 |
|
| Hypothesis Tests |
Introduction
Hypothesis tests are frequently performed for one and two samples. For one sample, you are often interested in whether a population characteristic such as the mean is equivalent to a certain value. For two samples, you may be interested in whether the true means are different. When you have paired data, you may be interested in whether the mean difference is zero.
Statistical hypothesis tests depend on a statistic designed to measure the degree of evidence for various alternative hypotheses. You compute the value of the statistic for your sample. If the value is improbable under the hypothesis you want to test, then you reject the hypothesis.
 |
Figure 8.1: Hypothesis Tests Menu
The Analyst Application enables you to perform hypothesis tests for means, proportions, and variances for either one or two samples.
The examples in this chapter demonstrate how you can use the Analyst Application to perform a one-sample t-test, a paired t-test, a two-sample test for proportions, and a two-sample test for variances. Additionally, the section "Discussion of Other Tests" provides information on other hypothesis tests you can perform with the Analyst Application.
|
|
Copyright © 2007 by SAS Institute Inc., Cary, NC, USA. All rights reserved.