Department evaluation guide
What the numbers tell you
Public evaluation · Current availability
A2C3 records program activity and combines it into totals. Use these results to guide outreach and choose topics to teach. Compare them with what your department knows locally.
Activity is not a count of people
- Starts and finishes: recorded program sessions. One person can start more than once.
- Estimated devices: browser profiles linked to a department, not identified people. Shared devices, different browsers, or cleared browser data can change this estimate.
- Responses: recorded answers or actions. Ten quiz answers do not mean ten people.
- All-program completion rate: finished sessions divided by starts across programs for the selected period and source.
- Quiz completion rate: completed quiz rounds divided by quiz rounds started. This quiz-only measure has a different scope from the all-program rate.
- A session is counted with the date, program, and QR code or link where it started.
When a session can finish
For reporting, a quiz attempt can finish within 2 hours. Other games can finish within 4 hours total. A Home Safety Survey can finish within 30 days. Recent starts may still be in progress, so the finish rate is a snapshot and may rise as sessions finish.
Matched participation dates
The Leadership Brief counts starts and finishes matched to the same original start date and source. If data coverage begins after the selected reporting period starts, the report shows the available matched window. Earlier activity is not included in those totals.
Small samples receive less detail
| Applicable observations | How to interpret the result |
|---|---|
| Below 5 | Small-sample results are suppressed. |
| 5–9 | Only very limited aggregate reporting is available. |
| 10–24 | Results are marked as a limited sample. |
| 25 or more | Standard aggregate reporting may be shown; this does not establish statistical significance. |
Rules apply to the relevant result group, not simply to all visits on the site. Question and scenario detail may remain unavailable when an overall total is larger. Read the report’s privacy notices. A hidden or blank value does not mean zero.
Each program measures something different
Quiz reports show correct and missed answers. Survey reports group structured responses and educational guidance statuses; they are not scores or inspection results. Scenario reports show recorded actions and outcomes. Compare like measures, program versions, date ranges, and campaign scopes.
Patterns are starting points for discussion
A frequently missed question may suggest a topic worth explaining. It does not establish a community-wide knowledge deficit. Campaign differences can reflect audience, timing, placement, repeated activity, or sample size. Completion does not prove learning, safer behavior, or reduced incidents. Topic heat displays show answer patterns, not a geographic fire-risk map.
Data quality and historical limitations
Knowledge Challenge activity through September 27, 2026 may understate answers, completions, scores, and rates because an older package rejected some events. Missing answers cannot be reconstructed. Use periods beginning September 28, 2026 for post-repair comparisons. Preserve any data-quality notices included with reports and exports.
Demonstration reports
Public samples use testing activity, not a measured community or department population. They illustrate the tools and are not for official reporting, grant documentation, or formal evaluation.
Read the full privacy and retention policy · Understand export rows and fields