A long-term test, at least three months, in the lowest level you actually live in. Everything else is a compromise, and it is worth knowing which compromise you are making.
Test for at least three months, ideally over the heating season, in the lowest level of the house you regularly use. Not the crawlspace you never enter. The basement office, the rec room, wherever people actually spend time.
The reason is that radon swings enormously from day to day with weather, wind, temperature and how often the doors have been open. A reading taken over three days tells you about those three days. Every guideline in the world is written as a long-term average because that is the only thing that describes a house.
A long-term alpha track kit is the usual choice. You buy it, hang it in the right place, leave it for three months or more, then post it to the lab and get a number back. It is genuinely straightforward and the result is as valid as anything else, provided you put it in the right place and leave it alone.
Follow the instructions about distance from walls, windows and heat, because those are the things that skew a result.
A C-NRPP certified measurement professional places the device, controls the conditions, and produces a report. It costs more. It matters most when the result has to satisfy somebody else, such as a buyer, a lender, or an employer, and when you want the reading defended by someone with a certification behind it.
Note that certification for measurement and for mitigation are separate. Our Calgary list separates them because the official registry does not.
A short-term test runs from two to seven days. It is not a good description of a house, and used casually it will mislead you in both directions.
It has one real use: a property transaction with a possession date that does not wait three months. In that situation a short test is what is available, and both sides should understand that the number is a prompt rather than a verdict. If it comes back high, that is grounds to act. If it comes back low, it is not proof the house is fine, and a long-term test afterwards is still worth doing.
Canada acts at 200 Bq/m³, the US EPA at about 148, and the World Health Organization sets a reference level of 100. The differences matter and we compare them here. There is no level below which risk is zero, so lower is better at any reading.
A long-term test needs at least three months and gives the most reliable picture, especially over the heating season. A short-term test takes two to seven days and should be treated as an indication rather than an answer.
Long-term, unless a deadline makes it impossible. Radon varies so much day to day that a short test can easily read half or double the true annual average. The only good reason to use a short test is a property deal on a clock.
A long-term do-it-yourself kit including lab analysis is usually in the region of fifty dollars. A certified professional placing and interpreting the test costs more and varies by company and by what is involved. It is a small enough job that quotes differ more than you would expect, so ask two or three.
Home inspectors often run short-term tests, which is a reasonable compromise given the timeline. Treat a high short-term result as a real signal and get quotes. Treat a low one as encouraging but not conclusive, and test properly once you are in the house.
In the lowest level you regularly occupy, at breathing height, away from exterior walls, windows, vents and heat sources, and somewhere it will not be moved or covered for three months. The kit instructions will be specific and they are worth following exactly.
Sources. Health Canada radon guidance and published measurement data; the 2024 Cross-Canada Survey of Radon; World Health Organization Handbook on Indoor Radon; United States Environmental Protection Agency; Canadian National Radon Proficiency Program. Costs described are general ranges, not quotes.