What It Is Used For
- To statistically determine if a sample represents a population or if two or more samples are from the same population.
When to Use It
- To confirm that an improvement is truly an improvement and not just common cause variation in the process.
- t-tests use sample means to look at the location of the process output relative to a known population or to another population (represented by another sample mean).
- F-tests are used to compare the variances of two samples to determine if the variation between the two populations they represent are likely to be equal.