For a supplied hypothesized sample mean and a supplied set of values, the Excel Z.TEST function calculates the onetailed probability value of the ZTest.
I.e. the function returns the probability that the supplied hypothesized sample mean is greater than the mean of the supplied data values.
The Z.TEST function is new to Excel 2010. However, this is simply a renamed version of the ZTEST function, that is available in earlier versions of Excel.
The format of the Z.Test function is :
where the function arguments are:
array    The set of values against which the hypothesized sample mean is to be tested 
x    The hypothesized sample mean 
[sigma]    An optional argument that represents the population standard deviation, if this is known. If omitted, the function uses the sample standard deviation. 
If you want to calculate the twotailed probability value of the ZTest, this can be done by using the Z.Test function, combined with the Excel Min function, as follows:
Column A of the above spreadsheet on the right contains an array of 12 data values. The mean of these values is 5.25.
Cells B1 and B2 of the example spreadsheet show the Excel Z.Test function used to calculate the the onetailed probability value of the ZTest for two different hypothesized sample means.
For the hypothesized sample mean 5.0, the onetailed probability value of the ZTest is calculated by the formula:
which gives the result 0.371103279.
For the hypothesized sample mean 6.0, the onetailed probability value of the ZTest is calculated by the formula:
which gives the result 0.838129187.
Note that in the above two examples, the [sigma] argument is omitted from the function. Therefore, the Z.Test function calculation uses the standard deviation of the supplied array.
Further information and examples of the Excel Z.Test function are provided on the Microsoft Office website.
If you get an error from the Excel Z.Test Function, this is likely to be one of the following :
#N/A    Occurs if the supplied array is empty  
#DIV/0!   
Occurs if either :


#VALUE!    Occurs if either the supplied x or the supplied [sigma] is nonnumeric 