A Panorama of Statistics

24 Statistical ‘laws’

When people speak of ‘the law of gravity’, they are generally referring to what is more exactly called ‘Newton’s Law of Universal Gravitation’. This law states that the gravitational force (that is, the mutual attraction) between any two physical bodies is directly proportional to the product of their individual masses and inversely proportional to the square of the distance between them.

Why would such a scientific relationship be called a ‘law’? An analogy, while imperfect, may be helpful. Think about the word ‘law’ as it is used in parliament.

A law is a rule of behaviour that parliament has agreed is binding on people everywhere in society. Parliamentarians agree on what behaviour should become law only after having clear evidence of the expected social benefits of the law. Similarly, a physical law is a rule of behaviour that scientists have agreed to regard as binding on physical matter everywhere in nature. Scientists agree on what behaviour of matter should be called a law only after having clear evidence of its major scientific importance.

To this italicised characteristic of a scientific law, we can add five more. When a scientific law represents a relationship between variables, that relationship can be expressed in simple terms: it relates the ‘response’ variable to just a few ‘stimulus’ variables. The relationship is usually causal: it implies not only a correlational connection between the stimulus variables and the response variable, ...