VAR.P
Calculates the variance based on the entire population. Modern replacement for the legacy VARP function.
VAR.P(number1, [number2], ...)Arguments
number1First numeric value or range representing the entire population.number2Optional. Up to 254 additional numbers, references, or arrays.(optional)
=VAR.P(10, 20, 30, 40, 50)Population variance of 5 values
- •Available since Excel 2010 — same result as the legacy VARP
- •Divides by n (no Bessel's correction)
- •Use VAR.S if your data is only a sample
- •Ignores text and logical values
- •Using VAR.P (population variance) when the data is actually just a sample of a larger population, which applies the wrong divisor (n instead of n-1) and slightly understates the true population variance
- •Interpreting VAR.P's result directly as a measure of spread in the same units as the data - variance is in squared units, so STDEV.P (its square root) is usually more intuitive to communicate
- •Forgetting that VAR.P ignores text and logical values entirely - if those need to be included in the calculation, a population-equivalent inclusive function like VARPA handles that instead
VAR.SCalculates variance for a sample rather than an entire population, using n-1 instead of n as the divisor.STDEV.PTakes the square root of variance, returning a value in the same units as the original data rather than squared units.VARPAIncludes text (as 0) and logical values when calculating population variance, unlike VAR.P which ignores them entirely.Should I use VAR.P or VAR.S?
Use VAR.P only when your data represents an entire population, not a sample - this is less common than you might think. For most real-world analysis involving a subset of data, VAR.S is the correct choice.
Is VAR.P the same as the legacy VARP function?
Yes, they produce identical results - VAR.P is simply the modern, more clearly named equivalent, kept alongside VARP mainly for consistency with other newer statistical function names.
Why is VAR.P's result so much larger than the values in my data?
Variance is expressed in squared units, so it naturally produces a larger number than the raw data. Take its square root (or use STDEV.P directly) for a more directly comparable measure of spread.
Need to translate a formula using VAR.P?
Use our translator to convert your complete formula