# Descriptive Categorical

Description

Categorical variables describe data that can be classified into distinct categories determined by a particular quality. Categorical data therefore fall into a fixed number of separate classes. The categories may follow no intrinsic order, in which case the variable is said to be nominal, or may have a numerical relationship, in which case the variable is ordinal. A nominal variable is said to be binary or dichotomous if it is limited to two categories. Examples of nominal variables are female/male, alive/dead. The term categorical variable may be used interchangeably with the terms qualitative variable and also nominal variable, which is thought of as purely categorical.

Frequency distributions represent the simplest way of summarizing purely categorical data. It refers to the set of frequencies, count of data points that have a particular quality, of each category of the variable. The proportion of points that fall into a category is called relative frequency, proportional frequency, or frequency percentage as denoted here.

Bar plots, in the purest form, are useful to represent the relationships between more than two variables in the form of vertical bars. Its purpose is to convey information in a way that can be understood fast and clearly.