ABSTRACT
GARY G. KOCH and GREGORY J. CARR University of North Carolina, Chapel Hill, North Carolina
Many studies in the pharmaceutical sciences are concerned with the relationships between categorical response variables which describe favorable or unfavorable outcome and one or more explanatory variables. The explanatory variables include experimental factors such as treatment in clinical trials and background characteristics such as age, sex, and baseline status of subjects; additional factors that often need to be taken into account are "center" in multicenter studies and "visit" in multivisit studies. Since the explanatory variables which are considered can have either a categorical (e.g., sex) or continuous (e.g., age) nature, the distinguishing feature of situations which require categorical data analyses is that the response variables are categorical.
