The paired samples t-test calculator compares two different sample means from the same sample (Gravetter and Walllnau, 2013). In this case, all subjects participate in all conditions of the test variable.
INSTRUCTIONS: Enter the following:
Paired Samples t-test: The calculator returns the following:
The computation for a paired samples t-test is similar to that of an independent samples t-test. Similar inputs are used, but it is important to remember they apply to the same sample, not two different samples. To demonstrate this, we will use the dataset "Comfortable TV viewing distance for college students" located. The data presented here can be input to the equation, which will return the means and standard deviations of both test samples (M and SD) and a summary of the results, including degrees of freedom (df), a critical t-value, your t-score, and the standard error (SE).
If you plug in the data values, you should get a return string that looks like this: Sample 1: M = 6, SD = 1.67. Sample 2: M = 5.17, SD = 2.23. Summary: df = 5, critical t-value = 2.015, t-value = -0.62, and SE value is 1.35. This tells us that the difference between the two means is not significant, because our t-value is not beyond the critical t-value. In context, according to this hypothetical data, college students do not find one TV viewing distance (8ft or 12 ft) more comfortable than another; they are roughly equally comfortable.
Gravetter, F. J., & Wallnau, L. B. (2013). Statistics for the Behavioral Sciences. Wadsworth, CA: Cengage Learning.
The example provided in the dataset was adapted from the textbook.