t-Test, Independent Samples

The Independent Samples t-Test

As noted, the independent samples t-test is widely used by communication researchers. The purpose of this test is to determine whether the means from two independent groups are significantly different from each other. It is important to remember that if this test is used, the two groups must be independent from one another. Groups can be considered independent when there is no overlap between the groups. For example, if one has a control group and a treatment group, participants cannot be in the control group at the same time that they are in the treatment group. Or, if we are grouping based on age, a participant cannot be 20 at the same time that she is 28. Groups can also be considered independent when we are collecting data from and comparing two different populations. For example, say that a researcher is hoping to compare the friendliness of current Wisconsin residents to the friendliness of current Oklahoma residents. Here, the participants are from two different populations (i.e., participant is either a current Wisconsin resident or a current Oklahoma resident), and the score for a participant from one group is not related in any way to the score for a participant from another group. In sum, for two groups to be considered independent—and thus, to be able to use the independent samples t-test—there must not be any naturally occurring or planned relationship between the members of the two groups. The participants within the groups are not matched in advance and there is no “before and after” type of testing used on just one group. When the above conditions are satisfied, we can assume the value for one group’s mean is not influenced or related to the value of the other group’s mean. Thus, the groups are independent.

...