Generate crossover study designs with proper sequence allocation and calculate sample sizes accounting for within-subject correlation. Supports Williams Latin Square and balanced designs with optional washout period planning.
Balanced for carryover effects — each treatment follows every other treatment equally
Get properly formatted citations for academic work •20k+ calculations in the past 30 days
StatsCalculators Team. (2026). Crossover Design Calculator. StatsCalculators. Retrieved May 20, 2026 from https://statscalculators.com/calculators/experimental-design/crossover-design-calculator
A crossover design is an experimental setup where each subject receives more than one treatment in sequence, separated by washout periods. Because each subject serves as their own control, between-subject variability is removed from the error term, dramatically increasing statistical power.
Each treatment follows every other treatment exactly once, controlling for first-order carryover effects.
All possible treatment sequences are included, providing complete balance across all orderings.
A washout period is a gap between treatment periods that allows the effect of the previous treatment to dissipate before the next treatment begins. Inadequate washout leads to carryover effects — where treatment A influences the observed response to treatment B.
The efficiency gain of a crossover design over a parallel design depends on the within-subject correlation (ρ) — how correlated repeated measurements on the same subject are. Higher correlation means greater efficiency.
Typical within-subject correlations range from 0.3–0.7 for most clinical and behavioral outcomes.
