I have a quick question on False Discovery Rates. Does the p-value algorithm mentioned in Benjamini and Hochberg's 1995 control the FDR at the alpha level when the test statistics are discrete or does it only work for the continuous case with independence and positive dependence? -- Thanks, Jim. You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.