The Dependent Samples t test is used to test for significant differences between two sets of numerical data produced by the same organisms or organisms that are matched on all relevant variables.
In one example, a group of people who attend a workshop may complete a pretest and a posttest. A Dependent Samples t test can be used to compare the mean differences between the pretest and the posttest.
An example of a Matched Pairs t test can be used to compare two groups of people in a reading method experiment. A relevant variable would be reading ability. A reading test could be used to identify people with similar scores. One member of the pair is then randomly assigned to a new reading method group and the matching person is then assigned to the traditional reading group. At the end of the study, a Matched Pairs t test can be used to compare mean scores for the groups.
When the same person produces two sets of scores, each person is their own control. Because of the level of control, there is less variation than would be the case when using different people in each group.
When the groups contain different unmatched people or subjects, the groups are considered independent samples and an Independent Samples t test is used.