This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 1 STA2023 Chapter 21 (More About Tests) Skip: Critical Values Again (pg. 539-540), Power (545-546), A Picture is WorthWords (547-548), Reducing Type I and Type II Errors (548-549) In chapter 20 you learned that one should reject H (and support H A ) only when the p-value is small. But, what is small enough? The level of significance ( ) is used to define a threshold for when a p-value is small enough. Note that this level of significance should be chosen before collecting your data and running your hypothesis test (page 538-539). If not, you might be accused of cheating. In everyday practice it is common to use alpha levels of 0.10, 0.05, or 0.01 p-value , Reject H and Support H A ( statistically significant test results) p-value > , Fail to Reject H and Do Not Support H A (results are not statistically significant ) Beware that the level of significance is often just an arbitrarily chosen threshold level. As such, in the real world, it should only be used as a guide in determining whether or not the test results are statistically significant. Recall that the previous chapter prompted you to always include the p-value in your conclusion statement of a hypothesis test. Perhaps now you see why the p-value is the true measure that defines the strength of the evidence against the null hypothesis (H ) and in favor of the alternative hypothesis (H A ). Activity 1: Lets say that we wanted to run a hypothesis test to determine if the percentage of all students currently passing this course (all instructors, not just me) is below 70%. H : p = 0.70 H A : p < 0.70 From a random sample of n = 87 students, it was found that only 54 are currently passing....
View Full Document
This note was uploaded on 11/22/2010 for the course STA 2023 taught by Professor Ripol during the Spring '08 term at University of Florida.
- Spring '08