T.DIST.RT in Collabora Office: Calculate a Student t Right-Tail Probability
Use Collabora T.DIST.RT for signed upper-tail t probabilities and distinguish them from two-sided magnitude tests.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: =T.DIST.RT(1;10) has independently expected result approximately 0.170446566. It is the probability above +1 under a t distribution with ten degrees of freedom.
Syntax and signed tail direction
=T.DIST.RT(Number; DegreesFreedom)
Number is the signed evaluation point, and DegreesFreedom identifies the t distribution. Both are required. RT means right tail, P(T>=x); it does not mean the smaller tail regardless of the sign of x.
The documented Collabora example uses positive one. Current source corroboration handles negative Number values by complementing the positive-magnitude upper-tail calculation. It approximately floors degrees of freedom and requires the effective count to be at least one.
A negative threshold can therefore have a right-tail probability above one half. Do not automatically take ABS(Number) unless the intended event is defined using a magnitude rather than the original signed value.
Example 1: build a signed upper-tail table
Put -2, -1, 0, 1 and 2 in A2:A6. Enter =T.DIST.RT(A2;10) in B2 and fill down. Independent probabilities are approximately 0.963305983, 0.829553434, 0.5, 0.170446566 and 0.036694017.
The right tail decreases as the threshold increases. Its value at -1 is about 0.829553434 because most of the distribution lies to the right of that negative point.
If both -1 and +1 produce the small value 0.170446566 in an imported calculation, inspect whether the sign was removed or a legacy magnitude convention was used.
Example 2: verify left/right complements
In C2 enter =T.DIST(A2;10;1) and fill down. Add B2+C2 in D2 and fill down; each independent target is 1.
At x=1, the left-tail value is approximately 0.829553434 and the right-tail value approximately 0.170446566. The density at the same point, approximately 0.230361989, is neither complementary probability.
The explicit cumulative flag in T.DIST matters. Function names alone are not enough to determine whether two numerical results can be added as complementary events.
Advanced application: calculate a one-sided threshold
For ten degrees of freedom, =T.INV(0.95;10) has independent target approximately 1.812461123. Passing that value to T.DIST.RT targets 0.05.
By comparison, the positive boundary of a symmetric two-sided 5% test is approximately 2.228138852, whose upper tail is 0.025. It is larger because the total 5% excluded probability is split between two tails.
Choose the direction and testing rule before looking at the data when using these calculations in a real analysis. The function does not decide which alternative hypothesis is appropriate.
Advanced application: relate a right tail to two tails
For a nonnegative magnitude x, =2*T.DIST.RT(x;10) equals the symmetric two-tail probability. At x=2 the independent result is approximately 0.073388035.
This relationship must not be applied blindly to negative x. For x=-2 the right-tail probability is about 0.963305983, so doubling it is not a valid probability. A two-sided magnitude calculation uses the positive magnitude two instead.
Keep the estimated effect direction in a separate cell even when a two-sided probability depends only on its absolute statistic. A small probability does not describe the effect’s practical importance.
Errors and troubleshooting
Use positive integer degrees in reproducible controls and verify whether the input is a signed t statistic or an absolute magnitude. Reversing the subtraction used to create the statistic changes the right-tail question, even though a symmetric two-tail result would be unchanged.
Direct right-tail evaluation avoids an explicit workbook subtraction from a cumulative value near one. That is a numerical motivation, not a native accuracy guarantee. Error displays, all coercions and extreme values remain unverified in the actual application. A probability conditional on a model is not a posterior probability that a hypothesis is true.
Availability and evidence
The exact function is documented in Collabora Office 26.04 desktop help. That documentation snapshot was retrieved on 2026-10-13 and hash-verified on October 15, 2026. It does not establish Online or mobile support. First introduction and last functioning release are unknown in this evidence.
Additional implementation statements come from Collabora’s current engine source, retrieved October 15, 2026. This moving-main source is corroboration, not a pinned desktop 26.04 binary or a native execution. No native Collabora Office run occurred. Worked outputs are independent mathematical expectations.
Formulas use English names and semicolon separators. The help identifies namespace COM.MICROSOFT.T.DIST.RT, outside its cited OpenFormula 1.3 standard; that does not establish identical behavior in another application.