PASS11帮助手册绝对是学习计算样本量最好的材料，它不仅仅是帮助如何使用PASS，更从理论等各方面详细说明了样本量计算的方法，过程，如果想学习样本量计算，大家可以下载。
简单向在家举个例子，在临床随机对照试验时，我们要采用非劣效试验，那么如何计算非劣效试验的样本量呢？
下面看本手册里的具体分析：
在Survival analysis下的logrank tests for non-inferiority下查看内容如下：

Introduction
This module computes the sample size and power for non-inferiority tests under the assumption
of proportional hazards. Accrual time and follow-up time are included among the parameters to
be set. The non-inferiority logrank test is used for data analysis.
Sometimes, the objective of a study is to show that an experimental therapy is not inferior to (no
worse than) the standard therapy. The experimental therapy may be cheaper, less toxic, or have
fewer side effects. Such studies are often called non-inferiority trials and have a one-sided
hypothesis.
Power and sample size calculations for the non-inferiority logrank test have been developed by
Jung et al. (2005), and we use their results. These calculations assume an underlying exponential
survival distribution with a uniform patient accrual pattern during the accrual period.

Technical Details
Test Statistic
Suppose a clinical trial consists of two independent groups. Designate group one as the standard
group with hazard rate h1 and sample size n1. Designate group two as the experimental group with
hazard rate h2 and sample size n2. The total sample size is N = n1+n2 . Usually, you would plan
to have n1=n2 .
Define the proportion of the total sample in each group as
Qi = ni/N, i=1,2
Individuals are recruited during an accrual period of R years (or months or days). They are
followed for an additional period of time until a total of T years is reached. Hence, the follow-up
period is T-R years. At the end of the study, the non-inferiority logrank test is conducted at
significance level α with power 1− β . Under the proportion hazards assumption, the hazard
ratio HR = h2 / h1 is constant across time.
For a given non-inferiority margin HR0 (>1) (the maximum ratio of clinical insignificance), the
statistical hypotheses tested are
H0 : HR ≥ HR0 vs. H1 : HR < HR0
.......
实际它把所有的计算过程都很详细的写了出来。