Ordinal Variable
- Categories follow a meaningful order or hierarchy.
- Differences between adjacent categories are not quantified.
- Common in surveys and social science measures of attitudes or attainment.
Definition
Section titled “Definition”An ordinal variable is a type of categorical variable in which the categories can be ranked or ordered in a specific way, meaning there is a clear hierarchy of the categories with one category being higher or lower than another.
Explanation
Section titled “Explanation”Ordinal variables indicate an order among categories but do not quantify the exact differences between those categories. Because only the order is meaningful, many statistical methods that assume continuous, normally distributed data are not appropriate for ordinal data. Instead, analysis typically relies on non-parametric approaches or tests designed for ranked data.
Examples
Section titled “Examples”Customer satisfaction
Section titled “Customer satisfaction”Categories: “very satisfied,” “satisfied,” “neutral,” “dissatisfied,” and “very dissatisfied.”
These categories can be ranked from the highest level of satisfaction to the lowest.
Educational attainment
Section titled “Educational attainment”Categories: “high school,” “associate’s degree,” “bachelor’s degree,” “master’s degree,” and “doctorate.”
These categories can be ranked from the lowest level of education to the highest.
Use cases
Section titled “Use cases”- Social science measurement of attitudes, opinions, and preferences.
- Market research to measure customer satisfaction and perceptions.
- Describing educational attainment in populations for policy analysis.
Notes or pitfalls
Section titled “Notes or pitfalls”- Ordinal variables do not measure the exact difference between categories; for example, the difference between “satisfied” and “neutral” may not equal the difference between “neutral” and “dissatisfied.”
- Certain parametric tests that assume continuous and normally distributed data are inappropriate for ordinal data; for example, a t-test or an ANOVA should not be used on ordinal data.
- Ordinal data is typically analyzed using non-parametric statistical tests, such as the chi-square test or the Mann-Whitney U test.
Related terms
Section titled “Related terms”- Categorical variable
- Non-parametric statistical tests
- t-test
- ANOVA
- Chi-square test
- Mann-Whitney U test