The generic term for various tests that are used to see if a hypothesis about data is statistically significant (that is, unlikely to happen by chance). Examples include the Student's t test used to see if the mean (μ) of two sets of data are different, and the Pearson correlation coefficient, used to test whether there is some form of relation between two variables in linear regression. Typically the test will be based on some form of theoretical distribution leading to a parametric test, but there are also nonparametric tests based purely on the order of data items, and also tests based on simulation or empirical data.
Used on pages 65, 66, 67, 86, 93
Also known as significance testing