NORM.S.INV
Returns the inverse of the standard normal cumulative distribution (mean = 0, standard deviation = 1). Replaces the legacy NORMSINV function.
NORM.S.INV(probability)Arguments
probabilityA probability corresponding to the standard normal distribution (between 0 and 1)
=NORM.S.INV(0.975)Critical value for a 95% two-sided confidence interval
=NORM.S.INV(0.5)Median of the standard normal distribution
=NORM.S.INV(0.05)5th percentile of the standard normal distribution
- •Returns #NUM! if probability ≤ 0 or ≥ 1
- •Equivalent to NORM.INV(probability, 0, 1)
- •Widely used in hypothesis testing and confidence intervals
- •Passing a probability of exactly 0 or 1 (or outside that range), which returns a #NUM! error since those extremes correspond to infinity in a true normal distribution
- •Using NORM.S.INV when the data doesn't actually represent a standard normal distribution - it always assumes mean 0 and standard deviation 1, so applying it to differently-scaled data produces an incorrect result
- •Confusing NORM.S.INV with NORM.INV - NORM.S.INV is specifically the standard (mean 0, standard deviation 1) case, while NORM.INV requires you to supply the actual mean and standard deviation
NORM.INVCalculates the same kind of inverse but for any mean and standard deviation, of which NORM.S.INV is the standard-normal special case.NORM.S.DISTPerforms the forward operation for the standard normal distribution - given a z-value, returns its cumulative probability, the inverse of what NORM.S.INV calculates.CONFIDENCE.NORMUses a similar underlying calculation to construct a confidence interval margin, a common practical application of NORM.S.INV.Why does NORM.S.INV return a #NUM! error?
The probability argument is at or beyond 0 or 1 - it must be strictly between those two values for a defined result.
Is NORM.S.INV just a simpler version of NORM.INV?
Yes, exactly - NORM.S.INV(probability) gives the identical result to NORM.INV(probability, 0, 1), just without needing to type the mean and standard deviation arguments each time.
What does NORM.S.INV(0.975) actually represent?
It returns approximately 1.96, the z-score below which 97.5% of a standard normal distribution falls - a value widely used as the critical value for 95% two-sided confidence intervals.
Need to translate a formula using NORM.S.INV?
Use our translator to convert your complete formula
