Paired T-Test Statistical Analysis

Use this endpoint when you have repeated measurements on the same subjects (within-subjects or pre/post design) and want to test whether the mean change is statistically significant. It is the correct choice over a two-sample t-test when the same individuals are measured twice, as pairing controls for individual variability. It also returns Cohen's dz, making it useful when you need both significance testing and effect size in one call. Prefer this over Mann-Whitney U when data is approximately normal and you want parametric results with confidence intervals.

• Mismatched array lengths between before and after parameters cause an error — both must have the same number of subjects in the same order
• Fewer than 2 pairs results in insufficient degrees of freedom to compute the test
• Non-numeric values in the before or after strings will cause a parsing error
• Alpha outside the range 0.001–0.2 is rejected as an invalid significance level
• Missing required before or after query parameters returns a 400 error
• Payment failure or insufficient USDC balance returns a 402 Payment Required response