A measure of the strength of evidence against the null hypothesis in statistical hypothesis testing, representing the probability of observing a test statistic as extreme as, or more extreme than, the one observed, assuming the null hypothesis is true.