The Problem of Carryover Effects (1 of 2)

---

Carryover effects can pose difficult problems for within-subject designs. Sometimes the problems are so severe that a within-subjects design is invalid and a between-subjects design must be used. For example, consider the study of incidental learning. In incidental learning, subjects are often presented with stimuli and asked to answer questions about them. They might be asked to count the number of letters in each word or to judge the pleasantness of each word. Subsequently, a surprise memory test is given. An experimenter wishing to test whether incidental learning was better when subjects counted the letters or made pleasantness judgments would not be able to use a within-subjects design. Once subjects had been given one memory test, the second memory test would not be a surprise.

It is possible to use a within- subjects design even if carryover effects are present as long as the carryover effects are (a) not severe and (b) are symmetric.