3.40 When a test statistic has a continuous distribution, the P-value has a null uniform dis-

tribution, P(P-value < α) = a for 0 < a < 1. For Fisher's exact test, explain why P(P-value < α) < α. [Hint: P(P-value < α) = E[P(P-value ≤ anı+, n+1, n)].]