Statistical tests assume that certain conditions are met, in order for the test to be accurate. The primary assumptions for the t-test are:
- Population data from which the sample data are drawn are normally distributed.
- The variances of the populations to be compared are equal.
Actually, empirical studies of the t-test have demonstrated that these assumptions can be violated to an amazing degree without substantial effect on the results.
For a discussion of the t-test, please see the following text, pages 319-323:
Hays, W. L. (1963). Statistics. New York: Holt, Rinehart and Winston.
Return to the discussion of the Posttest-Only Control Group Design.