ExcelFunctions.net

Search Site:

Related Function:

VARP Function
Variance & Sample Variance

The Variance is a statistical measure, that is commonly used across a set of values, to identify the amount that the values vary from the average value.

When your data set is a __sample__ of a population, (rather than an entire population),
you should use a slightly modified form of Variance, known as the **Sample Variance**.
The equation for this is :

where x is the average (statistical mean) of the set of values, and n is the number of values.

A full explanation can be found on the Wikipedia Variance page
VAR and VAR.S

In Excel 2010, the VAR function has been replaced by the VAR.S function, which has improved accuracy.

Although it has been replaced, the Var function is still available in Excel 2010 (stored in the list of compatibility functions), to allow compatibility with earlier versions of Excel.

The Excel VAR function returns the sample variance of a supplied set of values.

The format of the function is :

VAR( number1, [number2], ... )

Where the number arguments provide a minimum of 2 numerical values to the function. Each number argument can be a single value or an array of values.

You can enter up to 254 number arguments to the Var function in Excel 2007 and later versions of Excel, but you can only enter up to 30 number arguments in Excel 2003.

Note that the Var function calculates the __sample__ variance - i.e. the variance for a sample of a
population. If you are calculating the variance for an __entire__ population, you need to use the
Excel Varp or the Excel
Var.P function.

Note also, that the Var function ignores text values and logical values if these are supplied
as part of an array. However, if they are supplied directly to the function, text representations of
numbers and logical values __are__ interpreted as numbers. If you want a sample variance calculation
that does not ignore text and logical values that are supplied as a part of an array, consider using
the Vara function.

Imagine you wanted to find out the variance of the heights of adult males in London. It is not
realistic to measure the height of __all__ males, but you could take a sample of the population and
measure their heights.

The example spreadsheet on the right stores the measurements (in cm) of 3,000 adult males. The measured heights are stored in cells B3 - B1002, D3 - D1002 and F3 - F1002.

The variance of the heights of the sample group is calculated in cell H3 of the spreadsheet. The formula for this, as shown in the formula bar, is :

=VAR( B3:B1002, D3:D1002, F3:F1002 )

As shown in cell H3, the variance of the individual heights of the sample group is **9.261904762 cm**.

The example above shows the arguments to the Var function being input in the form of 3 cell ranges. However, you can also input figures directly, as individual numbers or number arrays.

For example, if you wanted to include two further heights, of 176cm and 177cm into the sample we could add these directly into the above function as follows:

Either as individual numbers:

=VAR( B3:B1002, D3:D1002, F3:F1002, 176, 177 )

Or, as an array of numbers:

=VAR( B3:B1002, D3:D1002, F3:F1002, {176,177} )

Further information and examples of the Excel Var function can be found on the Microsoft Office website.

If you get an error from the Excel Var Function, this is likely to be the #DIV/0! error:

Common Errors

#DIV/0! | - | Occurs if fewer than 2 numerical values have been supplied to the function |