PASS

Introduction
A fundamental tool in hypothesis testing is the F-test for testing the differences among several group means. It may also be used for analyzing more complex experimental designs. You specify an effect size, standard deviation, alpha, beta (or power), number of factor categories, and sample size. The program calculates the power using the non-central F distribution. What if analyses let you solve for any of the parameters.

PASS includes power analysis and sample size calculations for several experimental designs, including:

One-Way Design
Used when there is only one factor. For example, the factor might be temperature or dose level. Individual planned comparisons may also be entered and analyzed.

Factorial Design
Used when you have up to three factors in a factorial experimental design. This routine may also be used to analyze special designs such as the Latin Square design.

MANOVA
Used when want to analyze a design with up three factors and multiple responses per subject.

Randomized Block Design
Used when you have a randomized block design with up to two factors.

Multiple Comparisons
Used when you want to use multiple comparisons. Several type of multiple comparisons can be analyzed including pair-wise, treatments versus control, and multiple contrast.

Repeated Measures Design
Used when each subject is measured at least twice. The subjects may have been separated into groups. The analysis allows for the Geisser-Greenhouse correction to be taken into consideration. It allows you to specify various types of covariance patterns. This version allows for up to three grouping factors and up to three within-subject factors. It includes analysis for the F-test, Wilks Lambda, Hotelling-Lawley trace, and Pillai-Bartlett trace.

Simulation
Using Monte Carlo simulation, the power and precision of a test procedure can be evaluated when the data do not followed the normal distribution. This lets you investigate the impact of outliers, skewness, and other types of non-normality on your design.

[Back to PASS] [Back to PASS Procedure List]

System Requirements
PASS 2008, NCSS 2007, and GESS 2006 run on 32-bit and 64-bit systems with Windows 7, Vista, XP, 2000, NT, ME, 98, or 95. Requires pentium-class computer with at least 32 MB of RAM. Requires 200 MB of hard disk space. Requires Adobe Reader® version 7 or later to use the NCSS, PASS, and GESS Help Systems.

Accuracy
We at NCSS have put a great deal of effort into finding the most accurate algorithms possible. The programs have been tested and verified over and over, both by us and by our customers. Each routine has been verified against textbooks, journal articles, and, where possible, other software. NCSS is one of the most accurate statistical analysis programs available. NCSS calculates with seventeen-digit, double-precision accuracy.

Guarantee
If you are not completely satisfied with an NCSS product during the first 30 days for any reason, return the program for a full, prompt refund (excluding shipping).