This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Handout #10: Testing Hypotheses About Means STA215: Introductory Applied Statistics Dr. Jann-Huei Jinn The steps common to all tests of significance about population means are as follows: 1 . Null and Alternative Hypotheses: Null hypothesis is o H “ population mean μ equals to μ “, that is, o H : μ = μ Alternative hypothesis H a is one of following: μ > μ if we want to prove that population mean is greater than a given number μ < μ if we want to prove that population mean is less than a given number μ ≠ μ if we want to prove that population mean is different from a given number 2. Level of significance α : Choose a level (usually 0.01, 0.05, and 0.10) 3. Calculate test statistic: Example : one-sample t-test statistic : n s x t μ- = ( When σ is unknown ) 4. Find the p-value and make a decision: p-value is the amount of evidence to support o H . Smaller p-value means stronger evidence to reject the null hypothesis (refer to Table 13.1, page 556). 5. Draw a conclusion about the population parameter Reject H : if p-value ≤ α , thus stating that μ is not equal to (or larger, or smaller than)...
View Full Document
- Spring '11
- Statistics, Null hypothesis, Statistical hypothesis testing, H0, one-sample t-test statistic