Name: __________________________________ Date: ____________

SURVEYS & EXPERIMENTS

1. Near election time, a polling organization iii. In the above study, what does the
increases the sample size from about 1500 phrase ‘not statistically significant’
people to about 4000 people. The purpose mean?
of this is to
a. the difference was not enough to
a. reduce the bias of the result have an important impact on
b. increase the bias of the result teaching practice
c. reduce the variability of the result b. the students scored the same on
d. increase the variability of the result the achievement tests
e. increase the confidence interval c. the p-value was very small
d. the study was poorly designed
2. In an effort to study learning retention e. the observed difference could
among hyperactive students, an elementary plausibly have arisen by chance
school teacher separated all third grade
students in her school into two groups. 3. Which of the following is true about a
One group consisted of students who were double blind experiment?
previously identified as hyperactive. The
second group did not meet the definition a. neither the control group members
of hyperactive. Each student was given a nor the placebo group members know
reading achievement test and an arithmetic which group they are in
achievement test at the end of the school b. the evaluators do not know which
year. Each student was given the same test group has which participants
at the beginning of the next school year. c. both a and b are true
The differences in test scores for each d. either a or b, but not both, are true
student were compared. The differences e. neither a nor b is true; it means the
were not statistically significant. participants in the experiment are blind

i. Which of the choices below best 4. 16 mice and 16 rats that were considered
describes the above study? to be overweight were subjects in an
experiment to determine the effectiveness
a. a convenience sample of a new diet drug. 8 of the mice and 8 of
b. an observational study, not an the rats were chosen at random to receive
experiment the drug. The remaining mice and rats were
c. an experiment, but not a double given a placebo. The researchers who actu-
blind experiment ally worked with the animals did not know
d. a double blind experiment which group received the placebo. The
e. a stratified sample change in weight was measured for each
subject. The design of this experiment is
ii. In the above study, what is the
explanatory variable? a. completely randomized with one
factor, drug
a. the achievement tests b. completely randomized with one
b. the time of the year the tests factor, species
were taken c. randomized block, blocked by drug
c. the grade in school and species
d. whether a student was hyperactive d. randomized block, blocked by drug
e. whether the test was reading or e. randomized block, blocked by species
arithmetic

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 31
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS

5. To estimate the value of a population 9. A surveyor selects every tenth person

parameter, a statistician will most often entering a local shopping mall on Thursday
morning. Will this give a simple random
a. take a census sample of local shoppers?
b. take a sample
c. do an experiment a. no, because the surveyor did not
d. do an observational study use stratification
e. take a stratification b. yes, because each shopper has an
equal chance of being selected
6. Which of the following requires c. no, because some shoppers would
manipulation of the independent variable? be under-represented on Thursday
mornings
a. a survey d. yes, this is a systematic sample
b. an observational study e. no, the surveyor should have used
c. an experiment cluster sampling
d. a census
e. a sample 10. Which of the following statements is true?

7. Which of the following is not a potential I. Voluntary responses often over

source of bias in a survey? represent people with strong negative
opinions.
a. interviewer bias II. Questions with neutral wording
b. stratification generally do not have response bias.
c. choosing a non-random sample III. Stratified samples usually have
d. under representation of a population undercoverage bias.
e. asking the questions in a randomized
order a. I only
b. II only
8. A television network asked its viewers to
c. III only
vote for the best team in college football. It
d. I and II only
gave a telephone number that viewers
e. I and III only
could call to vote for their favorite team.
Each caller was charged $1.00 for their call
11. Consider the following situation. 40% of
which was donated to a national charity. Of
the students at a college are women. In a
the 23,000 people who responded, 53%
random sample of 60 students 31 are
voted for Team A. What does this survey
women. This is an example of
show?
a. sampling error
a. this was a legitimate sample since it
b. response bias
is so large
c. hidden bias
b. this survey was drawn from interested
d. human mistake
viewers who would be
e. voluntary error
knowledgeable about football
c. the results are unclear since the
location of the viewers is not known
d. the study is valid since the population
is at least 10 times as big as the sample
e. the survey is not valid because the
responses were voluntary
Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 32
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS

12. A newspaper reporter takes a straw poll on 14. In a stratified random sample,
the main shopping district regarding how
people feel about a city council decision to a. every member of the population has
build a new park. She interviews the first an equal chance of being selected
ten people that are willing to talk with her. b. every group has an equal chance of
being selected
i. What is the population of the c. every member of the population has
above poll? a known chance of being chosen
d. the population is divided into groups
a. all the residents of the town with similar characteristics, then a
b. all people who are shopping random sample is taken from each
that day group and combined with the others
c. all people who want a park e. every nth person is chosen
d. the ten people who talk to her
e. the city council 15. In a study on productivity, researchers
found that when music was played while
ii. What is the sample of the above poll? people were working, there was an increase
in productivity. However, when the music
a. all the residents of the town was no longer played, productivity again
b. all people who are shopping increased. Assuming that the workers
that day knew a study was in progress, this is an
c. all people who want a park example of
d. the ten people who talk to her
e. the city council a. Simpson’s paradox
b. sampling error
13. An education researcher is forming a ran- c. a poorly controlled experiment
dom sample of three students from the d. the placebo effect
eight listed below. e. single blind bias

1. Adam 2. Bart 3. Lisa 16. Which of the following is a good reason for
4. Holly 5. Sarah 6. Kiana randomly allocating units to treatments in
7. Vinh 8. Shravi an experiment?

a. it produces experimental groups that

Starting from the beginning of the
are similar in major respects
following list of random digits, which
b. it keeps the response variable within
group will be formed?
the bounds of the experiment
95457 53497 23894 57662 80181 33075
c. the independent variable is not known
95320 41475 84950
d. it creates a double blind experiment
e. it matches the population
a. Sarah, Holly, Kiana
b. Adam, Sara, Holly
17. If a design systematically favors an out-
c. Sarah, Holly, Vinh
come, it is called
d. Shravi, Sarah, Holly,
e. Lisa, Vinh, Kiana
a. bias
b. a treatment
c. a control
d. a cure
e. sampling error
Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 33
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS

18. Which of the following are true about an 21. A control is

experiment that uses matched pairs
as its design? a. understanding the idea behind the
treatment
I. Each subject of a pair receives a b. a group with which the effects of the
different treatment and any differences outside environment are controlled
in response is noted. c. a part of the experiment which helps
II. Each subject can only receive one one decide if the experiment was
treatment. affected by other sources
III. Matched pair is a form of blocking. d. a and c
e. b and c
a. I only
b. II only 22. Which is not a principle of experimental
c. III only design?
d. I and II only
e. I and III only a. control
b. sampling
19. What is the difference between an c. randomization
observational study and an experiment? d. replication
e. all of these are the principles
a. in an observational study, the most
important factor is keeping the 23. Which of these do not describe a matched
observer hidden; in an experiment, pair design?
the researcher can be seen
b. in an experiment, all the data is a. compares treatments
obtained by control, while in an b. it is not completely random
observational study, the data is c. replication is not necessary
obtained by very precise observations d. all three are part of a matched pair
c. in an observational study, no treatment e. none of these are part of a matched pair
is imposed; in an experiment the
subjects have some treatment 24. Which of these is a simple random sample?
imposed on them
d. in an experiment, the researcher a. assign numbers to each subject, then
chooses the treatment; in an use a random digits table to pick 10
observational study the subjects of them
choose the experiment b. pick the first 10 people that you see
e. there is no difference; they are both walking in the store
valid statistical research methods c. randomly selecting 1 subject out of
every group of 10
20. A treatment is d. separate subjects into groups with
similar traits, then pick one randomly
a. a part of an experiment that applies
from every group
something onto the subjects
e. place all the names on one piece of
b. a group which receives a placebo
paper and randomly pick 10 names
c. a group which gets the cure
d. an outcome of the experiment
e. the difference in two outcomes
between the group getting the drug
and the placebo
Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 34
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS

25. Sampling variability is ii. What is the population in the

survey above?
a. getting different answers due to bias
b. getting wrong answers due to bias a. the members of the school
c. getting unreliable answers due to bias newspaper
d. getting different answers due to b. all students of high school age
natural variability c. all students who enter school
e. getting unreliable answers due to through the main entrance
natural variability d. the ten students who volunteered
to answer the questions
26. What are confounded variables? e. all students at Cleveland HS

a. when more than one variable may iii. What is the sample in the survey
be affecting the response variable above?
b. when the response variable is affected
by two different explanatory variables a. the members of the school
c. variables that have a hidden effect on newspaper
the response variable b. all students of high school age
d. variables collected through a biased c. all students who enter school
sampling design through the main entrance
e. none of the above d. the ten students who volunteered
to answer the questions
27. What will happen if you send out question- e. all students at Cleveland HS
naires and ask the members of the club to
send in their responses? 29. A call-in poll conducted by a sports network
concluded that Americans believed that
a. nonresponse bias Notre Dame had the best football team in
b. response bias the country. It was later reported that 5,640
c. voluntary sample bias of the 7,880 calls for the poll came from the
d. undercoverage error offices of a company owned by one man, a
e. sampling error Notre Dame alumni. The results of this poll
are probably what?
28. In order to get an opinion of the students at
Cleveland HS, a reporter for the school a. surprising, but reliable, since it was
newspaper interviews the first 10 students conducted by a nationally recognized
who enter the school main entrance and organization
are willing to participate in the survey. b. surprising, but reliable, since everyone
had an equal chance of calling in as
i. What is the method of sampling used many times as he or she wanted to
above called? c. biased, but only slightly since the
sample size was quite large
a. simple random sample d. biased, understating the popularity
b. voluntary response of Notre Dame
c. a census e. biased, overstating the popularity
d. convenience sampling of Notre Dame
e. stratified random sample

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 35
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

1. The pet food company is trying out three new brands of cat food. Set up the procedure for
an experiment to determine which of the three new cat foods produces a shinier coat.

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 36
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

2. A certain city official wants to know the opinion of the public on the new city tax law.
Proposals are made for two types of surveys. In one setting, workers will be sent out to
major malls and every third person who wants to answer the survey will have the chance
to fill out a questionnaire. In the other setting, phone numbers will be generated by a random
number generator and workers will call residents, asking them to answer the survey. Discuss
the pros and cons of using either one of the two surveys.

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 37
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

3. A drug company wants to test a new drug that is supposed to reduce cholesterol levels.
Design an experiment for testing this drug in males and females ages 40 to 65.

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 38
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

4. You have been assigned to test the effectiveness of a new drug. Your supervisor wants to
see if there is a difference in the results based on two levels of dosages. The low dosage is
5 mg/kg of body weight. The high dose is 10 mg/kg of body weight. You have 15 mice to use
for the experiment and you also decide to use a placebo.

a. Describe what kind of study you would conduct. Explain the protocol you would use.

b. Use the row given from a random digits table to show how you would assign the mice
to the various treatment groups you chose in part a.
random digits table: 32744 74466 77190 56016 68326 76817 23540

c. List any lurking variables that might occur in this study.

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 39
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

5. A study of a new type of baby food used 40 randomly selected babies 2 through 8 months old.
Half of the group was fed the new food, while the other half was fed the old.

a. Is this an observational study or an experiment?

b. How should this study be blocked?

c. What is the reason for blocking within this study?

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 40
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

6. For each of the following questions, describe what is wrong with them and rewrite
when necessary.

a. Do you prefer the juicy hamburger at Hamburger Palace or the one in Hamburger Stop?

b. Senior citizens usually have bad vision. Do you think they should be allowed to drive?

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 41
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

7. What is wrong with collecting data in each of the following situations?

a. At a Democratic convention: Who will you vote for in the next presidential election?

b. At a mall in an affluent neighborhood: Do you think the gasoline tax is fair?

c. At a health club: What is your favorite type of food?

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 42
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

8. You have been engaged to determine whether there is a statistically significant difference
in the salaries of men and women at a large university. You have alphabetized lists of the
975 male faculty members and of the 630 female faculty members at the university.

Line
112 07817 28218 79856 92460 48978 92545 52751 41071 79167 98274 25389 33465
113 00544 15554 63346 85431 37506 05926 02380 33498 27797 84077 18802 15335
114 23648 52381 32299 28244 89847 29119 10425 02268 37641 54302 45237 93858
115 15640 50453 98416 23095 78022 14752 08755 75287 83056 35082 94180 86969
116 06766 79935 46361 78771 90133 15746 80427 26218 48184 83115 44931 18876

a. Explain how you would assign labels and use line 113 of the random number table
shown to choose a stratified random sample of 200 female and 200 male faculty members.

b. What are the labels of the first five males and the first five females in your sample?

c. Which group of faculty members, male or female, will have a narrower 95% confidence
interval for the mean salary?

d. If a preliminary sample has a sample standard deviation of $2,175, how many male
faculty members must be surveyed so that the mean salary confidence interval is within
$500 of the true mean at a 90% level?

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 43
Name: __________________________________ Date: ____________
SURVEYS & EXPERIMENTS • Free Response

9. A city contains 41 supermarkets. An inspector wants to check compliance with a new city
ordinance banning minors from working late hours. Because of the time required, he can
inspect only 10 markets. He decides to choose a stratified random sample and stratifies the
markets by sales volume. Stratum A consists of 4 large chain stores; the inspector decides
to inspect 3 of them. Stratum B consists of 12 smaller chain stores; 4 out of the 10 will be
inspected. Stratum C consists of 24 locally owned small stores; 3 of these 20 will be inspected.
Let "Yes" mean that the store is in compliance and "No" mean that it is not. The population
(unknown to the inspector, of course) is as follows.

Stratum Stratum Stratum

A B C
Store # Compliance Store # Compliance Store # Compliance Store # Compliance
1 Yes 1 No 1 Yes 13 Yes
2 Yes 2 Yes 2 Yes 14 Yes
3 No 3 No 3 No 15 No
4 Yes 4 No 4 Yes 16 Yes
5 Yes 5 Yes 5 No 17 Yes
6 No 6 No 6 No 18 No
7 Yes 7 No 19 No
8 No 8 Yes 20 No
9 No 9 No 21 Yes
10 Yes 10 No 22 Yes
11 Yes 11 Yes 23 No
12 No 12 No 24 No

Line
112 07817 28218 79856 92460 48978 92545 52751 41071 79167 98274 25389 33465
113 00544 15554 63346 85431 37506 05926 02380 33498 27797 84077 18802 15335
114 23648 52381 32299 28244 89847 29119 10425 02268 37641 54302 45237 93858
115 15640 50453 98416 23095 78022 14752 08755 75287 83056 35082 94180 86969
116 06766 79935 46361 78771 90133 15746 80427 26218 48184 83115 44931 18876

a. Use the random number table to choose a stratified random sample of size
10 allotted among the strata as described above. (Begin on line 112)

b. Use your sample results to estimate the proportion of the entire population
of stores that are in compliance with the ordinance.

c. Use the description of the population given above to find the true proportion
of stores in compliance. How does this compare with the estimate in part (b)?

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 44
Answer Key

Surveys & Experiments

1. c 5. b 11. a 16. a 22. b 28i. d
2i. b 6. c 12i. a 17. a 23. c ii. e
ii. d 7. e ii. d 18. a 24. a iii. d
iii. e 8. e 13. c 19. c 25. d 29. e
3. c 9. c 14. d 20. a 26. a
4. e 10. d 15. d 21. e 27. c

Surveys & Experiments • Free Response

1. As the cats are brought in to participate in the exper- c. Lurking variables might be the health status of the
iment, use a RDT to designate their group. Let digits mice prior to the experiment, the living conditions
1, 2, 3 get cat food #1, 4, 5, 6 get cat food #2, 7, 8, 9 of the different groups of mice, or a difference in the
get cat food #3, and ignore all zeros. Once a group is body weights. To avoid these, all mice should be
filled to the number of participants you are looking picked to be similar in those categories.
for, ignore the digits corresponding to that group. 5a. This is an experiment since baby food will be
Set up a scale for the coat’s shine and, after the administered to the participants.
experiment is finished, label each subject within b. The study should be blocked by weight.
each group with the appropriate number on the c. Since we are testing the food, we may want to look
scale. Make sure that all cats get equal treatment at the weight gain in infants within the experimental
to assure equivalency in all other life aspects. group. Children of different weight categories
2. The problem with the first survey is that asking at should therefore be separated to eliminate the bias
malls eliminates all those people who do not ever in weight gain.
go to malls. This is an undercoverage bias, which (Note: the answers may vary. Other blocks may
means that certain people in the population will not include age, gender, health status, etc. Each answer
be participating in the survey. Using phone numbers must contain an appropriate explanation.)
is a better approach, since phone numbers are gen- 6a. It is biased toward Hamburger Palace by mentioning
erated randomly, but it also has an undercoverage that their hamburgers are juicy. Rewrite: Which ham-
with those people who do not own a phone or do burger do you prefer, the one at Hamburger Palace
not answer it. or at Hamburger Stop?
3. Two groups must be created for this study, one b. It creates bias by stating that senior citizens usually
taking the new drug and the other taking the have bad vision. Rewrite: Do you think senior citi-
placebo. The groups also need to be blocked by zens should be allowed to drive?
gender, with the same number of males and females 7a. Only looks at the democrats who are active and will
assigned to each group. First take all the males, and most likely vote for a democrat.
using RDT, assign them to either of the two groups b. Only affluent people will be at the mall; they are
(even and odd digits can be used to choose the likely to have similar views on taxes.
group), then repeat the procedure for the females. c. People who attend a health club usually eat
The experiment must also be double blinded to healthier food than the general population.
reduce the bias. 8a. Number the male faculty members 1-975 and the
4a. This would be an experiment. For this study the female faculty members 1-630. To select male facul-
mice would be divided into three groups of five mice ty members, read the random numbers in groups of
each. One group would receive the low dosage, one three digits. Each time a number from 1 to 975
the high dosage, and one the placebo. Use RDT, dig- occurs, select the name associated with that num-
its 1, 2, 3 for the low dosage, 4, 5, 6 for the high ber. For female faculty members, read the random
dosage, and 7, 8, 9 for the placebo. Ignore all zeros. numbers in groups of three digits. Each time a num-
Once a group contains five mice, ignore all the digits ber occurs from 1 to 630, select the female faculty
corresponding to that group. member with that number.
b. Mouse # Group Mouse # Group b. Starting on line 113, the first five males chosen are
1 low 9 high 005, 441, 555, 463, and 346. Continuing on from that
2 low 10 plac
3 plac 11 plac point, the first five females will be 313, 605, 023, 349,
4 high 12 low and 215.
5 high 13 plac
6 plac 14 low c. Since the sample sizes are the same between the
7 high 15 low two groups, the 95% confidence interval will be the
8 high
same width, assuming the sample standard devi-
In this case, the last two mice automatically went ations are the same. Otherwise, the sample with
into the low group as the other two groups were the smaller sample standard deviation will have
filled. For example, the tenth digit was skipped the narrower confidence interval.
because it would have placed the mouse into the
high group, which already had 5 mice.

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 93
Answer Key

 (1.645)(2175) 
2 c. Only 19 of the 41 stores in the city are actually in
d. n =  = 51.02 ≈ 52 people compliance. The true proportion is 0.46, which is
 500  much lower than the estimate of 0.90.
9a. Beginning on line 112, the first three stores chosen
from Stratum A – large chain stores – are 1, 2, and 5.
Continuing on, the four stores chosen from the sec-
ond group are 10, 02, 03, and 05. The third sample
group consists of 04, 02, 21, and 08.
b. From the sample chosen in part a, 9 of the 10 stores
are in compliance with the city ordinance, which
gives a sample proportion of 0.90.

Probability & Discrete Distributions

1i. b ii. b 15. b 25. c ii. c 42. c
ii. a iii. b 16. d 26. d 34. a 43. e
iii. d iv. c 17. e 27. d 35. c 44i. c
2i. b 7. d 18i. b 28. b 36. a ii. e
ii. a 8. b ii. c 29. a 37i. b 45. b
3. e 9. a 19. e 30. a ii. d 46. a
4. d 10. c 20. c 31i. a 38i. a 47. c
5i. b 11. d 21. c ii. e ii. e 48. d
ii. c 12. e 22. a iii. b 39. c 49i. e
iii. b 13. d 23. d 32. e 40. c ii. c
6i. d 14. e 24. a 33i. d 41. e iii. d

Probability & Discrete Distributions • Free Response

1a. X (sum) 2 3 4 5 6 7 8 3. P(6) = 1/6

P(X) 1/16 2/16 3/16 4/16 3/16 2/16 1/16 P(6 four times) = (1/6)·(1/6)·(1/6)·(1/6) = 1/1296
1 2 3 4 3 2 1 P(not 6) = 5/6
b. 2 ? }} + 3 ? }} + 4 ? }} + 5 ? }} + 6 ? }} + 7 ? }} + 8 ? }} = 5 P(not 6 four times) = (5/6)·(5/6)·(5/6)·(5/6) =
16 16 16 16 16 16 16
625/1296 = 0.482
c. Variance = (2 2 5)2 ? }1} + (3 2 5)2 ? }2} + (4 2 5)2 ? }3} + a. The person who got four 6s in a row has a very
16 16 16
4 3 2 small chance of getting such results, so his die may
(5 2 5)2 ? }} + (6 2 5)2 ? }} + (7 2 5)2 ? }} + be loaded. There is still a probability that the die is
16 16 16
1 not loaded, but it is very small.
(8 2 5)2 ? }} = 2.5 b. The person who did not get any 6s has a large
16
chance of such occurrence so he probably does
Standard Deviation = 1.58 not have a loaded die.
d. P(4 at least once) = 1 – P(no 4 in three rolls) = 4a. The mean for the downtown locations is $12,250
2
 3 4069 with a s.d. of $73,374.
1−   = = 0.9934 b. 40% of the time, the suburban location shows
 16  4096
a profit.
2a. Expected Value = 1/0.03 = 33.3 c. The mean payoff for the suburban location is
It will take approximately 33 Saturdays to catch $21,500. It should be the chosen location.
the correct butterfly. d. (x)(0.20) + (95000)(0.20) + (-15000)(0.30) +
b. Saturday Probability of (-250000)(0.30) = 0
# catching on nth
Saturday x = $302,500
1 0.03 4 - 8.83
2 0.0291 5a. z = }} = -3.22; P(z , -3.22) = 0.0013, or 0.13%
1.5
3 0.0282
x-8.83
4 0.0274 b. Inverse z score = -1.88; -1.88 = }} ⇒ x = 6.01
5 0.0266 1.5
c. P = (0.95)(0.95)(0.95)(0.95)(0.95)(0.05) = 0.0387

Copyright © 2000 William K. Bradford Publishing Company. All rights reserved. Permission to reproduce this master is granted to registered purchasers for their classroom use. Printed in the United States of America. 94