

The mathematical analysis of experimental results by various methods that are designed to extract the maximum amount of useful and relevant information while at the same time taking into account the effects and potential errors produced by unknown or uncontrollable variables. The end result of any particular statistical treatment should be to reduce a mass of often incomprehensible raw data to a few figures that give an accurate summary of the results and usually an indication of how often such results could be expected to occur by pure chance. The application of statistical considerations to experimental design aims to ensure that results obtained from a necessarily limited sample can be seen to have general application.




