On the left panel, select "Procedures for Means", then select "Two Sample T Procedures". Select the Go toLink(Links to an external site.). On the left panel, select "Procedures for Means", then select "Two Go toLink(Links to an external site.) https://rconnect.byu.edu/Stat121App/ Sample T Procedures". Select the "Student Responses" dataset, then select "Married" as your categorical explanatory variable and "HoursSleep" as your quantitative response variable. Select "Yes" as Group 1 and "No" as Group 2, where "Yes" means a student is married and "No" means a student is not married. Generate boxplots of your data.

1. Calculate the standard deviation of the nightly amount of sleep for each group. Is the equal standard deviation condition met?

A.

Yes, both sample standard deviations are less than 2.

B.

Yes,0.85/0.99is less than 2.

C.

Yes, 0.99/0.85is less than 2.

D.

No, the sample standard deviations are not exactly equal.

2. Select "Proceed to Statistical Inference". Assuming the conditions are met, we want to test whether the amount of sleep is different between married and non-married BYU students. Calculate the test statistic and the p-value. Assuming that = 0.05, what is the appropriate conclusion we should make?

A.

Since the p-value is less than 0.05, we reject the null hypothesis and conclude that there is a significant difference in hours slept between married and non-married BYU students.

B.

Since the p-value is greater than 0.05, we reject the null hypothesis and conclude that there is a significant difference in hours slept between married and non-married BYU students.

C.

Since the p-value is less than 0.05, we fail to reject the null hypothesis and conclude that there is not a significant difference in hours slept between married and non-married BYU students.

D.

Since the p-value is greater than 0.05, we fail toreject the null hypothesis and conclude that there is not a significant difference in hours slept between married and non-married BYU students.

1. Use the slider bar to calculate a 95% confidence interval at the bottom of the computer output. If we wanted to use a confidence interval instead of a hypothesis test to see if there is a significant difference in sleep between married and non-married BYU students, would we make the same conclusion as your previous answer?

A.

No. The confidence interval does not contain 0, but we cannot make assumptions about whether married students or non-married students get more sleep.

B.

No. We cannot use confidence intervals to make conclusions relating to hypothesis tests.

C.

Yes. Because the confidence interval contains only positive values, that would imply that married students get more sleep than non-married students.

D.

Yes. The hypothesis test told us what our conclusion should be, so it doesn't matter what the confidence interval is.