Ordinal Logistic Regression is logistic regression using an ordinal variable. An ordinal variable is a categorical variable where order matters. A variable that takes on the values of “very bad,” “bad,” “mediocre,” “good,” “very good” would be an example.

The 1 to 5 or 1 to 7 scales that are often used in questionnaires are ordered categorical variables. For example questions in surveys similar to the following are very common:

Strongly Disagree | 1 | 2 | 3 | 4 | 5 | 6 | 7 | Strongly Agree |

When there are only two categories, ordinal logistic regression is the same as “standard” logistic regression where the dependent *Y* variable takes on only the values of 0 or 1.

Where there are more than two categories, standard logistic regression does not work and you need to fit the regression equation using methods specifically designed for on ordinal dependent *Y* variable. Look for methods called “ordered logistic regression” or “ordinal logistic regression” in the software that you are using.