Repeated T-Test

The Repeated T-Test is a parametric test for comparing the means of two variables from the same group.

Assumptions
The assumptions of this test are as follows:
 * 1) The difference between the two variables must be normally distributed.
 * 2) The variables must be at least at the interval level.

Non-Parametric Equivalent
The non-parametric equivalent of this test is the Wilcoxon test.

Example
Consider you have a group of participants and wish to compare the error rate of two versions of an interface. You conduct a study where each participant completes a task on both interfaces and you count the number of errors that they make. A Repeated T-Test could be conducted to see if the number of errors on the first version of the interface were lower than the number of errors on the second, deomonstrating whether or not Interface 1 had a statistically lower error rate.

In SPSS
asdf

Writing up
A paired-samples t-test was conducted to compare hours of sleep in caffeine and no caffeine conditions. There was a significant difference in the scores for caffeine (M=5.4, SD=1.14) and no caffeine (M=9.4, SD=1.14) conditions; t(4)=-5.66, p = 0.005. These results suggest that caffeine really does have an hours slept. Specifically, our results suggest that when humans consume caffeine, the number of hours they sleep decreases.