Variable types in statistics (categorical, continuous, ordinal, nominal)
Core concept
| Type | Definition | Example |
|---|---|---|
| Nominal (categorical) | Named groups, no order | Blood group, ethnicity, eye colour |
| Binary / dichotomous | Nominal with 2 levels | Sex, dead/alive, disease present/absent |
| Ordinal | Ordered, but intervals not equal | NYHA I-IV, mRS, pain 0-10, Likert, tumour grade |
| Discrete (count) | Numeric, whole numbers only | Number of exacerbations, seizures/month |
| Continuous | Any value in a range | Height, BP, creatinine, GFR |
- Interval vs ratio among continuous: ratio has a true zero (weight, kelvin) so ratios are meaningful; interval does not (degrees C, pH)
- The trap: ordinal data given numeric labels are still ordinal. "Mean NYHA class 2.4" is meaningless.
3 more sections, plus exam facts
Premium unlocks every note across every specialty, and the full exam fact library behind it.
Get premium access