## Hours Today: |

In statistics, it is important to recognize and identify the types of variables because this determines the measure of central tendency used to describe the data, as well as determining the statistical tests needed to analyze the data.

This is made up of categories**. **

If there are two groups, it is called a **binary variable**.

If there are more than two groups, it is a **nominal variable**.

Sometimes numbers are assigned to categorical variables, such as 0 for no and 1 for yes. But these numbers are just labels and cannot be used for analysis.

These variables are ordered, such as first, second, and third place. However, with ordinal data, the distance between ordered variables is not known. For example, how far second place is behind first place, and how far third place is behind second place is unknown. Limitations therefore exist on how you can analyze these variables.

These are variables that give a score for each person and can take on any value on the used measurement scale. True continuous variables can be measured to any precision, such as age. **Discrete variables** cannot be. For example, number of people is discrete because a half a person is meaningless.

Variables have equal distances between values are **interval variables. **

Variables with meaningful ratios of values (including a true and meaningful zero point) are **ratio variables. **

Self-report data such as Likert scales are often treated as interval variables but technically they are ordinal variables. There are pros and cons to treating Likert scales as interval data, but Likert data with more points, such as 11, helps to make the variables more like interval variables.