1.

The median is often a better representative of the central value of a data set when the data set:

Is bimodal.

Has a high standard deviation.

Is highly skewed.


Has no outliers

2.
The data in the Excel spreadsheet linked below provide information on the nutritional content (in
grams per serving) of some leading breakfast cereals.

For which nutrients is the mean nutrient content per serving greater than the median nutrient
content per serving? See attached excel document!


Proteins only.
 Complex carbohydrates only.
 Both nutrients.
 Neither nutrient.

3.
The histogram below plots the carbon monoxide (CO) emissions (in pounds/minute) of 40
different airplane models at take-off.

The distribution is best described as is:

Uniform.


Heteroskedastic.


Normal.


Skewed right.
 4.
The histogram below plots the carbon monoxide (CO) emissions (in pounds/minute) of 40
different airplane models at take-off.

Which of the following statements is the best inference that can be drawn from this histogram?

The mean amount of carbon monoxide emissions is greater than the median amount of carbon
monoxide emissions.


The mean amount of carbon monoxide emissions is less than the median amount of carbon
monoxide emissions.


The mean and median amounts of carbon monoxide emissions are about equal.


The relative sizes of the mean and median amounts of carbon monoxide emissions cannot be
inferred from the histogram.
 5.
In the data set shown below, the correlation coefficient of the two variables is:

-1.0.


-0.5


0.0


None of the above.

6.
The data in the Excel spreadsheet linked below give the ages and salaries of the chief executive
officers of 59 companies with sales between $5 million and $350 million.

The correlation between age and salary can be characterized as:

Strong and positive.



Strong and negative.



Weak and positive.



Weak and negative.

Salary
Age ($thousands)
53 145
43 621
33 262
45 208
46 362
55 424
41 339
55 736
36 291
45 58
55 498
50 643
49 390
47 332
69 750
51 368
48 659
62 234
45 396
37 300
50 343
50 536
50 543
58 217
53 298
57 1103
53 406
61 254
47 862
56 204
44 206
46 250
58 21
48 298
38 350
74 800
60 726
32 370
51 536
50 291
40 808
61 543
63 149
56 350
45 242
61 198
70 213
59 296
57 317
69 482
 44 155
56 802
50 200
56 282
43 573
48 388
52 250
62 396
48 572

7.
For a given set of data, the standard deviation measures:

The difference between the mean and the data point farthest from the mean.


The difference between the mean and the data point nearest to the mean.


The difference between the mean and the median.



None of the above.

8.
The data in the Excel spreadsheet linked below give the average exchange rates of four
currencies to the US dollar during October 2002.

Over this period, which currency was most volatile relative to the US dollar, as measured by the
coefficient of variation?

The Brazilian Real.



The Euro.

 The Japanese Yen.


The South Korean Won.

South
Brazilian Japanese Korean
Real per Yen per Euro per Won per
Date Dollar Dollar Dollar Dollar
10/1/2002 3.76 121.70 1.01 1224.70
10/2/2002 3.62 122.56 1.02 1224.50
10/3/2002 3.66 122.89 1.01 1230.60
10/4/2002 3.70 122.70 1.01 1230.40
10/5/2002 3.78 123.19 1.02 1229.40
10/6/2002 3.78 123.37 1.02 1229.40
10/7/2002 3.78 123.21 1.02 1229.40
10/8/2002 3.76 124.29 1.02 1237.70
10/9/2002 3.73 124.39 1.02 1254.30
10/10/2002 3.80 123.28 1.01 1247.30
10/11/2002 3.96 123.62 1.01 1258.20
10/12/2002 4.03 124.06 1.01 1257.30
10/13/2002 4.03 124.05 1.01 1257.30
10/14/2002 4.03 124.16 1.01 1257.30
10/15/2002 3.89 124.31 1.01 1259.00
10/16/2002 3.84 124.70 1.02 1260.20
10/17/2002 3.92 124.49 1.02 1247.20
10/18/2002 3.90 125.06 1.03 1249.30
10/19/2002 3.88 125.55 1.03 1241.00
10/20/2002 3.88 125.56 1.03 1241.00
10/21/2002 3.88 125.43 1.03 1241.00
10/22/2002 3.92 124.96 1.03 1237.00
10/23/2002 3.91 125.16 1.02 1247.00
10/24/2002 3.92 124.61 1.02 1228.90
10/25/2002 3.82 124.49 1.02 1228.90
10/26/2002 3.89 124.34 1.02 1228.60
10/27/2002 3.89 124.34 1.02 1228.60
10/28/2002 3.89 124.20 1.02 1228.60
10/29/2002 3.78 123.62 1.02 1233.70
10/30/2002 3.83 123.00 1.02 1224.40
10/31/2002 3.72 122.94 1.02 1234.50

9.
A national business magazine intends to survey its subscribers to determine who they think is the
"CEO of the Century." Subscribers are invited to complete an online survey.
 Based only on this information, which of the following can be inferred about the survey results?

They will be biased because the magazine's subscribers are unqualified to determine who the
CEO of the century is.


They will accurately reflect the opinion of the nation as a whole about who the CEO of the
century is.


They will be unbiased because the respondents will be self-selected.



They will be biased because subscribers who have access to the online survey are not
representative of the population of subscribers as a whole.

10.
A political consultant conducts a survey to determine what position the mayoral candidate she
works for should take on a proposed smoking ban in restaurants.

Which of the following survey questions will deliver an unbiased response?

Source


Should the city ban smoking in restaurants to protect our children from second-hand smoke?


Should tobacco smoke, a known cause of lung cancer, be banned from public spaces such as
restaurants?


Does the city have the right to restrict recreational activities, such as moderate consumption of
alcohol or tobacco, on the premises of privately-owned businesses?


None of the above.

11.
In a recently administered IQ test, the scores were distributed normally, with mean 100 and
standard deviation 15.

What proportion of the test takers scored between 70 and 130?



About 68%.


About 84%.


About 95%.


About 99.5%.

12.
Which of the following is not true about the Normal Distribution?

\


It is symmetric.


Its mean and median are equal.



It is completely described by its mean and its standard deviation.



It is bimodal.

13.
The histogram below shows the underlying distribution pattern of the results of a rolled die.
Suppose the die is rolled 50 times and the results are averaged. Suppose this process — rolling
50 dice and averaging their results — is repeated 100 times.

Which of the following best describes the distribution of these 100 averages?

Skewed right.


Skewed left.


Normal.


Uniform.

14.
A nutrition researcher wants to determine the mean fat content of hen's eggs. She collects a
sample of 40 eggs. She calculates a mean fat content of 23 grams, with a sample standard
deviation of 8 grams. From these statistics she calculates a 90% confidence interval of [20.9
grams, 25.1 grams].

What can the researcher do to decrease the width of the confidence interval?

Increase the confidence level.



Decrease the confidence level.



Decrease the sample size



None of the above.

15.
In a random sample of 321 senior citizens, 61 were found to own a home computer.

Based on this sample, the 95% confidence interval for the proportion of computer-owners among
senior citizens is:

[2.6%; 7.4%].


[13.4%; 24.6%].


[14.7%; 23.3%].


The answer cannot be determined from the information given.

16.
Preliminary estimates suggest that about 58% of students at a state university favor
implementing an honor code.

To obtain a 95% confidence interval for the proportion of all students at the university favoring
the honor code, what is the minimum sample size needed if the total width of the confidence
interval must be less than 5 percentage points (i.e., the confidence interval should extend at most
2.5 percentage points above and below the sample proportion)?

375.


264.


1,498.


The answer cannot be determined from the information given.

17.
In a survey of twelve Harbor Business School graduates, the mean starting salary was $93,000,
with a standard deviation of $17,000.

The 95% confidence interval for the average starting salary among all Harbor graduates is:

[$83,382; $102,618].


[$82,727; $103,327].


[$82,199; $103,801].

 [$59,000; $127,000].

7. In a survey of 53 randomly selected patrons of a shopping mall, the meanamount of currency

carried is $42, with a standard deviation of $78.

What isthe 95% confidence interval for the mean amount of currency carried by mallpatrons?

[$39.1; $44.9].

[$24.4; $59.6].

[$21.0; $63.0].

[$14.4; $69.6].

19.
A filling machine in a brewery is designed to fill bottles with 355 ml of hard cider. In practice,
however, volumes vary slightly from bottle to bottle. In a sample of 49 bottles, the mean volume
of cider is found to be 354 ml, with a standard deviation of 3.5 ml.

At a significance level of 0.01, which conclusion can the brewer draw?

The true mean volume of all bottles filled is 354 ml.



The machine is not filling bottles to an average volume of 355 ml.



There is not enough evidence to indicate that the machine is not filling bottles to an average
volume of 355 ml.


The machine is filling bottles to an average volume of 355 ml.

20.
To conduct a one-sided hypothesis test of the claim that houses located on corner lots (corner-lot
houses) have higher average selling prices than those located on non-corner lots, the following
alternative hypothesis should be used:

The average selling price of a corner-lot house is higher than it is commonly believed to be.


The average selling price of a corner-lot house is higher than the average selling price of all
houses.


The average selling price of a corner-lot house is the same as the average selling price of a house
not located on a corner lot.


The average selling price of a corner-lot house is higher than the average selling price of a house
not located on a corner lot.
Non-corner
Corner-lot Lot
House Price House Price
(in (in
$hundreds) $hundreds)
2150 2050
1999 2080
1800 2150
1375 1900
1250 1560
1110 1450
1139 1449
995 1270
900 1235
1695 1170
1553 1180
1300 1155
1020 995
1020 975
925 975
725 960
1299 860
1250 1250
1080 922
1050 899
835 850
805 876
750 890
773 870
1295 700
975 720
700 720
2100 749
600 731
1844 670
 699 2150
1330 1599
1129 1350
1050 1239
1000 1200
1030 1125
940 1100
874 1049
766 955
739 934
* 875
* 889
* 855
* 810
* 799
* 759
* 755
* 750
* 730
* 729
* 710
* 690
* 670
* 619
* 939
* 820
* 780
* 770
* 620
* 540
* 1070
* 725
* 660
* 580
* 1580
* 1160
* 1109
* 1050
* 1045
* 1020
* 975
* 950
* 920
* 945
* 872
* 870
 * 869

21.

The data in the Excel spreadsheet linked above indicate the selling prices of houses located on
corner lots ("corner-lot houses") and of houses not located on corner lots.

Conduct a one-sided hypothesis test of the claim that corner-lot houses have higher average
selling prices than those located on non-corner lots. Using a 99% confidence level, which of the
following statements do the data support?

Upscale, expensive neighborhoods have more street corners.

The average selling price of a corner-lot house is higher than that of the average house not
located on a corner lot.

The average selling price of a corner-lot house is no more than that of the average house not
located on a corner lot.

There is not enough evidence to support the claim that the average selling price of a corner-lot
house is higher than that of the average house not located on a corner lot.

22.

Two semiconductor factories are being compared to see if there is a difference in the average
defect rates of the chips they produce. In the first factory, 250 chips are sampled. In the second
factory, 350 chips are sampled. The proportions of defective chips are 4.0% and 6.0%,
respectively.

Using a confidence level of 95%, which of the following statements is supported by the data?

There is not sufficient evidence to show a significant difference in the average defect rates of the
two factories.


There is a significant difference in the average defect rates of the two factories.

The first factory's average defect rate is lower than the second factory's on 95 out of 100 days of
operation.

None of the above.

23.
The regression analysis below relates average annual per capita beef consumption (in pounds)
and the independent variable "average annual beef price" (in dollars per pound).

The coefficient on beef price tells us that:

For every price increase of $1, average beef consumption decreases by 9.31 pounds.


For every price increase of $1, average beef consumption increases by 9.31 pounds.


For every price increase $9.31, average beef consumption decreases by 1 pound.


For price increase of $9.31, average beef consumption increases by 1 pound.

24.
The regression analysis below relates average annual per capita beef consumption (in pounds)
and the independent variable "average annual beef price" (in dollars per pound).

In a year in which the average price of beef is at $3.51 per pound, we can expect average annual
per capita beef consumption to be approximately:


55.2 pounds


52.6 pounds


53.6 pounds


117.9 pounds

25.
The regression analysis below relates average annual per capita beef consumption (in pounds)
and the independent variable "average annual per capita pork consumption" (in pounds).

At what level is the coefficient of the independent variable pork consumption significant?

0.10.


0.05.


0.01.


None of the above.

 26.
The regression analysis below relates average annual per capita beef consumption (in pounds)
and the independent variable "average annual per capita pork consumption" (in pounds).

Which of the following statements is true?

Beef consumption can never be less than 65.09 pounds.



Beef consumption can never be greater than 65.09 pounds.



The y-intercept of the regression line is 65.09 pounds.



The x-intercept of the regression line is 65.09 pounds.

27.
The regression analysis at the bottom relates average annual per capita beef consumption (in
pounds) and the independent variables "average annual per capita pork consumption" (in pounds)
and "average annual beef price" (in dollars per pound).

Which of the independent variables is significant at the 0.01 level?



Beef price only.



Pork consumption only.



Both independent variables.



Neither independent variable.

28.
The regression analysis at the bottom relates average annual per capita beef consumption (in
pounds) and the independent variables "annual per capita pork consumption" (in pounds) and
"average annual beef price" (in dollars per pound).

The coefficient for beef price, -12, tells us that:

For every $1 increase in beef price, average beef consumption decreases by 12 lbs, not
controlling for pork consumption.


For every $12 drop in beef price, average beef consumption decreases by 1 lbs, not controlling
for pork consumption.


For every $1 increase in beef price, average beef consumption decreases by 12 lbs, controlling
for pork consumption, i.e. holding pork consumption constant.


For every $12 decrease in beef price, average beef consumption decreases by 1 lbs, controlling
for pork consumption, i.e. holding pork consumption constant.
 29.

The data in the Excel spreadsheet linked below give the seasonally adjusted value of total new
car sales (in millions of dollars) in the United States, total national wage and salary
disbursements (referred to here as "compensation") (in billions of dollars), and the employment
level in the non-agricultural sector (in thousands) for 44 consecutive quarters. An auto industry
executive wants to know how well she can predict new car sales two quarters in advance using
the current quarter's compensation data.

How many data points can she use in a regression analysis using the data provided?

41.

42.

43.

44.

Seasonally
Adjusted New Employment
Car Sales Compensation Level
Quarter ($millions) ($billions) (thousands)

1 46,787 1,193.3 87,973

2 44,256 1,217.3 90,021

3 41,523 1,254.4 90,120

4 42,367 1,284.7 91,180

 5 43,580 1,319.8 89,832

6 36,392 1,335.1 90,668

7 38,303 1,360.1 89,850

8 40,690 1,411.6 91,276

9 45,653 1,451.7 89,964

10 42,283 1,478.1 91,566

11 44,122 1,512.6 91,405

12 42,022 1,530.6 91,691

13 45,255 1,542.7 89,341

14 45,064 1,563.9 90,310

15 43,560 1,579.8 89,324

16 48,794 1,586.0 89,436

17 51,794 1,610.7 88,815

18 55,270 1,643.5 89,920

19 54,389 1,671.3 90,480

20 60,100 1,715.4 92,367

21 67,741 1,755.8 91,637

22 69,554 1,793.1 94,256

23 65,352 1,819.5 95,021

24 70,472 1,847.7 96,547

25 76,729 1,882.7 95,450

26 79,262 1,939.8 97,657

27 80,554 1,976.0 97,985

 28 74,859 2,012.8 99,383

29 79,916 2,044.1 98,213

30 82,091 2,069.8 99,850

31 87,893 2,097.9 99,877

32 86,040 2,128.4 101,169

33 81,814 2,163.2 99,922

34 84,304 2,211.9 102,186

35 85,062 2,265.1 102,657

36 81,125 2,325.0 104,522

37 92,036 2,358.7 103,445

38 92,467 2,405.6 106,050

39 89,461 2,452.2 105,963

40 91,508 2,505.1 107,644

41 95,508 2,560.7 106,425

42 95,871 2,572.9 108,741

43 96,594 2,586.6 108,897

44 88,836 2,612.7 110,260

30.
The Excel spreadsheet linked below contains the simple regressions of total new car sales (in
millions of dollars) on each of two independent variables: "compensation" (in billions of dollars)
and "employment level in the non-agricultural sector" (in thousands) .

Which of the following independent variables explains more than 90 percent of the observed
variation in new car sales?

Compensation only.


Employment level only.



Both independent variables.



Neither independent variable.

31.
The regression analysis below relates the value of new car sales (in millions of dollars) to
compensation (in billions of dollars) and the employment level in the non-agricultural sector (in
thousands) for 44 consecutive quarters.

Which of the two independent variables is statistically significant at the 0.05 level?

Compensation only.


Employment level only.



Both independent variables.



Neither independent variable.

 32.
The regression analysis below relates the value of new car sales (in millions of dollars) and the
independent variables "compensation" (in billions of dollars) and "employment level in the non-
agricultural sector" (in thousands) for 44 consecutive quarters. Compare this multiple regression
to the simple regressions with compensation and employment level as the respective independent
variables.

Which of the following is the likely culprit of the dramatic increase in the p-value for
employment level in the multiple regression?

Multicollinearity.


Heteroskedasticity.


Nonlinearity.


None of the above.

34.
A new blood pressure treatment is being tested. The regression analysis below describes the
relationship between the 41 test subjects' diastolic blood pressure and the dummy variable
"medication." When a test subject is taking the new drug, the value of medication is 1, when not,
the value of medication is 0.

Which of the following can be inferred from the regression analysis?

The medication has no statistically significant effect (at a 0.01 significance level).


The use of medication accounts for around 42% of the variation in diastolic blood pressure.


On average, test subjects taking the medication report a diastolic blood pressure level about 5
points lower than those not taking the medication.


None of the above.

Medication
Blood (1 = Yes,
Pressure 0 = No)
68 1
72 1
74 1
78 0
78 1
78 1
78 1
78 1
79 1
80 0
80 1
80 1
80 1
80 1
80 1
80 1
80 1
82 0
82 1
82 1
82 1
82 1
82 1
82 1
83 1
 84 1
84 1
84 1
86 1
86 1
88 0
88 0
88 0
88 0
88 1
90 0
90 0
90 1
90 1
90 1
92 0

35.

In a regression analysis, a residual plot is:

A scatter diagram that plots the values of the residuals against the values of the dependent
variable.

A scatter diagram that plots the values of the residuals against the values of an independent
variable.

A histogram that plots the frequency of certain value ranges of the residuals.

None of the above.

36.

In a regression analysis, if a new independent variable is added and R-squared increases and
adjusted R-squared decreases precipitously, what can be concluded?

The new independent variable improves the predictive power of the regression model.


The new independent variable does not improve the predictive power of the regression model.

The regression was performed incorrectly. It is impossible for R-squared to increase and adjusted
R-squared decrease simultaneously.

The new independent variable's coefficient is not significant at the 0.01 level.

37.
The table below displays data the First Bank of Silverhaven (FBS) has collected on the personal
savings accounts of its job-holding customers. The table includes data on the distribution of the
number of accounts held by Homeowners vs. Non-Homeowners, and by whether the customer is
Self-Employed or is Employed by a firm in which he or she does not have an ownership stake.

What is the probability that a given account-holder is self-employed?

15%


12%


3%


None of the above.

38.
The table below displays data the First Bank of Silverhaven (FBS) has collected on the personal
savings accounts of its job-holding customers. The table includes data on the distribution of the
number of accounts held by Homeowners vs. Non-Homeowners, and by whether the customer is
 Self-Employed or is Employed by a firm in which he or she does not have an ownership stake.

What is the conditional probability that an account-holder is employed by a firm in which he or

she does not have an ownership stake, given that the account-holder is a homeowner?

82.9%


68.2%


59.5%


None of the above.

39.
The table below displays data the First Bank of Silverhaven (FBS) has collected on the personal
savings accounts of its job-holding customers. The table includes data on the distribution of the
number of accounts held by Homeowners vs. Non-Homeowners, and by whether the customer is
Self-Employed or is Employed by a firm in which he or she does not have an ownership stake.

Which of the following statements is true?


Home ownership and employment status are statistically dependent.


The fact that a given personal account is held by a homeowner tells us nothing about the account
holder's employment status.


The fact that a given personal account is held by a self-employed person tells us nothing about
the account-holder's home ownership status.


None of the above.

40.
Electronics manufacturer SE must decide whether or not to invest in the development of a new
type of battery. If the development succeeds, the market for the battery may be large or small. If
it doesn't succeed, the development efforts may or may not generate minor innovations that
would offset some of the battery's development costs. The tree below summarizes the decision.

What is the expected monetary value of developing the new battery?



-$0.5 million


$0.9 million


$1.5 million

41.
Electronics manufacturer SE must decide whether or not to invest in the development of a new
type of battery. If the development succeeds, the market for the battery may be large or small. If
it doesn't succeed, the development efforts may or may not generate minor innovations that
would offset some of the battery's development costs. The tree below summarizes the decision.

The EMV of developing the new battery is $300,000. Based on EMV, SE should develop the
battery. If the manager chooses not to develop the battery, which of the following best describes
the manager's attitude towards this decision?


Risk averse.


Risk neutral.


Risk seeking.


Cowardly.

42.
Electronics manufacturer SE must decide whether or not to invest in the development of a new
type of battery. If the development succeeds, the market for the battery may be large or small. If
it doesn't succeed, the development efforts may or may not generate minor innovations that
would offset some of the battery's development costs. The tree below summarizes the decision.

The EMV of developing the new battery is $300,000. Based on EMV, SE should develop the
battery. For what values of p = Prob[success] does developing the battery have a lower EMV
than not developing the battery?


p < 25%

p > 25%

p < 75%


None of the above.

43.
Electronics manufacturer SE must decide whether or not to invest in the development of a new
type of battery. If the development succeeds, the market for the battery may be large or small. If
 it doesn't succeed, the development efforts may or may not generate minor innovations that
would offset some of the battery's development costs. The tree below summarizes the decision.

The EMV of developing the new battery is $300,000. Based on EMV, SE should develop the
battery. Given that development is successful, for what endpoint values for a large market does
developing the battery have a lower EMV than not developing the battery?

Less than $1.2 million.



Greater than $1.2 million.



Greater than $0.75 million.



None of the above.

 44.
The manager of the Eris Shoe Company must decide whether or not to contract a controversial
sports celebrity as its spokesperson. The new spokesperson's value to Eris depends heavily on
consumers' perception of him. An initial decision analysis based on available data reveals that
the expected monetary value of contracting the new spokesperson is $260,000. For $50,000 Eris
can engage a market research firm that will help Eris learn more about how consumers might
react to the celebrity. The EMV of buying this sample information (assuming it is free) for this
decision is $300,000.

The tree below summarizes Eris's decision. Based on EMV analysis, Eris's manager should:

Hire the research firm.



Not hire the research firm, but contract the new spokesperson.


Not hire the research firm and not contract the new spokesperson.


The answer cannot be determined from the information provided.