2.
A study was done to determine whether there is a relationship between snoring and the
risk of heart disease.
Among 1105 snorers in the study, 85 had heart disease, while only
24 of 1379 non-snorers had heart disease.
a.
We have summarized data here.
Use it to determine a 95% confidence interval that
estimates p1-p2 = difference in proportions of non-snorers and snorers that have
heart disease.
Hint: use Minitab!

b.
Based on this confidence interval, can we infer that the population proportions with
heart disease differ for non-snorers and snorers?
Activity 3: Linking the Two-proportion Z procedure with the Chi-square test and inference
about the population relative risk.
1.Revisit Activity 1 where we compared the population proportion of mean with tattoos in the group ‘has ear piercing’ with the group ‘no ear piercing’. Was the result statistically significant?
2.Would the result have still been statistically significant if we had used a two-sided hypothesis?
3.What if we had performed a chi-square test for a relationship between the two variables ‘has tattoo’ and ‘has ear piercing’, would our result have been significant?
4.What is the relative risk that indicates no difference in risks between the two groups? Recall that the relative risk = (risk in group 1) / (risk in group 2).
5.Based on the results in parts 2 and 3, do you think that a 95% confidence interval for the population relative risk would include the value 1, when comparing men with and without ear piercings?
Open the data set
UCDavis2
in Minitab. We need to assume that the sample is representative of a larger
population of students because the sample was not randomly selected, but was a convenience sample of
students in a statistics class.

#### You've reached the end of your free preview.

Want to read all 6 pages?

- Spring '08
- BARROSO,JOAOR
- Statistics, Normal Distribution, Statistical hypothesis testing