A formal test, using probability and sampling distributions to decide which of two conflicting hypotheses should be accepted. Sometimes referred to as a significance test (where the degree of error is emphasised).
A statistical test of whether a claim is supported by a set of data. The problem is broken down into two hypotheses - the null and the alternative. The null hypothesis is not rejected unless there is substantial evidence against it (innocent until proven guilty!). The logic is that a test statistic is calculated assuming H0 is true. The observed value of the test statistic is compared with its known distribution under H0 and the probability of obtaining that value or more extreme if H0 is true (the p-value) is found. If this probability is very small,it indicates that the assumption that H0 is true is unlikely and so we reject H0. If the p-value is large, we say we do not have enough evidence to reject H0.
a way of using statistical data to test a theory or answer a question