Collabora Office NORMDIST: Normal Probabilities and Density
Use NORMDIST’s optional cumulative flag, standardize values and distinguish its syntax from mandatory-flag NORM.DIST. Scoped to the new Linux desktop source build.
Verification status: See the evidence and test scope described in this guide. How formula examples are checked
NORMDIST evaluates a normal distribution with an explicit mean and standard deviation. Its optional fourth argument chooses lower-tail cumulative probability or density. When that argument is omitted, the checked legacy evaluator selects the CDF.
Syntax and the dotted contract
=NORMDIST(Number; Mean; StDev [; C])
Number is the value x, Mean is mu and StDev is a positive sigma. C omitted or TRUE gives P(X≤x); FALSE gives the density at x. Use Boolean flags explicitly when clarity matters.
The related NORM.DIST(Number;Mean;StDev;Cumulative) calls the same evaluator with a four-argument minimum. Legacy NORMDIST(115;100;15) is therefore not a drop-in four-argument dotted call. These are distinct signatures even where the mathematical result agrees.
Step by step: one standard deviation above the mean
Enter A2 115, B2 100 and C2 15, then calculate =NORMDIST(A2;B2;C2).
=NORMDIST(115;100;15)
Standardization gives (115-100)/15 = 1. The independent lower-tail probability is approximately 0.8413447461, or 84.13447461%. It means the stipulated normal model places that fraction at or below 115; it does not establish that an actual dataset follows the model.
=NORM.DIST(115;100;15;TRUE) has the same ordinary-input source/math target. At the mean, =NORMDIST(100;100;15) has expected 0.5, reflecting symmetry.
Density mode includes the scale factor
=NORMDIST(115;100;15;FALSE)
The density is phi(1)/15, approximately 0.01613138163, where phi is the standard normal density. At the mean, =NORMDIST(100;100;15;FALSE) is approximately 0.02659615203.
Dividing by sigma is essential. The density at standardized z=1 is not the same numeric quantity as the density at x=115 measured in the original units. Density has inverse units of x and is not a point probability. A continuous distribution assigns probability zero to one exact point; probabilities over ranges are areas or differences of CDF values.
Reproduce the help’s rounded values
The published examples use x=70, mean 63 and sigma 5. Standardization is (70-63)/5 = 1.4.
| Formula | Independent value | Help’s rounded display |
|---|---|---|
=NORMDIST(70;63;5;FALSE) |
0.02994549313 | 0.03 |
=NORMDIST(70;63;5;TRUE) |
0.9192433408 | 0.92 |
The two outputs answer different questions. Displaying the density as a percentage would not turn it into the probability of observing exactly 70. The extra digits above come from independent mathematics, not a captured application result or an accuracy guarantee.
Advanced: intervals and upper tails
For the mean-100, sigma-15 model, the probability between 85 and 115 is the CDF at 115 minus the CDF at 85. Symmetry gives 0.8413447461-0.1586552539, approximately 0.6826894921.
The probability above 115 is 1-0.8413447461, approximately 0.1586552539. This ordinary example is numerically well behaved, but subtracting a CDF extremely close to 1 can lose relative precision. No extreme-tail accuracy is promised by these controls.
A formula for the central interval is =NORMDIST(115;100;15)-NORMDIST(85;100;15). Use the same mean and standard deviation for both endpoints. Subtracting values from two different distributions would not give the probability of one interval in either model.
Errors and model checks
The source rejects sigma≤0 with internal IllegalArgument, as in =NORMDIST(100;100;0). Standard deviation is not variance: entering 225 instead of 15 changes the model’s scale substantially. Negative x is allowed mathematically; it is a location on the real line, not an invalid negative probability input.
Keep units consistent between x, mean and sigma. The function evaluates supplied parameters and does not estimate them or test normality. This guide does not infer every text/empty-slot coercion, localized error message or statistical decision rule from the scalar examples.
Product, version and verification
This guide is scoped to the new Collabora Office desktop for Linux, package source tag coda-26.04.4.3-1, commit 81339d176148d0d5c81ff05d82540f58f8aaa117. The 26.04 NORMDIST help, exact evaluator and dispatch source were checked October 18, 2026. The Linux package manifest provides the recorded product mapping; it is mutable and is not proof of an installed binary.
No Collabora runtime or calculation engine was executed. All worked values are independent mathematical controls or explicit source predictions; vendor display rounding is identified separately. No screenshots, performance benchmarks or cross-platform tests are claimed. This evidence does not certify Classic, Online, CODE, mobile, Windows or macOS.
“Legacy” distinguishes the function’s spelling and argument contract. The checked registration and current help do not establish that it is retired or deprecated. Its first Collabora introduction release is unknown; upstream release history and file-conversion compatibility are not inferred.