Covariance helps to measure the relationship between two data sets by calculating the average of the products of deviations for each data point pair. This can be used to analyse correlations between variables, such as whether greater education is linked to higher earnings. Covariance can be a useful tool for analysing relationships between different components of a dataset.

`COVAR(array1, array2)`

- array1 - a range or array of integer values
- array2 - a range or array of integer values

`=COVAR(A2:A6, B2:B6)`

The function returns the covariance of the product of deviations in each data point pair in the arguments. In the case of the example given, the result of the covariance is 5.2.

`=COVAR(A2:A6, B2:B6)`

This function returns an error value of #DIV/0! if either array1 or array2 is empty.

The COVAR function is a Sourcetable feature introduced in 2000. It calculates the correlation between two sets of values, ignoring any text or logical values.

- The COVARIANCE.P function calculates the population covariance using the average of products of deviations for each data point pair in two data sets
- Covariance is used to determine the relationship between two data sets, and can be used to examine whether greater income accompanies greater levels of education
- COVARIANCE.P returns the population covariance, and will return a #N/A error if array1 and array2 differ in the number of data points they have

The COVAR function returns the covariance of two sets of numbers. This is used to determine the relationship between two data sets.

The COVAR function requires both arrays to be of numbers or be named arrays with arrays of numbers. Text, logical values, and empty cells are ignored.

Use the COVARIANCE.P and COVARIANCE.S functions instead of COVAR. These are newer functions with better accuracy than COVAR.

Drop CSV