PASS

Introduction
The t-test is used to compare two means. With PASS you can specify the effect size (hypothesized difference between the population means), the standard deviation(s), the type-I and type-II error rates, and the sample size(s). The noncentral-t distribution is used in the power calculations.

PASS offers power analysis and sample size calculations for several different t-tests, including:

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.

One-Sample T-Test
Used to test whether a sample mean is significantly different from a population value. Note that a finite population correction is available when studying small populations.

Paired T-Test
Used to test whether the difference in the means of two paired variables is significantly different from zero.

Confidence Interval
Used to estimate the sample size required for confidence limits of the mean with a specified precision.

Two-Sample T-Test
Used to test whether two sample means are statistically different. Both the variances and the sample sizes can be unequal.

Hotelling’s T-Squared
Used to test whether a vector of means are statistically different. Both one and two group tests are available

Mann-Whitney Test and Wilcoxon Test
Nonparametric tests used when the assumptions of the corresponding t-test are not met.

Equivalence
Used to test whether one treatment is equivalent to another treatment in terms of their means.

Non-Inferiority
Used to test whether one treatment is no worse than another treatment.

Group-Sequential T-Test
Used in clinical trials when interim statistical tests will be conducted before the trial is completed to determine if the trial should be stopped early. The alpha spending functions described by Lan and DeMets are implemented.

Cluster Randomization
Used in place of the two-sample t-test when each observation is itself a cluster of several individuals.

[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).