Methodology & data sources

Radon-level data

County radon levels come from the CDC National Environmental Public Health Tracking Network (EPHT) — specifically the lab-reported radon test measures, which aggregate residential radon tests analyzed by laboratories into county-level statistics. We publish four figures per county:

• Percent of tests at or above 4 pCi/L — the share of pre-mitigation tests that met or exceeded the EPA action level. This is the headline number.
• Median and mean radon level (pCi/L), where reported.
• Maximum recorded reading, and the number of tests (n) behind every figure.

Vintage: the CDC lab dataset covers a rolling ten-year window; the current national block is 2008–2017. We label this on every page. Radon levels change slowly, but a 2008–2017 average is a long-run picture, not this year's snapshot. Where a state publishes fresher data (see below), we use that instead and say so.

The n-floor: why we hide some counties

A county where only three homes were ever tested can show “100% over the action level” — which is technically true and statistically meaningless. We apply a minimum of 50 tests before a county gets its own page or appears in a “worst counties” ranking. The ZIP lookup will still show low-count data (people want an answer for their address), but it flags any county under 50 tests as limited data. We never hide the n.

Fresher state datasets

Several states publish their own radon test data, often more recent and more granular than the CDC's national block. Where we use one, the page cites that state agency directly. Our first deep-data state is Pennsylvania, whose Department of Environmental Protection publishes individual test records from 1986 to the present via the Pennsylvania Open Data portal. Some high-radon states (Iowa, the Dakotas) publish nothing beyond the CDC baseline — we say so rather than pretending otherwise.

Cost data

Radon mitigation prices on this site come from real, published contractor pricing and licensed-installer quotes — not synthetic cost models. Where we cite a typical price (for example, ~$1,500 for a standard sub-slab system), it reflects prices verified with working NRPP-certified mitigation contractors and state radon programs. When state-specific pricing is thin, we say so and give the national range rather than inventing a precise-looking number. We would rather be honestly approximate than falsely exact.

What "% over 4 pCi/L" means

The EPA sets an action level of 4 pCi/L (picocuries per liter). At or above it, the EPA recommends installing a mitigation system. There is no known safe level of radon; the action level is the threshold for action, not a line between safe and dangerous. When we say “46% of tests came back over the action level,” we mean 46% of that county's tested homes read at or above 4 pCi/L.

Updates

We refresh the CDC baseline when a new ten-year block publishes, and state datasets on their own cadences (Pennsylvania's is effectively continuous). Each page shows the data vintage it's built from.

Full attribution

Centers for Disease Control and Prevention, National Environmental Public Health Tracking Network. Accessed July 2026. ephtracking.cdc.gov.

Pennsylvania Department of Environmental Protection, radon test results, 1986–present, via data.pa.gov. State datasets are cited on their respective pages. U.S. Census Bureau ZCTA-to-county relationship files (2020) and GeoNames postal data underlie the ZIP lookup.

Questions about a specific figure? Look up your ZIP or start with the state data hub.