Correlation Tests Two-Sample Independent Test
The Two-Sample Independent Test for Correlation is used to test if differences exist in correlations from two independent bivariate populations from which the samples were drawn.
The following hypotheses may be tested:
Where and are the population correlations from which the samples were drawn.
- The samples has been randomly drawn from two independent bivariate (two-dimensional) populations (Critical)
- The populations may be modeled with a bivariate normal distribution (Not Critical, if unimodal)
- The correlations represent the degree to which a linear relationship exists between two variables (Critical)
The p-value is flagged with an asterisk (*) when p <= alpha.