Chapter 4

Introduction to Probability

Solutions

1. Number of experimental outcomes = (3)(2)(4) = 24

2.

ABC ACE BCD BEF

ABD ACF BCE CDE

ABE ADE BCF CDF

ABF ADF BDE CEF

ACD AEF BDF DEF

6!
P36  (6)(5)(4) 120
3. (6  3)!

BDF BFD DBF DFB FBD FDB

4. a.

1st Toss 2nd Toss 3rd Toss

H (H,H,H)
T
H (H,H,T)
T
H (H,T,H)
H
T
(H,T,T)

T H (T,H,H)
T
H (T,H,T)
T
H (T,T,H)
T
(T,T,T)

b. Let: H be head and T be tail:

 (H,H,H) (T,H,H)

(H,H,T) (T,H,T)

(H,T,H) (T,T,H)

(H,T,T) (T,T,T)

c. The outcomes are equally likely, so the probability of each outcome is 1/8.

5. P(Ei) = 1/5 for i = 1, 2, 3, 4, 5

P(Ei) ≥ 0 for i = 1, 2, 3, 4, 5

P(E1) + P(E2) + P(E3) + P(E4) + P(E5) = 1/5 + 1/5 + 1/5 + 1/5 + 1/5 = 1

The classical method was used.

6. P(E1) = .40, P(E2) = .26, P(E3) = .34

The relative frequency method was used.

7. No. Requirement (4.4) is not satisfied; the probabilities do not sum to 1.

P(E1) + P(E2) + P(E3) + P(E4) = .10 + .15 + .40 + .20 = .85

8. a. There are four outcomes possible for this two-step experiment; planning

commission positive—council approves; planning commission positive—council

disapproves; planning commission negative—council approves; and planning

commission negative—council disapproves.

b. Let p = positive, n = negative, a = approves, and d = disapproves

 Planning Commission Council
(p, a)
a

d
p
(p, d) .

n
(n, a)
a

(n, d)

9.

10. a.

Programmer Total Lines of Code Number of Lines of Code Probability

Written Requiring Edits

Liwei 23,789 4,589 0.1929

Andrew 17,962 2,780 0.1548

Jaime 31,025 12,080 0.3894

Sherae 26,050 3,780 0.1451

Binny 19,586 1,890 0.0965

Roger 24,786 4,005 0.1616

 Dong-Gil 24,030 5,785 0.2407

Alex 14,780 1,052 0.0712

Jay 30,875 3,872 0.1254

Vivek 21,546 4,125 0.1915

b. Probability = 4589 / 23789 = 0.1929

c. Probability = 1 – 3780 / 26050 = 1 – 0.1451 = 0.8549

d. The lowest probability is Alex at 0.0712; the highest probability

is Jaime at 0.3894.

11. a. P(adult in the Tri-State Region smokes) = 118/(118+492) = 118/610 = .1934

b. P(adult in KY smokes) = 47/223 = .2108

P(adult in IN smokes) = 32/166 = .1928

P(adult in OH smokes) = 39/221 = .1765

For these three states, Ohio has the lowest estimated probability of an adult being

a smoker.

12. Initially a probability of .20 would be assigned if selection is equally likely. Data do

not appear to confirm the belief of equal consumer preference. For example, using the

relative frequency method we would assign a probability of 5/100 = .05 to the design

1 outcome, .15 to design 2, .30 to design 3, .40 to design 4, and .10 to design 5.

13. a. Step 1—Use the counting rule for combinations:

(696)= 5 ! 69−5
69 !
¿
! ¿=
(69)(68)(67)(66)(65)
(5)(4)(3)(2)(1)
=11,238,513

Step 2—There are 26 ways to select the red Powerball from digits 1 to 26.

Total number of Powerball lottery outcomes:

 (11,238,513) x (26) = 292,201,338

b. Probability of winning the lottery: 1 chance in 292,201,338

= 1/(292,201,338) = .00000000342

14. a. P(E2) = ¼

b. P(any 2 outcomes) = 1/4 + 1/4 = 1/2

c. P(any 3 outcomes) = 1/4 + 1/4 + 1/4 = 3/4

15. a. S = {ace of clubs, ace of diamonds, ace of hearts, ace of spades}

b. S = {2 of clubs, 3 of clubs, . . . , 10 of clubs, J of clubs, Q of clubs, K of clubs, A of

clubs}

c. There are 12; jack, queen, or king in each of the four suits.

d. For a: 4/52 = 1/13 = .08

For b: 13/52 = 1/4 = .25

For c: 12/52 = .23

16. a. (6)(6) = 36 sample points

b.

Die 2
1 2 3 4 5 6

1 2 3 4 5 6 7

2 3 4 5 6 7 8 Total for Both .

3 4 5 6 7 8 9
Die 1

4 5 6 7 8 9 10

5 6 7 8 9 10 11

6 7 8 9 10 11 12
 c. 6/36 = 1/6

d. 10/36 = 5/18

e. No. P(odd) = 18/36 = P(even) = 18/36 or 1/2 for both.

f. Classical. A probability of 1/36 is assigned to each experimental outcome.

17. a. (4,6), (4,7), (4,8)

b. .05 + .10 + .15 = .30

c. (2,8), (3,8), (4,8)

d. .05 + .05 + .15 = .25

e. .15

18. a. Let C = corporate headquarters located in California:

= 53/500 = .106

b. Let N = corporate headquarters located in New York

T = corporate headquarters located in Texas

P(N) = 50/500 = .100

P(T) = 52/500 = .104

Located in California, New York, or Texas:

c. Let A = corporate headquarters located in one of the eight states

Total number of companies with corporate headquarters in the eight states = 283

P(A) = 283/500 = .566

More than half the Fortune 500 companies have corporate headquartered located

in these eight states.

19. a. Probability = 131/(134 + 131 + 2) = 131/(267) = .4906

 b. Probability = (432)/(293 + 432 + 8) = 432/733 = .5894

c. Probability = (134 + 293) /(267 + 733) = 427/1,000 = .427

d. Older respondents appear to be less concerned about global warming being a

threat in their lifetime than are younger respondents.

20. a.

Age

Experimental Financially Number of Probability

Outcome Independent Responses

E1 16 to 20 191 191/944 = 0.2023

E2 21 to 24 467 467/944 = 0.4947

E3 25 to 27 244 244/944 = 0.2585

E4 28 or older 42 42/944 = 0.0445

944

b.

c.

d. The probability of being financially independent before age 25, .6970, seems high

given the general economic conditions. The teenagers who responded to this

survey may have had unrealistic expectations about becoming financially

independent at a relatively young age.

1416
21. a. P ( pole , post )= =.1358
10426

896
b. P ( guard rail ) = =.0859
10426

c. The fixed object that is least likely to be involved in a fatal collision is a bridge
 231
with probability P ( bridge )= =.0222
10426

d. The fixed object that is most likely to be involved in a fatal collision is a

2585
shrubbery or tree with probability P ( shrubbery , tree ) = =.2479
10426

22. a. P(A) = .40, P(B) = .40, P(C) = .60

b. P(A  B) = P(E1, E2, E3, E4) = .80. Yes, P(A  B) = P(A) + P(B).

c. Ac = {E3, E4, E5} Cc = {E1, E4} P(Ac) = .60 P(Cc) = .40

d. A  Bc = {E1, E2, E5} P(A  Bc) = .60

e. P(B  C) = P(E2, E3, E4, E5) = .80

23. a. P(A) = P(E1) + P(E4) + P(E6) = .05 + .25 + .10 = .40

P(B) = P(E2) + P(E4) + P(E7) = .20 + .25 + .05 = .50

P(C) = P(E2) + P(E3) + P(E5) + P(E7) = .20 + .20 + .15 + .05 = .60

b. A  B = {E1, E2, E4, E6, E7}

P(A  B) = P(E1) + P(E2) + P(E4) + P(E6) + P(E7)

= .05 + .20 + .25 + .10 + .05 = .65

c. A  B = {E4} P(A  B) = P(E4) = .25

d. Yes, they are mutually exclusive.

e. Bc = {E1, E3, E5, E6};

P(Bc) = P(E1) + P(E3) + P(E5) + P(E6)

= .05 + .20 + .15 + .10 = .50

24. Let E = experience exceeded expectations

M = experience met expectations

a. Percentage of respondents that said their experience exceeded expectations

= 100 – (4 + 26 + 65) = 5%
 P(E) = .05

b. P(M  E) = P(M) + P(E) = .65 + .05 = .70

25. Let F = Uses Facebook

L = Uses LinkedIn

a. P ( F ∪ L )=P ( F )+ P ( L )−P ( F ∩ L ) =.68+.25−.22=.71

b. 1−P ( F ∪ L )=1−.71=.29

26. a. Let D = Domestic Equity Fund

P(D) = 16/25 = .64

b. Let A = 4- or 5-star rating

Thirteen funds were rated 3 star or less; thus, 25 – 13 = 12 funds must be 4 star or 5

star.

P(A) = 12/25 = .48

c. Seven Domestic Equity funds were rated 4 star, and two were rated 5 star. Thus,

nine funds were Domestic Equity funds and were rated 4 star or 5 star:

P(D  A) = 9/25 = .36

d. P(D  A) = P(D) + P(A) - P(D  A)

= .64 + .48 - .36 = .76

27. Let A = the event that a randomly selected U.S. adult uses social media

B = the event that a randomly selected U.S. adult is age 18–29

a. P(A) = 1 - .35 = .65

b. P(B) = 1 – .78 = .22

c. The question asks for the probability of A intersected with B.

From the addition law,

 So,

28. Let: B = rented a car for business reasons

P = rented a car for personal reasons

a. P(B  P) = P(B) + P(P) - P(B  P)

= .54 + .458 - .30 = .698

b. P(Neither) = 1 – .698 = .302

29. P(E) =

a. P(R) =

P(D) =

b. Yes; P(E  D) = 0

c. Probability =

d. Let F denote the event that a student who applies for early admission is deferred

and later admitted during the regular admission process.

Events E and F are mutually exclusive and the addition law applies.

P(E  F) = P(E) + P(F)

P(E) = .3623 from part (a)

Of the 964 early applicants who were deferred, we expect 18%, or .18(964)

students, to be admitted during the regular admission process. Thus, for the total

of 2,851 early admission applicants:

P(F) =
 P(E  F) = P(E) + P(F) = .3623 + .0609 = .4232

Note: .18(964) = 173.52. Some students may round this to 174 students. If

rounding is done, the answer becomes .4233. Either approach is acceptable.

30. a.

b.

c. No because P(A | B)  P(A)

31. a. P(A  B) = 0

b.

c. No. P(A | B) ≠ P(A);  the events, although mutually exclusive, are not

independent.

d. Mutually exclusive events are dependent.

32. a. Dividing each entry in the table by 500 yields the following (rounding to two

digits):

Yes No Totals

Men 0.210 0.282 0.492

Women 0.186 0.322 0.508

Totals 0.396 0.604 1.000

Let M = 18-34-year-old man,

W = 18-34-year-old woman,

Y = responded yes,

N = responded no

b. P(M) = .492, P(W) = .508

 P(Y) = .396, P(N) = .604

c. P(Y|M) = .210/.492 = .4268

d. P(Y|W) = .186/.508 = .3661

e. P(Y) = .396/1 = .396

f. P(M) = .492 in the sample. Yes, this seems like a good representative sample

based on gender.

33. a.

Undergraduate Major

Business Engineering Other

Intended Enrollment Status Full-Time 0.270 0.151 0.192

Part-Time 0.115 0.123 0.149

Totals 0.385 0.274 0.341

b. P(Business) = 0.385, P(Engineering) = 0.274, and P(Other) = 0.341, so Business is

the undergraduate major that produces the most potential MBA students.

c.

d.

e. Let A denote the event that student intends to attend classes full time in pursuit of

an MBA degree, and let B denote the event that the student was an undergraduate

Business major. Are events A and B independent? Justify your answer.

For independence, we must have that ; from the joint

probability table in part (a) of this problem, we have

 P(A) = 0.613

P(B) = 0.385

So

P ( A ) P ( B )= (.613 ) (.385)=.236

But

Because , the events are not independent.

34. a. Let O = flight arrives on time

L = flight arrives late

J = Jet Blue flight

N = United flight

U = US Airways flight

Given:

P(O | J) = .768 P(O | N) = .715 P(O | U) = .822

P(J) = .30 P(N) = .32 P(U) = .38

Joint probabilities using the multiplication law

With the marginal probabilities P(J) = .30, P(N) = .32, and P(U) = .38 given, the joint

probability table can then be shown as follows.

On Time Late Total

 Jet Blue .2304 .0696 .30

United .2288 .0912 .32

US Airways .3124 .0676 .38

Total .7716 .2284 1.00

b. Using the joint probability table, the probability of an on-time flight is the

marginal probability

P(O) = .2304 + .2288 + .3124 = .7716

c. Because US Airways has the highest percentage of flights into terminal C, US

Airways with P(U) = .38 is the most likely airline for Flight 1382.

d. From the joint probability table, P(L) = .2284:

Most likely airline for Flight 1382 is now United with a probability of .3992. US

Airways is now the least likely airline for this flight with a probability of .2961.

35. a. The total sample size is 1,010. Dividing each entry by 1,010 provides the following

joint probability table.

I Am My Spouse We Are Equal

Husband .2752 .1257 .1010 .5020

Wife .2871 .1099 .1010 .4980

 .5624 .2356 .2020

Let I = I am

S = My spouse

E = We are equal

H = Husband

W = Wife

b. Using the marginal probabilities, P(I) = .5642, P(S) = .2356 and P(E) = .2020. “I

am” is the most likely response. It is over twice as likely as either “My spouse” or

“We are equal.”

c.

d.

e.

f.

36. a. We have that P(Make the Shot) = .93 for each foul shot, so the probability that the

player will make two consecutive foul shots is that P(Make the Shot) P(Make the

Shot) = (.93)(.93) = .8649.

b. There are three unique ways in which the player can make at least one shot: He can

make the first shot and miss the second shot, miss the first shot and make the

second shot, or make both shots. Because the event “Miss the Shot” is the

complement of the event “Make the Shot,” P(Miss the Shot) = 1 – P(Make the

Shot) = 1 – .93 = .07. Thus:

 P(Make the Shot) P(Miss the Shot) = (.93)(.07) = .0651

P(Miss the Shot) P(Make the Shot) = (.07)(.93) = .0651

P(Make the Shot) P(Make the Shot) = (.93)(.93) = .8649

.9951

c. We can find this probability in two ways. We can calculate the probability directly:

P(Miss the Shot) P(Miss the Shot) = (.07)(.07) = .0049

Or we can recognize that the event “Miss both Shots” is the complement of the

event “Make at Least One of the Two Shots,” so

P(Miss the Shot) P(Miss the Shot) = 1 – .9951 = .0049

d. For the player who makes 58% of his free throws, we have:

P(Make the Shot) = .58 for each foul shot,

so the probability that this player will make two consecutive foul shots is

P(Make the Shot) P(Make the Shot) = (.58)(.58) = .3364.

Again, there are three unique ways that this player can make at least one shot: He

can make the first shot and miss the second shot, miss the first shot and make the

second shot, or make both shots. Because the event “Miss the Shot” is the

complement of the event “Make the Shot” P(Miss the Shot) = 1 – P(Make the

Shot) = 1 – .58 = .42. Thus,

P(Make the Shot) P(Miss the Shot) = (.58)(.42) = .2436

P(Miss the Shot) P(Make the Shot) = (.42)(.58) = .2436

P(Make the Shot) P(Make the Shot) = (.58)(.58) = .3364

.8236

We can again find the probability the 58% free-throw shooter will miss both shots
 in two ways. We can calculate the probability directly:

P(Miss the Shot) P(Miss the Shot) = (.42)(.42) = .1764

Or we can recognize that the event “Miss Both Shots” is the complement of the

event “Make at Least One of the Two Shots,” so

P(Miss the Shot) P(Miss the Shot) = 1 – .9951 = .1764

Intentionally fouling the 58% free-throw shooter is a better strategy than

intentionally fouling the 93% shooter.

37. Let C = person could give up cell phone

T = person could give up television

a. P ( C ) =.48

P(T ∩C ) .31
b. P ( T|C )= = =.65
P (C) .48

C
P(T ∩C ) .38
c. P ( T|C ) =
C
= =.73 (Note: C C is the complement of C, or in other
P(C C ) .52

words, person could not give up cell phone.)

d. The probability a person could give up television if they could not give up a cell

phone is higher than the probability a person could give up television if they could

up a cell phone.

38. Let Y = has a college degree

N = does not have a college degree

D = a delinquent student loan

a. From the table,

b. From the table,

c.
 d.

e. Individuals who obtained a college degree have a .3810 probability of a delinquent

student loan, whereas individuals who dropped out without obtaining a college degree

have a .5862 probability of a delinquent student loan. Not obtaining a college degree

will lead to a greater probability of struggling to payback the student loan and will

likely lead to financial problems in the future.

39. a. Yes, because P(A1  A2) = 0

b. P(A1  B) = P(A1)P(B | A1) = .40(.20) = .08

P(A2  B) = P(A2)P(B | A2) = .60(.05) = .03

c. P(B) = P(A1  B) + P(A2  B) = .08 + .03 = .11

d.

40. a. P(B  A1) = P(A1)P(B | A1) = (.20)(.50) = .10

P(B  A2) = P(A2)P(B | A2) = (.50)(.40) = .20

P(B  A3) = P(A3)P(B | A3) = (.30)(.30) = .09

b.

c.

Events P(Ai) P(B | Ai) P(Ai  B) P(Ai | B)

A1 .20 .50 .10 .26

A2 .50 .40 .20 .51

A3 .30 .30 .09 .23

 1.00 .39 1.00

41. S1 = successful, S2 = not successful, and B = request received for additional

information.

a. P(S1) = .50

b. P(B | S1) = .75

c.

42. M = missed payment

D1 = customer defaults

D2 = customer does not default

P(D1) = .05 P(D2) = .95 P(M | D2) = .2 P(M | D1) = 1

a.

b. Yes, the probability of default is greater than .20.

43. Let C = event male patient has prostate cancer

+ = positive PSA test for prostate cancer

– = negative PSA test for prostate cancer

P ( C ) =.02 P ( +¿ CC ) =.75 P (−¿ C ) =.20

a. We are interested in finding P ¿. According to Bayes’ rule:

P¿

b. We are interested in finding P ¿. According to Bayes’ rule:

P¿

c. Here we repeat the calculations from parts a and b, but now P ( C ) =.40 . We are

interested in finding P ¿ and P ¿. According to Bayes’ rule:

 P ¿P ¿

d. The probability of having prostate cancer when the PSA test is positive in part a is

only .0213. The difference between P ¿ and P ¿ in parts a and b is relatively small.

The probability of having prostate cancer when the PSA test is positive is .0213

compared to .0161 when the PSA test is negative. When the prior probability

increases in part c, P ¿ increases to .4156, but P ¿ is also relatively high at .3478.

Results similar to these are why the guidelines for when males are tested for

prostate cancer and how they are treated if they test positive have changed in

recent years.

44. M = the current visitor to the ParFore website is a male

F = the current visitor to the ParFore website is a female

D = a visitor to the ParFore website previously visited the Dillard website

a. Using past history, P(F) = .40.

b. P(M) = .60, , and

The revised probability that the current visitor is a female is .6667.

ParFore should display the special offer that appeals to female visitors.

45. Let A1 = American under 18

A2 = American age 18 to 64

A3 = American age 65 or older

U = Uninsured

a. P ( A 3 ) =1−P ( A 1 )−P ( A 2) =1−.228−.614=.158

b.

The probability that a randomly selected person in the United States is older than
 65 is .158. But if it is known that the person is uninsured, the probability of that

person being older than 65 falls to .019 because most people 65 and older have

health insurance.

46. a. 422 + 181 + 80 + 121 + 201 = 1005 respondents

b. Most frequent response a day or less; probability = 422/1005 = .4199

c. 201/1005 = .20

d. Responses of two days, three days, and four or more days = 181 + 80 + 121 = 382

Probability = 382/1005 = .3801

47. a. (2)(2) = 4

b. Let, S = successful

U = unsuccessful

O Bon
il ds E
S
1

U
S
E2

U
S E3

U
E4

c. O = {E1, E2}

M = {E1, E3}

d. O  M = {E1, E2, E3}

 e. O  M = {E1}

f. No, because O  M has a sample point.

48. a. 0.5029

b. 0.5758

c. No, from part a we have P(F) = 0.5029, and from part b we have P(A|F) = 0.5758.

Because P(F) ≠ P(A|F), events A and F are not independent.

49. Let I = treatment-caused injury

D = death from injury

N = injury caused by negligence

M = malpractice claim filed

$ = payment made in claim

We are given

P(I) = 0.04,

P(N | I) = 0.25,

P(D | I) = 1/7,

P(M | N) = 1/7.5 = 0.1333,

and

P($ | M) = 0.50

a. P(N) = P(N | I) P(I) + P(N | Ic) P(Ic)

= (0.25)(0.04) + (0)(0.96) = 0.01

b. P(D) = P(D | I) P(I) + P(D | Ic) P(Ic)

= (1/7)(0.04) + (0)(0.96) = 0.006

c. P(M) = P(M | N) P(N) + P(M | Nc) P(Nc)

= (0.1333)(0.01) + (0)(0.99) = 0.001333

 P($) = P($ | M) P(M) + P($ | Mc) P(Mc)

= (0.5)(0.001333) + (0)(0.9987) = 0.00067

50. a. Probability of the event = P(average) + P(above average) + P(excellent)

11 14 13
 
= 50 50 50 = .22 + .28 + .26 = .76

b Probability of the event = P(poor) + P(below average)

4 8
 .24
= 50 50

51. a.

Household Income
Level of Education Under $25,000 $25,000 to $49,999 $50,000 to $99,999 $100,000 and over Total
High School graduate 0.1505 0.1519 0.1438 0.0530 0.4993
Bachelor's degree 0.0378 0.0634 0.1168 0.1191 0.3371
Master's degree 0.0104 0.0184 0.0460 0.0624 0.1371
Doctoral degree 0.0012 0.0024 0.0064 0.0164 0.0265
Total 0.2000 0.2361 0.3130 0.2509 1.0000

b. This is the sum of two marginal probabilities.

P(master's degree or more education) = .1371 + .0265 = .1636

c. This is a conditional probability.

d. This is a marginal probability.

P(Under 25) = .2000

e. This is a conditional probability.

f. No. is not equal to P(100 or More) = .2508.

 Household income is not independent of education level. Individuals with a high

school degree have a lower probability of having a higher household income.

52. a.

Yes No Total

23 and Under .1026 .0996 .2022

24 - 26 .1482 .1878 .3360

27 - 30 .0917 .1328 .2245

31 - 35 .0327 .0956 .1283

36 and Over .0253 .0837 .1090

Total .4005 .5995 1.0000 .

b. .2022

c. .2245 + .1283 + .1090 = .4618

d. .4005

53. a. P(24 to 26 | Yes) = .1482/.4005 = .3700

b. P(Yes | 36 and over) = .0253/.1090 = .2321

c. .1026 + .1482 + .1878 + .0917 + .0327 + .0253 = .5883

d. P(31 or more | No) = (.0956 + .0837)/.5995 = .2991

e. No, because the conditional probabilities do not all equal the marginal

probabilities. For instance,

P(24 to 26 | Yes) = .3700 

P(24 to 26) = .3360

54. a. .1485+.2273+.4008 = .7766

b.

c. P ( 50+¿ NOT OKAY )=P ¿ ¿

d. The attitude about this practice is not independent of the age of the respondent.

We can show this in several ways. One example is to use the result from part (b).

We have

and

If the attitude about this practice were independent of the age of the respondent,

we would expect these probabilities to be equal. Because these probabilities are

not equal, the data suggest the attitude about this practice is not independent of the

age of the respondent.

e. Respondents in the 50+ age category are far more likely to say this practice is not

okay than are respondents in the 18–29 age category:

55. a.

We have P(B | S) > P(B).

Yes, continue the ad because it increases the probability of a purchase.

 b Estimate the company’s market share at 20%. Continuing the advertisement should

increase the market share because P(B | S) = .30.

The second ad has a bigger effect.

56. a. P(A) = 200/800 = .25

b. P(B) = 100/800 = .125

c. P(A  B) = 10/800 = .0125

d. P(A | B) = P(A  B)/P(B) = .0125/.125 = .10

e. No, P(A | B) ≠ P(A) = .25

57. Let A = lost time accident in current year

B = lost time accident previous year

Given:

P(B) = .06, P(A) = .05, P(A | B) = .15

a. P(A  B) = P(A | B)P(B) = .15(.06) = .009

b. P(A  B) = P(A) + P(B) - P(A  B) = .06 + .05 - .009 = .101 or 10.1%

58. a. Let = student studied abroad

= student did not study abroad

F = female student

M = male student

P(A1) = .095

P(A2) = =1 .095 = .905

P(F | A1) = .60

P(F | A2) = .49

 Tabular computations

Event P(Ai) P(F|Ai) P(Ai∩F) P(Ai|F)

A1 .095 .60 .0570 .1139

A2 .905 .49 .4435 .8861

P(F) = .5005

P(A1|F) = .1139

b.

Events P(Ai) P(M|Ai) P(Ai∩M) P(Ai|M)

A1 .095 .40 .0380 .0761

A2 .905 .51 .4615 .9239

P(M) = .4995

P(A1|M) = .0761

c. From the preceding, P(F) = .5005 and P(M) = .4995, so almost 50:50 female and

male full-time students.

59. a. P(Oil) = .50 + .20 = .70

b Let S = Soil test results

Events P(Ai) P(S | Ai) P(Ai  S) P(Ai | S)

High quality (A1) .50 .20 .10 .31

Medium quality (A2) .20 .80 .16 .50

No oil (A3) .30 .20 .06 .19

 1.00 P(S) = .32 1.00

P(Oil) = .81

which is good. Probabilities, however, now favor medium quality rather than high

quality oil.

60. a.

If a message includes the word shipping!, the probability the message is spam is high

(.7910), so the message should be flagged as spam.

A message that includes the word today! is more likely to be spam. P(spam|
 today!) is higher than P(spam|here!) because P(today!|spam) is larger than

P(here!|spam) and P(today!|ham) = P(here!|ham) meaning that today! occurs

more often in unwanted messages (spam) than here!, and just as often in

legitimate messages (ham). Therefore, it is easier to distinguish spam from ham in

messages that include today!.

b It is easier to distinguish spam from ham when a word occurs more often in

unwanted messages (spam), less often in legitimate messages (ham), or both.

Case Solutions

Case Problem 1 - Hamilton County Judges

The data in the table provides the basis for the analysis. We provide notes as a guide to

answering questions 1 through 5.

1. The conditional probabilities of cases being appealed in the three courts are given in

the three Total rows in the table. For Common Pleas Court, the probability of an

appeal is .0401; for Domestic Relations Court, the probability of an appeal is .00348;

and for Municipal Court, the probability of an appeal is .00461. Appeals are much

more likely in Common Pleas Court, but even there only 1 in 25 cases is appealed.

The unconditional probability of an appeal across all three courts is

(1762 + 106 + 500)/(43,945 + 30,499 + 108,464) = .0129

2. The probability of a case being appealed for each judge is given in column 5 of the

table. Judges Winkler, Panioto, and Grady have the lowest probability of appeal for

Common Pleas, Domestic Relations, and Municipal Courts, respectively.

3. The probability of a case being reversed for each judge is given in column 7 of the

table. Judges Winkler, Panioto, and Grady and Hair have the lowest probability of

reversal for Common Pleas, Domestic Relations, and Municipal Courts, respectively.

These are the probabilities for reversal for all cases disposed of, not just the ones

appealed.
4. The probability of a reversal given an appeal for each judge is given in column 9 of

the table. Judges Nurre, Panioto, and Grady and Hair have the lowest probability of

reversal for Common Pleas, Domestic Relations, and Municipal Courts, respectively.

5. We describe here how the Cincinnati Enquirer used these data to rank the judges.

Other approaches may also be valid, but a rationale should be provided. The

newspaper provided rankings for each judge within each court on percentage of cases

appealed, percentage of cases reversed, and percentage of appealed cases reversed.

Those rankings were the same as those we have computed based on probabilities in

columns 6, 8, and 10 of the table. Then they summed the three rankings to come up

with a total ranking for each judge. We provide those total ranks in column 11 of the

table. Judge Winkler is the highest ranked judge in Common Pleas Court, Judge

Panioto is the highest ranked judge in Domestic Relations Court, and Judge Grady is

the highest ranked judge in Municipal Court.

Common Pleas Court

Domestic Relations Court

Total Cases Disposed

Appealed Cases

Reversed Cases

Reversal Given

Sum of Ranks
Probability of

Probability of

Probability of
Conditional
Reversal
Appeal

Appeal
Rank

Rank

Rank
Judge
Penelope Cunningham 2729 7 1 0.00257 2 0.00037 2 0.14286 2 6
Patrick Dinkelacker 6001 19 4 0.00317 3 0.00067 3 0.21053 4 10
Deborah Gaines 8799 48 9 0.00546 4 0.00102 4 0.18750 3 11
Ronald Panioto 12970 32 3 0.00247 1 0.00023 1 0.09375 1 3
Total 30499 106 17 0.00348 0.00056 0.16038
Total Cases Disposed

Appealed Cases

Reversed Cases

Reversal Given

Sum of Ranks
Probability of

Probability of

Probability of
Conditional
Reversal
Appeal

Appeal
Rank

Rank

Rank
Judge
Fred Cartolano 3037 137 12 0.04511 14 0.00395 6 0.08759 5 25
Thomas Crush 3372 119 10 0.03529 4 0.00297 4 0.08403 4 12
Patrick Dinkelacker 1258 44 8 0.03498 3 0.00636 12 0.18182 14 29
Timothy Hogan 1954 60 7 0.03071 2 0.00358 5 0.11667 9 16
Robert Kraft 3138 127 7 0.04047 10 0.00223 3 0.05512 2 15
William Mathews 2264 91 18 0.04019 7 0.00795 15 0.19780 16 38
William Morrissey 3032 121 22 0.03991 6 0.00726 14 0.18182 14 34
Norbert Nadel 2959 131 20 0.04427 13 0.00676 13 0.15267 12 38
Arthur Ney, Jr. 3219 125 14 0.03883 5 0.00435 9 0.11200 8 22
Richard Niehaus 3353 137 16 0.04086 11 0.00477 10 0.11679 10 31
Thomas Nurre 3000 121 6 0.04033 8 0.00200 2 0.04959 1 11
John O'Connor 2969 129 12 0.04345 12 0.00404 7 0.09302 6 25
Robert Ruehlman 3205 145 18 0.04524 15 0.00562 11 0.12414 11 37
J. Howard Sundermann Jr. 955 60 10 0.06283 16 0.01047 16 0.16667 13 45
Ann Marie Tracey 3141 127 13 0.04043 9 0.00414 8 0.10236 7 24
Ralph Winkler 3089 88 6 0.02849 1 0.00194 1 0.06818 3 5
Total 43945 1762 199 0.0401 0.00453 0.11294

Municipal Court
 Total Cases Disposed

Appealed Cases

Reversed Cases

Reversal Given

Sum of Ranks
Probability of

Probability of

Probability of
Conditional
Reversal
Appeal

Appeal
Rank

Rank

Rank
Judge
Mike Allen 6149 43 4 0.00699 20 0.00065 7 0.09302 4 31
Nadine Allen 7812 34 6 0.00435 9 0.00077 11 0.17647 10 30
Timothy Black 7954 41 6 0.00515 12 0.00075 10 0.14634 6 28
David Davis 7736 43 5 0.00556 15 0.00065 6 0.11628 5 26
Leslie Isaiah Gaines 5282 35 13 0.00663 19 0.00246 20 0.37143 18 57
Karla Grady 5253 6 0 0.00114 1 0.00000 1 0.00000 1 3
Deidra Hair 2532 5 0 0.00197 3 0.00000 1 0.00000 1 5
Dennis Helmick 7900 29 5 0.00367 6 0.00063 5 0.17241 9 20
Timothy Hogan 2308 13 2 0.00563 17 0.00087 13 0.15385 7 37
James Patrick Kenney 2798 6 1 0.00214 4 0.00036 4 0.16667 8 16
Joseph Luebbers 4698 25 8 0.00532 14 0.00170 18 0.32000 16 48
William Mallory 8277 38 9 0.00459 11 0.00109 14 0.23684 14 39
Melba Marsh 8219 34 7 0.00414 8 0.00085 12 0.20588 13 33
Beth Mattingly 2971 13 1 0.00438 10 0.00034 3 0.07692 3 16
Albert Mestemaker 4975 28 9 0.00563 16 0.00181 19 0.32143 17 52
Mark Painter 2239 7 3 0.00313 5 0.00134 16 0.42857 19 40
Jack Rosen 7790 41 13 0.00526 13 0.00167 17 0.31707 15 45
Mark Schweikert 5403 33 6 0.00611 18 0.00111 15 0.18182 11 44
David Stockdale 5371 22 4 0.00410 7 0.00074 9 0.18182 11 27
John A. West 2797 4 2 0.00143 2 0.00072 8 0.50000 20 30
Total 108464 500 104 0.00461 0.00096 0.20800
Case Problem 2 - Rob’s Market

There are 278/1,000 records with no purchase, so the probability = .278.

We have the following table based on bread and jelly.

Column Labels

Count of Jelly Row Labels Grape None Strawberry Grand Total

None 32 325 18 375

Wheat 10 132 22 164

White 306 117 38 461

Grand Total 348 574 78 1000

We have the following table based on bread and peanut butter.

Column Labels

Count of Peanut Butter Row Labels Creamy Natural None Grand Total

None 25 31 319 375

Wheat 17 16 131 164

White 186 54 221 461

Grand Total 228 101 671 1000

P(creamy | white bread) = 186/461 = .403

P(natural | white bread) = 54/461 = .117

P(creamy | wheat bread) = 17/164 = .104

P(natural | wheat bread) = 16/164 = .098

Of the 1000 records, 372 are white-grape-creamy, so that the probability of this

purchase is .372.
Summary There is a .278 chance that a random customer purchases none of these

three product types. So the probability that a customer purchased at least one of these

three types of products is 1 – .278 = .722. Also, only 188/1000 = .188, or 18.8%, have

purchased all three products (bread, jelly, and peanut butter). Therefore, there does

seem to be an opportunity to use coupons and other marketing efforts to encourage

consumers who have purchased at least one but not all three of the products to

purchase the product(s) they are not currently purchasing.

The probability of white–grape–creamy is relatively high (.372), so

encouraging customers who have purchased one or two of these products to purchase

the other(s) appears to be a promising strategy. Note, too, that customers who do not

purchase bread or who purchase wheat bread tend not to purchase peanut butter or

jelly (see low counts for each table in the wheat and none rows). If Mr. White believes

RM can change the behavior of these customers, then they may represent an

opportunity for increased sales of jelly or peanut butter or both. If Mr. White believes

these customers’ tastes for and attitudes toward jelly or peanut butter are entrenched,

then promotional activities for jelly or peanut butter or both toward this consumer

segment will likely be fruitless.