Homework 3 STAT 350 – Fall 2015 1. Researchers waited outside a bar. They started with the 3rd person who came out of the bar (this number was randomly selected) and then stopped every 10th person who came out of the bar and asked whether he or she thought drinking and driving was a serious problem. They hoped to generalize the results to all adults. Identify the following for this scenario. a. b. c. d. e. The target population The study population The sampling frame The sampling method Potential sources of bias and/or non-sampling errors (if any exist) 2. A company packaging snack foods maintains quality control by randomly selecting 10 cases from each day’s production and weighing the bags. Then they open one bag from each case and inspect the contents. a. b. c. d. The study population The sampling frame The sampling method Potential sources of bias and/or non-sampling errors (if any exist) 3. A simple random sample of 300 employees is selected from a company list containing the names of 5,000 full-time employees in order to evaluate job satisfaction. a. Give an example of possible coverage error. b. What could you do to protect against the error discussed in part a? c. Give an example of possible non-response error. d. Give an example of possible sampling error. e. Give an example of possible measurement error. 4. The General Education Requirement at Winona State University requires that students graduating from WSU complete 2 credit hours of courses in physical development and wellness. Suppose the faculty senate wanted to know what proportion of WSU students thought this was a necessary requirement. Suppose a professor on the faculty senate offered to conduct a survey study, and she did so by selecting a simple random sample of students who were majoring in her department. She sent them an email asking for them to respond to an online survey question. 1 Homework 3 STAT 350 – Fall 2015 a. Discuss the extent to which coverage error is likely to occur in this study. b. Discuss the extent to which non-response error is likely to occur in this study. c. Which of the above errors do you think will be more serious? Explain your reasoning. d. How should this professor have selected respondents for this study? Propose an improved sampling plan. 5. Suppose a state has 10 universities, 25 four-year colleges, and 50 community colleges, each of which offers multiple sections of an introductory statistics class each year. Researchers want to conduct a survey of students taking introductory statistics in the state. Explain a method for collecting each of the following types of samples. a. Describe the target population. b. Describe the sampling frame. c. Describe how you could draw a simple random sample. d. Describe how a stratified random sample could be drawn. What are the advantages/disadvantages of using this method? e. Describe how a cluster random sample of 50 persons could be drawn. What are the advantages/disadvantages of using this method? f. Even if the “best” of the above methods is used, what kind of errors might still arise that might make the sample statistics poor measures of the population’s characteristics? 6. Find an example of a poll based on a self-selected convenience sample. Report the wording of the questions and the number who chose each response. Comment on whether or not you think the results can be extended to any population. 7. A survey question will be asked to determine whether people think smoking should be banned on campus. a. Write a version of the question that is as neutral and unbiased as possible. b. Write a version of the question that is likely to get people to respond that smoking should be forbidden on campus. c. Write a version of the question that is likely to get people to respond that smoking should be allowed on campus. 2 Homework 3 STAT 350 – Fall 2015 8. Write open-ended questions to collect the following data. Your questions should be written to maximize validity and reliability. a. The respondent’s opinion on the quantity of homework assigned throughout the semester. b. The respondent’s degree of preference for different musical styles c. The respondent’s confidence in his or her swimming ability 9. Write closed questions to collect the following data. Again, your questions should be written to maximize validity and reliability. a. The respondent’s opinion on the quantity of homework assigned throughout the semester. b. The respondent’s degree of preference for different musical styles c. The respondent’s confidence in his or her swimming ability 3