A paired-samples t-test is used to determine whether two means collected from the same sample differ significantly.

A paired-samples t-test is used to determine whether two means collected from the same sample differ significantly.

• A. A test comparing means from two independent groups.
• B. A nonparametric test for paired data.
• C. A test for correlation between two variables.
• D. A test using the t-statistic for differences between means from related observations.

Answer: (D)

Pairing means we’re dealing with related observations, like measurements on the same people before and after a treatment. The paired-samples t-test asks whether the average of those difference scores is different from zero. It does this by computing a t-statistic from the mean difference, the standard deviation of the differences, and the number of paired observations, with degrees of freedom equal to n minus one. This is a parametric test assuming the differences are roughly normally distributed. So this option fits best because it describes using the t-statistic for differences between means from related observations. It wouldn’t describe independent groups, a nonparametric approach, or a test of association.

Pairing means we’re dealing with related observations, like measurements on the same people before and after a treatment. The paired-samples t-test asks whether the average of those difference scores is different from zero. It does this by computing a t-statistic from the mean difference, the standard deviation of the differences, and the number of paired observations, with degrees of freedom equal to n minus one. This is a parametric test assuming the differences are roughly normally distributed. So this option fits best because it describes using the t-statistic for differences between means from related observations. It wouldn’t describe independent groups, a nonparametric approach, or a test of association.