StatTools : Paired Difference Explained

Links : Home Index (Subjects) Contact StatTools

Related Links:
Paired Difference Programs Page
Sample Size for Mean of Paired Difference Explanations and Tables Page

Introduction Parametric Paired t Test Nonparametric Wilcoxon PSRT Permutation Test References
The paired difference is a powerful and commonly used model in clinical research and quality control. By examining the difference outcomes from different causes in the same individual, variations between individuals are very much reduced, so a smaller difference can be detected by a small sample.

In clinical research, questions such as whether husbands are older than wives from the same family, whether boys are heavier than girls in non-identical twins, whether the effects of two different treatments for the same condition in the same patient, all use the paired difference model.

Paired difference is also commonly used in quality control, each pair evaluating a measurement against a standard. Questions such as whether the waiting time for operations exceed a benchmark, whether blood loss in operations exceeds that expected, also used the paired difference model.

StatTools provides three programs for evaluating paired difference, the parametric paired t test and 95% confidence interval of the paired difference, and the nonparametric Wilcoxon Paired Signed Rank Test, and the Permutation Test. The algorithms for calculation are in the Paired Difference Programs Page , and each test is discussed in its own panel.