PASS 2025 Update

Update Release Notes

This page lists the changes that have been made to PASS 2025 since it was released.

Version: 25.0.6

Released: June 15, 2026

1. Corrected a typo in the documentation for the Mixed Models Tests for Two Means in a 3-Level Hierarchical Design (Level-2 Randomization) procedure. The formula for f2 was noted incorrectly as f2 = 1 + Mρ1 – Mρ2. The formula has been updated to the correct formula: f2 = 1 + (M – 1)ρ1 – Mρ2. This error did not affect software calculations, which were (and still are) based on the correct formula.

2. Corrected an error in the Logrank Tests with Proportional Hazards (Schoenfeld and Wu) procedure. When solving for sample size with Group Allocation set to “Enter percentage in Group 1, solve for N1 and N2” and if Percent in Group 1 was greater than 50, the calculation would not find a solution.

3. Corrected and clarified documentation and Option Info messages related to the dispersion parameter in five Negative Binomial procedures (Non-Inferiority Tests for the Ratio of Two Negative Binomial Rates, Superiority by a Margin Tests for the Ratio of Two Negative Binomial Rates, Equivalence Tests for the Ratio of Two Negative Binomial Rates, Non-Inferiority Tests for Vaccine Efficacy using the Ratio of Two Negative Binomial Rates, and Superiority by a Margin Tests for Vaccine Efficacy using the Ratio of Two Negative Binomial Rates) where Negative Binomial dispersion was mis-identified as ϕ instead of κ. The Negative Binomial dispersion parameter is now identified as κ. The change did not affect calculations, which were (and still are) correct.

Version: 25.0.5

Released: February 12, 2026

1. Corrected and clarified the mouse-over message for βᴍ in the Tests of Mediation Effect in Logistic Regression procedure.

2. Corrected a "Value Out of Range" error in the Historical Controls procedures that occurred when running with Windows language settings that use comma as the decimal separator.

Version: 25.0.4

Released: January 21, 2026

1. Corrected errors in the Non-Inferiority Tests for the Difference Between Two Correlated Proportions (Conditional McNemar's Test) and Superiority by a Margin Tests for the Difference Between Two Correlated Proportions (Conditional McNemar's Test) procedures. The Min(P11) and Max(P11) formulas and resulting values were incorrect. The calculations have been adjusted to align with Lu & Bean (1995). This error did not affect actual sample size and power calculations for given values of P11, nor when using the Lu & Bean Midpoint. It did effect cases where the value for P11 was computed using the Range, since the Min, Max, and resulting Range were incorrect in some cases. Also, cases where P11 is greater than or equal to its maximum value now correctly result in no power and sample size calculations.

2. Corrected an error in the Non-Inferiority, Superiority by a Margin, and Non-Zero Null Tests for the Difference Between Two Proportions procedures. When Power Calculation Method was set to Binomial Enumeration and Input Type was set to Mixture, the power and sample size for the Farrington & Manning and Miettinen & Nurminen Likelihood Score Tests were calculated incorrectly.

3. Corrected a minor typo in the documentation for the Tests for Two Means Assuming Equal Variances using a Bayesian Approach procedure. On page 123-2, the formula for θ = … was replaced with θ² = … . Also, clarification regarding α in the formulas was added.

4. Corrected a minor typo in the documentation for the Confidence Intervals for Kappa procedure. The confidence interval formula incorrectly referenced SD instead of SE.

Version: 25.0.3

Released: July 11, 2025

1. Corrected a display issue with the Procedure Window. In some cases when the Procedure Window was scrolled, changing an option in a dropdown box would cause the window inputs to misalign and appear crammed at the top of the window. This has been corrected.

2. Corrected an error in Logrank Tests with Non-Proportional Hazards (Lakatos and Wu), Weighted Logrank Tests with Non-Proportional Hazards (Lakatos and Wu), and Logrank Tests with Proportional Hazards (Freedman and Wu). In all three procedures, when the hazard ratio was greater than 1, the power was computed incorrectly, resulting in no output when solving for sample size.

Version: 25.0.2

Released: March 10, 2025

1. Corrected an error in the Tests for Two Groups using the Win-Ratio Composite Endpoint procedure. When solving for sample size with Group Allocation set to "Enter N1, solve for N2", the resulting group sample size (N2) was too large. This has been corrected.

2. Corrected an error that caused Message Boxes to appear behind other windows in some cases, so that they could not be seen.

Version: 25.0.1

Released: January 20, 2025

1. Initial Release of PASS 2025