Glossary » within-subjects design
within-subjects design
a study designed to make a comparison of 2 or more designs and that compares them by having each user try each design, measuring their performance for each design. The most common problem is that people learn over time, and so they frequently perform better on the 2nd design than the first, regardless of the relative quality of the designs. Thus, it’s important to counterbalance such a study, having half the subjects start with one design and half start with the other, in order to avoid having the order effects give an advantage to one design.