COVAR Function

Excel Functions › Compatibility

All versions (legacy) Compatibility

The Excel COVAR function returns the population covariance — the average of the products of paired deviations — measuring how two variables move together. It is the legacy name of COVARIANCE.P, introduced in Excel 2010.


Quick answer:
=COVAR({3,2,4,5,6},{9,7,12,15,17}) population covariance = 5.2

Syntax

=COVAR(array1, array2)
ArgumentDescription
array1RequiredThe first range or array of integer values.
array2RequiredThe second range or array of integer values, the same size as array1.

How to use it

COVAR returns the population covariance: it pairs each value of array1 with its partner in array2, multiplies their deviations from the respective means, and averages over all n pairs.

=COVAR({3,2,4,5,6},{9,7,12,15,17}) // population covariance = 5.2

A positive result means the variables tend to rise together; a negative result means one rises as the other falls. Because COVAR divides by n (not n−1), its modern equivalent is COVARIANCE.P; for the sample version use COVARIANCE.S.

Use COVARIANCE.P instead: =COVARIANCE.P({3,2,4,5,6},{9,7,12,15,17}) in Excel 2010+ returns the same 5.2. The 2010 release also added COVARIANCE.S for the sample covariance (divides by n−1).

Try it: interactive demo

Live demo

Pick a COVAR example to see the formula and its result.

Result:

Practice workbook

📊
Download the free COVAR practice workbook
Every example on this page, ready to open in Excel — plus practice challenges with answers on a separate tab. No sign-up required.

Frequently asked questions

What is the modern replacement for COVAR?
COVARIANCE.P, added in Excel 2010, which computes the same population covariance. =COVARIANCE.P(a,b) equals =COVAR(a,b).
Is COVAR the population or sample covariance?
Population — it divides the sum of deviation products by n. For the sample covariance (divide by n−1) use the newer COVARIANCE.S.
What does the sign of covariance mean?
Positive covariance means the two variables tend to increase together; negative means one tends to increase as the other decreases; near zero means little linear relationship.
How is covariance different from correlation?
Correlation is covariance scaled to a −1 to 1 range, so it is unit-free and comparable across datasets. Covariance keeps the original units and has no fixed bounds.

Master functions like this in one day

This page covers one function. Our Excel Formulas and Functions class covers the 30 that matter most — live, hands-on, taught by professionals in Dallas–Fort Worth, Houston, Austin, Oklahoma City, Denver, or online.

See the Formulas & Functions Class

Related functions: COVARIANCE.P · CONFIDENCE · FTEST · CHITEST · COVARIANCE.S