CONFIDENCE.T Function (LibreOffice Calc)
The CONFIDENCE.T function in LibreOffice Calc returns the margin of error for a confidence interval using the Student t-distribution. This guide explains syntax, interpretation, examples, errors, and best practices.
Compatibility
▾| Excel | ✔ |
| Gnumeric | ✔ |
| Google_sheets | ✔ |
| Libreoffice | ✔ |
| Numbers | ✖ |
| Onlyoffice | ✔ |
| Openoffice | ✖ |
| Wps | ✔ |
| Zoho | ✔ |
What the CONFIDENCE.T Function Does â–¾
- Computes the margin of error using the t-distribution
- Correct for small samples or unknown population SD
- Uses sample standard deviation
- Works across sheets
The confidence interval is:
[ \bar{x} \pm \text{CONFIDENCE.T} ]
Syntax â–¾
CONFIDENCE.T(alpha; standard_dev; size)
Where:
alpha— significance level (e.g., 0.05 for 95% confidence)standard_dev— sample standard deviation (STDEV.S)size— sample size
Use CONFIDENCE.NORM only when population SD is known or sample size is large.
Interpretation of Alpha â–¾
| Alpha | Confidence Level |
|---|---|
| 0.10 | 90% |
| 0.05 | 95% |
| 0.01 | 99% |
Basic Examples â–¾
95% confidence interval margin of error
=CONFIDENCE.T(0.05; STDEV.S(A1:A30); COUNT(A1:A30))
99% confidence interval
=CONFIDENCE.T(0.01; STDEV.S(A1:A20); COUNT(A1:A20))
Across sheets
=CONFIDENCE.T(0.05; Sheet1.B1; Sheet2.C1)
Advanced Examples â–¾
Full confidence interval (lower and upper bounds)
Lower bound:
=AVERAGE(A1:A30) - CONFIDENCE.T(0.05; STDEV.S(A1:A30); COUNT(A1:A30))
Upper bound:
=AVERAGE(A1:A30) + CONFIDENCE.T(0.05; STDEV.S(A1:A30); COUNT(A1:A30))
Confidence interval ignoring errors
=CONFIDENCE.T(0.05; STDEV.S(IF(ISNUMBER(A1:A100); A1:A100)); COUNT(IF(ISNUMBER(A1:A100); A1:A100)))
(Confirm with Ctrl+Shift+Enter in older Calc.)
Confidence interval after removing outliers
=CONFIDENCE.T(0.05; STDEV.S(FILTER(A1:A100; A1:A100<1000)); COUNT(FILTER(A1:A100; A1:A100<1000)))
Confidence interval for small-sample experiments
=CONFIDENCE.T(0.05; STDEV.S(ExperimentData); COUNT(ExperimentData))
Confidence interval for standardized data
=CONFIDENCE.T(0.05; 1; COUNT(A1:A30))
(Standardized SD = 1)
How CONFIDENCE.T Calculates Margin of Error â–¾
- Compute t-critical value:
[ t = t^{-1}(1 - \alpha/2, , n - 1) ]
- Compute standard error:
[ SE = \frac{s}{\sqrt{n}} ]
- Compute margin of error:
[ ME = t \cdot SE ]
Where:
- ( s ) = sample standard deviation
- ( n ) = sample size
Common Errors and Fixes â–¾
Err:502 — Invalid argument
Occurs when:
- Alpha ≤ 0 or ≥ 1
- Standard deviation ≤ 0
- Sample size < 2
Err:504 — Parameter error
Occurs when:
- Semicolons are incorrect
- Range references malformed
Margin of error seems too large
Possible causes:
- Very small sample size
- Large sample SD
- High confidence level (small alpha)
Best Practices â–¾
- Use CONFIDENCE.T for small samples or unknown population SD
- Use CONFIDENCE.NORM only for large samples or known SD
- Remove outliers before computing SD
- Use named ranges for cleaner formulas
- Report both margin of error and interval bounds
- Use T.TEST and T.DIST for deeper inferential analysis
CONFIDENCE.T gives you the statistically correct margin of error for real-world data — where population standard deviation is almost never known.