8.3 Sample Standard Deviation Comparisons with the F-Test

The F-Test is used to compare two variances to determine if they are significantly different. The way ShotStat is written, the values to be compared are the standard deviations, which are internally converted to the variances. The F-Test Window in ShotStat displays input boxes for Samples A and B:

Lot Number

(optional) Free format description the user may wish to use to identify the sample(s)

Standard Deviation

(required) standard deviation (square root of variance) for each sample

Number

(required) the number of items/measurements made for each sample; for example, a five round group has a Number of 5. Larger numbers result in more reliable tests.

Under the Sample A and Sample B areas, ShotStat displays an input box for:

Significance Level

(required) entered as a percentage, the desired acceptable chance for ShotStat reporting that the samples are different when they are actually, in fact, the same. This must be between 0.001 and 45, but generally, a value of 10, 5 or 2.5 will be used. The default is 5%.

There are three command options available to the user:

Compute

performs the test; upon completion of the test, ShotStat displays a message stating when the samples are probably the same or probably different.

Cancel

cancels the F-Test Tool and closes the F-Test Window; all entered data are lost

Print

after a test is performed, the option to print the current data and results exists. ShotStat sends the output to the default printer for the local computer. The ``Print'' command button is not visible until a test is performed that can be printed.