Section 8.3:

For each of the following problems, use Minitab to compute the test statistic and \(P\)-value as we did in class. You do not need to compute the test statistics by hand.

You should:

Print the session window.
Annotate the printout with the null and alternative hypotheses.
Test the null hypothesis using the \(P\)-value reported by Minitab for the 2-sample \(T\)-test.
State your conclusion in terms of the original claim. Do not just state 'reject' / 'do not reject'.
- For example, for Problem 8.3.5, your conclusion should be in terms of the claim that the population of heavy users of marijuana has a lower mean score of memory recall than the light users of marijuana.

For a refresher on how to perform a \(T\)-test with two independent samples using Minitab, see this Minitab documentation. Hint: For these problems, you want to enter the summarized data.

Important: Be sure you choose the correct alternative hypothesis to test the claim in the Options... dialog box.

5, 21

Additional Problem on Assessing Normality:

Use this Minitab data set from Triola & Triola to answer the following questions. This is Data Set 1 from Appendix B of Triola & Triola, which consists the health exam results for 40 men and 40 women from the National Health and Nutrition Examination Survey. The worksheet with male data is called MHEALTH and the worksheet with female data is called FHEALTH. Use this data set to perform the following steps:

Determine if the cholesterol levels in the male subjects are approximately normal.
- Plot a frequency histogram of the cholesterol levels with a normal density histogram overlayed.
- Plot a probability plot comparing the observed cholesterol levels to the expected cholesterol levels under the normal density histogram.
- Test for normality using one of the Anderson-Darling or Kolmogorov-Smirnov tests.
- Based on these three results, state your conclusion about whether the cholesterol levels are normally distributed.
Repeat the above steps to determine whether cholesterol levels in the female subjects are approximately normal.
For the above steps, print out the plots. You do not need to print out the session window or the worksheets for the above steps.
Regardless of your conclusion about the normality for the cholesterol levels in men and women (since \(n > 30\) for both samples), test the claim that the average cholesterol levels between men and women differ at the \(\alpha = 0.01\) significance level.
Hint: Remember that you can get the necessary summary statistics from each of the two samples using Stat > Basic Statistics > Display Descriptive Statistics....
- Print out the session window.
- Annotate the printout with the null and alternative hypotheses.
- Test the null hypothesis using the P-value reported by Minitab
- State your conclusion as it relates to the original claim, not just as 'reject the null' / 'do not reject the null'.