The Excel T.Test function calculates the probability associated with the Student's T Test, which is commonly used for identifying whether two data sets are likely to have come from the same two underlying populations with the same mean.
The T.TEST function is new to Excel 2010. However, this is simply an updated version of the Ttest function, which is available in earlier versions of Excel.
The syntax of the T.Test function is:
where the function arguments are:
array1    The first data set.  
array2    The second data set (must have the same length as array1).  
tails    The number of tails for the distribution. This can be either:
 
type    An integer that represents the type of ttest. This can be either:

A  B  

1  4  8 
2  5  3 
3  2  7 
4  5  3 
5  8  5 
6  9  2 
7  3  2 
8  2  7 
9  3  9 
10  8  4 
11  9  4 
12  5  7 
Columns A and B of the above spreadsheet on the right contain two arrays of data.
The probability associated with the Student's paired ttest with a onetailed distribution, for the two arrays of data can be calculated using the Excel T.Test function as follows:
This gives the result 0.449070689.
The probability associated with the Student's paired ttest with a twotailed distribution, for the same two arrays of data, is calculated as follows:
This gives the result 0.898141379.
Note that the probability associated with the twotailed TTest is double that of the onetailed TTest.
For further information and examples of the Excel T.Test function, see the Microsoft Office website.
If you get an error from the Excel T.Test Function, this is likely to be one of the following:
#N/A    Occurs if the two supplied arrays have different lengths. 
#NUM!    Occurs if either:

#VALUE!    Occurs if either the supplied tails argument or the supplied type argument is nonnumeric. 