ABSTRACT
The use of the dependent t-test is appropriate when the samples are not independent but represent repeated measures of the same subjects, who have been measured twice. An interesting feature of the pre–post or paired two-sample case is that authors can test to see whether two means differ significantly by turning it into a one-sample case. This is called the direct difference method. The hypothesis testing way to decide whether to accept or reject the null hypothesis is to conduct what is called a z-test. For the one-sample case, authors examined the difference between a sample mean and a hypothesized population mean, and for the two-sample independent t-test, authors examined the difference between two independent sample means. For the t-test to be an accurate comparison between two means, certain statistical assumptions must be met.
