A paired t-test indicates whether the mean difference between two numeric columns appears to differ from zero. If observations in two columns are related — for example, test scores from the same student across time — a paired t-test is a more powerful alternative to the unpaired t-test. The paired t-test is equivalent to a repeated-measures ANOVA test, which is a generalization of the paired t-test to more than two columns.

To perform a paired t-test:

- In the Summary view, use the confidence slider to choose a desired confidence level
- Command-click two numeric columns of interest on the left
- Check the box labeled
*Repeated measures*below the summary picture - Select the button below the Covariate view labeled
*t-test* - The result of the paired t-test will appear in the Bottom Line

See also:

- Performing a Wilcoxon signed-rank test of the median difference between two numeric columns
- Performing an unpaired t-test of two numeric columns’ means
- Performing a t-test of a numeric column’s mean across two categories
- Performing a repeated measures ANOVA test of three or more numeric columns’ means
- Performing a Shapiro-Wilk test on the normality of differences between two columns
- Formatting descriptive and inferential statistics
- About the Summary view