Identifying types of variable

Quantitative analysis p. 306

  1. Ethnic origin’ has more than two categories, but they cannot be put in any meaningful order. It is therefore a nominal (or categorical) variable.
  2. ‘Length of sentence following conviction (in years)’ has many categories which can be put in order: 0, 1, 2, 3, etc. The distances between categories are equal (one year) and are expressed in whole numbers only. It is therefore an interval/ration variable with discrete data.
  3. ‘University educated’ can either be ‘yes’ or ‘no’. There are only two possible options, so it is a binary (or dichotomous) variable.
  4. ‘Number of times arrested’ has categories of ‘0’, ‘1-3’, ‘4-10’, ’11-20’ and ‘more than 20’ as possibilities. There are more than two orderable categories, but the distances between them are unequal. It is therefore an ordinal variable.
  5. ‘Blood alcohol level as measured at the roadside (in milligrams per 100ml)’ has a range of possible values (certainly more than two). The values can be ordered and are equally spaced. The values need not be expressed in whole numbers. Blood alcohol levels are often given to one or two decimal places. Hence this is an interval/ratio variable with continuous data.
Back to top