8 - STAT E-50 - Introduction to Statistics Hypothesis...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
STAT E-50 - Introduction to Statistics Hypothesis Testing Hypotheses ± The null hypothesis: H 0 : parameter = value ± The alternative hypothesis: H a : parameter = value we would accept if we reject the null hypothesis Plan ± Check the conditions. ± Specify the model you will use to test the null hypothesis. ± Specify the parameter of interest .
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Page 2 Mechanics ± Calculate a test statistic from the data. ± Obtain a P-value = the probability that the observed value of the test statistic (or a more extreme value) could occur if the null model were correct. ± If the P-value is small enough, we will reject the null hypothesis. Decision and Conclusion ± Statistical conclusion: state the reason you have decided reject or fail to reject the null hypothesis. ± Conclusion in context: write a complete, clear statement explaining what you conclude.
Background image of page 2
Page 3 Questions about statistical inference to think about when looking at data: 1. Am I surprised? (Should I reject the null hypothesis?) 2. How surprised am I? (What’s the P-value?) 3. What would not surprise me? (Write a confidence interval for the parameter.) A confidence interval proposes a range of plausible values for that parameter, and should almost always accompany a hypothesis test.
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Page 4 Inferences about two proportions Independence assumption: Within each group, the data should be based on results for independent individuals.
Background image of page 4
Page 5 Conditions: Randomization condition 10% condition Success/failure condition for each sample Independent samples condition The two groups we are sampling must be independent of each other
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
The Sampling Distribution Model for the Difference Between Two Independent Proportions 12 ˆˆ pp is modeled by a Normal model with: Provided that the sampled values are independent, the samples are independent, and the sample sizes are large enough, the sampling distribution of Page 6 μ p = mean p () 11 22 pq and standard deviation p p nn −= + SD
Background image of page 6
12 pp When the conditions are met, the confidence interval for the difference of two proportions is: () ( ) * ˆˆ zS E −± × Page 7 11 22 pq SE p p nn −= + where we find the standard error of the difference, from the observed proportions. The value of z* depends on the level of confidence that you specify.
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
1. Researchers at the National Cancer Institute released results of a study that examined the effects of weed-killing herbicides on house pets. (Associated Press, Sept. 4, 1991). Dogs, some of whom were from homes where the herbicide was used on a regular basis, were examined for the presence of malignant lymphoma. Use the following data to estimate the difference between the proportion of exposed dogs that develop lymphoma and the proportion of unexposed dogs that develop lymphoma: Group Sample Size Number with Lymphoma ˆ p Exposed 827 473 Unexposed 130 19 Check the conditions: Randomization condition 10% condition Success/failure condition for each sample Independent samples condition Page 8
Background image of page 8
1. Researchers at the National Cancer Institute released results of a study that examined the effects of weed-killing herbicides on house pets.
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/20/2008 for the course STAT 50 taught by Professor Weinstein during the Spring '08 term at Harvard.

Page1 / 59

8 - STAT E-50 - Introduction to Statistics Hypothesis...

This preview shows document pages 1 - 10. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online