?Xr15-52 A critical issue for service companies is how many customers cancel. Some

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. Consider a multinomial experiment involving n = 300 trials and k = 5 cells. The observed frequencies resulting from the experiment are shown in the accompanying table, and the null hypothesis to be tested is as follows: \(H_{0}: \quad p_{1}=.1, p_{2}=.2, p_{3}=.3, p_{4}=.2, p_{5}=.2\) Test the hypothesis at the 1% significance level. Cell 1 2 3 4 5 Frequency 24 64 84 72 56 Text Transcription: H_0: p_1=.1, p_2=.2, p_3=.3, p_4=.2, p_5=.2

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. Repeat Exercise 15.1 with the following frequencies: Cell 1 2 3 4 5 Frequency 12 32 42 36 28

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. Repeat Exercise 15.1 with the following frequencies: Cell 1 2 3 4 5 Frequency 6 16 21 18 14

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. Review the results of Exercises 15.1–15.3. What is the effect of decreasing the sample size?

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. Consider a multinomial experiment involving n = 150 trials and k = 4 cells. The observed frequencies resulting from the experiment are shown in the accompanying table, and the null hypothesis to be tested is as follows: \(H_{0}: \quad p_{1}=.3, p_{2}=.3, p_{3}=.2, p_{4}=.2\) Cell 1 2 3 4 Frequency 38 50 38 24 Text Transcription: H_0: p_1=.3, p_2=.3, p_3=.2, p_4=.2

Developing an Understanding of Statistical Concepts Exercises 15.1–15.6 are “what-if” analyses designed to determine what happens to the test statistic of the goodness-of-fit test when elements of the statistical inference change. These problems can be solved manually or using Excel’s CHITEST. For Exercise 15.5, retest the hypotheses, assuming that the experiment involved twice as many trials (n = 300) and that the observed frequencies were twice as high as before, as shown here. Cell 1 2 3 4 Frequency 76 100 76 48

Exercises 15.7–15.21 require the use of a computer and software. Use a 5% significance level unless specified otherwise. The answers to Exercises 15.7–15.16 may be calculated manually. See Appendix A for the sample statistics. Xr15-07 The results of a multinomial experiment with k = 5 were recorded. Each outcome is identified by the numbers 1 to 5. Test to determine whether there is enough evidence to infer that the proportions of outcomes differ.

Exercises 15.7–15.21 require the use of a computer and software. Use a 5% significance level unless specified otherwise. The answers to Exercises 15.7–15.16 may be calculated manually. See Appendix A for the sample statistics. Xr15-08 A multinomial experiment was conducted with k = 4. Each outcome is stored as an integer from 1 to 4 and the results of a survey were recorded. Test the following hypotheses. \(H_{0}: p_{1}=.15, p_{2}=.40, p_{3}=.35, p_{4}=.10\) \(H_{1}\) : At least one \(p_{i}\) is not equal to its specified value Text Transcription: H_0: p_1=.15, p_2=.40, p_3=.35, p_4=.10 H_1 p_i

Exercises 15.7–15.21 require the use of a computer and software. Use a 5% significance level unless specified otherwise. The answers to Exercises 15.7–15.16 may be calculated manually. See Appendix A for the sample statistics. Xr15-09 To determine whether a single die is balanced, or fair, the die was rolled 600 times. Is there sufficient evidence to allow you to conclude that the die is not fair?

Applications Xr15-10 Grades assigned by an economics instructor have historically followed a symmetrical distribution: 5% A’s, 25% B’s, 40% C’s, 25% D’s, and 5% F’s. This year, a sample of 150 grades was drawn and the grades (1 = A, 2 = B, 3 = C, 4 = D, and 5 = F) were recorded. Can you conclude, at the 10% level of significance, that this year’s grades are distributed differently from grades in the past?

Applications Xr15-11 Pat Statsdud is about to write a multiple- choice exam but as usual knows absolutely nothing. Pat plans to guess one of the five choices. Pat has been given one of the professor’s previous exams with the correct answers marked. The correct choices were recorded where 1 = (a), 2 = (b), 3 = (c), 4 = (d), and 5 = (e). Help Pat determine whether this professor does not randomly distribute the correct answer over the five choices? If this is true, how does it affect Pat’s strategy?

Applications Xr15-12 Financial managers are interested in the speed with which customers who make purchases on credit pay their bills. In addition to calculating the average number of days that unpaid bills (called accounts receivable) remain outstanding, they often prepare an aging schedule. An aging schedule classifies outstanding accounts receivable according to the time that has elapsed since billing and records the proportion of accounts receivable belonging to each classification. A large firm has determined its aging schedule for the past 5 years. These results are shown in the accompanying table. During the past few months, however, the economy has taken a downturn. The company would like to know whether the recession has affected the aging schedule. A random sample of 250 accounts receivable was drawn and each account was classified as follows: 1 = 0?14 days outstanding 2 = 15?29 days outstanding 3 = 30?59 days outstanding 4 = 60 or more days outstanding Determine whether the aging schedule has changed.

Applications Xr15-13 License records in a county reveal that 15% of cars are subcompacts (1), 25% are compacts (2), 40% are midsize (3), and the rest are an assortment of other styles and models (4). A random sample of accidents involving cars licensed in the county was drawn. The type of car was recorded using the codes in parentheses. Can we infer that certain sizes of cars are involved in a higher than expected percentage of accidents?

Applications Xr15-14 In an election held last year that was contested by three parties. Party A captured 31% of the vote, party B garnered 51%, and party C received the remaining votes. A survey of 1,200 voters asked each to identify the party that they would vote for in the next election. These results were recorded where 1 = party A, 2 = party B, and 3 = party C. Can we infer at the 10% significance level that voter support has changed since the election?

Applications Xr15-15 Pitch trackers allow baseball fans watching televised games to see whether an umpire made the correct call. After tracking all pitches over an entire season the following probabilities were recorded. 1. Pitch was in the strike zone and the umpire called a ball 5.28% 2. Pitch was in the strike zone and the umpire called it a strike 34.72% 3. Pitch was outside the strike zone and the umpire called it a ball 51.00% 4. Pitch was outside the strike zone and the umpire called it a strike 9.00% A new umpire has just started and the pitch tracker has recorded each outcome using these codes. Is there evidence at the 5% significance level to infer that this umpire differs in his calls?

Applications Xr15-16 Refer to Exercise 15.15. After an entire season of pitches the following proportions for all pitches were observed. 1. Pitch in the strike zone 40.0% 2. Pitch misses strike zone upper left 14.4% 3. Pitch misses the strike zone lower left 17.8% 4. Pitch misses strike zone upper right 9.5% 5. Pitch misses strike zone lower right 18.3% A rookie pitcher starts his first game and the pitch outcome was recorded using the codes. Is there enough evidence at the 5% significance level to conclude that the rookie differs from all other pitchers?

Applications Xr15-17 An aging pitcher has recorded the following pitch distribution over his long career from 1995 to 2014. 1. Fast ball 52% 2. Curve ball 19% 3. Slider 15% 4. Changeup 14% However, undergoing a surgery to repair a tendon in 2015 the pitches for the first 5 games in 2016 were recorded using the codes. Is there evidence to conclude that his pitching distribution has changed?

Applications Xr15-18 From January 1 to December 31, 2011, there were 5,086 bank robberies in the United States. The percentage of that total for each day of the week is listed next. The number of robberies per day for the first two months of 2016 was recorded. Can we infer that the distribution of bank robberies per day has changed? (Source: FBI) 1. Monday 17% 2. Tuesday 18% 3. Wednesday 17% 4. Thursday 18% 5. Friday 21% 6. Saturday and Sunday 9%

Applications Xr15-19 Using the data supplied by NOAA National Weather Service, 19,450 tornadoes struck the United States between 2000 and 2014. The percentage for each month is listed here. January 2.4% February 3.0% March 6.7% April 16.9% May 21.2% June 17.4% July 8.0% August 5.7% September 5.9% October 4.9% November 4.9% December 2.8% The monthly number of tornadoes for 2015 was recorded. Is there enough statistical evidence at the 5% significance level to infer that the percentages have changed?

Market Segmentation Market segmentation was introduced in Section 12-4, where a statistical technique was used to estimate the size of a segment. In Chapters 13 and 14, statistical procedures were applied to determine whether market segments differ in their purchases of products and services. Exercise 15.20 requires you to apply the chi-squared goodness-of-fit test to determine whether the relative sizes of segments have changed. Xr12-131* Refer to Exercise 12.131 where the statistics practitioner estimated the size of market segments based on education among California adults. Suppose that census figures from 10 years ago showed the education levels and the proportions of California adults, as follows: Level Proportion 1. Did not complete high school .23 2. Completed high school only .40 3. Some college or university .15 4. College or university graduate .22 Determine whether there has been a change in these proportions.

GSS2014* Test to determine whether there is sufficient evidence that the General Social Survey in 2014 overrepresented at least one race (RACE).

GSS2014* Can we infer that the General Social Survey in 2014 overrepresented at least one category of marital status (MARITAL)?

GSS2014* Can we infer that the General Social Survey in 2014 overrepresented at least one education category (DEGREE)?

SCF2013:\All* Test to determine whether there is sufficient evidence that the Survey of Consumer Finances in 2013 overrepresented at least one race (RACE).

SCF2013:\All* Can we infer that the Survey of Consumer Finances in 2013 overrepresented at least one education category (EDCL)?

Developing an Understanding of Statistical Concepts Conduct a test to determine whether the two classifications L and M are independent, using the data in the accompanying cross-classification table. (Use \(\alpha\)= .05.) Text Transcription: alpha

Developing an Understanding of Statistical Concepts Repeat Exercise 15.26 using the following table:

Developing an Understanding of Statistical Concepts Repeat Exercise 15.26 using the following table:

Developing an Understanding of Statistical Concepts Review the results of Exercises 15.26–15.28. What is the effect of decreasing the sample size?

Developing an Understanding of Statistical Concepts Conduct a test to determine whether the two classifications R and C are independent, using the data in the accompanying cross-classification table. (Use \(\alpha\)= .10.) Text Transcription: alpha

Applications Use a 5% significance level unless specified otherwise. The trustee of a company’s pension plan has solicited the opinions of a sample of the company’s employees about a proposed revision of the plan. A breakdown of the responses is shown in the accompanying table. Is there enough evidence to infer that the responses differ between the three groups of employees?

Applications Use a 5% significance level unless specified otherwise. The operations manager of a company that manufactures shirts wants to determine whether there are differences in the quality of workmanship among the three daily shifts. She randomly selects 600 recently made shirts and carefully inspects them. Each shirt is classified as either perfect or flawed, and the shift that produced it is also recorded. The accompanying table summarizes the number of shirts that fell into each cell. Do these data provide sufficient evidence to infer that there are differences in quality between the three shifts?

Applications Use a 5% significance level unless specified otherwise. Xr15-51 A Gallup survey asked a random sample of federal government and private sector workers to judge their well-being. The responses are 1 = thriving, 2 = struggling, 3 = suffering. Is there enough evidence to conclude that government and private sector workers differ in their well-being? One of the issues that came up in a recent national election (and is likely to arise in many future elections) is how to deal with a sluggish economy. Specifically, should governments cut spending, raise taxes, inflate the economy (by printing more money) or do none of the above and let the deficit rise? And as with most other issues, politicians need to know which parts of the electorate support these options. Suppose that a random sample of 1,000 people was asked which option they support and their political affiliations. The possible responses to the question about political affiliation were Democrat, Republican, and Independent (which included a variety of political persuasions). The responses are summarized in the accompanying table. Do these results allow us to conclude at the 1% significance level that political affiliation affects support for the economic options?

Applications Use a 5% significance level unless specified otherwise. Econetics Research Corporation, a well-known Montreal-based consulting firm, wants to test how it can influence the proportion of questionnaires returned from surveys. In the belief that the inclusion of an inducement to respond may be important, the firm sends out 1,000 questionnaires: Two hundred promise to send respondents a summary of the survey results, 300 indicate that 20 respondents (selected by lottery) will be awarded gifts, and 500 are accompanied by no inducements. Of these, 80 questionnaires promising a summary, 100 questionnaires offering gifts, and 120 questionnaires offering no inducements are returned. What can you conclude from these results?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xm02-04 (Example 2.4 revisited) A major North American city has four competing newspapers: the Globe and Mail (G&M), Post, Sun, and Star. To help design advertising campaigns, the advertising managers of the newspapers need to know which segments of the newspaper market are reading their papers. A survey was conducted to analyze the relationship between newspapers read and occupation. A sample of newspaper readers was asked to report which newspaper they read: Globe and Mail (1) Post (2), Star (3), Sun (4), and to indicate whether they were blue-collar workers (1), white-collar workers (2), or professionals (3). Can we infer that occupation and newspaper are related?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-36 To determine the actual side effects, pharmaceutical companies often conduct studies that compare the side effects of their drug versus the side effects of a placebo. One such study examined the side effects of a new cold remedy. A random sample of 250 people was given the cold remedy and another 250 were given a placebo that looked like the cold remedy. These responses were recorded as 1 = Headache, 2 = Drowsiness, 3 = Stomach upset, 4 = No side effect Do these data provide enough evidence to infer that the reported side effects differ between the cold remedy and the placebo?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr02-45 (Exercise 2.45 revisited) Is there brand loyalty among car owners in their purchases of gasoline? To help answer the question, a random sample of car owners was asked to record the brand of gasoline in their last two purchases: 1 = Exxon, 2 = Amoco, 3 = Texaco, 4 = Other. Can we conclude that there is brand loyalty in gasoline purchases?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-38 During the past decade, many cigarette smokers have attempted to quit. Unfortunately, nicotine is highly addictive. Smokers use a large number of different methods to help them quit. These include nicotine patches, hypnosis, and various forms of therapy. A researcher for the Addiction Research Council wanted to determine why some people quit while others attempted to quit but failed. He surveyed 1,000 people who planned to quit smoking. He determined their educational level and whether they continued to smoke 1 year later. Educational level was recorded in the following way: 1 = Did not finish high school 2 = High school graduate 3 = University or college graduate 4 = Completed a postgraduate degree A continuing smoker was recorded as 1; a quitter was recorded as 2. Can we infer that the amount of education is a factor in determining whether a smoker will quit?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-39 Because television audiences of newscasts tend to be older (and because older people suffer from a variety of medical ailments), pharmaceutical companies’ advertising often appears on national news on the three networks (ABC, CBS, and NBC). To determine how effective the ads are a survey was undertaken. Adults over 50 were asked about their primary sources of news. The responses are 1. ABC News 2. CBS News 3. NBC News 4. Newspapers 5. Radio 6. None of the above Each person was also asked whether they suffer from heartburn, and if so, what remedy they take. The answers were recorded as follows: 1. Do not suffer from heartburn 2. Suffer from heartburn but take no remedy 3. Suffer from heartburn and take an over-thecounter remedy (e.g., Tums, Gavoscol) 4. Suffer from heartburn and take a prescription pill (e.g., Nexium) Is there a relationship between an adult’s source of news and his or her heartburn condition?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr02-44 (Exercise 2.44 revisited) The associate dean of a business school was looking for ways to improve the quality of the applicants to its MBA program. In particular, she wanted to know whether the undergraduate degree of applicants differed among her school and the three nearby universities with MBA programs. She sampled 100 applicants of her program and an equal number from each of the other universities. She recorded their undergraduate degrees (1 = BA, 2 = BEng, 3 = BBA, 4 = other) as well as universities (codes 1, 2, 3, and 4). Do these data provide sufficient evidence to infer that undergraduate degree and the university each person applied are related?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-41 The relationship between drug companies and medical researchers is under scrutiny because of possible conflict of interest. The issue that started the controversy was a 1995 case control study that suggested that the use of calcium-channel blockers to treat hypertension led to an increase risk of heart disease. This led to an intense debate both in technical journals and in the press. Researchers writing in the New England Journal of Medicine (“Conflict of Interest in the Debate over Calcium Channel Antagonists,” January 8, 1998, p. 101) looked at the 70 reports that appeared during 1996–1997, classifying them as favorable, neutral, or critical toward the drugs. The researchers then contacted the authors of the reports and questioned them about financial ties to drug companies. The results were recorded in the following way: Column 1: Results of the scientific study; 1 = favorable, 2 = neutral, 3 = critical Column 2: 1 = financial ties to drug companies, 2 = no ties to drug companies Do these data allow us to infer that the research findings for calcium-channel blockers are affected by whether the research is funded by drug companies?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-42 After a thorough analysis of the market, a publisher of business and economics statistics books has divided the market into three general approaches to teach applied statistics. These are (1) use of a computer and statistical software with no manual calculations, (2) traditional teaching of concepts and solution of problems by hand, and (3) mathematical approach with emphasis on derivations and proofs. The publisher wanted to know whether this market could be segmented on the basis of the educational background of the instructor. As a result, the statistics editor organized a survey that asked 195 professors of business and economics statistics to report their approach to teaching and which one of the following categories represents their highest degree: 1. Business (MBA or Ph.D. in business) 2. Economics 3. Mathematics or engineering 4. Other a. Can the editor infer that there are differences in type of degree among the three teaching approaches? If so, how can the editor use this information? b. Suppose that you work in the marketing department of a textbook publisher. Prepare a report for the editor that describes this analysis.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-43 Every year, there are more than 300,000 robberies in the United States. A researcher took a random sample of robberies in 2000, 2005, and 2010, and recorded the weapon used (1 = Firearm, 2 = Knife or other cutting instrument, 3 = Other, 4 = No weapon). Is there sufficient evidence to infer that the frequency of the use of weapons in robberies differed over the three years?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-44 Refer to Exercise 12.139. The engineering team took samples of bridges in 2004, 2006, 2008, and 2010 and recorded whether each bridge was structurally deficient, functionally obsolete, or structurally sound. These three categories were recorded as 1, 2, and 3, respectively. Do these data allow us to conclude that the deteriorating condition of American bridges has changed over the years? Source: Adapted from Statistical Abstract of the United States, 2012, Table 1090.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-45 Household types are categorized in the following way: 1. Married couple with children, 2. Married couple without children, 3. Single parent, 4. One person, 5. Other. Random samples of families in the United States, Canada, and the United Kingdom were drawn and the household types recorded. Is the sufficient evidence to infer that there are differences in household types between the three countries? Source: Adapted from Statistical Abstract of the United States, 2012, Table 1338.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-46 Refer to Exercise 15.45. Random samples from Denmark, Ireland, the Netherlands, and Sweden were drawn. Is the sufficient evidence to infer that there are differences in household types between the four countries?

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-47 A statistics practitioner took random samples from Canada, Australia, New Zealand, and the United Kingdom, and classified each person as either obese (2) or not (1). Can we conclude from these data that there are differences in obesity rates between the four Commonwealth nations? Source: Adapted from Statistical Abstract of the United States 2012, Table 1342.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-48 To measure the extent of cigarette smoking around the world, random samples of adults in Denmark, Finland, Norway, and Sweden were drawn. Each was asked whether he or she smoked (2 = Yes, 1 = No). Can we conclude that there are differences in smoking between the four Scandinavian countries? Source: Adapted from Statistical Abstract of the United States 2012, Table 1343.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-49 Refer to Exercise 15.48. The survey was performed in Canada, Australia, New Zealand, and the United Kingdom. Is there enough evidence to infer that there are differences in adult cigarette smoking between the four Commonwealth countries? Source: Adapted from Statistical Abstract of the United States 2012, Table 1343.

Exercises 15.35–15.50 require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level unless specified otherwise. Xr15-50 In 2013, the Supreme Court of the United States ruled on a California law that banned same-sex marriage. An important element of that decision was public opinion. In March, Public Policy Polling conducted a survey of Florida voters and asked each to identify themselves as either: 1. Democrat, 2. Republican, or 3. Independent, and to choose one of the following: 1. Gay couples should be allowed to marry legally. 2. Gay couples should be allowed to form civil unions but not marry. 3. There should be no legal recognition of a gay couple’s relationship. The results were recorded. Is there sufficient evidence to infer that the three political persuasions differ?

Xr15-51 A Gallup survey asked a random sample of federal government and private sector workers to judge their well-being. The responses are 1 = thriving, 2 = struggling, 3 = suffering. Is there enough evidence to conclude that government and private sector workers differ in their well-being?

Xr15-52 A critical issue for service companies is how many customers cancel. Some wireless carriers lose an average of 3% of their subscribers each month. Should companies spend more effort getting new customers or trying to win back old customers who left? Researchers have argued that it is easier and cheaper to lure back customers. One telecom firm tested four win-back offers. Their offers and their costs are listed next. 1. Discount offer: $20 off for 6 months. Cost $120 2. Upgrade offer: a $35 movie channel for 3 months. Cost $105 3. Bundled offer: $20 off for 6 months, plus a $35 movie channel free for 3 months. Cost $225 4. Tailored offer: Customers who left over price get the discount; customers who left over service get the upgrade. Cost $120/$105 A random sample of customers who left was drawn and one-quarter received one type of offer. The responses were recorded where 1 = took the offer and 2 = did not take the offer. Can we infer at the 5% significance level that there are differences in success rates between the four win-back offers?

Use a 5% significance level. GSS2014* An important element in the business of politics is to know who supports you and who doesn’t. Is there sufficient evidence to infer that men and women (SEX: 1 = Male, 2 = Female) differ in their support of the political groups (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican)?

Use a 5% significance level. GSS2014* Do foreign-born Americans differ from Americans born in the United States (BORN: 1 = Born in the United States, 2 = Born elsewhere) in their support of political groups (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican)?

Use a 5% significance level. GSS2014* Is it a myth that men are more conservative and women more liberal? Conduct a test to determine whether men and women (SEX: 1 = Male, 2 = Female) differ in their political views (POLVIEWS3: 1 = Liberal, 2 = Moderate, 3 = Conservative).

Use a 5% significance level. GSS2014* Do Americans who work for themselves (WRKSLF: 1 = Self-employed, 2 = Someone else) support the same political groups (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican) as Americans who work for someone else? Conduct a statistical test to answer the question.

Use a 5% significance level. GSS2014* Are government workers more or less conservative than people who work in the private sector? Is there sufficient evidence to conclude that government workers (WRKGOVT: 1 = Government, 2 = Someone else) differ from private-sector workers in terms of their political views (POLVIEWS3: 1 = Liberal, 2 = Moderate, 3 = Conservative)?

Use a 5% significance level. GSS2006* GSS2008* GSS2010* GSS2012* GSS2014* The issue of gun control in the United States is one that is often debated particularly during elections. The question arises, What does the public think about the issue and does support vary from year to year? Test to determine whether there is enough evidence to conclude that support for gun laws (GUNLAW: 1 = Favor, 2 = Oppose) varied from year to year.

Use a 5% significance level. GSS2014* In the last two decades, an increasing proportion of women have entered the workforce. Determine whether there is enough evidence to conclude that men and women (SEX) differ in their work status (WRKSTAT).

Use a 5% significance level. GSS2014* Is there sufficient evidence to infer that support for capital punishment (CAPPUN: 1 = Favor, 2 = Oppose) is related to political affiliation (PARTYID3: 1 = Democrat, 2 = Republican, 3 = Independent)?

Use a 5% significance level. GSS2014* Immigration has become a hot-button issue in American politics. One question that arises is: Are immigrants’ educational attainments different from those born in the United States? Conduct a test to answer the question (BORN: 1 = U.S., 2 = outside U.S.; DEGREE: Highest degree completed of respondent, 0 = Left high school, 1 = High school, 2 = Junior college, 3 = Bachelor’s degree, 4 = Graduate degree).

Use a 5% significance level. GSS2014* Is there sufficient evidence to infer that there are differences in marital status between people born in the United States and those born outside the United States (BORN: 1 = U.S., 2 = outside U.S.; MARITAL: Marital status: 1. Married, 2. Widowed, 3. Divorced, 4. Separated, 5. Never married)?

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* MARRIED: 1 = Married, 2 = Not married.

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Turned down for credit in the previous 5 years (TURNDOWN: 0 = No, 1 = Yes)

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has declared bankruptcy in the previous 5 years (BNKRUPTLAST5: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has debt (HDEBT: Household has any debt: 0 = No, 1 = Yes). Exercises 15.67–15.71 examine the issue of race among middle-class households (RACE: 1 = White, non-Hispanic, 2 = Black/African American, 3 = Hispanic, 5 = Other). For each variable, test to determine whether there is sufficient evidence to conclude that differences exist between the four races.

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* House ownership (HOUSECL: 1 = Owns, 2 = Does not own).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has been turned down for credit in the previous 5 years (TURNDOWN: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has declared bankruptcy in the previous 5 years (BNKRUPTLAST5: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has incurred debt (HDEBT: Household has any debt: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has at least one late payment in the previous year (LATE: 0 = No, 1 = Yes). Exercise 15.72–15.78 examine whether there are differences between the middle-class men and women with respect to several financial-related variables (HHSEX: 1 = Male, 2 = Female). For each variable, test to determine whether there is enough evidence to conclude that men and women differ.

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Education category of head of household EDCL: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Industry classification for head of household (INDCAT: 1 = Mining 1 construction 1 manufacturing, 2 = Transportation 1 communications 1 utilities and sanitary services 1 wholesale trade 1 finance, insurance and real estate, 4 = Agriculture 1 retail trade 1 services 1 public transportation).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household overall expenses over last 12 months (EXPENSHILO: 1 = Unusually high, 2 = Unusually low, 3 = Normal).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has been turned down for credit in the previous 5 years (TURNDOWN: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has declared bankruptcy in the previous 5 years (BNKRUPTLAST5: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has incurred debt (HDEBT: Household has any debt: 0 = No, 1 = Yes).

Conduct all tests at the 5% significance level. The following exercises are based on the middle-class respondents of the 2013 survey. Exercises 15.63–15.66 address differences in educational attainment (EDCL Education category of head of household: 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree) with respect to several demographic and financial variables. For each variable, test to determine whether there are differences between the levels of education. SCF2013:\MC* Household has at least one late payment in the previous year (LATE: 0 = No, 1 = Yes).

Suppose that a random sample of 100 observations was drawn from a population. After calculating the mean and standard deviation, each observation was standardized and the number of observations in each of the following intervals was counted. Can we infer at the 5% significance level that the data were not drawn from a normal population?

A random sample of 50 observations yielded the following frequencies for the standardized intervals: Can we infer that the data are not normal? (Use \(\alpha\)= .10.) Text Transcription: alpha

The following exercises require the use of a computer and software. Xr12-31 Refer to Exercise 12.31. Test at the 10% significance level to determine whether the amount of time spent working at part-time jobs is normally distributed. If there is evidence of nonnormality, is the t-test invalid?

The following exercises require the use of a computer and software. Xr12-37 The t-test in Exercise 12.37 requires that the costs of prescriptions is normally distributed. Conduct a test with = .05 to determine whether the required condition is unsatisfied. If there is enough evidence to conclude that the requirement is not satisfied, does this indicate that the t-test is invalid?

The following exercises require the use of a computer and software. Xr13-25 Exercise 13.25 required you to conduct a t-test of the difference between two means. Each samples productivity data are required to be normally distributed. Is that required condition violated? Test with \(\alpha\) = .05. Text Transcription: alpha

The following exercises require the use of a computer and software. Xr13-26 Exercise 13.26 asked you to conduct a t-test of the difference between two means (reaction times). Test to determine whether there is enough evidence to infer that the reaction times are not normally distributed. A 5% significance level is judged to be suitable.

The following exercises require the use of a computer and software. Xr13-115 In Exercise 13.115, you performed a test of the mean matched pairs difference. Test with a 10% significance level to determine whether the normality requirement is violated.

Use a 5% significance level, unless specified otherwise. An organization dedicated to ensuring fairness in television game shows is investigating Wheel of Fortune. In this show, three contestants are required to solve puzzles by selecting letters. Each contestant gets to select the first letter and continues selecting until he or she chooses a letter that is not in the hidden word, phrase, or name. The order of contestants is random. However, contestant 1 gets to start game 1, contestant 2 starts game 2, and so on. The contestant who wins the most money is declared the winner and he or she is given an opportunity to win a grand prize. Usually, more than three games are played per show, and as a result it appears that contestant 1 has an advantage: Contestant 1 will start two games, whereas contestant 3 will usually start only one game. To see whether this is the case, a random sample of 30 shows was taken and the starting position of the winning contestant for each show was recorded. These are shown in the following table: Starting position 1 2 3 Number of wins 14 10 6 Do the tabulated results allow us to conclude that the game is unfair?

Use a 5% significance level, unless specified otherwise. It has been estimated that employee absenteeism costs North American companies more than $100 billion per year. As a first step in addressing the rising cost of absenteeism, the personnel department of a large corporation recorded the weekdays during which individuals in a sample of 362 absentees were away over the past several months. Do these data suggest that absenteeism is higher on some days of the week than on others?

Use a 5% significance level, unless specified otherwise. Suppose that the personnel department in Exercise 15.87 continued its investigation by categorizing absentees according to the shift on which they worked, as shown in the accompanying table. Is there sufficient evidence at the 10% significance level of a relationship between the days on which employees are absent and the shift on which the employees work?

Use a 5% significance level, unless specified otherwise. A management behavior analyst has been studying the relationship between male/female supervisory structures in the workplace and the level of employees’ job satisfaction. The results of a recent survey are shown in the accompanying table. Is there sufficient evidence to infer that the level of job satisfaction depends on the boss/employee gender relationship?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-90 Stress is a serious medical problem that costs businesses and government billions of dollars annually. As a result, it is important to determine the causes and possible cures. It would be helpful to know whether the causes are universal or do they vary from country to country. In a survey, American and Canadian adults were asked to report their primary source of stress in their lives. The responses are: 1 = Job, 2 = Finances, 3 = Health 4 = Family life, 5 = Other The data were recorded using these codes plus 1 = American and 2 = Canadian. Do these data provide sufficient evidence to conclude that Americans and Canadians differ in their sources of stress?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-91 More than 3,000 Americans quit smoking each day. Because nicotine is one of the most addictive drugs, quitting smoking is a difficult and frustrating task. It usually takes several tries before success is achieved. There are various methods, including cold turkey, nicotine patch, hypnosis, and group therapy sessions. In an experiment to determine how these methods differ, a random sample of smokers who have decided to quit is selected. Each smoker has chosen one of the methods listed above. After one year, the respondents report whether they have quit (1 = Yes, 2 = No) and which method they used (1 = Cold turkey, 2 = Nicotine patch, 3 = Hypnosis, 4 = Group therapy sessions). Is there sufficient evidence to conclude that the four methods differ in their success?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-92 A newspaper publisher trying to pinpoint his market’s characteristics wondered whether the way people read a newspaper is related to the reader’s educational level. A survey asked adult readers which section of the paper they read first and asked to report their highest educational level. These data were recorded (column 1 = First section read where 1 = Front page, 2 = Sports, 3 = Editorial, and 4 = Other) and column 2 = Educational level where 1 = Did not complete high school, 2 = High school graduate, 3 = University or college graduate, and 4 = Postgraduate degree). What do these data tell the publisher about how educational level affects the way adults read the newspaper?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-93 Every week, the Florida Lottery draws six numbers between 1 and 49. Lottery ticket buyers are naturally interested in whether certain numbers are drawn more frequently than others. To assist players, the Sun-Sentinel publishes the number of times each of the 49 numbers has been drawn in the past 52 weeks. The numbers and the frequency with which each occurred were recorded. a. If the numbers are drawn from a uniform distribution, what is the expected frequency for each number? b. Can we infer that the data were not generated from a uniform distribution?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-94* Clinical depression is a serious disorder that affects millions of people. Depression often leads to alcohol as a means of easing the pain. A Gallup survey attempted to study the relationship between depression and alcohol. A random sample of adults was drawn and after a series of question each respondent was identified as a 1 = Nondrinker, 2 = moderate drinker, 3 = heavy drinker. Additionally, each respondent was asked whether they had ever been diagnosed as clinically depressed at some time in their lives (1 = Yes, 2 = No). Is there enough evidence to conclude that alcohol and depression are related?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Refer to Exercise 15.94. Each respondent was also asked whether they are currently depressed (1 = Yes, 2 = No). Is there sufficient evidence to infer that alcohol and current depression are related?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-96 Gallup asked in a recent survey conducted around the world, “In this country, are you satisfied or dissatisfied with your freedom to choose what you do with your life?” The responses are 1 = Satisfied, 2 = Dissatisfied. The results for 1 = Australia, 2 = Canada, 3 = New Zealand, and the 4 = United States were recorded. Can we infer that there are differences between the four countries in their satisfaction with the freedom in their countries?

The following exercises require the use of a computer and software. The answers may be calculated manually. See Appendix A for the sample statistics. Use a 5% significance level, unless specified otherwise. Xr15-97 In a series of annual surveys of residents of the United States between 2009 and 2013 Gallup asked, “Is corruption widespread throughout the government in this country or not”? The responses (1 = Yes and 2 = No) were recorded. Is there sufficient evidence to conclude that residents’ perceived corruption in the government fluctuated over the 5 year period?

In Section 15-4, we showed how to test for normality. However, we can use the same process to test for any other distribution. Xr15-98 A scientist believes that the gender of a child is a binomial random variable with probability = .5 for a boy and .5 for a girl. To help test her belief, she randomly samples 100 families with five children. She records the number of boys. Can the scientist infer that the number of boys in families with five children is not a binomial random variable with p = .5? (Hint: Find the probability of X = 0, 1, 2, 3, 4, and 5 from a binomial distribution with n = 5 and p = .5.)

In Section 15-4, we showed how to test for normality. However, we can use the same process to test for any other distribution. Xr15-99 Given the high cost of medical care, research that points the way to avoid illness is welcome. Previously performed research tells us that stress affects the immune system. Two scientists at Carnegie Mellon Hospital in Pittsburgh asked 114 healthy adults about their social circles; they were asked to list every group they had contact with at least once every 2 weeks— family, coworkers, neighbors, friends, religious groups, and community groups. Participants also reported negative life events over the past year, events such as death of a friend or relative, divorce, or job-related problems. The participants were divided into four groups: Group 1: Highly social and highly stressed Group 2: Not highly social and highly stressed Group 3: Highly social and not highly stressed Group 4: Not highly social and not highly stressed Each individual was classified in this way. In addition, whether each person contracted a cold over the next 12 weeks was recorded (1 = Cold, 2 = No cold). Can we infer that there are differences between the four groups in terms of contracting a cold?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr12-105* Exercise 12.105 described the problem of a looming shortage of professors, possibly made worse by professors desiring to retire before the age of 65. A survey asked a random sample of professors whether he or she intended to retire before 65. The responses are 1 = No and 2 = Yes. In addition, the survey asked to which faculty did the professor belong (1 = Arts, 2 = Science, 3 = Business, 4 = Engineering, 5 = Other). Do these provide sufficient evidence to infer that whether a professor wishes to retire is related to the faculty?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr12-110* Refer to Exercise 12.110. Determine whether there is enough evidence to infer that there are differences in the choice of Christmas tree between the three age categories.

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr12-111* Exercise 12.111 described a study to determine whether viewers (older than 50) of the network news had contacted their physician to ask about one of the prescription drugs advertised during the newscast. The responses (1 = No, 2 = Yes) were recorded. Also recorded were which of the three networks they normally watch (1 = ABC, 2 = CBS, 3 = NBC). Can we conclude that there are differences in responses between the three network news shows?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr13-164* Exercise 13.164 described a survey of adults wherein, on the basis of several probing questions, each was classified as either a member of the health conscious group (code = 1) or not (code = 2) and whether he or she buys Special X (1 = No, 2 = Yes). Additionally, his or her educational attainment (1 = Did not finish high school, 2 = Finished high school, 3 = Finished college or university, 4 = Postgraduate degree) was recorded. a. Do the data allow the surveyor to conclude that there differences in educational attainment between those who do and those who do not belong to the health-conscious group? b. Can we infer that there is a relationship between the four educational groups and whether or not a person buys Special X?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xm12-05* Example 12.5 described exit polls wherein people are asked whether they voted for the Democrat or Republican candidate for president. The surveyors also record gender (1 = Female, 2 = Male), educational attainment (1 = Did not finish high school, 2 = Completed high school, 3 = Completed college or university, 4 = Postgraduate degree), and income level (1 = Under $25,000, 2 = $25,000 to $49,999, 3 = $50,000 to $75,000, 4 = over $75,000). a. Is there sufficient evidence to infer that voting and gender are related? b. Do the data allow the conclusion that voting and educational level are related? c. Can we infer that voting and income are related?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr02-65* Exercise 2.65 described a survey of the business school graduates undertaken by a university placement office. The respondents reported (among other questions) gender (1 = Female, 2 = Male) and area of employment (1= Accounting, 2 = Finance, 3 = General management, 4 = Marketing/sales, 5 = Other). Can we infer from the data that female and male graduates differ in their areas of employment?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr02-46 Exercise 2.46 asked the question: Are you more likely to smoke if your parents smoke? To shed light on the issue, a sample of 20- to 40-yearold people was asked whether they smoked and whether their parents smoked. The results are stored the following way: Smoke: 1 = Do not smoke, 2 = Smoke and Parent: 1 = Neither parent smoked, 2 = Father smoked, 3 = Mother smoked, 4 = Both parents smoked. Test to determine whether there is enough evidence to infer that parents’ smoking and their children smoking are related.

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr02-33* Exercises 2.33 and 2.49 described a survey that took a random sample of 285 graduating students and asked each to report which of the following is their favorite light beer: 1 =Bud Light, 2 = Busch Light, 3 = Coors Light, 4 = Michelob Light, 5 = Miller Lite, 6 = Natural Light, and 7 = other brands. Also recorded was the gender (1 = Male and 2 = Female). Do these data allow us to conclude that males and females differ in their preference for light beer?

The following exercises employ data files associated with examples and exercises seen earlier in this book. Xr02-64 Exercise 2.64 described the survey the Red Lobster Restaurant chain conducts to monitor the performance of individual restaurants. One of the questions asks customers to rate the overall quality of their last visit. The listed responses are Poor (1), Fair (2), Good (3), Very good (4), and Excellent (5). The survey also asks respondents whether their children accompanied them (1= Yes and 2 = No) to the restaurant. Is there sufficient evidence to infer that ratings and whether children accompanied the customer are related?

Market Segmentation In Section 12-4 and in Chapters 13 and 14, we described how marketing managers use statistical analyses to estimate the size of market segments and to determine whether there are differences between segments. The following exercises require the application of the chi-squared test of a contingency table to determine whether market segments differ with respect to some nominal variable. Xr12-132* Exercise 12.132 described the market segments defined by JC Penney. Another question included in the questionnaire that classified the women surveyed was asked whether each worked outside the home. The responses were: 1. No 2. Part-time job 3. Full-time job These data plus the classifications (1 = Conservative, 2 = Traditional, 3 = Contemporary) were recorded. Can we infer from these data that there are differences in employment status between the three market segments?

Market Segmentation In Section 12-4 and in Chapters 13 and 14, we described how marketing managers use statistical analyses to estimate the size of market segments and to determine whether there are differences between segments. The following exercises require the application of the chi-squared test of a contingency table to determine whether market segments differ with respect to some nominal variable. Xr12-132* Refer to Exercise 12.132. The women in the survey were also asked to define value by identifying what they considered to be the most important attribute of value. The responses are: 1. Price 2. Quality 3. Fashion The responses and the classifications of segments (1 = Conservative, 2 = Traditional, 3 = Contemporary) were recorded. Do these data allow us to infer that there are differences in the definition of value between the three market segments?

Market Segmentation In Section 12-4 and in Chapters 13 and 14, we described how marketing managers use statistical analyses to estimate the size of market segments and to determine whether there are differences between segments. The following exercises require the application of the chi-squared test of a contingency table to determine whether market segments differ with respect to some nominal variable. Xm12-06* Refer to Example 12.6. In segmenting the breakfast cereal market, a food manufacturer uses health and diet consciousness as the segmentation variable. Four segments are developed: 1. Concerned about eating healthy foods 2. Concerned primarily about weight 3. Concerned about health because of illness 4. Unconcerned A survey was undertaken and each person was asked how often they ate a healthy breakfast (defined as cereal and/or fruit). The responses are: 1. Never 2. Seldom 3. Often 4. Always The responses and the market segments of each respondent were recorded. Can we infer that there are differences in frequency of healthy breakfasts between the market segments?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-01 An analysis of the applicants of all MBA programs in North America reveals that the proportions of each type of undergraduate degree are as follows: The director of Wilfrid Laurier University’s (WLU’s) MBA program recorded the undergraduate degree of the applicants for this year using the codes in parentheses. Do these data indicate that applicants to WLU’s MBA program are different in terms of their undergraduate degrees from the population of MBA applicants?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-02 The experiment to determine the effect of taking a preparatory course to improve SAT scores in Exercise A13.16 was criticized by other statisticians. They argued that the first test would provide a valuable learning experience that would produce a higher test score from the second exam even without the preparatory course. Consequently, another experiment was performed. Forty students wrote the SAT without taking any preparatory course. At the next scheduled exam (3 months later), these same students took the exam again (again with no preparatory course). The scores for both exams were recorded in columns 1 (first test scores) and 2 (second test scores). Can we infer that repeating the SAT produces higher exam scores even without the preparatory course?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-03 How does dieting affect the brain? This question was addressed by researchers in Australia. The experiment used 40 middle-age women in Adelaide, Australia; half were on a diet and half were not (National Post, December 1, 2003). The mental arithmetic part of the experiment required the participants to add two three-digit numbers. The amount of time taken to solve the 48 problems was recorded. The participants were given another test that required them to repeat a string of five letters they had been told 10 seconds earlier. They were asked to repeat the test with five words told to them 10 seconds earlier. The data were recorded in the following way: Column 1: Identification number Column 2: 1 = dieting, 2 = not dieting Column 3: Time to solve 48 problems (seconds) Column 4: Repeat string of 5 letters (1 = no, 2 = yes) Column 5: Repeat string of 5 words (1 = no, 2 = yes) Is there sufficient evidence to infer that dieting adversely affects the brain?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-04 A small but important part of a university library’s budget is the amount collected in fines on overdue books. Last year, a library collected $75,652.75 in fine payments; however, the head librarian suspects that some employees are not bothering to collect the fines on overdue books. In an effort to learn more about the situation, she asked a sample of 400 students (out of a total student population of 50,000) how many books they had returned late to the library in the previous 12 months. They were also asked how many days overdue the books had been. The results indicated that the total number of days overdue ranged from 0 to 55 days. The number of days overdue was recorded. a. Estimate with 95% confidence the average number of days overdue for all 50,000 students at the university. b. If the fine is 25 cents per day, estimate the amount that should be collected annually. Should the librarian conclude that not all the fines were collected?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-05 An apple juice manufacturer has developed a new product—a liquid concentrate that produces 1 liter of apple juice when mixed with water. The product has several attractive features. First, it is more convenient than bottled apple juice, which is the way apple juice is currently sold. Second, because the apple juice that is sold in cans is actually made from concentrate, the quality of the new product is at least as high as that of bottled apple juice. Third, the cost of the new product is slightly lower than that of bottled apple juice. The marketing manager has to decide how to market the new product. She can create advertising that emphasizes convenience, quality, or price. To facilitate a decision, she conducts an experiment in three different small cities. In one city, she launches the product with advertising stressing the convenience of the liquid concentrate (e.g., easy to carry from store to home and takes up less room in the freezer). In the second city, the advertisements emphasize the quality of the product (“average” shoppers are depicted discussing how good the apple juice tastes). Advertising that highlights the relatively low cost of the liquid concentrate is used in the third city. The number of packages sold weekly is recorded for the 20 weeks following the beginning of the campaign. The marketing manager wants to know whether differences in sales exist between the three advertising strategies. (We will assume that except for the type of advertising, the three cities are identical.)

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-06 Mutual funds are a popular way of investing in the stock market. A financial analyst wanted to determine the effect income had on ownership of mutual funds and whether the relationship had changed from four years earlier. She took a random sample of adults 25 years of age and older and asked each person whether he or she owned mutual funds (No = 1 and Yes = 2) and to report the annual household income. The categories are 1. Less than $25,000 2. $25,000 to $34,999 3. $35,000 to $49,999 4. $50,000 to $74,999 5. $75,000 to $100,000 6. More than $100,000 Can we infer from the data that household income and ownership of mutual funds are related?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-07 Refer to Exercise A15.5. Suppose that in addition to varying the marketing strategy, the manufacturer also decided to advertise in one of the two media that are available: television and newspapers. As a consequence, the experiment was repeated in the following way. Six different small cities were selected. In city 1, the marketing emphasized convenience, and all the advertising was conducted on television. In city 2, marketing also emphasized convenience, but all the advertising was conducted in the daily newspaper. Quality was emphasized in cities 3 and 4. City 3 learned about the product from television commercials, and city 4 saw newspaper advertising. Price was the marketing emphasis in cities 5 and 6. City 5 saw television commercials, and city 6 saw newspaper advertisements. In each city, the weekly sales for each of 10 weeks were recorded. What conclusions can be drawn from these data?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-08 After a recent study, researchers reported on the effects of folic acid on the occurrence of spina bifida-a birth defect in which there is incomplete formation of the spine. A sample of 2,000 women who gave birth to children with spina bifida and who were planning another pregnancy was recruited. Before attempting to get pregnant again, half the sample was given regular doses of folic acid, and the other half was given a placebo. After 18 months, researchers recorded the result for each woman: 1 = birth to normal baby, 2 = birth to baby with spina bifida, 3 = not pregnant or no baby yet delivered. Can we infer that folic acid reduces the incidence of spina bifida in newborn babies?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-09 Slow play of golfers is a serious problem for golf clubs. Slow play results in fewer rounds of golf and less profits for public course owners. To examine this problem, a random sample of British and American golf courses was selected. The amount of time taken (in minutes) was recorded for a random sample of British and American golfers. Can we conclude that British golfers play golf in less time than do American golfers?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-10 The United States and Canada (among others) are countries in which a significant proportion of citizens are immigrants. Many arrive in North America with few assets but quickly adapt to a changed economic environment. The question often arises, How quickly do immigrants increase their standard of living? A study initiated by Statistics Canada surveyed three different types of families: 1. Immigrants who arrived before 1976 2. Immigrants who came to Canada after 1986 3. Canadian-born families The survey measured family wealth, which includes houses, cars, income, and savings and recorded the results (in $1,000s). Can we infer that differences exist between the three groups? If so, what are those differences?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-11 During the decade of the 1980s, professional baseball thrived in North America. However, in the 1990s attendance dropped, and the number of television viewers also decreased. To examine the popularity of baseball relative to other sports, surveys were performed. In 1985 and again in 1992, a Harris Poll asked a random sample of 500 people to name their favorite sport. The results, which were published in the Wall Street Journal (July 6, 1993), were recorded in the following way: favorite sport (1 = professional football, 2 = baseball, 3 = professional basketball, 4 = college basketball, 5 = college football, 6 = golf, 7 = auto racing, 8 = tennis, and 9 = other); year (1 = 1985, 2 = 1992). Do these results indicate that North Americans changed their favorite sport between 1985 and 1992?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-12 In an attempt to learn more about traffic congestion in a large North American city, the number of cars passing through intersections was determined (National Post, October 18, 2006). The number of cars was counted in 5-minute samples throughout several days. The counts for one busy intersection were recorded. Estimate with 95% confidence the mean number of cars in 5 minutes. Use the result to estimate the counts for a 24-hour day.

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-13 Organizations that sponsor various leisure activities need to know the number of people who wish to participate. Bureaucrats need to know the number because many organizations apply for government grants to pay the costs. The U.S. National Endowment for the Arts conducts surveys of American adults to acquire this type of information. One part of the survey asked a random sample of adults whether they participated in exercise programs. The responses (1 = yes and 2 = no) were recorded. A recent census reveals that there are 205.9 million adults in the United States. Estimate with 95% confidence the number of American adults who participate in exercise programs.

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-14 Low back pain is a common medical problem that sometimes results in disability and absence from work. Any method of treatment that decreases absence would be welcome by individuals and insurance companies. A randomized control study (published in Annals of Internal Medicine, January 2004) was undertaken to determine whether an alternate form of treatment is effective. The study examined 134 workers who were absent from work because of low back pain. Half the sample was assigned to graded activity, a physical exercise program designed to stimulate rapid return to work. The other half was assigned to the usual care, which involves mostly rest. For each worker, the number of days absent from work because of low back pain in the following 6 months was recorded. Do these data provide sufficient evidence to infer that the graded activity is effective?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. XrA15-15 Clinical depression is a serious and sometimes debilitating disease. It is often treated by antidepressants such as Prozac and Zoloft. Recent studies may indicate another possible remedy. Researchers took a random sample of people who are clinically depressed and divided them into three groups. The first group was treated with antidepressants and light therapy, the second was treated with a placebo and light therapy, and the third group treated with a placebo. Whether the patient showed improvement (code = 1) or not (code = 2) and the group number were recorded. Can we infer that there are differences between the three groups?

We remind you that we do not specify significance levels in the exercise that follow. Choose your own. How well do airlines keep to their schedules? To help answer this question, an economist conducted a survey of 780 takeoffs in the United States and determined that 77.4% of them departed on time (defined as a departure that is within 15 minutes of its scheduled time). There were 7,140,596 flight departures in the United States in 2005. Estimate with 95% confidence the total number of on-time departures.

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* It seems obvious that political persuasion affects beliefs about what the government can and cannot do. However, do the data support this belief? Is there enough statistical evidence to infer that differences exist between the three political groups (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican) with respect to the following question, “Should government reduce income differences between rich and poor (EQWLTH: 1 = Government should reduce differences; 2, 3, 4, 5, 6, 7 = No government action)?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Respondents were asked the following question. “Does Earth go around the Sun or does the Sun go around Earth (EARTHSUN: 1 = Earth around Sun, 2 = Sun around Earth, 8 = Don’t know, 9 = No answer)? The correct answer: Earth around Sun. Is there sufficient evidence to infer that Republicans get the correct answer more frequently than do Democrats?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Are Americans optimistic that their children will do better financially than themselves? The survey asked respondents, “When your children are at your age will their standard of living be…. Can we infer that the majority of Americans believe that their children will do better than themselves (KIDSSOL: 1 = Much better, 2 = Somewhat better, 3 = About the same, 4 = Somewhat worse, 5 = Much worse)?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Are married couples postponing bearing children? One way to measure this is to determine how old people are when their first child is born. Estimate with 95% confidence the average age of Americans when their first child is born (AGEKDBRN).

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Estimate the number of women who have never married (SEX: 2 = Female) and MARITAL: 5 = Never married).

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* It is often argued that the reason that men earn more than women is that men work harder and longer. Examine the issue by determining whether there is sufficient evidence to conclude that men work longer hours (HRS1) than women (SEX: 1 = Male, 2 = Female).

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* The survey asked respondents, “Taking all things together, how would you describe your marriage? Is there enough evidence to conclude that more than half of all marriages are very happy (HAPMAR: 1 = Very happy)?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* The survey asked respondents the following question. “Should government improve standard of living of poor people?” The responses were HELPPOOR: 1 = Government act; 2, 3, 4, 5 = People should help themselves. Is there sufficient evidence to conclude that the five categories of educational attainment differ in their responses to the question (DEGREE: 0 = Left high school, 1 = High school, 2 = Junior college, 3 = Bachelor’s degree, 4 = Graduate)?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Did the aftereffects of the recession that began in 2008 end in 2012? One way to judge is to determine how secure the jobs of workers were. Estimate the number of people who believed that it is very likely that they will lose their job or be laid off (JOBLOSE: 1 = Very likely).

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Does the race of an individual affect whether he or she is likely to be self-employed? Can we conclude that differences in whether an individual works for himself or herself (WRKSLF: 1 = Self-employed, 2 = Someone else) exists between the races (RACE: 1 = White, 2 = Black, 3 = Other)?

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Do jobs in the public sector require more education (EDUC) than the ones in the private sector (WRKGOVT: 1 = Government, 2 = Private)? Conduct a test to answer the question.

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* How much education (EDUC) do Democrats, Independents, and Republicans have (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican)? Are they about the same or do they differ? Conduct a test to answer the question.

Conduct all tests at the 5% significance level. Use a 95%confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Do teenage and adult children living with their parents contribute to household income by holding down full- or part-time jobs? And is it more likely that they do so for affluent than for less-affluent families? To answer the question, test to determine whether the data allow us to conclude that there are differences in the number of family members earning money (EARNRS) between the four classes (CLASS: 1 = Lower class, 2 = Working class, 3 = Middle class, 4 = Upper class).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Estimate the number of men (SEX: 1 = Male) who work for themselves (WRKSLF: 1 = Self-employed).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* Capital punishment for murderers exists in most U.S. states. However, a few states ban this form of punishment. Politicians often need to know which members of the public support and which oppose. Can we conclude from the data that there is a difference between Democrats, Republicans, and Independents (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican) in terms of support for capital punishment (CAPPUN)?

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. In 2012, there were 221,963,000 Americans aged 21 years or more of whom 115,219,000 were female and 106,744,000 were male. GSS2012* In Chapter 2, we used a graphical technique and data from the American National Election Survey to attempt to determine whether men and women differ in their political affiliation. Use a suitable statistical inference technique to determine whether there is sufficient evidence to infer that men and women (SEX: 1 = Male, 2 = Female) differ in their political affiliations (PARTYID3: 1 = Democrat, 2 = Independent, 3 = Republican).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. CF2010:\MC* Does home ownership (HOUSECL: 1 = Own, 2 = Not) affect how much is spent on food away from home (FOODAWAY)? Perform a statistical analysis to determine whether there is enough evidence to infer that middle-class home owners spend less on food away than do middleclass households who rent.

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\UC The upper class in the 2010 survey had household net worth between $1,345,975 and $7,402,095. In 2010, the interest paid by money market deposit and money market mutual funds paid very low interest. Estimate the mean invested in these accounts by households in the upper class (MMA).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\MC* Who goes deeper into debt (DEBT) among heads of middle-class households, men or women (HHSEX: 1 = Male, 2 = Female)? Conduct a test to determine whether there is enough evidence to conclude that male heads of households go deeper into debt.

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\UC* One way to judge how well upperclass households are doing this year as opposed to other years is to determine what their normal income is. Estimate the normal income for upper-class households (NORMINC).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* Do more educated heads of households have less debt? Conduct a test to determine whether there are differences in household debt (DEBT) between the four categories of education (EDCL 1 = No high school diploma, 2 = High school diploma, 3 = Some college, 4 = College degree.).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* The nuclear family is defined as a pair of adults and their children. Is this going out of style? Estimate the proportion of nuclear family households (FAMSTRUCT: 4 = Married or living with partner + children).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* One survey question asked in which industry was the head of the household employed in. One of the categories was mining, construction, or manufacturing, all of which may be in decline. Estimate the number of people employed in one of these industries (INDCAT: 1).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* Between student loans, credit card balances, vehicle loans, and mortgages almost all household have some debt. Estimate the number of Americans with no household debt (HDEBT: 0).

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* Do banks and financial institutions turn down female heads (HHSEX: 1 = Male, 2 = Female) of households more frequently than male heads of households’ application for credit (TURNDOWN: 1 = Turned down)? Use an appropriate statistical method to determine whether there is enough evidence to answer the question affirmatively.

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* One of the questions asked respondents whether their household had been late with any debt payment in the previous year (LATE: 0 = No, 1 = Yes). Is there enough evidence to conclude that male heads of households are less likely than female heads (HHSEX: 1 = Male, 2 = Female) to be late with a debt payment?

Conduct all tests at the 5% significance level. Use a 95% confidence level for estimates. According to the U.S. Census, there were 220,958,853 adults in the United States in 2010. SCF2010:\All* Does one’s occupation affect whether the household has been turned down for credit in the previous 5 years (TURNDOWN: 1 = Yes, 2 = No)? Is there enough evidence to conclude that there are differences between the three working categories (OCCAT2: 1 = Managerial/professional, 2= Technical/sales/services, 3 = Other)?