Bivariate inductive statistics - binary variables - independent groups

Fisher's exact test

This test was designed to test the hypothesis that the two column percentages in a 2-by-2 contingency table are equal. Contingency tables are tables of counts of the joint distribution of two variables that each has only a few distinct values. If column percentages are equal, it means that row and column variable are independent. The calculation of probability levels is made by calculating how many of the possible tables that may be constructed from the marginal totals given in this table support the alternative hypothesis. Fisher's exact test is especially useful when sample sizes are small (even zero in some cells) and the Chi-square test is not appropriate.

Example: Is there relationship between taking some drug and improvement of patient's condition?