Fire Safety Science Digital Archive

Statistics is a natural tool for any experimenter to minimize the number of tests without loosing the wanted information, and to ensure that the results are statistically valid. A proper test design also makes it easier to interpret data and to produce a mathematical model of the response variable being studied. This document contains a quick introduction to the fields of statistical planning and analysis of data, and shows that this is a way to save time for a research scientist dealing with fire experiments. The theory is illustrated with a few concrete, simple examples from Cone. Calorimeter testing.