1 / 7

Ch. 27 More tests for averages

Ch. 27 More tests for averages. New test: compare 2 averages from 2 samples Box A Box B average of sample A average of sample B SD of sample A SD of sample B a =SE for avg. of sample A b =SE for avg. of sample B. Null hypothesis: Average of box A = Average of box B

trygg
Download Presentation

Ch. 27 More tests for averages

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Ch. 27 More tests for averages • New test: compare 2 averages from 2 samples Box A Box B • average of sample A average of sample B • SD of sample A SD of sample B • a=SE for avg. of sample A b=SE for avg. of sample B

  2. Null hypothesis: Average of box A = Average of box B • Expected: average of sample A – average of sample B = 0 • Observed: average of sample A – average of sample B • SE for difference of averages =

  3. Example 1 Ch.27 B #3 p.509 • Example 2 Ch. 27 B #4 p.509 • This same method can be used for proportions and percentiles. • Example 3 Ch. 27 B #8 p. 510

  4. Summary • SE for difference = • This is true for averages, proportions, and percentiles. • Called a 2-sample t-test or z-test. • When do we use this test? • Sample with replacement • 2 independent samples

  5. Examples of when to use this test: • Ch. 26 C #7 • Tax returns: Sample 2 numbers per return – the difference is on the ticket – NOT independent. • Roll die 100 times and compare the percent of 1’s to percent of 3’s. • Experiments • Randomized controlled experiment • Compare 2 groups

  6. This violates the assumption that we have 2 independent samples. • The 2 samples are not independent • The sampling is without replacement • However, we assume that the effects of these 2 violations cancel each other out. • Illustration using a box model: Each ticket has 2 numbers, 1 for result if in treatment group and 1 for result if in control group.

  7. Draw a sample of tickets without replacement, observe to responses that relate to the treatment. • Draw a sample without replacement from the remaining tickets, observe the response that relate to the control. • Example 4: Ch. 27 C #2 p.513

More Related