For this Application Assignment, you complete three problems in which you calculate confidence intervals with two samples.
The difficult part is deciding whether the differences you observe in two groups are statistically significant. The confidence intervals can help you make that decision. When comparing the means of two groups, subtract one mean from the other. If they are the same, the answer will be zero. Remember that with the 95% confidence interval there is a range of values that contain the true mean in 95 of 100 samples. If zero is part of that range then the difference of the two means is not statistically significant. But what happens when you are comparing odds as in odds ratios? Since this is a ratio, divide the odds of one group by the odds of the other. If they are the same the answer is one. Again, remember that with the 95% confidence interval there is a range of values that contain the true mean in 95 of 100 samples. Now, however, if one is part of that range the ratio of the two odds is not statistically significant. Keep this in mind as you complete this week's Application Assignment.
To complete
this Application Assignment, answer the following problems:
Problem 5.1a:
Given the data set
Paired_t.sav, use SPSS to calculate the 95% confidence interval of the difference of means of
Sys_BP_1
and
Sys_BP_2. Interpret the output.
Problem 5.1b:
Given the data set
Independent_t.sav, use SPSS to calculate the 95% confidence interval of the difference of means of
BMI as grouped by gender. Interpret the output.
Problem 5.2:
Using the following 2x2 table, calculate the odds ratio and its 95% confidence interval. Interpret the confidence interval.
Exposure to Asbestos
|
Disease yes (Cancer case)
|
Disease no (control)
|
Total
|
Yes
|
160
|
80
|
240
|
No
|
20
|
100
|
120
|
Total
|
180
|
180
|
360
|
Copy and paste (or use the
File/Export
option in SPSS) any required data charts or summaries into a Word document, or export the data output as a Word/RTF document. Save your Application Assignment as a ".doc"