Collabora Office NORMSDIST: Standard Normal Lower-Tail Probabilities

=NORMSDIST(Number)

Number is z, not a probability and not an unstandardized measurement. This legacy signature has no mean, standard deviation or density flag. The related dotted form is NORM.S.DIST(Number;Cumulative), where the second argument is required. Its FALSE mode can return density; NORMSDIST itself has no corresponding mode.

GAUSS has another distinct convention: the help states GAUSS(z)=NORMSDIST(z)-0.5. Thus NORMSDIST(0) is 0.5, whereas GAUSS(0) is zero. Do not treat those two names as interchangeable lower-tail probabilities.

Enter A2 1 and calculate =NORMSDIST(A2). The complete literal form is:

=NORMSDIST(1)

Independent standard normal calculation gives approximately 0.8413447461. The help rounds the value to 0.84, describing the area to the left of 1. Formatting the full value as a percentage gives about 84.13447461%.

=NORM.S.DIST(1;TRUE) has the same source/math target with an explicit cumulative flag. It is a paired comparison under a separate signature, not evidence that the two names accept identical argument lists.

At z=0, =NORMSDIST(0) has expected result 0.5. At z=-1, the expected value is 0.1586552539. A negative input is a valid location on the standardized axis; the returned lower-tail probability remains nonnegative.

For ordinary finite values, the mathematical identity Phi(-z)=1-Phi(z) provides a useful check. The values at ±1 sum to 1, as do those at ±2. These independent numerical controls do not certify every extreme-tail digit in the application’s approximation.

The CDF increases with z. If a purported lower-tail result decreases as a positive threshold grows, check whether an upper-tail expression or a density function was used instead.

For a fictional normal model with mean 63, standard deviation 5 and measurement 70, standardize to (70-63)/5 = 1.4. The expression =NORMSDIST((70-63)/5) has independent target 0.9192433408, matching the scalar NORMSDIST(1.4) control.

Feeding 70 directly into NORMSDIST would ask about a location 70 standard deviations above zero, which is a different question. The function does not infer a mean or scale from nearby cells. Mean, measurement and standard deviation must be expressed in compatible units, and the standard deviation used for this transformation must be positive.

The probability between -1 and 1 is NORMSDIST(1)-NORMSDIST(-1), approximately 0.6826894921. The probability above 2 is 1-NORMSDIST(2), approximately 0.02275013195. For a continuous distribution, including or excluding the exact finite endpoint does not change that mathematical probability.

These are probability calculations under a stated model, not a recommendation to accept or reject a hypothesis or a claim that observed data are normally distributed. A two-sided statistical procedure requires its own assumptions and definition; doubling an arbitrary tail is not a universal decision rule.

Do not append FALSE to NORMSDIST expecting density. Use the documented two-argument dotted form when that is the required quantity. Likewise, GAUSS can be negative for negative z because it measures a signed difference from one-half; NORMSDIST is a full lower-tail CDF.

The exact source calls its normal integral helper. No extra input-range restriction to [0,1] is present for z, and no exhaustive text, blank or nonfinite error behavior is claimed here. Ordinary complement formulas can lose relative precision in extreme tails; these examples remain moderate and independently calculable. Results are mathematical expectations and source traces, not runtime observations.

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 NORMSDIST 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.

• NORM.S.DIST
• GAUSS
• STANDARDIZE