworn personnel across supervisory and command ranks. The question a hearing officer asks is never whether the agency intended to be fair. It is whether the agency can prove, document by document, that the process measured what the rank requires and treated every candidate by the same standard. Intent does not survive cross-examination. Documentation does.

Where challenged processes actually break

The job analysis is stale or missing. The exam tests the job as it existed a decade ago, or as a neighboring agency defined it, or as a test vendor's generic template assumed it. If the job analysis for the rank was not conducted or updated for this cycle, every downstream component inherits the defect. Under the Uniform Guidelines on Employee Selection Procedures, content validity starts with the job as it exists today, not as anyone remembers it.

The cut-off score is tradition, not method. Seventy percent is the most litigated number in public safety testing. Asked why the passing score is 70, agencies too often answer that it has always been 70. A cut-off set by custom rather than by a documented standard-setting method, such as a properly conducted Angoff procedure with qualified subject matter experts, is an open door for an appeal.

Assessors scored differently and nobody noticed. One assessment center panel scores generously, another severely. Without behaviorally anchored rating scales, structured assessor training, and rater-consistency monitoring during the process, candidates with identical performances receive different scores depending on which room they walked into. That fact pattern will be reconstructed at the hearing.

Adverse impact was checked after certification, or never. The four-fifths analysis belongs before the list is certified, while the agency can still examine and address what the numbers show. Discovering a disparity for the first time in litigation discovery means the agency analyzed its process at the same time as opposing counsel did.

The file cannot reconstruct the decision. The single most common failure. The process may have been sound, but the record cannot prove it: no documented rationale for the weighting, no SME panel roster, no scoring sheets retained, no chain of custody on the exams. A process that cannot be reconstructed cannot be defended, no matter how well it was run.

What survives

Processes that survive phase two share one design principle: every component is built for the hearing it might face. The job analysis is current, rank-specific, and signed off by incumbents and supervisors. Every exam item traces to a task or competency in that analysis. The cut-off has a named method, a documented panel, and a written rationale. Assessors are trained against anchored scales and monitored for consistency, and the assessment is recorded where the format allows. Adverse impact analysis runs before certification, with results and any responses documented. And the whole record is assembled as it happens, not reconstructed after a demand letter arrives.

None of this slows a process down once it is built into the design. All of it is expensive to retrofit after the list posts.

Five questions before your next promotional cycle

1. Is the job analysis for each tested rank current for this cycle, and can you produce the documentation behind it?
2. Can you state the method, the panel, and the written rationale behind every cut-off score?
3. Are assessors trained against behaviorally anchored scales, and is rater consistency monitored while scoring is underway?
4. Will adverse impact analysis run before the list is certified, and who reviews the result?
5. If a hearing officer asked for the complete record tomorrow, could you produce it without writing anything new?

The answer to all five should be yes before the announcement goes out. Each no is a finding waiting for an appellant to discover it.

Closing

The promotional list is the visible deliverable. The chief signs it. The candidates frame it. The grievance, the merit board appeal, and the consent decree that can follow focus on everything underneath it. Build the process for the day the list posts, and phase two becomes routine instead of existential.

← All insights