GAUSS in ONLYOFFICE Spreadsheet Editor

• Introduced: Original introduction release is not established by the checked sources.
• Available: Listed in current unversioned ONLYOFFICE help, checked October 18, 2026.
• Implementation inspected: Official SDKJS v10.0.0.169, commit 3679e0375a758d7dc59d31337a45461be3525ed8. The snapshot does not prove identical behavior in every installed editor.
• Retired or deprecated: These sources do not establish retirement or formal deprecation. A legacy name alone does not mean the function has been removed.
• Testing: Numerical examples are independent mathematical expectations, and implementation discrepancies are labeled source predictions. No ONLYOFFICE application or calculation-engine execution was performed.

The examples use English function names and commas. Use your locale’s argument separator if different. Retain full precision in formulas; round only the displayed result. The examples describe function mechanics under the stated model, rather than establishing that your observations follow that model.

=GAUSS(z)

One numeric argument is required. z is a standardized value: the measurement minus its model mean, divided by its positive standard deviation. The function returns the standard-normal CDF minus 0.5. It therefore measures a signed area from zero to z, rather than the entire lower tail.

For a positive z the result is positive; for a negative z it is negative. GAUSS(-1) is approximately −0.341344746069, not a negative probability assigned to an event. Adding 0.5 converts that signed quantity to the ordinary lower-tail probability, approximately 0.158655253931.

The mathematical relation is GAUSS(z) = erf(z/SQRT(2))/2. Its values lie between −0.5 and 0.5 for finite real z. This function is not the density PHI, and it is unrelated to returning the gamma special function despite the similar-looking names.

1. Put -1, 0, 1 and 2 in A2:A5.
2. Enter =GAUSS(A2) in B2 and copy down.
3. Enter =0.5+B2 in C2 and copy down to obtain lower-tail probabilities.
4. Label B as “signed area from zero” and C as “lower-tail probability.”

The independent checks use the mathematical error function, rather than executing the SDKJS approximation. The symmetry check is especially useful: GAUSS(z) + GAUSS(−z) should be near zero, allowing for floating-point rounding. This tests the interpretation as signed area, not the assumption that your observations are normally distributed.

Suppose a process is modeled with mean 50 units and standard deviation 4 units. A value of 54 has z = (54-50)/4 = 1. On a fresh sheet place 54 in A2, 50 in B2 and 4 in C2. =0.5+GAUSS((A2-B2)/C2) has independent expectation 0.841344746069, the modeled proportion at or below 54.

For the interval from 46 to 54, use =GAUSS((54-50)/4)-GAUSS((46-50)/4). The independent result is 0.682689492137. The 0.5 terms cancel in an interval difference. For a symmetric interval within two standard deviations, =2*GAUSS(2) gives approximately 0.954499736104.

The model’s mean and standard deviation must use the same physical units as the measurement. A nonpositive standard deviation is invalid for standardization even though GAUSS itself only receives the computed z. The formula does not estimate parameters, assess independence, or check whether a normal model is suitable.

For moderate positive z, an upper-tail expression is =0.5-GAUSS(z). At z = 1 it gives approximately 0.158655253931. For large positive z, subtraction near 0.5 can lose relative precision in the tiny tail. Likewise, 0.5+GAUSS(z) may lose precision far into the negative tail. The inspected source itself notes cancellation concerns when combining GAUSS with 0.5 for sufficiently negative arguments and has a separate normal-integral helper. Do not assume this simple expression is the best route for extreme-tail calculations.

The wrapper converts the argument to a number and propagates conversion errors. If its numerical helper yields NaN, it returns internal not_numeric, conventionally #NUM!. It does not impose a positive-z restriction: negative z is a normal and useful input. Use genuine numeric cells; the conversion behavior of arbitrary text, ranges and arrays is outside the scalar examples.

The inspected helper uses piecewise polynomial and asymptotic approximations. The listed values are independent expectations to the displayed precision, not evidence that every extreme input achieves that accuracy in an application. Format signed areas as numbers if percentage formatting would confuse readers into treating negative signed area as an event probability.

Recalculate representative formulas when moving a workbook between editors. Matching names do not guarantee identical argument coercion, array behavior, numerical precision or error handling. The scalar formulas here do not establish automatic spill behavior or a general import/export guarantee.

• STANDARDIZE
• EXP

• ONLYOFFICE GAUSS help, checked October 18, 2026.
• Official SDKJS implementation, 7688–7720, tag v10.0.0.169 and exact commit above.