Question

For this project, you will make decisions about how two parameters (proportions or means) compare using hypothesis tests, and you will estimate the difference between the two parameters using confidence intervals. For each confidence, report the following: the confidence interval limits rounded as directed (from StatCrunch) An interpretation of the confidence interval (e.g. "I am 95% confident ... .") Here is a template for reporting the answer for a sample confidence interval problem. Sample problem: Find a 90% confidence interval for the difference between the proportion of all males who use their smart phones to watch movies and the proportion of all females who do. Interval Limits: 0.036 to 0.264 Interpretation: "I am 90% confident the proportion of males who use their smartphones to watch movies is 3.6% to 26.4% higher than the proportion of females who do." Note, if the interval goes from a positive number to a negative number and therefore contains the number 0, then you are confident it plausible the two parameters being compares are equal to each other (its plausible the difference between the parameters is zero). For each hypothesis test, report the following: The null hypothesis, H0 The alternative hypothesis, H1 The test statistic rounded to the nearest hundredth (from StatCrunch) The P-value for the test (from StatCrunch) The formal decision (Reject H0 or Fail to reject H0) The conclusion of the test in non-technical terms Here is a template for reporting the answer for a sample hypothesis test problem. Sample problem: Use a 10% significance level to test the claim the proportion of all males who use their smart phones to watch movies is greater than the proportion of all females who do. Assume p1 = the proportion of males who use their smartphones to watch movies and p2 = the proportion of females who use their smartphones to watch movies. Null hypothesis: H0: p1 - p2 = 0 Alternative Hypothesis: H1: p1 - p2 > 0 Test Statistic: z = 2.15 P-value: 0.016 Decision: Reject the null hypothesis Conclusion: There is enough evidence to conclude the proportion of all males who use their smartphones to watch movies is greater than the proportion of females who do. The resources on the course home page will provide you more information on these types of hypothesis tests and confidence intervals and how to use StatCrunch to find test statistics, P-values, and confidence intervals. ******************************************************************************************* To complete this project, use the class data to complete some two sample inference. Type your answers into a Microsoft Word or rich text format document making sure you clearly report your answers for each problem and upload your document to submit your work. Use the Grapevine Online Statistics Data File in StatCrunch shared by user sgrapevine. Assume this data is representative of all online students. Make sure you use StatCrunch to find all confidence intervals, test statistics, and P-values. You will have to use the Where option to separate the data by gender. Each problem is worth a total of 6 points. For each confidence interval earn 4 points for reporting the correct interval limits and 2 points for correctly interpreting the interval. For each hypothesis test, earn 1 point for each correct component. You can also earn one point for rounding as directed for a total of 55 points. Note, this is not a team assignment. You must submit your own original work!

9) Use a 1% significance level to test the claim there is no difference between the hours of sleep on week nights and the hours of sleep on weekend nights for online students. (Hint: this is a matched pairs test)

Answer 1