Within-subjects design
Also known as: Repeated-measures design, Within vs between subjects
Every participant experiences every condition, serving as their own control.
What it means
A within-subjects design exposes each participant to all levels of the manipulated variable, so comparisons are made within individuals rather than between separate groups. Because each person is their own control, the design removes between-person variability and gains statistical power, often needing far fewer participants than a between-subjects study. The trade-off is exposure to order and carryover effects — practice, fatigue, and contamination across conditions — which must be tamed by counterbalancing. It is contrasted with between-subjects designs, where each participant sees only one condition, eliminating carryover at the cost of power.
Examples
Measuring each person's reaction time under both quiet and noisy conditions, rather than assigning quiet to one group and noise to another.
A blind taste test in which every drinker rates all four coffees, so one person's general fussiness cannot be mistaken for a difference between the blends.
A crossover trial where each patient takes the drug for eight weeks and the placebo for eight weeks, with half the group getting each order first to cancel out sequence effects.
First described in Classical experimental design.