Comparing two sample means t tests.

Abstract

The t test is used to test for differences in means or to test a criterion measure between two groups of scores. Whether an investigator designs a study where the subjects' scores from one group are independent of the scores in the other group (independent t test), the basic assumptions of the test are identical. Hence, the subjects' scores from each group are assumed to be normally distributed and the variance of the two groups of scores are assumed to be homogeneous. The basic differences between the independent t test and the correlated t test are in calculating the t statistic and the df and in the method in which subjects are assigned to each group. If an investigator uses the same subjects in each group, then the correlated t test must be used. If subjects are matched or paired by some related variable, such as age, height, or weight, then the correlated t test should be used. In all other cases, the independent t test would be appropriate.