Planning the Sample’s Size
摘要
In Chap. 2 , we examined various methods that can be used to select the individuals that will be in a sample. Another aspect of taking a sample is to decide how many observational units the sample should contain. What we would like to do is to take a sample large enough to allow rejection of a false null hypothesis. In other words, our sample should have sufficient statistical power. On the other hand, collecting data for large samples is costly. Therefore, we do not want to take a larger sample than is necessary to meet our needs.