This preview shows page 1. Sign up to view the full content.
Unformatted text preview: hw.2Pop.Prop.Zonly.doc Homework: Two Population Proportions using Z only Revised 11/16/10 Company 1 (population 1) uses only email to interact with employees, and some proportion of these
company 1 employees are aware of a policy change  they would find out only via email. Company 2
(population 2) uses no email at all to interact with employees, and some proportion of these company 2
employees are aware of a policy change  they would find out not via email but instead via some other
way.
In a sample of 200 from company 1, 150 were aware of the policy change. In a sample of 100 from
company 2, 86 were aware of the policy change.
Using Z, assume s^ p = 0.049
p^
1 and use this as the denominator without modification. 2 CI: Estimate the difference between p1 and p2 by using a 95% confidence interval.
1.
This is a CI for what?
2.
Construct the CI.
3.
Give interpretation
HYPOTHESIS TEST.
The test is this: is there a difference in the population proportions of employee awareness in the two
companies? Of course, one has to look at the statement of the problem to see if any difference pertains to a
onetailed test or a twotailed test. (The 2 companies use different methods to make employees aware.)
Recall the 5steps are: Hypos; CV(s); Decision Rule; Decision; Conclusion in terms of the problem.
Use the regular method and 5 steps to test at α=0.05 whether the population proportion aware at
company 1 is smaller than the population proportion at company 2. (Of course, even though we are
^^
testing that p1 < p2 , the hypotheses must be stated in terms of a difference because p1  p2 is the
sampling distribution we're using.
Therefore the hypotheses have to be in this form:
p1  p2 [null: ≥ 0, or ≤ 0, or alternate: > 0, or < 0]
Always put p1 first: p1 < p2 NOT p2 > p1 .
Provide a fully completed graph. Pay particular attention to the conclusion in terms of the problem.
2.
Determine the pvalue and show where it is on the graph.
3.
Do the test using STS and label the graph. (No need to recompute the sample Z.)
4.
Look at the email or the website for instructions about whether to turn in this HW.
a.
If it is to be turned in, then it must be turned in at the beginning of class [even if you are late],
and stapled. Not accepted after class nor without being stapled.
b.
If it is not to be turned in, then, of course, just do it on your own. 1. end ...
View
Full
Document
This note was uploaded on 01/19/2012 for the course GEOL 123 taught by Professor Kim during the Winter '11 term at Bowling Green.
 Winter '11
 kim

Click to edit the document details