Dyr og Data

Statistical thinking — types of data

Gavin Simpson

Aarhus University

Mona Larsen

Aarhus University

Wednesday, 19 August 2026

Random variables

A random variable is a variable whose values are not known for certain before a sample is taken

Discrete
  • Take only a finite set of values
  • Does an animal have a disease?
  • Number of treatments with antibiotics
  • Sex
Continuous
  • Take any value
  • Penguin flipper length
  • Milk yield of a cow at a milking

Measurement level

A further classification of random variables relates to the scale on which we measure them. Some discrete variables can be nominal or ordinal

  • A nominal scale is for unordered categories
    • Species
    • Sex
    • Religion
  • An ordinal scale is where there is some implicit ordering to the categories
    • Pain on a scale from 1–10
    • Behaviour

Measurement level

Continuous variables can be interval or ratio scaled. Discrete counts can also be considered ratio scaled variables.

  • Interval scaled variables are those with equal intervals along the scale
    • Temperature (˚C)
    • As no fixed zero value, ratios can not be calculated
    • 5˚C is not half the temperature of 10˚C
  • Ratio scaled variables
    • Have a defined 0 point
    • Temperature (K)