TDIST 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.

With nonnegative x, mode 1 returns P(T≥x). Mode 2 returns P(|T|≥x), twice the upper tail because the t distribution is symmetric. Neither setting is the left-tail CDF. A two-tail value concerns unusually large magnitudes in either direction. Function reference.

Student t reference distributions arise in several statistical methods. For a one-sample mean test under an independent normal-sample model, the statistic is (sample mean−hypothesized mean)/(sample SD/SQRT(n)), with n−1 degrees of freedom. The function itself only evaluates a distribution tail.

=TDIST(Number; DegreesFreedom; Mode)

All three arguments are required. Number must be nonnegative. The degree count and Mode are floored; afterward the degree count must be at least 1 and Mode must be exactly 1 or 2. There is no omitted-mode default. Release evaluator.

Use an integer mode deliberately. Although 2.9 floors to 2 in the pinned evaluator, it is not a meaningful third kind of test. Similarly, the relevant degree count comes from the model rather than from arbitrary rounding of a sample count.

The following independently calculated results are rounded to twelve decimal places. They do not represent a new native Calc test run.

Suppose n=11, sample mean=52, sample SD=3 and the hypothesized mean is 50. Put those four values in A2:D2. In E2 enter =(B2-D2)/(C2/SQRT(A2)), giving approximately 2.211083193570. In F2 enter =TDIST(ABS(E2);A2-1;2).

The expected two-sided tail probability is approximately 0.051466173011. This calculation assumes the stated independent normal-sample setting; it does not establish those conditions or justify deleting inconvenient observations.

ABS is appropriate here because a two-sided question treats equally large positive and negative departures alike. For a one-sided question, the sign matters. Automatically taking ABS would discard the direction the test was supposed to examine.

The old TDIST interface rejects a negative first argument. To evaluate the upper tail of a signed t statistic in E2, use:

=IF(E2>=0;TDIST(E2;10;1);1-TDIST(-E2;10;1))

At E2=−2 the expected upper tail is 0.963305982615; at +2 it is 0.036694017385. Choose the alternative direction before observing the data. A lower-tail question reverses which probability is wanted.

For x=2 and df=10, =TDIST(2;10;2) equals =2*TDIST(2;10;1). The probability between −2 and +2 is =1-TDIST(2;10;2), approximately 0.926611965229. These identities are useful orientation checks.

Negative Number, a floored degree count below 1, or a floored mode other than 1 or 2 follows the invalid-argument branch. At Number=0 the upper tail is 0.5 and the two-tail probability is 1. Tail probability is dimensionless and should not be formatted as a measurement in the units of the original data.

The examples check ordinary numeric cases and the documented/source contract. They do not certify every text-conversion rule, locale or extreme tail. Preserve the mode when transferring a formula; a two-argument inverse or another application’s cumulative t function is not an interchangeable call.

• Function-specific Apache OpenOffice 4.1.16 implementation

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