Significance and importance: some common misapprehensions about statistics

Abstract

There are many common misapprehensions about statistics that occur in the literature. We are sure that the three misapprehensions we deal with in this short review are widespread. They concern:

• 1)what P values mean;
• 2)what an insignificant result means, and what it does not mean; the question of the ‘power’ of a statistical test;
• 3)the difference between importance and statistical significance.

We produce no formulae or recipes for dealing with particular situations, instead we concentrate on the commonsense use of simple statistics. We emphasise that if the use of any but the simplest statistics is intended, it is much better to get proper statistical help before starting experiments, rather than afterwards. Copyright © 2009 John Wiley & Sons, Ltd.