Skip to content

Dixon's Q Test Calculator

Whether that one odd measurement can be discarded.

Work out Dixon's Q Test. Whether that one odd measurement can be discarded. Written for the problem set you are stuck on.

Written and maintained by Mohit PatelLast checked August 4, 2026How we build these

Between 3 and 10 values, comma separated. The test only works in that range.

Q

0.842

Above the 0.71 critical value — 12 can be rejected

Q0.8421
Critical value at 95%, n = 50.710
Suspect value12
Gap to nearest neighbour1.6
Full range1.9
VerdictReject the outlier

Q is the gap between the suspect value and its nearest neighbour, divided by the total range. Above the critical value the point can be discarded at 95% confidence. The test assumes exactly one outlier in a normally distributed set, and it applies only to the highest or lowest value. Running it repeatedly to strip several points is not what it tests — that manufactures a precision the data never had, and it is why journals ask you to report rejected points rather than quietly removing them. A failed Q test is not permission to ignore a result either. An outlier that keeps appearing is usually telling you something about the method.

How the Dixon's Q Test Calculator works

Paste three to ten measurements and this identifies the suspect value, computes Q and compares it against the 95% critical value. It also says plainly what the test does not license — repeated application to strip several points is not what it tests.

Also known as: q test calculator · outlier test calculator · how to reject an outlier · q test critical values

Frequently asked questions

How do I do a Q test?

Sort the data, take the gap between the suspect value and its nearest neighbour, and divide by the full range. If Q exceeds the critical value for your sample size, the point can be rejected at 95% confidence.

What are the critical values?

At 95% confidence: 0.970 for three points, 0.829 for four, 0.710 for five, 0.625 for six, 0.568 for seven, 0.526 for eight, 0.493 for nine and 0.466 for ten.

Can I run the test twice to remove two outliers?

No. It assumes a single outlier in a normal distribution. Applying it repeatedly manufactures precision the data does not have, which is why journals ask for rejected points to be reported rather than quietly removed.

What if I have more than ten measurements?

Use Grubbs' test, which is tabulated for larger sets and is generally preferred in modern practice. Dixon's Q is only tabulated to ten.

Put this calculator on your own site

Free to use, on any site, commercial or not. Paste this where you want it to appear. It is a plain iframe, so it works in WordPress, Squarespace, Wix, Webflow, Ghost and anything else that accepts HTML.

The one-line version
<iframe src="https://www.thecalclibrary.com/embed/dixons-q-test-calculator" width="100%" height="640" style="border:1px solid #e2e8f0;border-radius:12px" loading="lazy" title="Dixon's Q Test Calculator"></iframe>
<p style="font:13px/1.5 system-ui,sans-serif;margin:6px 0 0;color:#64748b">Powered by <a href="https://www.thecalclibrary.com/dixons-q-test-calculator" style="color:#64748b">Dixon's Q Test Calculator</a> from The Calc Library</p>

The only condition is that the credit line below the frame stays in place. That one line is what pays for the tool being free — it is how anyone else finds it.

Related calculators