ABSTRACT
A categorical variable is a variable that can take on one of a countable, but usually fixed, number of values. The term discrete variable is synonymous with categorical variable. Continuous—or scale—variables such as blood pressure or a person’s height, can attain any real value within their range. A contingency table also called a cross-tabulation, is a table in matrix format that displays the observed counts of categorical variables. The effect estimate is always associated with uncertainty. A confidence interval quantifies this uncertainty by giving a range of plausible values of the effect measure given the data. The nominal coverage, also called the confidence level or the confidence coefficient, is usually set to 95%, which interpret as the amount of confidence have that the true but unknown effect measure is contained in the interval. The link function specified the expectation of the response variable as a linear function of the explanatory variables.
