Within-Subjects Study

An experimental design in which every participant experiences every condition being compared, so each person acts as their own control. Within-subjects studies increase statistical power with smaller samples and remove between-person variance, but must counterbalance order (e.g. using a Latin square) to control for learning, fatigue, and carryover effects. They are common in accessibility HCI because recruiting larger numbers of participants with specific disabilities is often difficult; the trade-off is that participants see the compared tools back-to-back, which can bias ratings toward the novel or recently experienced condition.