TINV Function (OpenOffice Calc)

• Introduced: First release not established. The OpenOffice.org 3.3 function appendix documents this name; that is presence evidence, not its introduction date.
• Supported in: Apache OpenOffice 4.1.16, with no final supported release established.
• Scope: Desktop Calc; English names and semicolon separators are used below. The numeric contract refers to the cited release.
• Checked: October 14, 2026. Retained official documentation/source and independent mathematics checked; no new Apache OpenOffice Calc execution is claimed.

TINV solves TDIST(t;df;2)=p for a nonnegative t. The input is the combined probability in both tails. Thus p=0.05 gives the central 95% reference interval’s positive boundary; it does not request the lower fifth percentile. Function reference.

A smaller positive tail probability produces a larger cutoff. This inverse operation is different from taking 1/TDIST. When moving a formula between programs, verify whether its inverse function uses one tail, two tails or a left-tail cumulative probability.

=TINV(Number; DegreesFreedom)

Both arguments are required. Number must satisfy 0<p≤1. The degree count is floored and must be at least 1. Apache OpenOffice 4.1.16 also requires floored degrees of freedom below 100,000. A value of exactly 100,000 is rejected by this wrapper, even though the mathematical t distribution exists there. The guards are explicit in the pinned evaluator.

At p=1 the central reference interval shrinks to zero and the cutoff is 0. At p=0 the finite cutoff does not exist; the evaluator rejects that endpoint rather than returning a usable large number. Permitted inputs can separately encounter a numerical-convergence failure.

These independent numerical references are rounded to twelve decimal places. They are not new application observations.

For an independent sample from a normal population with n=11 and sample SD=3, a two-sided 95% mean interval uses =TINV(0.05;10)*3/SQRT(11). The expected margin is approximately 2.015427423505.

If the sample mean is 52, the bounds are approximately 49.984572576495 and 54.015427423505. Use a sample SD, ordinarily calculated with STDEV or its supported sample-SD equivalent, and the count of the same numeric observations. A population SD known independently leads to a different normal-critical-value calculation.

In a worksheet, keep Alpha, n, sample SD and sample mean in separate labeled cells. Retain the unrounded TINV result while computing the margin; round the final reported interval only after calculation.

For an upper-tail probability 0<q≤0.5, use =TINV(2*q;df). Thus =TINV(0.10;10) is approximately 1.812461122812, the upper 5% cutoff. Entering 0.05 directly instead gives the more extreme two-tail 5% cutoff.

Check the two-tail orientation with =TDIST(TINV(0.05;10);10;2), expected approximately 0.05. Changing the final mode to 1 gives approximately 0.025. An interval covering the central 90% uses TINV(0.10;df), not TINV(0.90;df).

A t cutoff is dimensionless. Multiplying it by a standard error creates a margin in the original measurement units. The interval concerns a population mean under its sampling model, not 95% of individual observations or future outcomes.

Negative or zero probabilities, probabilities above 1 and invalid degree counts follow the argument guards; non-convergence is a separate numerical condition. Avoid masking all errors with a zero margin. Model assumptions, missing observations, text conversion and extreme-tail numerical precision require their own checks rather than being inferred from a successful ordinary example.

• Function-specific Apache OpenOffice 4.1.16 implementation
• Two-tail dispatch or inverse helper

The examples were recomputed independently against the retained release contract. Native import, locale and extreme-input behavior is not newly certified.