Measure the effect of an intervention or change over time by comparing the same participants before and after.
The Paired Samples t-Test (also called the dependent samples or repeated measures t-test) compares two related sets of measurements from the same participants. Unlike the independent t-test — which compares separate groups — the paired version looks at within-person change. By subtracting each participant's pre-score from their post-score and testing whether the mean difference is significantly different from zero, the test effectively controls for individual differences that would inflate error variance.
This design is common in clinical research (blood pressure before vs. after medication), educational psychology (pre-test vs. post-test knowledge), sports science (performance before vs. after training), and any before-after or matched-pairs design. The paired approach is statistically more powerful than an independent t-test when the two measurements are correlated — which is almost always the case with repeated measures.
StatRyx handles the entire paired t-test analysis: it aligns your two measurement columns, computes difference scores, checks normality with Shapiro-Wilk, calculates the t-statistic and p-value on the difference scores, computes Cohen's d for dependent samples, generates a confidence interval for the mean difference, and formats everything in APA 7 notation.
| Measure | M | SD | t | df | p | d |
|---|---|---|---|---|---|---|
| Pre-intervention | 21.7 | 6.4 | 8.29 | 44 | < .001 | 1.24 |
| Post-intervention | 14.3 | 5.1 | ||||
| Mean difference | 7.4 | 5.96 |
A paired samples t-test indicated that depression scores were significantly lower after the 8-week CBT intervention (M = 14.3, SD = 5.1) compared to before (M = 21.7, SD = 6.4), t(44) = 8.29, p < .001, d = 1.24, 95% CI for difference [5.72, 9.08].
Yes. You can run it without an account (3 free analyses), and a free account adds 10 more — a one-time allowance, no payment needed. Every result includes the APA 7 write-up. Paid plans (Student from $9/month) add more analyses and Word, PDF and Excel export.
No coding is required. Just describe your two time points in plain English and StatRyx handles all the calculation, assumption checking, and result formatting automatically.
For detecting a medium effect (d = 0.5) at 80% power and α = .05, you need approximately 34 pairs. The paired design is more efficient than the independent t-test because it controls for individual differences, so you typically need fewer participants.
For paired data, Cohen's d is calculated as the mean difference divided by the standard deviation of the difference scores — not the pooled SD of the two groups. This version reflects within-person change and is often larger than the corresponding independent-samples d for the same data.
If Shapiro-Wilk indicates non-normality and your sample is small (fewer than ~30 pairs), StatRyx will recommend the Wilcoxon Signed-Rank Test as the nonparametric alternative. With larger samples, the Central Limit Theorem makes the t-test robust to moderate departures from normality.
Run a Paired Samples t-Test on your data →