In many studies using preschool children as subjects, “missing” data are a problem. For example, a study that measured the effect of length of exposure to material on learning was hampered by the fact...

In many studies using preschool children as subjects, “missing” data are a problem. For example, a study that measured the effect of length of exposure to material on learning was hampered by the fact that the small children fell asleep during the period of exposure, thereby resulting in unbalanced data. The results of one such experiment are shown in Table 11.23. The measurement was based on a “recognition” value, which consists of the number of objects that can be associated with words. The factors were (1) the length of time of exposure and (2) the medium used to educate the children.

(a) Using the dummy variable model, test for differences in time of exposure, medium used, and interaction. Explain your results.

(b) Do you think that the pattern of missing data is related to the factors? Explain. How does this affect the analysis?