Solutions to End-of-Section and Chapter Review Problems

75

CHAPTER 3

3.1

(a) (b) (c)

Mean = 6 Median = 7 There is no mode. Range = 7 Variance = 8.5 Interquartile range = 5.5 Standard deviation = 2.915 Coefficient of variation = (2.915/6)100% = 48.59% Since the mean is less than the median, the distribution is left-skewed. Mean = 7 Median = 7 Mode = 7 Range = 9 Variance = 10.8 Interquartile range = 5 Standard deviation = 3.286 Coefficient of variation = (3.286/7)100% = 46.94% Since the mean equals the median, the distribution is symmetrical. Mean = 6 Median = 7 Mode = 7 Range = 12 Variance = 16 Interquartile range = 6 Standard deviation = 4 Coefficient of variation = (4/6)100% = 66.67% Since the mean is less than the median, the distribution is left-skewed. Mean = 2 Median = 7 Mode = 7 Range = 17 Variance = 62 Interquartile range = 14.5 Standard deviation = 7.874 Coefficient of variation = (7.874/2)100% = 393.7% Since the mean is less than the median, the distribution is left-skewed. Set 1 Set 2 Mean 4 14 Median 3 13 Mode 2 12 The data values in Set 2 are each 10 more than the corresponding values in Set 1. The measures of central tendency for Set 2 are all 10 more than the comparable statistics for Set 1. Set 1 Set 2 Range 8 8 Interquartile range 3 3 Variance 8.33* 8.33* Standard deviation 2.89* 2.89* Coefficient of variation 72.17% 20.62% *Note: Slight differences are due to rounding. Since the mean is greater than the median for each data set, the distributions are both rightskewed. Because the data values in Set 2 are each 10 more than the corresponding values in Set 1, the measures of spread among the data values remain the same across the two sets, with the exception of the coefficient of variation. The coefficients of variation are different because the sample standard deviation is divided by the sets mean; in the case of Set 2, the mean is 10 more than the mean for Set 1, resulting in a larger denominator and a smaller coefficient. Set 2 is a reflection of Set 1 simply shifted up the scale 10 units, so the distributions are also reflections of each other.

3.2

(a) (b) (c)

3.3

(a) (b) (c)

3.4

(a) (b) (c)

3.5

(a)

(b)-(c) (d)

(e) (f)

76

Chapter 3: Numerical Descriptive Measures Generally stated, when a second data set is an additive shift from an original set, the measures of central tendency for the second set are equal to the comparable measures for the original set plus the value, or distance, of the shift; the measures of spread for the second set are equal to the corresponding measures for the original set, with the exception of the coefficient of variation; the shape of the second distribution will be a reflection of the shape of the original distribution.
1/ 2

3.5 (g) cont.

3.6 3.7

RG = ( 1 + 0.1) ( 1 + 0.3)
(a)

1 = 19.58%

(b)

(c)

Grade X Grade Y Mean 575 575.4 Median 575 575 Standard deviation 6.40 2.07 If quality is measured by the average inner diameter, Grade X tires provide slightly better quality because Xs mean and median are both equal to the expected value, 575 mm. If, however, quality is measured by consistency, Grade Y provides better quality because, even though Ys mean is only slightly larger than the mean for Grade X, Ys standard deviation is much smaller. The range in values for Grade Y is 5 mm compared to the range in values for Grade X which is 16 mm. Grade X Grade Y, Altered Mean 575 577.4 Median 575 575 Standard deviation 6.40 6.11 In the event the fifth Y tire measures 588 mm rather than 578 mm, Ys average inner diameter becomes 577.4 mm, which is larger than Xs average inner diameter, and Ys standard deviation swells from 2.07 mm to 6.11 mm. In this case, Xs tires are providing better quality in terms of the average inner diameter with only slightly more variation among the tires than Ys. The arithmetic mean will be larger than the median. The distribution of new home prices will very likely be right-skewed. The relatively few very expensive new home prices in the upper tail of the positively skewed distribution will heavily influence the mean price and, hence, results in overstating the central tendency of new home prices.

3.8

(a) (b) (c)

3.9

(a)

Mean = 16.43 median = 13 mode = 11 range = 47 variance = 113.36 standard deviation = 10.65 (b) . The average P/E ratio for the 30 stocks traded on the New York Stock Exchange is 16.43. The P/E ratio that occurs most is 11 while the middle ranked P/E ratio is 13. The average of the squared differences between all the P/E ratio in the sample and the sample average P/E ratio is 113.36.

3.10 (a) (b)

= 32 %. The average return of the ten largest internet mutual funds for the first four months of 2000 is -32%. 2 = 107.53 = 10.37 . The average of the squared differences between all the returns for the ten largest internet mutual funds and the population mean is 107.53. The square-root of this is 10.37. The empirical rule suggests that the majority of the returns, 2 out of 3, are within 10.37% around -32%.
X = 44.69 . median = 46.25 There is no mode. range = 60 S 2 = 370.4241 S = 19.25

3.11 (a) (b)

Solutions to End-of-Section and Chapter Review Problems

77

(c)

The average percentage of the total code written for a software system that is part of the reuse data base is 44.69%. The middle ranked percentage of reuse is 46.25. There is no particular reuse rate that occurs more than once. The average of the squared differences between all the reuse rates in the sample and the sample mean is 370.4241. According to the empirical rule, roughly 2 out of 3 reuse rates in the population will scatter within 19.25% of 44.69%.

3.12 (a)

Mean = 473.46 Median = 451 There is no mode. The median seems to be better descriptive measures of the data, since they are closer to the observed values than is the mean. (b) The manufacturer should be interested in characterizing the average number of hours its batteries power a flashlight both for purposes of quality control as well as competitive market value. (c) Range = 785 Variance = 44,422.44 Standard deviation = 210.77 (d) Six times the standard deviation is approximately 1,265, which is not equal to the range for this data set (785). The range is 785/210.77 = 3.7 times the standard deviation. The one very large observation (1049) makes both measures large. (e) From the manufacturers viewpoint, the worst measure would be to compute the percentage of batteries that last over 400 hours (8/13 = .61). The median (451) and the mean (473.5) are both over 400, and would be better measures for the manufacturer to use in advertisements. (f)-(g) Original Data Altered Data Mean 473.46 550.38 Median 451 492 Mode none none Range 785 1,078 Variance 44,422.44 99,435.26 Standard deviation 210.77 315.33 All values are significantly higher for the altered data than they are for the original data. The median seems to be better descriptive measures of both the original data as well as the altered data, since it is closer to their respective observed values than is the mean for the sets. Six times the standard deviation for the altered set is approximately 1,892, which is much larger than its range (1078). The range is 1078/315.33 = 3.4 times the standard deviation. The two very large observations in the altered data set (1049 and 1342) make both measures large. From the manufacturers viewpoint, the worst measure remains the percentage of batteries that last over 400 hours (9/13 = .69). The median (492) and the mean (550.38) are both well over 400, and would be better measures for the manufacturer to use in advertisements. The shape of the distribution of the original data is right-skewed, since the mean is larger than the median. The shape of the distribution of the altered data set is right-skewed as well, since its mean is also larger than its median.

(h) (i)

78

Chapter 3: Numerical Descriptive Measures

3.13 (a)

Mean = 4.287 Median = 4.5 Q1 = 3.20 Q3 = 5.55 Variance = 2.683 Standard deviation = 1.638 Range = 6.08 Interquartile range = 2.35 Coefficient of variation = 38.21% (b) Since the mean is less than the median, the distribution is left-skewed. (c) The mean and median are both under 5 minutes and the distribution is left-skewed, meaning that there are more unusually low observations than there are high observations. But six of the 15 bank customers sampled (or 40%) had wait times in excess of 5 minutes. So, although the customer is more likely to be served in less than 5 minutes, the manager may have been overconfident in responding that the customer would almost certainly not wait longer than 5 minutes for service. (d) According to the empirical rule, between 90% to 95% of the observations will fall within 2 standard deviation around the mean. Hence, using a cutoff of 4.287 + 2 ( 1.638 ) = 7.563 minutes will result in roughly a 5% chance of having to pay this small payment of gift. 3.14 (a) Mean = 7.11 Median = 6.68 Q1 = 5.64 Q3 = 8.73 Variance = 4.336 Standard deviation = 2.082 Range = 6.67 Interquartile range = 3.09 Coefficient of variation = 29.27% (b) Since the mean is greater than the median, the distribution is right-skewed. (c) The mean and median are both well over 5 minutes and the distribution is right-skewed, meaning that there are more unusually high observations than low. Further, 13 of the 15 bank customers sampled (or 86.7%) had wait times in excess of 5 minutes. So, the customer is more likely to experience a wait time in excess of 5 minutes. The manager overstated the banks service record in responding that the customer would almost certainly not wait longer than 5 minutes for service. (d) According to the empirical rule, between 90% to 95% of the observations will fall within 2 standard deviation around the mean. Hence, using a cutoff of 7.11 + 2 ( 2.082 ) = 11.274 minutes will result in roughly a 5% chance of having to provide this small payment of gift. (e) The sample reported in Problem 3.13 above was taken from a bank located in a commercial district of a city and was taken between 12 and 1 p.m. The sample reported in this problem was taken from a bank located in a residential area and was taken on Friday evening between 5 and 7 p.m. Customer service needs are likely to be quite different between the two locales, which may influence the length of time required to serve each customer. Differences in time of day and day of week may also influence the mix of customer service needs. Additional information on what week and what month the samples were taken would be useful, as would the customer base and the number of employees for each branch. 3.15 X = 3.16 (a) (b) (c)

( 2.98% + 0.13% )
2

= 1.555%

RG = ( 1 + 2.98% ) ( 1 + 0.13% )
1/ 5 1/ 5

1/ 2

1 = 1.545%

RG = ( 1 + 0.158 ) ( 1 + 0.032 ) ( 1 + 0.115 ) ( 1 + 0.401) ( 1 + 0.516 )

1 = 23.13% 1 = 10.94%

RG = ( 1 0.348 ) ( 1 + 0.072 ) ( 1 + 0.302 ) ( 1 + 0.195 ) ( 1 + 0.545 )

The Australian and New Zealand Banking Group LTD has twice as high the rate of return for those 5 years using the geometric mean.
1/ 8

0.089 + )( + )( +( 0.0 3.17 RG = ( 1 0.289) ( 1 + ) ( 1 0.226 1 0.041 1 0.161 1 0.064 1 29) ( 1 0.022 ) )+ ( )

+ 1 2.426% =

Solutions to End-of-Section and Chapter Review Problems 3.18 (a) RG (S&P 500) = (1 +0.195 1+ 0.267 1) ( 0.310 + ) 0.203 +1 0.341 = 1) + 1 ( ( )( )( ) (b) RG (S ) = ( ( 1 0.274 ) (1 0.045 ) (1+ 0.001 ) (1 0.202 ) 1 0.763 + (+
1/ 5

79

1/5

26.2%

1 ) ) = 8.02% 1/ 5 RG ( MSFT ) = ( ( 1 + 0.684) ( 1 + 1.146) ( 1 + 0.564 ( 1 + 0.883 ( 1 + 0.436 ) 1 = 72.52% ) ) ) 1/ 5 RG (BMG ) = ( 1 0.493 1 0.286 ) ( 0.160 ) 0.186 1 0.224 =1 ) 28.11% 1 1 ( ( )( )( )

3.19 (a) (b)

Five-number summary: 2 3 7 8.5 9

10

(c) 3.20 (a) (b)

The distribution is left-skewed. There is a longer distance from Q1 to Q2 than from Q2 to Q3, confirming our conclusion that the data are left-skewed. Five-number summary: 3 4 7 9 12

10

15

(c) 3.21 (a) (b)

The distribution is almost symmetrical. The data set is almost symmetrical since the median line almost divides the box in half. Five-number summary: 0 3 7 9 12

- 10

10

20

(c) 3.22 (a) (b)

The distribution is left-skewed. The box-and-whisker plot shows a longer left box from Q1 to Q2 than from Q2 to Q3, visually confirming our conclusion that the data are left-skewed. Five-number summary: 8 6.5 7 8 9

- 10

-5

10

(c)

The distribution is left-skewed. The box-and-whisker plot shows a longer left box from Q1 to Q2 than from Q2 to Q3, visually confirming our conclusion that the data are left-skewed.

80

Chapter 3: Numerical Descriptive Measures Five-number summary: 264 307.5 451 553.5 1,049
* 0 500 1000 1500

3.23 (a) (b)

(c) 3.24 (a) (b)

The distribution is right-skewed. Because the data set is small, one very large value (1,049) skews the distribution to the right. Minimum = 6, 1st quartile = 11, median = 13, 3rd quartile =19 and maximum = 53 The distribution of P/E ratio is skewed to the right.
Box-and-whisker Plot

PE Ratio

10

20

30

40

50

3.25 (a) (b)

Five-number summary: 15 20 22 26 30 The distribution of bounced-check fee is skewed slightly to the right.
Box-and-whisker Plot

Bounced-Check Fee

10

15

20

25

30

(c) Five-number summary: 0 5 7 10 12

Solutions to End-of-Section and Chapter Review Problems

81

3.25 cont.

(d)

Box-and-whisker Plot

Monthly Service Charge

-1

11

13

(e)

The distribution is skewed to the left. The central tendency of the bounced-check fee is substantially higher than that of monthly service fee. While the distribution of the bounced-check fee is quite symmetrical, the distribution of monthly service fee is skewed more to the left with a few banks charging very low or no monthly service fee. Minimum = 43, 1st quartile = 48, median = 51, 3rd quartile = 55 and maximum = 59

3.26 (a) (b)

Box-and-whisker Plot

Chicken

40

45

50

55

60

65

(c)

The distribution is only slightly skewed to the left. Minimum = 45, 1st quartile = 49.5, median = 56, 3rd quartile = 57 and maximum = 60.

82

Chapter 3: Numerical Descriptive Measures

3.26 (d) cont.

Box-and-whisker Plot

Burger

40

45

50

55

60

65

The distribution is skewed to the left. 3.27 (a) (b)

Box-and-whisker Plot

minimum = -50.5, 1st quartile = -41, median = -28.9, 3rd quartile = -24.6 and maximum = -15

Return

-60

-50

-40

-30

-20

The distribution is skewed to the left with more funds suffering substantial loss.

Solutions to End-of-Section and Chapter Review Problems

83

3.28 (a) (b)

minimum = 15, 1st quartile = 25, median =46.25, 3rd quartile = 62.5 and maximum = 75

Box-and-whisker Plot

Reuse

10

30

50

70

The distribution is skewed to the left. 3.29 (a) (b) Five-number summary: 0.38 3.2 4.5 5.55 6.46

Box-and-whisker Plot

Waiting Time

(c)

The distribution is skewed to the left. Five-number summary: 3.82 5.64 6.68 8.73 10.49

84

Chapter 3: Numerical Descriptive Measures

3.29 (d) cont.

Box-and-whisker Plot

Waiting Time

10

12

(e)

The distribution is skewed slightly to the right. The central tendency of the waiting time for the bank branch located in the commercial district of a city is lower than that of the branch located in the residential area. There are a few longer than normal waiting time for the branch located in the residential area whereas there are a few exceptionally short waiting time for the branch located in the commercial area. Population Mean = 6 Population Variance, 2 = 9.4 Population Mean = 6 Population Standard deviation, = 1.67 Population Standard Deviation, = 3.066 Population Variance, 2 = 2.8

3.30 (a) (b) 3.31 (a) (b) 3.32 (a)

(b) (c) (d) (e)

Stem-and-leaf of Quarterly Sales Tax Receipts 5 3 6 57 7 3568 8 04679 9 02335689 10 00123345567 11 011125668 12 555789 13 00 14 5 15 1 Mean = 10.28 2 = 2.177 = 1.475 (1) 58% (2) 80%(3) 96% These percentages are lower than the empirical rule would suggest.

Solutions to End-of-Section and Chapter Review Problems

85

3.33 (a) (b) (c) (d) (e) (f)

67% 90%-95% not calculable 75% 88.89%

4 to + 4 or 1.2 to 55.2
Stem-and-leaf for Top Fifth Ten Thousand $ Thousand $ 8 0349 9 112344677899 10 0223344677 11 000112366667 12 033479 13 122 14 3779

3.34 (a) Stem-and-leaf for Bottom Fifth Thousand $ Hundred $ 5 2 6 24447 7 23457 8 1169 9 002334699 10 01456778 11 01125 12 02457 13 1133 14 22368 15 7 Bottom Fifth Top Fifth (b) = 10.309 = = 2.574 = (c) 2 = 6.626 2 = (d) Bottom Fifth: 1 (1) 2 (2) 3 (3)

110.338 16.638 276.831 Upper bound 12.883 15.457 18.031

Percent of values In Interval 30/50 = 60% 49/50 = 98% 50/50 = 100% Percent of values Top Fifth: Lower bound Upper bound In Interval 1 (1) 93.700 126.976 35/50 = 70% 2 (2) 77.062 143.614 47/50 = 94% 3 (3) 60.424 160.252 50/50 = 100% (e) Only 60% of the top fifth of average income for U.S. families with children fell within one standard deviation of the population mean. According to the empirical rule, approximately two-thirds or 67% of the data values should be within one standard deviation of the mean. All other values (90-95% within 2 and 100% within 3 ) fall within the values predicted by the empirical rule. Lower bound 7.735 5.161 2.587

86

Chapter 3: Numerical Descriptive Measures The District of Columbia represents both the lowest value in the poorest fifth of the distribution and the highest value in the richest fifth of the average incomes considered. By removing the District of Columbia, the average income for the top fifth rises from $10,309 to $10,409, and the average income for the bottom fifth drops from $110,338 to $109,555. With the removal of an extreme value from each set, the standard deviations both drop noticeably, from two, and three standard deviations on either side of the mean are also altered significantly, as Bottom Fifth: Lower bound Upper bound 1 (1) 7.910 12.908 2 (2) 5.411 15.407 3 (3) 2.912 17.906 Top Fifth: Lower bound Upper bound 1 (1) 93.709 125.400 2 (2) 77.864 141.245 3 (3) 62.019 157.090 Population Mean = 4.53 The average year-to-date return of the DJ-US real estimate investment group is 4.53 percentage points. Population variance = 172.1171 Population Standard Deviation = 13.12 The average squared distances between all year-to-date returns and the population mean is 172.1171. The majority of the year-to-date returns are within 13.12 percentage point around 4.53. Since the population standard deviation is about three times the value of the population mean, there is considerable variability in the YTD returns of the REITs in the DJ-US real estate investment group as majority of the YTD returns are within 13.12 percentage points around 4.53.

3.34 (f) cont.

3.35 (a) (b)

(c)

3.36 (a)

Population mean = 5.027 The average annual return for the ten largest money market funds for the period ending March 31, 1999 is 5.027 percentage points. (b) Population variance = 0.017321 Population standard deviation = 0.1316 The average of the squared differences between all the annual return and the population average return is 0.01732. The majority of the annual returns are within 0.1316 percentage point around 5.027. (c) There is not a lot of variability in the annual returns of the ten largest money market funds. SSX = 217.64, SSY = 1958.73, SSXY = 652.91.

3.37 (a)

r=
(b) 3.38 (a)

SSXY SSX SSY

652.91 217.64 1958.73

= +1.0

There is a perfect positive linear relationship between X and Y because all the points lie exactly on a straight line with a positive slope. According to the article, the NASDAQ and U.S. investment in information technology during the nineties were positively correlated. The NASDAQ is a weighted average of mostly technology stocks and, hence, the rapid increase in investments of U.S. businesses in computers and information technology will only help boost the performance of stocks of businesses involved in information technology. Quarterly net earnings for Baltimore Gas & Electric Company will be negatively correlated to the average temperature of that period. The lower the average temperature,

3.39 (a)

Solutions to End-of-Section and Chapter Review Problems

87

(b)

(c) 3.40 (a) (b) 3.41 (a)

the higher will be the demand for gas or electric for heating appliances, which translate into higher earnings for the company. Investors might use the average temperature in a particular quarter to predict the stock prices of Baltimore Gas & Electric Company since stock prices are positively correlated to earnings as well. When the average temperature of a particular quarter is exceptionally low, one will predict that the stock prices of the company will increase in the following quarter. The relationship between stock prices and average temperature should be relevant only for a subset of natural resource stocks that are closely related to the heating industry. -0.1641 The low coefficient of correlation between price and energy cost indicates that there is a very week negative linear relationship between the two if there is any.

Scatter Diagram
350 300 250 Calories 200 150 100 50 0 0 5 10 Fat (%) 15 20

(b) 3.42 (a) (b)

There appears to be a positive relationship between calories and fat content. The coefficient of correlation of 0.9864 is very close to +1, which indicates that there is a very high linear positive relationship between weight of mail and the number of orders.

r = 0.9864

3.43 We should look for ways to describe the typical value, the variation, and the distribution of the data within a range. 3.44 Central tendency or location refers to the fact that most sets of data show a distinct tendency to group or cluster about a certain central point.

88

Chapter 3: Numerical Descriptive Measures

3.45 The arithmetic mean is a simple average of all the values, but is subject to the effect of extreme values. The median is the middle ranked value, but varies more from sample to sample than the arithmetic mean, although it is less susceptible to extreme values. The mode is the most common value, but is extremely variable from sample to sample. 3.46 Measures of central tendency measure a central point in a set of data. Measures of noncentral tendency such as quartiles, measure other points in the distribution. 3.47 Variation is the amount of dispersion, or spread, in the data. 3.48 The range is a simple measure, but only measures differences between the extremes. The interquartile range measures differences in the center fifty percent of the data. The standard deviation measures variation around the mean, and is the only measures that takes into account each observation. 3.49 The empirical rule relates the mean and standard deviation to the percentage of values that will fall within a certain number of standard deviation of the mean. 3.50 Shape is the manner in which the data are distributed. 3.51 The box-and-whisker plot and five number summary allow one to determine the approximate shape of the distribution. In addition, the box-and-whisker plot allows the comparison of several groups in terms of their location, dispersion, and shape. 3.52 (a) (b) (c) (d) (e) mean = 5.5014 median = 5.515 first quartile = 5.44 third quartile = 5.57 range = 0.52 interquartile range = 0.13 variance = 0.0112 standard deviation = 0.106 coefficient of variation = 1.924% The average weight of the tea bags in the sample is 5.5014 grams while the middle ranked weight is 5.515. The company should be concerned about the central tendency because that is where majority of the weight will scatter around. The average of the squared differences between the weights in the sample and the sample mean is 0.0112 whereas the square-root of it is 0.106 grams. The difference between the lightest and the heaviest tea bags in the sample is 0.52. 50% of the tea bags in the sample weight between 5.44 and 5.57 grams. According to the empirical rule, about 67% of the tea bags produced will have weight that falls within 0.106 grams around 5.5014 grams. The company producing the tea bags should be concerned about the variation because tea bags will not weigh exactly the same due to various factors in the production process, e.g. temperature and humidity inside the factory, differences in the density of the tea, etc. Having some idea about the amount of variation will enable the company to adjust the production process accordingly.

Solutions to End-of-Section and Chapter Review Problems

89

3.52 (f) cont.

Box-and-whisker Plot

Teabags

5.2

5.4

5.6

5.8

(g) (h) (i)

The data is slightly left skewed. On average, the weight of the teabags is quite close to the target of 5.5 grams. Even though the average weight is close to the target weight of 5.5 grams, the standard deviation of 0.106 indicates that about 75% of the teabags will fall within 0.212 grams around the target weight of 5.5 grams. The interquartile range of 0.13 also indicates that half of the teabags in the sample fall in an interval 0.13 grams around the median weight of 5.515 grams. The process can be adjusted to reduce the variation of the weight around the target mean. mean = 43.89 median = 45 1st quartile = 18 3rd quartile = 63 range = 76 interquartile range = 45 variance = 639.2564 standard deviation = 25.28 coefficient of variation = 57.61%

3.53 (a) (b) (c) (d)

Box-and-whisker Plot

Time

10

30

50

70

90

90

Chapter 3: Numerical Descriptive Measures The distribution is skewed to the right because there are a few policies that require exceptionally long period to be approved. The average approval process takes 43.89 days with 50% of the policies being approved in less than 45 days. 50% of the applications are approved between 18 and 63 days. About 67% of the applications are approved between 18.6 to 69.2 days. Mean = 43.04 Median = 28.5 Q1 = 14 Q3 = 54 Range = 164 Interquartile range = 40 Variance = 1,757.79 Standard deviation = 41.926 Coefficient of variation = 97.41% Box-and-whisker plot for Days to Resolve Complaints
** * 0 50 100 * 150 * 200

3.53 (e) cont. (f)

3.54 (a) (b) (c) (d)

(e) (f)

The distribution is right-skewed. Half of all customer complaints that year were resolved in less than a month (median = 28.5 days), 75% of them within 54 days. There were five complaints that were particularly difficult to settle which brought the overall average up to 43 days. No complaint took longer than 165 days to resolve. The data represent a sample of the population of attendance at all N.Y. Mets' games.
Summary Statistics Yes Mean 29586.53 Standard Error 1819.199 Median 29438 Mode 21981 Standard Deviation 7929.706 Sample Variance 62880242 Kurtosis -0.89911 Skewness 0.064708 Range 27327 Minimum 16248 Maximum 43575 Sum 562144 Count 19 No 20971.57 1124.711 18484 18484 8784.278 77163533 3.543064 1.788166 42046 12001 54047 1279266 61

3.55 (a) (b)

(c)

The average attendance is 8615 more when there is a promotion than when there is not and the variation of attendance when there is a promotion is also smaller than that when there is not a promotion. (d) There are many factors that can cause variation in the paid attendance. Some of them are, e.g., weather condition, time and day of the game, which is the visiting team, etc.

Solutions to End-of-Section and Chapter Review Problems

91

3.56 (a)

(b) (c)

mean = 8.42, median = 8.42, range = 0.186 and standard deviation = 0.046. On average, the width is 8.42 inches. The width of the middle ranked observation is 8.42. The difference between the largest and smallest width is 0.186 and majority of the widths fall between 0.046 inches around the mean of 8.42 inches. Minimum = 8.312, 1st quartile = 8.404, median = 8.42, 3rd quartile = 8.459 and maximum = 8.498
B x n -w is e P t o -a d h k r lo

W th id

8 .2

8 .3

8 .4

8 .5

8 .6

(d) 3.57 (a) (b)

Even though the median equals to the mean, the distribution is not symmetrical but skewed to the left. All the troughs fall within the limit of 8.31 and 8.61 inches. mean = 1723.4 median = 1735 range = 348 standard deviation = 89.55 The average force required to break the insulators in the sample is 1723.4 pounds. The middle ranked breaking force is 1735 pounds. The differences between the smallest and largest breaking force is 348 pounds. Roughly about 67% of the insulators will have breaking force that falls within 89.55 pounds of 1723.4 pounds. Five-number summary: 1522 1662 1735 1784 1870.

(c) (d)

B ox-and-w hisker Plot

Force

1500

1600

1700

1800

1900

(e)

The distribution is skewed to the left. All the observations in the sample have breaking force that is greater than 1500 pounds and, hence, will fulfill the company's requirement.

3.58 (a), (b), (c)

92

Chapter 3: Numerical Descriptive Measures

(d)

(e) (f)

Times to clear problems at both central offices are right-skewed. Times to clear problems for Office I are less dispersed about the mean than times to clear problems for Office II, even though the average for Office I times is higher (2.214) than that for Office II (2.012). If the value 7.55 were incorrectly recorded as 27.55, the mean would be one minute higher (from 2.012 to 3.012) and the standard deviation would be over 3 times as large (from 1.892 to 5.936). Plant A 9.382 8.515 Plant B 11.354 11.96

(g)

3.59 (a), (b), (c) Mean Median

Solutions to End-of-Section and Chapter Review Problems

93

(d)

Q1 7.29 Q3 11.42 Range 17.2 Interquartile range 4.13 Variance 15.981 Standard deviation 3.998 Coefficient of variation 42.61% Box-and-whisker plot for Plant A
*

6.25 14.25 23.42 8 26.277 5.126 45.15%

10

20

30

Box-and-whisker plot for Plant B

* 0 10 20 30

(e) (f)

The distribution of processing times for Plant A are right-skewed. The distribution of processing times for Plant B are left-skewed. Processing times for Plants A and B are quite different. Plant B has a greater range of processing times, much more dispersion among data values, a higher median, a higher value for the third quartile, and a greater extreme value than Plant A.

3.60 (a), (b), (c) Minimum First Quartile Median Third Quartile Maximum Mean Mode Standard Deviation Sample Variance Range Interquartile range Coefficient of Variations Cost (per ounce) Calories Fiber (grams) Sugar (grams) 0.10 50.00 5.00 0.00 0.13 135.00 5.00 6.00 0.17 190.00 6.00 11.00 0.20 200.00 8.00 17.50 0.27 210.00 13.00 23.00 0.17 165.76 6.91 11.39 0.15 200.00 5.00 18.00 0.05 51.78 2.40 6.65 0.00 2681.44 5.77 44.25 0.17 160.00 8.00 23.00 0.07 65.00 3.00 11.50 27% 31% 35% 58%

94

Chapter 3: Numerical Descriptive Measures

3.60 (d) cont.

Box-and-whisker Plot (Cost per ounce)

0.05

0.1

0.15

0.2

0.25

0.3

Box-and-whisker Plot (Calories)

40

90

140

190

Box-and-whisker Plot (Fiber)

10

15

Solutions to End-of-Section and Chapter Review Problems

95

3.60 (d) cont.

Box-and-whisker Plot (Sugar)

-5

10

15

20

25

(e)

The distributions for cost per ounce and fiber content are skewed to the right while the distribution for calories is skewed to the left. The distribution for sugar content is quite symmetrical. Cost Per Serving Dog Dog Cat Cat Dry Canned Dry Canned Mean 0.5486 1.9896 0.2700 1.0953 Standard Error 0.0284 0.2964 0.0202 0.1365 Median 0.5400 1.5550 0.2600 0.8400 Mode 0.6200 #N/A 0.2600 #N/A Standard Deviation 0.1504 1.4520 0.1030 0.5949 Sample Variance 0.0226 2.1083 0.0106 0.3540 Range 0.6900 7.4100 0.4600 2.0900 Minimum 0.2200 0.9300 0.1400 0.5200 Maximum 0.9100 8.3400 0.6000 2.6100 1st Quartile 0.4900 1.3900 0.2000 0.6800 3rd Quartile 0.6500 1.9600 0.2900 1.3900 Interquartile Range 0.1600 0.5700 0.0900 0.7100 C.V. 0.2741 0.7298 0.3815 0.5432

3.61 (a), (b), (c)

96

Chapter 3: Numerical Descriptive Measures

3.61 (d) cont.

Box-and-whisker Plot 9 8 7 6 5 4 3 2 1 0
(e) (f)
Dog (Dry) Dog (Canned) Cat (Dry) Cat (Canned)

For cost per serving, all but dry dog food are skewed to the right. The skewness is particularly severe for canned dog food. Canned cat food also skewed significant to the right. On average, canned dog food is more expensive than canned cat food. Dry dog food is also more expensive than dry cat food on average.

3.62 (a)
miles per gallon fuel tank capacity length wheelbase width turning circle requirement weight luggage capacity front shoulder room front leg room front head room rear shoulder room rear leg room rear head room mean 21.292 17.883 187.594 107.632 70.915 39.849 3502.311 25.231 56.472 41.255 4.288 54.745 28.363 3.024 median 21.000 17.400 188.000 107.000 70.000 40.000 3395.000 16.000 56.000 41.000 4.500 56.000 28.500 3.000

Solutions to End-of-Section and Chapter Review Problems

97

3.62 (b)

cont.
1st quartile miles per gallon fuel tank capacity length wheelbase width turning circle requirement weight luggage capacity front shoulder room front leg room front head room rear shoulder room rear leg room rear head room 19 15.8 178 103 68 38 3105 14 54 40.5 3.5 52.5 27 2 range miles per gallon fuel tank capacity length wheelbase width turning circle requirement weight luggage capacity front shoulder room front leg room front head room rear shoulder room rear leg room rear head room 18.00 17.90 74.00 46.00 14.00 15.00 3430.00 66.00 12.00 5.00 6.50 58.00 11.50 6.00 3rd quartile 23 19 195 112 73 41 3890 34.5 58.5 42 5 57 30 4 variance 14.21 11.74 156.09 49.21 10.48 7.83 401263.89 337.37 9.25 1.04 1.64 50.56 4.83 1.88 standard deviation 3.769 3.426 12.494 7.015 3.237 2.797 633.454 18.368 3.042 1.022 1.280 7.110 2.197 1.370 coefficient of variation 0.1770 0.1916 0.0666 0.0652 0.0456 0.0702 0.1809 0.7280 0.0539 0.0248 0.2985 0.1299 0.0775 0.4532

(c)
interquartile range 4 3.2 17 9 5 3 785 20.5 4.5 1.5 1.5 4.5 3 2

98 3.62 cont.

Chapter 3: Numerical Descriptive Measures (d)

Box-and-w hisker Plot

50

40

Front Leg Room (inches)

Turning Circle (feet)

30

Rear Leg Room (inches) 20

Capacity (gal) 10

Mileage (mpg)

Front Head Room (inches)

Rear Head Room (inches)

-10

Solutions to End-of-Section and Chapter Review Problems

99

3.62 cont.

(d)

Box-and-w hisker Plot

200

150

Length (inches)

100 Wheel Base (inches)

Width (inches) Front Shoulder (inches)

50

Luggage (cu. ft.) 0

Rear Shoulder (inches)

100

Chapter 3: Numerical Descriptive Measures

3.62 (d) cont.

Box-and-whisker Plot 6410 5910 5410 4910 4410 3910 3410 2910 2410 Weight (lbs)

(e) Skewness miles per gallon fuel tank capacity length wheelbase width turning circle requirement weight luggage capacity front shoulder room front leg room front head room rear shoulder room rear leg room rear head room somewhat symmetrical right somewhat symmetrical right right right right right slightly to the right somewhat symmetrical slightly to the right left symmetrical symmetrical

Solutions to End-of-Section and Chapter Review Problems

101

3.63 (a), (b), (c) Food Dcor Service Price Food Dcor Service Price (NYC) (NYC) (NYC) (NYC) (LI) (LI) (LI) (LI) Minimum 15.00 12.00 15.00 19.00 15.00 9.00 15.00 18.00 First Quartile 18.00 15.00 17.00 29.00 18.00 14.00 17.00 26.00 Median 19.50 17.00 18.00 36.00 21.50 16.00 18.50 29.00 Third Quartile 22.00 20.00 19.00 43.00 23.00 20.00 21.00 37.00 Maximum 28.00 26.00 26.00 75.00 26.00 25.00 24.00 50.00 Mean 20.30 17.50 18.34 37.72 20.62 16.66 18.94 31.36 Median 19.50 17.00 18.00 36.00 21.50 16.00 18.50 29.00 Mode 18.00 15.00 17.00 34.00 22.00 14.00 17.00 26.00 Standard Deviation 3.19 3.21 2.57 11.15 3.00 3.74 2.30 8.26 Sample Variance 10.17 10.30 6.60 124.37 9.02 13.98 5.28 68.24 Range 13.00 14.00 11.00 56.00 11.00 16.00 9.00 32.00 Interquartile range 4.00 5.00 2.00 14.00 5.00 6.00 4.00 11.00 Coefficient of Variation 0.16 0.18 0.14 0.30 0.15 0.22 0.12 0.26 (d)

Box-and-whisker Plot 80 70 60 50 40 30 20
Food (NYC) Price (NY) Dcor (NYC) Service (NYC) Food (LI) Dcor (LI) Service (LI) Price (LI)

10 0
(e) (f)

The distribution for all the variables are skewed to the right with varying degree with the only exception of prince in New York City, which is skewed slightly to the left. There is no noticeable difference in terms of the rating of food, dcor, and service between New York City and Long Island restaurants. The price of restaurants in New York City is on average higher than the price of restaurants in Long Island. The most expensive restaurant in New York City costs almost $30 more than the most expensive restaurant in Long Island.

102

Chapter 3: Numerical Descriptive Measures The Springville Herald Case

SH3.1 (a)

(1) X = 39.685 (2) Median = 18 (3) Q1 = 5 (4) Q3 = 52 (5) Range = 236 (8) Interquartile range = 47 (9) s2 = 2,671.4 (10) s = 51.686 (11) c.v. = 130.24% Five- Number Summary: 1

18

52

237

(b)

The data are highly variable with right skewness caused by some jobs that access a large number of data cartridges (The mean, midrange, and midhinge are substantially above the median). Twenty-five percent of the jobs use less than five cartridges, 50% use less than 18 cartridges, awhile 75% use less than 52 cartridges. The data have a range of 236, with the most typical fifty percent of the jobs varying by 47 cartridges. The standard deviation is high, and is more than 100% of the mean. The right skewness is clear from the box-and-whisker plot which shows a large distance from the third quartile to the largest number of cartridges accessed.