T.DIST.2T in Google Sheets: Two-Tailed t Probabilities
Calculate two-tailed Student t probabilities with T.DIST.2T in Google Sheets, handle negative test statistics correctly and distinguish outside-tail from central area.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
Quick answer: =T.DIST.2T(2,9) has expected result 0.0765528238. It returns the total probability beyond -2 and +2 for a t distribution with 9 degrees of freedom.
Syntax and documented domain
=T.DIST.2T(x, degrees_freedom)
The Google T.DIST.2T reference requires both arguments. It documents a numeric error for negative x or degrees of freedom below 1. Use a nonnegative threshold magnitude and valid degrees of freedom.
x: magnitude of the t threshold.degrees_freedom: degrees of freedom for the distribution.
There is no cumulative flag or separate tails selector in this function. The two-tailed choice is built into its name. The reviewed page does not establish fractional-degrees-of-freedom handling, so the examples use whole numbers.
Google states that T.DIST.2T corresponds to TDIST with its tails argument set to 2. This does not make it interchangeable with a left-tail CDF or a probability density.
Worked example: compare threshold magnitudes
Enter 0, 1, 2 and 3 in A2:A5. Put 9 in D1. In B2, enter and fill down:
=T.DIST.2T(A2,$D$1)
| Threshold magnitude | Expected combined outside-tail probability |
|---|---|
| 0 | 1 |
| 1 | 0.3434363961 |
| 2 | 0.0765528238 |
| 3 | 0.0149563639 |
The outside-tail probability decreases as the threshold moves farther from zero. At zero, the two tails together cover the entire continuous distribution, giving probability 1.
For threshold 2, approximately 7.66% of the distribution lies at least that far from zero in either direction. This is not the probability between -2 and +2; that central probability is the complement.
Independent mathematical checks
The t distribution is symmetric. For a nonnegative x, the combined outside probability is twice the right-tail probability at x. For x = 2 and 9 degrees of freedom, the right-tail area is approximately 0.0382764119, and doubling it gives 0.0765528238.
NIST’s t-distribution reference identifies the 1-degree-of-freedom case as a Cauchy distribution. That provides the simple check:
=T.DIST.2T(1,1)
Expected result: 0.5. Each outside tail has area 0.25. Google’s own printed examples also independently reproduce the expected two-tailed values for (1.96,60) and (1,2).
These are mathematical and documentation checks, not observations from a native Sheets session.
Advanced application: use a signed test statistic safely
Suppose E2 contains a previously calculated signed statistic -2, and E3 contains 9 degrees of freedom. For a two-sided test, use its magnitude:
=T.DIST.2T(ABS(E2),E3)
Expected result: 0.0765528238. Passing -2 directly is outside Google’s documented domain. Taking ABS is appropriate here because the two-sided question treats equal departures in either direction symmetrically.
Do not automatically apply ABS for a directional one-sided test. That would discard the sign needed to decide which tail corresponds to the prespecified alternative.
If the test’s assumptions and statistic construction are valid, this outside-tail area can serve as a two-sided p-value. It is not the probability that the null hypothesis is true, nor a measure of practical effect size.
Central area and a rejection rule
For the probability between -2 and +2:
=1-T.DIST.2T(2,9)
Expected result: 0.9234471762. The complement converts outside area into central area.
For an illustrative 5% decision threshold, put 0.05 in G1 and use:
=IF(T.DIST.2T(ABS(E2),E3)<$G$1,"Below alpha","Not below alpha")
With E2 = -2 and E3 = 9, the expected label is Not below alpha. Choose the test and significance level before examining the result; changing them to obtain a preferred conclusion invalidates the simple interpretation.
Errors, limitations and version scope
Check negative x, degrees of freedom below 1, numeric input types and whether the question is genuinely two-sided. The reviewed page explicitly documents the first two domain failures but does not enumerate all text/blank coercion or fractional-input cases.
Degrees of freedom are not always sample size minus one; they depend on the statistical procedure. A wrong degree count can produce a plausible probability without a spreadsheet error.
Current Google Sheets documents T.DIST.2T as a replacement for the two-tail TDIST form. That statement does not establish that TDIST has stopped working. Original introduction date and first or last functioning numbered releases were not established.
NOT_TESTED_IN_GOOGLE_SHEETS. Google’s individual page and NIST mathematics were inspected on October 17, 2026. Original tails, complements and decision outcomes were independently calculated. No native execution or screenshot is claimed.