How our practice scores work
Practice score from our model — not a CritiCall result. Agencies set their own cut-offs, and most never show candidates a score.
Version 0.1 · 2026-10-23
What Biddle and hiring agencies publish
- The CritiCall software scores itself, automatically, across almost every part of the test.
- Each hiring agency sets its own cut-off score. Biddle publishes no universal passing score and no pass rate.
- The agency owns the scores, and most agencies never show candidates a result.
- Emergency messages allow 15 seconds each; a wrong agency or a late answer earns no credit.
- Practice items before each module do not count. Skipping is penalized unless the instructions say otherwise.
- Keyboarding is scored in words per minute. Data entry is described as speed and accuracy, with no published unit.
- Percentage bars agencies publish: 70% (Windsor); 75% overall (Pinal, Navajo); 80% (Vermont, Collin County, Santa Fe County, Austin overall, Wichita Falls).
- Typing bars agencies publish: 30 wpm (Newton, Cambridge); 35 wpm (Pinal, Navajo, Vermont, Collin County, Austin call takers); 40 wpm (Austin dispatchers, Windsor, RCMP).
- Agencies combine results in different ways: every module must pass (RCMP British Columbia, Colorado), one overall percentage (Pinal, Navajo, Austin), rank order (Arizona DPS, Colorado), or data entry pass/fail beside the other modules (an Arizona DPS posting).
- Biddle publishes no per-module item counts, time limits, data-entry bars or keyboarding error rule.
Our model
Timings (our timings)
Biddle publishes no per-module timings or item counts. Ours are set from the session lengths agencies publish.
- 10 minutes — Each non-data-entry module. Newton and Cambridge's 2-hour maximum divided by their 11 modules (10.9 minutes, rounded down).
- 16 minutes — Multitask data entry (with emergency messages). Scottsdale's 50 minutes divided by its 3 modules (16.7 minutes, rounded down).
- 5 minutes — Keyboarding. The only typing-test length an agency publishes: Winnipeg's separate typing test, not CritiCall.
Items per module, the emergency-message rate, memory display times and call length have no published basis. We set them, and every screen that shows one labels it ours. Timed audio plays once.
Scores
- Percentage modules: correct answers divided by items shown. An unanswered item counts as wrong.
- Emergency messages: messages answered with the right agency within 15 seconds, divided by messages shown. Late and wrong answers are reported separately.
- Data entry: net keystrokes per hour (characters in fields that end exactly right, per active hour) plus field accuracy (exact fields divided by required fields). Our measure; no agency publishes one.
- Keyboarding: net wpm = gross wpm minus 2 for each uncorrected error — the rule Winnipeg publishes for its typing test. CritiCall's own rule is unpublished.
Bands
Percentage modules and the emergency-message overlay fall in exactly one band, anchored to the bars agencies publish. Agencies publish no separate bar for emergency messages, so the same range is used.
- 80–100% Above the bar
- 70–79.9% In the range
- 0–69.9% Below the range
Agencies we found publish 70% (Windsor) to 80% (Vermont, Collin County, Santa Fe County, Austin).
- Clears 40 wpm
- Clears 35 wpm
- Clears 30 wpm
- Below 30 wpm
Typing bands follow the published typing bars: 30 wpm (Newton, Cambridge), 35 wpm (Pinal, Navajo, Vermont, Collin County, Austin call takers), 40 wpm (Austin dispatchers, Windsor, RCMP).
Net keystrokes per hour — our measure. No agency we found publishes a data-entry bar.
Overall result
- Weakest-module view, for agencies where every module must pass: the lowest band among your percentage modules, plus your keyboarding band.
- Overall view, for agencies that use one overall percentage: the unweighted average of your percentage modules, in the same bands.
- Data-entry results are shown beside the overall percentage, never inside it.
Percentile
Your percentile compares a run with completed, non-practice runs of the same module version on this site. It appears once 30 candidates have run the module; runs that replayed audio are left out.
The labels on your results
Every results screen opens with: “Practice score from our model — not a CritiCall result. Agencies set their own cut-offs, and most never show candidates a score.”
The only band words we use:
- Below the range
- In the range
- Above the bar
- Clears 40 wpm
- Clears 35 wpm
- Clears 30 wpm
- Below 30 wpm