Scales of Measurement

The four scales of measurement — nominal, ordinal, interval, and ratio (sometimes remembered as NOIR) — were proposed by psychologist Stanley Stevens in 1946 and remain foundational in research methodology. Nominal scale is the simplest — items are named or categorized but not ordered. Examples: blood type, diagnosis, treatment group. You can count frequencies and calculate mode, but not mean or median. Ordinal scale adds rank or order. Examples: pain severity (mild, moderate, severe), satisfaction ratings, tumor stage. You can determine median and rank-order, but you cannot assume equal spacing between categories. Interval scale has equal intervals between values but no meaningful zero point. The classic example is temperature in Celsius or Fahrenheit — the difference between 20°C and 30°C is the same as between 30°C and 40°C, but 0°C does not mean 'no temperature.' You can calculate means and standard deviations, but ratios are not meaningful (40°C is not 'twice as hot' as 20°C). Ratio scale has equal intervals AND a true zero point. Examples: height (0 cm = no height), weight (0 kg = no weight), blood pressure (0 mmHg = no pressure), and most laboratory values. You can calculate means, standard deviations, AND meaningful ratios (80 kg is twice 40 kg). In clinical research, most continuous variables are ratio-scale (weight, BP, lab values), and most categorical variables are nominal or ordinal. True interval-scale data is relatively rare. The practical distinction that matters most is between nominal/ordinal (non-parametric tests) and interval/ratio (parametric tests).