ABSTRACT
Statistical Guide: As you know from Exercises 23 through 25, we use at test to determine the statistical significance of the difference between two means. In those exercises, individuals in one group were not paired or matched with individuals in the other group; thus, you were conducting t tests for independent data. When there is pairing or matching, we conduct a t test for dependent (or correlated) data. For example, if we use twins in an experiment with one member of each pair of twins randomly assigned to the experimental group and the other one assigned to the control group, pairs of individuals are matched across groups. Another kind of pairing is done when we give two tests to the same individuals (such as a pretest and a posttest) and wish to test for the significance of the difference. While the computational procedures are different for an independent and a dependent t test, the purpose and interpretation of the two types oft tests are the same.2
