The source publishes one row per sample taken at a fixture, so every figure below counts fixture samples. A sample is counted above the action level here when its measured value is at or above 5 ppb.
- Samples collected
- 20,366
- Non-detects
- 7,673 — reported below the laboratory’s reporting limit. A non-detect is a censored observation, not a zero and not the limit, so it counts in the samples-collected denominator and is excluded from the measured one.
- Measured results
- 12,693
7,673 non-detects + 12,693 measured = 20,366 samples collected
22.2% (4,531 of 20,366 samples collected) — fixtures at or above 5 ppb, the state's action level
35.7% (4,531 of 12,693 measured samples) — fixtures at or above 5 ppb, the state's action level
Both figures count the same 4,531 results at or above 5 ppb. They differ only in what that count is divided by, and dropping the rows with no measured value can only raise the rate — for a state reporting mostly non-detects it raises it by more than an order of magnitude. So we publish both denominators, and neither figure is ever shown alone.
85.4% (351 of 411 schools with published results) — schools at or above 5 ppb, the state's action level
The two rates above count fixture samples; this one counts schools with at least one sample at or above the level. The gap between them is itself informative, and the two units must never be mixed inside one comparison.
Sampling period: 2001-03-13 to 2019-04-09.
Testing is required by Washington's own school testing requirement, which this survey records as in force; the survey has not established which programme produced the rows published here.