ABSTRACT

The researcher must rely on observational studies to provide evidence of epidemiological and statistical associations between a possible risk factor and a disease. In choosing an appropriate statistical method for categorical data analysis, one should consider the measurement scale of the response variable as well as the independent variable. Cox indicated that the statistical advantage of the odds ratio is that it can be estimated from any of the study designs. The chapter reviews methods that have been proposed for comparing the overall proportion of successes among individuals in several groups when the clusters are the sampling units. In epidemiological research, it is important that the collected data are translated into interpretable results that can be easily communicated to clinicians. There are two broad categories of investigative studies that produce statistical data: the first is designed controlled experiments and the second is observational studies.