ABSTRACT
Introduction to Random Effects To this point, we have considered models appropriate for the situation in which all factors have levels selected because of the interest of the investigator in those particular levels. Such factors are said to be fixed because the same, fixed levels would be included in replications of the study. If one were interested in comparing the relative effectiveness of behavior modification and psychoanalytic approaches to treating agoraphobia, for example, then only experiments containing implementations of these two particular approaches would be relevant to the question of interest.
