Why can a paired test detect a small change that an independent-sample test misses?
StatisticsLevel: IntermediateMean and paired inference
Why can a paired test detect a small change that an independent-sample test misses?
Hints
- Expand the variance of a difference.
- Shared subject variation can cancel.
Answer
It analyzes within-subject differences. Positive before/after covariance reduces Var(A−B)=Var(A)+Var(B)−2Cov(A,B), potentially reducing standard error substantially.
Source and licence: SciPy — statistical estimators and hypothesis tests · BSD-3-Clause · answer: Quantoros editorial