CONFIDENCE.T Function (LibreOffice Calc)

Math Intermediate LibreOffice Calc Introduced in LibreOffice 4.0
confidence-interval statistics t-distribution inferential-statistics margin-of-error

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

â–¾

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

  1. Compute t-critical value:

[ t = t^{-1}(1 - \alpha/2, , n - 1) ]

  1. Compute standard error:

[ SE = \frac{s}{\sqrt{n}} ]

  1. 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.

Copyright 2026. All rights reserved.