stat 2220 practice tt1 uofm

Upload: favelax895

Post on 11-Feb-2018

235 views

Category:

Documents


1 download

TRANSCRIPT

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    1/21

    STAT 2220 - Practice Term Test 1

    1. The stemplot below displays the number of points scored by the Toronto Raptors foreach of their 41 home games in the 2012/2013 NBA season.

    8 1 2 3 4 5 8 8 8 89 0 1 2 2 3 3 4 5 5 6 6 7 7 7 8 8 8 9

    10 0 0 1 2 2 2 3 5 5 6 8 911 41213 3

    What is the interquartile range for the number of points scored by the Raptors in thesegames?

    (A) 10 (B) 10.5 (C) 11 (D) 11.5 (E) 12

    2. The value (in $) of the last ten receipts at a gas station are shown below, with the lowestvalues (receipts below $10) denoted as L:

    30 L 20 56 35 L 15 70 23 50

    What is the value of the first quartile for these data?

    (A) 15 (B) 20 (C) 21.5 (D) 15 + L

    2 (E)

    20 + L

    2

    3. Consider the following variables:

    Area code (Vancouver 604, Edmonton 780, Calgary 403, etc.)

    Weight class of a professional boxer (lightweight, middleweight, heavyweight, etc.)

    Office number of a Statistics professor in Machray Hall

    These three variables are, respectively:

    (A) categorical and ordinal, categorical and ordinal, categorical and nominal

    (B) categorical and nominal, categorical and ordinal, categorical and nominal

    (C) categorical and nominal, categorical and nominal, categorical and ordinal

    (D) categorical and ordinal, categorical and nominal, quantitative

    (E) categorical and nominal, categorical and ordinal, categorical and ordinal

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    2/21

    The next twoquestions (4 and 5) refer to the following:

    The sport of boxing divides its athletes into different weight classes in order to make thecompetition more fair. The side-by-side quantile box plots shown below display the weights(in pounds) of a random sample of 50 Cruiserweight boxers and 49 Heavyweight boxers.

    320

    300

    280

    260

    240

    220

    200

    Cruiserweight Heavyweight

    The five number summaries for the two weight classes are shown below:

    Cruiserweight: 204 220 226 230 250Heavyweight: 270 295 304 312 320

    4. Which of the following statements is/are true?

    I. Any Cruiserweight with a weight less than 205 pounds or greater than 245 pounds

    would be considered an outlier.

    II. There are about 37 Cruiserweights in the sample who weigh more than 220 pounds.

    III. The distribution of weights for the Heavyweights is skewed to the right.

    (A) I (B) II (C) I and II (D) II and III (E) I, II and III

    Page 2

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    3/21

    5. What is the median of all 99 boxers in the sample (Cruiserweights and Heavyweights)combined?

    (A) 250 (B) 265 (C) 270 (D) 285 (E) 260

    The next twoquestions (6 and 7) refer to the following:

    The frequency distribution shown below displays the daily saturated fat intake (in grams)for a sample of people.

    Saturated Fat Intake Frequency10 20 1820 30 1930 40 2040 50 850 60 4

    60 70 370 80 1

    6. The distribution of saturated fat intake is:

    (A) skewed to the left and so the mean is greater than the median.

    (B) skewed to the right and so the mean is greater than the median.

    (C) skewed to the left and so the median is greater than the mean.

    (D) skewed to the right and so the median is greater than the mean.(E) approximately symmetric and so the mean and median are approximately equal.

    7. Which interval contains the median saturated fat intake for the sample?

    (A) 20 30

    (B) 30 40

    (C) 40 50

    (D) 50 60

    (E) impossible to determine with the information given.

    Page 3

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    4/21

    The next twoquestions (8 and 9) refer to the following:

    The average number of goals per game for the 30 National Hockey League teams andthe 29 American Hockey League teams for the 2009/10 season are displayed in the side-by-side basic (quantile) box plots shown below:

    NHL AHL

    4.5

    4.0

    3.5

    3.0

    2.5

    2.0

    Some summary statistics for the two distributions are shown below:

    min Q1 Me Q3 max mean std. dev.NHL 2.51 2.65 2.74 2.90 3.88 2.85 0.29AHL 2.41 2.60 2.93 3.05 4.28 2.92 0.38

    8. Which of the following statements is/are true?

    I. The interquartile range for the AHL is greater than that for the NHL.

    II. The distribution of goals per game for NHL teams is skewed to the right.III. About 7 AHL teams scored more than 3.05 goals per game.

    (A) I (B) I and II (C) I and III (D) II and III (E) I, II and III

    Page 4

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    5/21

    9. The NHL consists of 24 teams from the US and 6 Canadian teams. The mean numberof goals per game for Canadian teams was 2.74. What was the mean number of goalsfor the US teams?

    (A) 2.88 (B) 2.90 (C) 2.92 (D) 2.94 (E) 2.96

    10. A father drops his three children off at the movie theatre and gives them each $25 tospend. After each of the children buys a movie ticket for $9, what is the standarddeviation of the amounts of money the children have left?

    (A) $3.00 (B) $2.12 (C) $4.00 (D) $2.83 (E) $0.00

    11. Consider the following data set:

    3 4 9 0 7

    What is the standard deviation for this data set?

    (A) 4.69 (B) 27.5 (C) 3.36 (D) 22.0 (E) 5.24

    12. Which of the following sample measures is resistant to the effect of outliers?

    (A) mean

    (B) range(C) first quartile

    (D) standard deviation

    (E) correlation

    13. We would like to make a histogram (with vertical bars) of the number of vehicle theftslast year in all North American cities with population over one million people. Thehorizontal and vertical axes represent, respectively:

    (A) city and number of thefts.

    (B) population and frequency.

    (C) population and number of thefts.

    (D) number of thefts and frequency.

    (E) number of thefts and population.

    Page 5

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    6/21

    14. A small graduate class of three students writes a math test. The student who finishedwriting the fastest got the highest score in the class. The student who finished secondgot the second highest score, and the student who took the longest to write the test gotthe lowest score. Let Xbe the time it takes for a student to write the test and let Y bethe students test score. What can be said about the correlationrbetweenXandY for

    this class?

    (A) There is a perfect negative linear relationship betweenX and Y, and so r= 1.

    (B) The correlation betweenX and Yis negative, but not necessarily equal to 1.

    (C) There is no linear relationship betweenX and Y, and so r= 0.

    (D) There is perfect positive linear relationship betweenX and Y, and so r= 1.

    (E) The correlation betweenX and Y is positive, but no necessarily equal to 1.

    15. Which of the following statements about the least squares regression line is (are) true?

    I. The slope of the least squares regression line always has the same sign as thecorrelation.

    II. The least squares regression line is the line that minimizes the sum of residuals.

    III. The least squares regression line is the line that maximizes the value of the corre-lation.

    (A) I (B) II (C) I and II (D) I and III (E) I, II and III

    16. We want to calculate the correlation r between two variables X and Y. Which of thefollowing conditions are necessary for r to be a meaningful measure of association?

    I. X and Yare both quantitative variables.

    II. The relationship betweenX and Y is linear.

    III. Xis an explanatory variable and Yis a response variable.

    (A) I (B) II (C) I and II (D) I and III (E) I, II and III

    17. Three friends took the same university course. Each of them scored 25% lower on thefinal exam than they did on the midterm test (i.e., each students mark was only 75%of their midterm mark). What is the correlation between midterm score and final scorefor these three friends?

    (A) 0.25 (B) 1 (C) 0.75 (D) 1 (E) 0.25

    Page 6

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    7/21

    18. Can a students midterm score be used to predict his final exam score? The midtermscoresX(out of 50) and the final exam scoresY(out of 100) are measured on a sample ofstudents in a large class. The least squares regression line is calculated to be y= 8+1.8x.What would be the equation of the least squares regression line if we had instead recordedthe students midterm score as a percentage (i.e., out of 100)?

    (A) y= 8 + 0.9x

    (B) y= 16 + 3.6x

    (C) y= 8 + 1.8x

    (D) y= 16 + 0.9x

    (E) y= 8 + 3.6x

    19. Data are collected for some explanatory variableXand some response variable Yand the

    correlationr and the least-squares regression line are calculated. Which of the followingstatements is false?

    (A) The correlation tells us the direction of the linear relationship betweenX and Y.

    (B) The correlation measures the strength of the linear relationship betweenXand Y.

    (C) The slope of the regression line tells us the direction of the linear relationship be-tween X and Y.

    (D) The slope of the regression line measures the strength of the linear relationshipbetween Xand Y.

    (E) If the correlation is equal to 0, then the slope must be equal to 0.

    Page 7

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    8/21

    The next fourquestions (20 to 23) refer to the following:

    We would like to know if the amount of sodium can be predicted by the amount of sugarin breakfast cereals. A random sample of 20 popular breakfast cereals was collected andthe equation of the regression line is calculated to be y= 0.2440.007x.

    20. What is the correct interpretation of the slope of the least squares regression line?

    (A) When the sugar content increase by 1 gram, we predict sodium content will increaseby 0.007 grams.

    (B) When the sodium content increase by 1 gram, we predict sugar content will decreaseby 0.007.

    (C) When the sugar content increases by 0.007 grams, we predict sodium content willdecrease by 1 gram.

    (D) When the sugar content increases by 1 gram, we predict sodium content will decreaseby 0.007 grams.

    (E) When the sodium content increases by 0.007 grams, we predict sugar content willdecrease by 1 gram.

    21. Fruit loops cereal has a sodium content of 0.13 grams and a sugar content of 13 grams.What is the value of the residual for Fruit Loops?

    (A) 0.153 (B) 0.023 (C) 0.086 (D) 0.153 (E) 0.023

    22. From the data, we calculate that 18.8% of the variation in sodium content can be ac-counted for by its regression on sugar content. What is the value of correlation r?

    (A) 0.188 (B) 0.434 (C) 0.035 (D) 0.188 (E) 0.434

    Page 8

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    9/21

    23. Suppose we measured both sugar content and sodium content in milligrams instead ofgrams (1 g = 1000 mg) and we recalculated the regression line and the correlation.Which of the following statements is true?

    (A) The intercept would change but the slope and correlation would remain the same.

    (B) The slope would change but the intercept and the correlation would remain thesame.

    (C) The intercept and the slope would change but the correlation would remain thesame.

    (D) The slope and the correlation would change but the intercept would remain thesame.

    (E) The intercept, the slope and the correlation would all remain the same.

    24. A local research firm often uses newspaper advertisements to invite individuals to vol-unteer for various experiments the firm is planning to conduct. One such study was todetermine if a newly formulated cream had any beneficial effect on individuals sufferingwith psoriasis (a skin condition). Psoriasis patients were invited to participate in a studywith volunteers to be given either the new cream of the most popular cream currentlyon the market, with similar appearance and feel as the new cream. What conclusionscan be drawn from such a study?

    (A) Useful information can be gained from studies such as this, provided that treatmentsare randomly assigned to the individuals.

    (B) Because there is bias associated with the selection of subjects, nothing can be con-cluded.

    (C) In such a study there is great potential for the presence of confounding or lurkingvariables, so little information can be gained.

    (D) Little if anything can be concluded because this is an observational study.

    (E) Because no placebo group is included, little will be learned from such studies.

    Page 9

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    10/21

    25. Which of the following studies uses a matched pairs design?

    I. Which cola tastes better Pepsi or Coke? In a blind taste test, volunteers wereasked to taste each of the colas (in random order) and to score the taste of eachof them out of 10. The colas were compared by tabulating the differences in scores

    for each volunteer.II. Do students perform better on multiple-choice of long-answer exams? A math

    professor teaches two sections of an introductory course. He randomly chooses oneof the sections to write a multiple-choice exam and the other to write a long-answerexam. Scores for the two sections will be compared.

    III. Is it easier to remember words or images? Ten set of twins are available to partici-pate in a study. One twin is randomly assigned to study a list of 50 words and theother studies a page with 50 images. After 30 minutes, they are asked to list asmany words or images as they can remember. The number of correct answers willbe compared for each set of twins.

    (A) I (B) I and II (C) II and III (D) I and III (E) I, II and III

    26. An amateur gardener planted a different variety of tomatoes this year than he did lastyear. He put in six plants the previous year and six plants this year using the same partof the garden. The average yield was 11.3 pounds per plant in the previous year and14.5 pounds per plant using the new variety this year. This is an example of:

    (A) a completely randomized design.

    (B) a multistage design, because two years were involved.

    (C) a matched pairs design.

    (D) an observational study.

    (E) a randomized block design with two blocks.

    Page 10

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    11/21

    27. An experiment is being conducted to study the effectiveness of different brands of sun-screen and SPF (sun protection factor) levels. Volunteers will be randomly assigned toapply either Coppertone or Ombrelle sunscreen, with an SPF level of either 30 or 60.Subjects will spend eight hours outside in the sun, and the degree of sunburn (if any)will be compared for all treatments. What is/are the factor(s) in this experiment?

    (A) degree of sunburn

    (B) sunscreen brand and SPF level

    (C) Coppertone, Ombrelle, SPF 30, SPF 60

    (D) Coppertone/SPF 30, Ombrelle/SPF 30, Coppertone/SPF 60, Ombrelle/SPF 60

    (E) time spent outside

    28. An experiment is being conducted to compare the effects of five different diet plans on

    the weight loss of overweight adults. Fifty men and fifty women volunteer to participatein the experiment. It is believed that males and females may respond differently to thevarious treatments, and so a randomized block design is used. The experiment shoulduse:

    (A) ten blocks five with 10 males and fife with 10 females.

    (B) ten blocks each with 5 males and 5 females.

    (C) five blocks each with 10 males and 10 females.

    (D) two blocks one with all 50 males and one with all 50 females.

    (E) two blocks each with 25 males and 25 females.

    29. How is randomization used in a matched pairs experiment?

    (A) to select pairs to participate in the study

    (B) to select which two treatments will be examined in the study

    (C) to place individuals in pairs

    (D) to assign each of the two treatments within each pair

    (E) all of the above

    Page 11

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    12/21

    The next twoquestion (30 and 31) refer to the following:

    An experiment is being conduct to examine the effectiveness of a new migraine headachemedication. One hundred patients suffering from migraine headaches volunteer to par-ticipate in the experiment. A doctor places all patients names in a hat and selects 50of them to receive a pill containing the actual medication and the other 50 to receivea sugar pill (one with no medication) that looks the same as the pill received by thefirst group. The doctor will give the pill to nurse (who will not know whether it is theactual medication or a sugar pill), who will then give it to the patient (who will not betold which group they are in). The reported pain relief for the two groups will then becompared.

    30. How many of the following are used in this experiment?

    randomization

    a placebo

    repition

    double blinding

    a control group

    (A) 1 (B) 2 (C) 3 (D) 4 (E) 5

    31. Suppose that most of the patients receiving the actual medication report significant pain

    relief, while most of the patients receiving the sugar pill report no relief. Can we concludethat the medication was likely the cause of the pain relief?

    (A) No, because we dont know how bad each patients migraines were before the study.

    (B) No, because the difference in response is likely due to the placebo effect.

    (C) No, because the patients participating in the sudsy were not selected randomly.

    (D) No, because not all patients in the medication group reported significant pain relief.

    (E) Yes, because this was a properly designed experiment.

    Page 12

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    13/21

    32. A doctor would like to examine how smoking cigarettes and drinking alcohol affecta persons weight. During routine physical examinations, the doctor asks her patientswhether they smoke and drink. She examines the files of 50 patients who smoke cigarettesand drink alcohol and records their weights. She does the same for 50 patients whoneither smoke nor drink. She calculates that the average weight of those who smoke and

    drink is significantly higher than those who dont. Which of the following statements ittrue?

    (A) It is clear that at least on of the behaviours (smoking or drinking) causes weightgain.

    (B) It is clear that both smoking and drinking cause weight gain.

    (C) Little if anything can be concluded because the subjects were not randomly selectedto participate in the study.

    (D) Little if anything can be concluded because the subjects did not volunteer to par-ticipate in the study.

    (E) Little if anything can be concluded because this is an observational study and theeffect of smoking and drinking on weight gain are confounded.

    33. A new formula for a cough syrup is being tested. It is given to 500 people suffering fromcolds. Within four days, 450 of the patients report that their cold is gone. Which of thefollowing statements is true?

    (A) The new cold medication has been shown to be effective, as 90% of patients colds

    went away.(B) The cold medication has not been shown to be effective, because not all patients

    colds went away.

    (C) The cold medication has not been shown to be effective, because most patients whogot better probably did so because of the placebo effective.

    (D) The cold medication has not been shown to be effective because the experiment wasnot conducted properly. A control group should have been included, consisting ofpatients with colds who were given a placebo.

    (E) The cold medication has not been shown to be effective because the experiment wasnot conducted properly. A control group should have been included, consisting ofpatients without colds who were given the same cough syrup.

    Page 13

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    14/21

    The next twoquestion (34 and 35) refer to the following:

    Former Manitoba Premier Gary Doer was recently appointed as Canadian Ambassadorto the United States. His office commissions a telephone survey in order to study therelations between the people of the two countries. A simple random sample of 20 peopleis contacted from each of the 50 American states. Respondents are asked whether theyhave a favourable opinion of Canadians.

    34. The population of interest in this study is:

    (A) all Canadians

    (B) all Americans

    (C) all Canadians and all Americans

    (D) all Americans with a favourable opinion of Canadians

    (E) the 50 American states

    35. The resulting sample of 100 Americans is a:

    (A) stratified sample.

    (B) multistage sample.

    (C) simple random sample.

    (D) randomized block sample.

    (E) systematic sample.

    36. A simple random sample of sizen is the only type of sample that guarantees that:

    (A) every individual in the population has a known chance of being selected into thesample.

    (B) every individual in the population has an equal chance of being selected into thesample.

    (C) every group ofnindividuals has an equal chance of being selected into the sample.

    (D) results will not be biased.

    (E) all of the above.

    Page 14

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    15/21

    37. A stratified random sample is conducted by:

    (A) grouping the subjects using some criteria, then randomly selecting subjects fromeach group to interview.

    (B) randomly selecting subjects to interview, then grouping the subjects using somecriteria.

    (C) grouping the subjects using some criteria, then randomly selecting entire groups tointerview.

    (D) randomly selecting subjects to interview, then randomly selecting groups to inter-view.

    (E) randomly grouping the subjects, then randomly selecting the subjects from eachgroup to interview.

    38. Consider the following three situations:

    I. Jim phoned a number he saw on the 6:00 news to respond to a survey question.

    II. Mary hung up on the phone when she heard the person calling was from a pollingfirm.

    III. Matthew had no chance of begin selected in the phone survey because he doesnthave a phone.

    The types of bias present in these three examples are, respectively:

    (A) voluntary response, undercoverage, nonresponse(B) convenience sampling, nonresponse, undercoverage

    (C) voluntary response, nonresponse, undercoverage

    (D) convenience sampling, undercoverage, undercoverage

    (E) voluntary response, nonresponse, nonresponse

    Page 15

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    16/21

    39. Which of the following statements is false?

    (A) A voluntary response sample is not reliable, as individuals who feel strongly aboutthe issue are overrepresented.

    (B) Undercoverage occurs when some individuals in the population have no chance ofbeing selected into the sample.

    (C) In multistage sampling, individuals in the population dont always have the samechange to be selected into the sample.

    (D) Nonresponse is a frequent source of bias in telephone surveys.

    (E) A convenience sample is reliable as long as the sample size is large.

    40. In a particular election, 40% of voters voted for the NDP, 35% voted for the Liberalsand 25% voted for the Conservatives. If we take a random sample of two voters, what

    is the probability that they voted for different parties?

    (A) 0.345 (B) 0.465 (C) 0.575 (D) 0.655 (E) 0.715

    41. Event A has a probability of 0.2 of occurring. Event B has a probability of 0.5 ofoccurring. IfAand B are mutually exclusive, then P(Aand B) is equal to:

    (A) 0.0 (B) 0.1 (C) 0.2 (D) 0.6 (E) 0.7

    42. A hockey player scores a goal in 43% of his games. His team wins 65% of their games.In 31% of the teams games, the player scores a goal and the team wins. What is theprobability that the team wins if the player does not score a goal?

    (A) 0.3595 (B) 0.4625 (C) 0.4985 (D) 0.5965 (E) 0.6345

    43. You roll three fair dice. What is the probability that the total number of points facingup is at least 17?

    (A) 1216

    (B) 2216

    (C) 3216

    (D) 4216

    (E) 5216

    Page 16

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    17/21

    The next fourquestions (44 to 47) refer to the following:

    The manager of a gas station and convenience store compiles the following information:

    60% of customers purchase gasoline (G).

    35% of customers purchase a drink (D).

    37% of customers purchase cigarettes (C).

    21% of customers purchase gas and a drink.

    23% of customers purchase gas and cigarettes.

    16% of customers purchase a drink and cigarettes.

    9% of customers purchase all three items.

    44. What is the probability that a customer purchases a drink or cigarettes?

    (A) 0.59 (B) 0.13 (C) 0.72 (D) 0.94 (E) 0.56

    45. Which of the following statements is true?

    (A) G and D are independent.

    (B) G and C are mutually exclusive.

    (C) D and C are independent.

    (D) G and D are mutually exclusive.(E) G and C are independent.

    46. What is P(Gc C), i.e. the probability that a customer buys cigarettes but not gaso-line?

    (A) 0.14 (B) 0.19 (C) 0.15 (D) 0.23 (E) 0.17

    47. If a customer buys a drink, what is the probability that he or she also buys cigarettes?

    (A) 0.4571 (B) 0.4324 (C) 0.4987 (D) 0.4162 (E) 0.4298

    Page 17

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    18/21

    48. Answer each of the following questions:

    (a) If a variableY can be written as a linear function of a variableX, i.e., ify = a+bx,then show that y= a + bx and sy =|b|sx.

    (b) A phone company sells prepaid long distance phone cards. For each call, the cus-

    tomer is charged a base fee of $1.49, plus $0.25 per minute. The mean time fora sample of calls is calculated to be 12 minutes, and the standard deviation is 7minutes. What are the mean and standard deviation of charges for this sample ofcalls?

    49. The timeYit takes (in seconds) for a chemical reaction to take place depends on (amongother things) the temperature X(in C) to which the chemical solution is exposed. Tentrials are conducted at various temperatures and the reaction time is recorded for eachtrial. A scatterplot reveals that a linear relationship is a reasonable assumption. Theleast squares regression line is calculated to be y= 87.20.5x. It is also observed that90% of the variation in time is accounted for by its regression on temperature.

    (a) What is the value of the correlation between temperature and time?

    (b) Interpret the meaning of the slope of the least squares regression line.

    (c) For one of the trials, a temperature of 70Cwas used, and a reaction time of 54seconds was observed. What is the value of the residual for this trial?

    50. We measure the outdoor temperatures (in C) of a random sample of 18 summer daysin both Winnipeg and Calgary. The five-number summaries for the two cities are shown

    below:

    City min Q1 Me Q3 maxWinnipeg 15 24 26 28 34Calgary 12 19 24 27 30

    (a) Do you expect the mean temperature for Calgary to be less than or greater thanthe median? Why?

    (b) The temperatures for Winnipeg are shown below:

    15 16 19 23 24 24 24 25 26

    26 27 27 27 28 29 31 32 34Create an outlier box plot for this data set. What is the shape of the distributionif we exclude outliers from consideration?

    Page 18

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    19/21

    51. The owner of an ice cream truck would like to know if the temperature affects the amountof ice cream he sells. The temperature (in C) and ice cream sales (in $) for a sample ofsix days from the past summer are recorded. The sample means and standard deviationsare shown below:

    mean std.devTemperature 24.00 4.86

    Sales 310.00 123.98

    A scatterplot of the data reveals that a linear relationship between Temperature andSales is an appropriate assumption. The equation of the least squares regression line iscalculated to be y= 231.92 + 22.58x.

    (a) Find the value of the correlation between Temperature and Sales.

    (b) Interpret the slope of the least squares regression line in this example.

    (c) One day, the temperature was 27Cand the ice cream sales were $350. Find the

    value of the residual for this day. What does the sign of the residual tell us?

    (d) What percentage of the variation in Sales can be accounted for by its regression onTemperature?

    52. Gordon and his wife Michelle are avid gardeners who take pride in maintaining a healthygreen lawn. Over the course of the last summer, Gordon was in charge of maintainingthe grass in the back yard and Michelle cared for the lawn in the front yard. Gordon usedone brand of fertilizer (Brand A) and watered the grass once a week. Michelle used adifferent fertilizer (Brand B) and watered the grass twice a week. At the end of summer,it was apparent that the grass in the front yard looked much greener and healthier thanthe grass in the back yard. Michelle argues that this is clear evidence that Fertilizer B isbetter than Fertilizer A and that watering the grass twice a week is better than wateringit just once a week. There are many flaws in Michelles reasoning. One problem withthis conclusion is that the front yard gets more sunlight than the back yard.

    (a) Explain some other problems with Michelles conclusion.

    (b) You are asked to help Gordon and Michelle conduct an experiment together nextsummer to determine the optimum growing conditions for their grass. (The couple iswilling to sacrifice the appearance of their lawn for one summer in order to conducta proper experiment).

    (i) What type of experimental design would you propose?

    (ii) Identify the factors, factor levels, treatments and response variable.

    (c) Explain to the couple how the experiment should be conducted. Be sure to explainthe role of randomization in your experiment.

    Page 19

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    20/21

    53. An automobile manufacturer would like to examine the effect of speed and type of tireon the stopping distance of its vehicles in winter driving conditions. Vehicles will beoutfitted with either regular tires or snow tires, and will drive at a speed of either 40,60 or 80 km/h. The stopping distances will then be compared. Each combination offactor levels will be tested on three vehicles. The manufacturer believes that the effect

    of speed and type of tire may differ for its different models of car, so the experiment isconducted separately for each model of car they make. Identify each of the following inthis experiment:

    (a) Factor(s)

    (b) Treatment(s)

    (c) Response Variable(s)

    (d) Blocking Variable(s)

    54. The Winnipeg Symphony Orchestra is performing one night at the Centennial ConcertHall. A random sample of 22 audience members is selected. Their ages are ordered andare shown below:

    19 28 33 36 42 45 48 50 52 52 5355 58 59 60 61 61 65 68 70 72 79

    The five-number summary for this data set is:

    19 45 54 61 79

    (a) Construct a modified (outlier) boxplot for this data set. What is the shape of thedistribution of ages?

    (b) The same evening, Elton John is performing at MTS Centre. A random sample of17 audience members is selected. Their ages are recorded and shown below:

    17 18 23 25 29 30 33 37 37 41 42 46 49 50 54 54 62

    Create a back-to-back stem plot of the ages of audience members at the symphonyand the Elton John concert. Would it be more appropriate to use the five-numbersummary or the mean and standard deviation to summarize the distribution of agesat the Elton John concert? Explain.

    Page 20

  • 7/23/2019 STAT 2220 Practice TT1 UofM

    21/21

    STAT 2220 Practice Term Test 1 - Solutions

    Question Answer Question Answer

    1 D 26 D2 A 27 B

    3 E 28 D4 C 29 D5 A 30 E6 B 31 E7 A 32 E8 E 33 D9 A 34 B

    10 E 35 A11 E 36 C12 C 37 A13 D 38 C

    14 B 39 E15 A 40 D16 C 41 A17 D 42 D18 A 43 D19 D 44 E20 D 45 A21 E 46 A22 E 47 A23 A

    24 A25 D

    Page 21