Contact
12 programs
Explore the site

Collabora Office CHIDIST: Right-tail chi-square probability

Statistical Intermediate Collabora Office Calc

Calculate a chi-square survival probability with legacy CHIDIST, correctly interpret the result and preserve its negative-input distinction.

Verification status: See the evidence and test scope described in this guide. How formula examples are checked

Quick answer: =CHIDIST(4;2) independently gives approximately 0.1353352832, the probability that a chi-square random variable with two degrees of freedom is at least 4. CHIDIST returns a right-tail probability, not the chi-square statistic, a density or the probability that a hypothesis is true.

Syntax and distribution direction

=CHIDIST(Number; DegreesFreedom)

Number is the statistic or distribution coordinate. DegreesFreedom identifies the chi-square distribution. The inspected evaluator floors DegreesFreedom and requires the resulting value to be at least 1. The official help describes this function in a hypothesis-testing context, but its “hypothesis confirmed” wording should not be read as a posterior probability for the hypothesis.

For a valid chi-square statistic x, the returned value is P(X≥x) under the specified reference distribution. A smaller right-tail value indicates a more extreme large statistic under that model. Whether that model is appropriate depends on the statistical procedure and its assumptions, not on CHIDIST alone.

Worked example: a hand-checkable two-degree distribution

Enter B2 = 4 and B3 = 2. Use =CHIDIST(B2;B3).

For two degrees of freedom, the survival function has the simple form exp(−x/2). At x = 4, the independent calculation is exp(−2), approximately 0.135335283236613. A percentage display rounds this to about 13.5335%.

The corresponding left-tail probability is 1−exp(−2), approximately 0.8646647168. These two probabilities add to 1. The density at the same point is instead exp(−2)/2, approximately 0.0676676416. Confusing any two of these quantities changes the interpretation substantially.

CHISQDIST provides the left CDF or density using its own optional flag. CHIDIST has no cumulative flag because its output is already the right tail.

Advanced use: start from observed and expected counts

Consider a simple goodness-of-fit illustration with observed counts 6, 14, 10 and expected counts 10, 10, 10. The totals both equal 30. The Pearson statistic is:

(6−10)^2/10 + (14−10)^2/10 + (10−10)^2/10 = 3.2.

If the expected category proportions were specified beforehand and no additional model parameters were estimated, three categories give two degrees of freedom. The reference right-tail probability is exp(−3.2/2), approximately 0.2018965180; the corresponding expression is =CHIDIST(3.2;2).

This example demonstrates the mechanics, not an automatic test recommendation. Expected counts must be positive, categories and observations must satisfy the chosen test’s assumptions, and parameter estimation can change degrees of freedom. CHIDIST does not compute the statistic from a data range or inspect those assumptions. A probability of about 0.202 is not a 20.2% probability that the proposed model is true.

For an interval of chi-square values, subtract right tails in the correct order: =CHIDIST(2;2)-CHIDIST(6;2) independently gives 0.3180923728, from exp(−1)−exp(−3). Reversing that subtraction would incorrectly produce a negative interval probability.

Source-specific boundaries and compatibility

The pinned legacy helper returns 1 for every x ≤ 0, so =CHIDIST(-1;2) predicts 1. A negative chi-square statistic is not a meaningful Pearson statistic; this boundary extension should not hide an upstream data or calculation error.

=CHIDIST(4;2.9) uses floored degrees of freedom 2 and predicts the same result as the main example. =CHIDIST(4;0.9) floors to zero and follows internal IllegalArgument. Use integer degrees of freedom derived from the actual procedure rather than relying on flooring to repair input.

CHISQ.DIST.RT shares a helper but is dispatched with a different flag: its wrapper rejects negative x, while CHIDIST permits the extension described above. They are therefore not interchangeable contracts over all inputs. See ScChiDist and GetChiDist and the right-tail helper.

These examples use elementary independent calculations. Native UI error presentation, export handling and extreme-tail numerical accuracy have not been measured.

Product scope, compatibility and evidence

This guide covers the new Collabora Office desktop for Linux, using the Flathub source tag coda-26.04.4.3-1 and commit 81339d176148d0d5c81ff05d82540f58f8aaa117. The official desktop release notes and Linux package manifest identify that product/build. The package manifest is a dated observation from October 18, 2026, rather than an installed-binary check.

The 26.04 function help and exact evaluator source were inspected on October 18, 2026. Expected examples were independently calculated; no Collabora application was run. Expected scalar results were derived from the documented contracts, inspected source and independent calculations; native output formatting and error localization have not been tested. Classic, Online/CODE, other operating systems and file-format round trips require their own checks.

These are the documented legacy contracts. A similar dotted name does not establish identical argument order, defaults or boundary behavior. The first Collabora release introducing this function is not established by the reviewed sources. No cited retirement or deprecation notice was found; “legacy” here describes the function contract, not a claim that the function has been removed. File-format round trips require separate verification.