TDIST Function (Gnumeric)
Calculate one-tail and two-tail Student t probabilities in Gnumeric, with negative-statistic handling and fractional degrees of freedom.
Verification status: Selected formulas were recalculated with Gnumeric 1.12.57 using ssconvert and retained workbook results. Gnumeric 1.12.62 was source-checked only; this does not establish every edge case. How formula examples are checked
Quick answer: =TDIST(1,10,1) gives approximately 0.170446566151.
Availability and version history
- Introduced: Gnumeric 0.25, explicitly named in its release notes.
- Supported in: Gnumeric 1.12.62, confirmed by its function registration. No retirement is established.
- Scope: Desktop Gnumeric with the Statistics Functions plugin. Tables use English comma-separated formulas.
- Checked: October 14, 2026. Tables were recalculated in 1.12.57 and finite reference values independently checked. Version 1.12.62 was source-checked, not executed.
What the function calculates
With tails=1, TDIST returns the right-tail probability P(T≥x), including negative x. With tails=2 and x≥0, it returns P(|T|≥x), twice the upper tail at x. These are different questions. At x=0, one-tail probability is 0.5 and two-tail probability is 1. A negative one-tail statistic therefore need not have a small p-value. Official function reference.
Syntax and arguments
=TDIST(x,dof,tails)
| Argument | Meaning |
|---|---|
| x | t statistic; must be nonnegative in two-tail mode. |
| dof | Degrees of freedom, at least 1; fractional values are retained. |
| tails | Exactly 1 for a right-tail probability or 2 for the two-tail probability at nonnegative x. |
All arguments shown are required. The version-pinned help and evaluator establish the Gnumeric contract.
Reproducible reference cases
The table records the tested release’s outputs, with finite mathematical references rounded to twelve decimal places. Keep full precision when using a result in a later calculation.
| Formula | Result in Gnumeric 1.12.57, approximately |
|---|---|
=TDIST(0,10,1) |
0.5 |
=TDIST(0,10,2) |
1 |
=TDIST(1,10,1) |
0.170446566151 |
=TDIST(1,10,2) |
0.340893132302 |
=TDIST(-1,10,1) |
0.829553433849 |
=TDIST(2,10,2) |
0.073388034771 |
=TDIST(2,5.5,1) |
0.048377195116 |
=TDIST(1,1,1) |
0.25 |
Practical worksheet
For a pre-specified two-sided t test whose justified statistic is 2 with 10 degrees of freedom, put 2, 10 and 2 in A2:C2. =TDIST(A2,B2,C2) returns approximately 0.073388034771. For a one-sided upper alternative, changing C2 to 1 gives half that value. Choose the direction and number of tails before inspecting the data; switching afterward changes the test.
Advanced examples and independent checks
For a lower-tail probability at a signed statistic t, symmetry gives =TDIST(-t,dof,1). For a two-sided probability from a signed statistic use =TDIST(ABS(t),dof,2). For example, t=−1 and dof=10 give an upper-tail probability of approximately 0.829553433849, a lower-tail probability of 0.170446566151, and a two-sided probability of 0.340893132302.
Gnumeric retains fractional degrees of freedom. This is useful when a justified model, such as an unequal-variance approximation, produces a noninteger value. =TDIST(2,5.5,1) therefore does not mean “use five degrees of freedom.” The degree count must still come from the statistical method, not be adjusted to obtain a desired result.
Input boundaries and common mistakes
The tails argument must be exactly 1 or 2; 1.5 is not rounded into a valid selector. Two-tail mode rejects negative x, whereas one-tail mode accepts it. Degrees of freedom below 1 are rejected without flooring accepted fractional values. TDIST only evaluates a distribution tail; it does not inspect sample independence, normality, variance assumptions or the way the statistic was constructed.
These additional outcomes were observed in the same native recalculation. F5 was genuinely empty; it is not an omitted argument. Missing inputs still need deliberate validation, even if they are converted to a numeric zero.
| Formula | Observed in 1.12.57 |
|---|---|
=TDIST(-1,10,2) |
#NUM! |
=TDIST(1,0.9,1) |
#NUM! |
=TDIST(1,10,1.5) |
#NUM! |
=TDIST(F5,10,1) |
0.5 |
The worked calculations do not certify every locale, imported text value, extreme parameter or later binary version. Shared numerical helpers can change even when a function’s wrapper remains the same.
Sources and verification
The reference and boundary tables use recorded 1.12.57 command-line recalculation. Further model examples were independently calculated from their stated inputs; they are not additional native test claims.