Fisher's exact test from a Bayes perspective

Fisher's exact test is commonly used to test the null hypothesis H-0 of independence or homogeneity in 2 x 2 contingency tables. This article presents a closed formula for the Bayes factor for testing independence or homogeneity in 2 x 2 contingency tables under a flexible family of spike-and-slab testing priors. Bayes factors are also compared numerically to Fisher's exact p-values for 2 x 2 contingency tables with equal row totals. Numerical results show that a large percentage of tables with 0.01 < p-value <= 0.05 have more evidence in favor of H-0 than H-1 from the Bayesian perspective.