Chapter 5: Discrete Distributions 1

Chapter 5
Discrete Distributions

LEARNING OBJECTIVES

The overall learning objective of Chapter 5 is to help you understand a category of probability
distributions that produces only discrete outcomes, thereby enabling you to:

1. Distinguish between discrete random variables and continuous random variables.

2. Know how to determine the mean and variance of a discrete distribution.

3. Identify the type of statistical experiments that can be described by the binomial
distribution and know how to work such problems.

4. Decide when to use the Poisson distribution in analyzing statistical experiments and
know how to work such problems.
.
5. Decide when binomial distribution problems can be approximated by the Poisson
distribution and know how to work such problems.

6. Decide when to use the hypergeometric distribution and know how to work such
problems

CHAPTER TEACHING STRATEGY

Chapters 5 and 6 introduce the student to several statistical distributions. It is important

to differentiate between the discrete distributions of chapter 5 and the continuous distributions of
chapter 6.

The approach taken in presenting the binomial distribution is to build on techniques

presented in chapter 4. It can be helpful to take the time to apply the law of multiplication for
independent events to a problem and demonstrate to students that sequence is important. From
there, the student will more easily understand that by using combinations, one can more quickly
determine the number of sequences and weigh the probability of obtaining a single sequence by
that number. In a sense, we are developing the binomial formula through an inductive process.
 Chapter 5: Discrete Distributions 2

Thus, the binomial formula becomes more of a summary device than a statistical "trick". The
binomial tables presented in this text are noncumulative. This makes it easier for the student to
recognize that the table is but a listing of a series of binomial formula computations. In addition,
it lends itself more readily to the graphing of a binomial distribution.

It is important to differentiate applications of the Poisson distribution from binomial

distribution problems. It is often difficult for students to determine which type of distribution to
apply to a problem. The Poisson distribution applies to rare occurrences over some interval. The
parameters involved in the binomial distribution (n and p) are different from the parameter
(Lambda) of a Poisson distribution.

It is sometimes difficult for students to know how to handle Poisson problems where the
interval for the problem is different than the interval for which Lambda was developed. If they
can view Lambda as a long run average which can be appropriately adjusted for various
intervals, then students can be more successful with these types of problems.

Solving for the mean and standard deviation of binomial distributions prepares the
students for chapter 6 where the normal distribution is sometimes used to solve binomial
distribution problems. In addition, graphing binomial and Poisson distributions affords the
student the opportunity to visualize the meaning and impact of a particular set of parameters for a
distribution.

It can be emphasized that the hypergeometric distribution is an exact distribution.

However, it is cumbersome to determine probabilities using the hypergeometric formula
particularly when computing cumulative probabilities. The hypergeometric distribution can be
presented as a fall-back position to be used when the binomial distribution should not be applied
because of the non independence of trials and size of sample.

CHAPTER OUTLINE

5.1 Discrete Versus Continuous Distributions

5.2 Describing a Discrete Distribution

Mean, Variance, and Standard Deviation of Discrete Distributions
Mean or Expected Value
Variance and Standard Deviation of a Discrete Distribution

5.3 Binomial Distribution

Solving a Binomial Problem
Using the Binomial Table
Using the Computer to Produce a Binomial Distribution
Mean and Standard Deviation of the Binomial Distribution
Graphing Binomial Distributions
 Chapter 5: Discrete Distributions 3

5.4 Poisson Distribution

Working Poisson Problems by Formula
Using the Poisson Tables
Mean and Standard Deviation of a Poisson Distribution
Graphing Poisson Distributions
Using the Computer to Generate Poisson Distributions
Approximating Binomial Problems by the Poisson Distribution

5.5 Hypergeometric Distribution

Using the Computer to Solve for Hypergeometric Distribution

Probabilities

KEY TERMS

Binomial Distribution Hypergeometric Distribution

Continuous Distributions Lambda (λ)
Continuous Random Variables Mean, or Expected Value
Discrete Distributions Poisson Distribution
Discrete Random Variables Random Variable

SOLUTIONS TO PROBLEMS IN CHAPTER 5

5.1 x P(x) x·P(x) (x-µ)2 (x-µ)2·P(x)

1 .238 .238 2.775556 0.6605823

2 .290 .580 0.443556 0.1286312
3 .177 .531 0.111556 0.0197454
4 .158 .632 1.779556 0.2811700
5 .137 .685 5.447556 0.7463152

µ = [x·P(x)] = 2.666 σ2 = (x-µ)2·P(x) = 1.836444

σ = 1.836444 = 1.355155
 Chapter 5: Discrete Distributions 4

5.2 x P(x) x·P(x) (x-µ)2 (x-µ)2·P(x)

0 .103 .000 7.573504 0.780071

1 .118 .118 3.069504 0.362201
2 .246 .492 0.565504 0.139114
3 .229 .687 0.061504 0.014084
4 .138 .552 1.557504 0.214936
5 .094 .470 5.053504 0.475029
6 .071 .426 10.549500 0.749015
7 .001 .007 18.045500 0.018046

µ = [x·P(x)] = 2.752 σ2 = (x-µ)2·P(x) = 2.752496

σ = 2.752496 = 1.6591

5.3 x P(x) x·P(x) (x-µ)2 (x-µ)2·P(x)

0 .461 .000 0.913936 0.421324

1 .285 .285 0.001936 0.000552
2 .129 .258 1.089936 0.140602
3 .087 .261 4.177936 0.363480
4 .038 .152 9.265936 0.352106

E(x)=µ= [x·P(x)]= 0.956 σ2 = (x-µ)2·P(x) = 1.278064

σ = 1.278064 = 1.1305

5.4 x P(x) x·P(x) (x-µ)2 (x-µ)2·P(x)

0 .262 .000 1.4424 0.37791

1 .393 .393 0.0404 0.01588
2 .246 .492 0.6384 0.15705
3 .082 .246 3.2364 0.26538
4 .015 .060 7.8344 0.11752
5 .002 .010 14.4324 0.02886
6 .000 .000 23.0304 0.00000

µ = [x·P(x)] = 1.201 σ2 = (x-µ)2·P(x) = 0.96260

σ = .96260 = .98112
 Chapter 5: Discrete Distributions 5

5.5
a) n=4 p = .10 q = .90
3 1
P(x=3) = 4C3(.10) (.90) = 4(.001)(.90) = .0036

b) n=7 p = .80 q = .20

P(x=4) = 7C4(.80)4(.20)3 = 35(.4096)(.008) = .1147

c) n = 10 p = .60 q = .40

P(x > 7) = P(x=7) + P(x=8) + P(x=9) + P(x=10) =

7 3 8 2 9 1 10 0
10C7(.60) (.40) + 10C8(.60) (.40) + 10C9(.60) (.40) +10C10(.60) (.40) =

120(.0280)(.064) + 45(.0168)(.16) + 10(.0101)(.40) + 1(.0060)(1) =

.2150 + .1209 + .0403 + .0060 = .3822

d) n = 12 p = .45 q = .55

P(5 < x < 7) = P(x=5) + P(x=6) + P(x=7) =

5 7 6 6 7 5
12C5(.45) (.55) + 12C6(.45) (.55) + 12C7(.45) (.55) =

792(.0185)(.0152) + 924(.0083)(.0277) + 792(.0037)(.0503) =

.2225 + .2124 + .1489 = .5838

5.6 By Table A.2:

a) n = 20 p = .50

P(x=12) = .120

b) n = 20 p = .30

P(x > 8) = P(x=9) + P(x=10) + P(x=11) + ...+ P(x=20) =

 Chapter 5: Discrete Distributions 6

.065 + .031 + .012 + .004 + .001 + .000 = .113

c) n = 20 p = .70

P(x < 12) = P(x=11) + P(x=10) + P(x=9) + ... + (Px=0) =

.065 + .031 + .012 + .004 + .001 + .000 = .113

d) n = 20 p = .90

P(x < 16) = P(x=16) + P(x=15) + P(x=14) + ...+ P(x=0) =

.090 + .032 + .009 + .002 + .000 = .133

e) n = 15 p = .40

P(4 < x < 9) =

P(x=4) + P(x=5) + P(x=6) + P(x=7) + P(x=8) + P(x=9) =

.127 + .186 + .207 + .177 + .118 + .061 = .876

f) n = 10 p = .60

P(x > 7) = P(x=7) + P(x=8) + P(x=9) + P(x=10) =

.215 + .122 + .040 + .006 = .382

5.7

a) n = 20 p = .70 q = .30

µ = n⋅p = 20(.70) = 14

σ = n ⋅ p ⋅ q = 20(.70)(.30) = 4.2 = 2.05

b) n = 70 p = .35 q = .65

µ = n⋅p = 70(.35) = 24.5

 Chapter 5: Discrete Distributions 7

σ = n ⋅ p ⋅ q = 70(.35)(.65) = 15.925 = 3.99

c) n = 100 p = .50 q = .50

µ = n⋅p = 100(.50) = 50

σ = n ⋅ p ⋅ q = 100(.50)(.50) = 25 = 5

5.8
a) n = 6 p = .70 x Prob
0 .001
1 .010
2 .060
3 .185
4 .324
5 .303
6 .118

b) n = 20 p = .50 x Prob
0 .000
1 .000
2 .000
3 .001
4 .005
5 .015
6 .037
7 .074
 Chapter 5: Discrete Distributions 8

8 .120
9 .160
10 .176
11 .160
12 .120
13 .074
14 .037
15 .015
16 .005
17 .001
18 .000
19 .000
20 .000

c) n=8 p = .80 x Prob

0 .000
1 .000
2 .001
3 .009
4 .046
5 .147
6 .294
7 .336
8 .168
 Chapter 5: Discrete Distributions 9

5.9

a) n = 20 p = .78 x = 14

20C14 (.78)14(.22)6 = 38,760(.030855)(.00011338) = .1356

b) n = 20 p = .75 x = 20

20C20 (.75)20(.25)0 = (1)(.0031712)(1) = .0032

c) n = 20 p = .70 x < 12

Use table A.2:

P(x=0) + P(x=1) + . . . + P(x=11)=

.000 + .000 + .000 + .000 + .000 + .000 + .000 +

.001 + .004 + .012 + .031 + .065 = .113

5.10 n = 16 p = .40

P(x > 9): from Table A.2:

x Prob
9 .084
10 .039
11 .014
12 .004
13 .001
.142
 Chapter 5: Discrete Distributions 10

P(3 < x < 6):

x Prob

3 .047
4 .101
5 .162
6 .198
.508

n = 13 p = .88

P(x = 10) = 13C10(.88)10(.12)3 = 286(.278500976)(.001728) = .1376

P(x = 13) = 13C13(.88)13(.12)0 = (1)(.1897906171)(1) = .1898

Expected Value = µ = n p = 13(.88) = 11.44

5.11 n = 25 P = .60

a) x > 15

P(x > 15) = P(x = 15) + P(x = 16) + · · · + P(x = 25)

Using Table A.2 n = 25, p = .80

x Prob

15 .161
16 .151
17 .120
18 .080
19 .044
20 .020
21 .007
22 .002

.585

b) x > 20

from a): P(x > 20) = P(x = 21) + P(x = 22) + P(x = 23) +

P(x = 24) + P(x = 25) =

.007 + .002 + .000 + .000 + .000 = .009

 Chapter 5: Discrete Distributions 11

c) P(x < 10)

from Table A.2, x Prob.

9 .009
8 .003
7 .001
<6 .000
.013

5.12

The highest probability values are for x = 15, 16, 14, 17, 13, 18, and 12.
The expected value is 25(.60) = 15. The standard deviation is 2.45.
15 + 2(2.45) = 15 + 4.90 gives a range that goes from 10.10 to 19.90. From
table A.2, the sum of the probabilities of the values in this range (11 through
19) is .936 or 93.6% of the values which compares quite favorably with the
95% suggested by the empirical rule.

5.13
n = 15 p = .20
5 10
a) P(x = 5) = 15C5(.20) (.80) =

3003(.00032)(.1073742) = .1032

b) P(x > 9): Using Table A.2

P(x = 10) + P(x = 11) + . . . + P(x = 15) =

 Chapter 5: Discrete Distributions 12

.000 + .000 + . . . + .000 = .000

0 15
c) P(x = 0) = 15C0(.20) (.80) =

(1)(1)(.035184) = .0352

d) P(4 < x < 7): Using Table A.2

P(x = 4) + P(x = 5) + P(x = 6) + P(x = 7) =

.188 + .103 + .043 + .014 = .348

e)

5.14 n = 18

a) p =.30 µ = 18(.30) = 5.4

p = .34 µ = 18(.34) = 6.12

b) P(x > 8) n = 18 p = .30

from Table A.2

x Prob
8 .081
9 .039
10 .015
11 .005
12 .001
.141
 Chapter 5: Discrete Distributions 13

c) n = 18 p = .34

P(2 < x < 4) = P(x = 2) + P(x = 3) + P(x = 4) =

2 16 3 15 4 14
18C2(.34) (.66) + 18C3(.34) (.66) + 18C4(.34) (.66) =

.0229 + .0630 + .1217 = .2076

d) n = 18 p = .30 x=0
0 18
18C0(.30) (.70) = .00163

n = 18 p = .34 x=0
0 18
18C0(.34) (.66) = .00056

The probability that none are in the $500,000 to $1,000,000 is higher because
there is a smaller percentage in that category which is closer to zero.

5.15 a) Prob(x=5λ = 2.3)=

(2.35)(e-2.3) = (64.36343)(.1002588) = .0538

5! (120)

b) Prob(x=2λ = 3.9) =

(3.92)(e-3.9) = (15.21)(.02024) = .1539

2! (2)

c) Prob(x < 3λ = 4.1) =

Prob(x=3) + Prob(x=2) + Prob(x=1) + Prob(x=0) =

(4.13)(e-4.1) = (68.921)(.016574) = .1904

3! 6

+ (4.12)(e-4.1) = (16.81)(.016573) = .1393

2! 2

+ (4.11)(e-4.1) = (4.1)(.016573) = .0679

1! 1
 Chapter 5: Discrete Distributions 14

+ (4.10)(e-4.1) = (1)(.016573) = .0166

0! 1

.1904 + .1393 + .0679 + .0166 = .4142

d) Prob(x=0λ = 2.7) =

(2.70)(e-2.7) = (1)(.0672) = .0672

0! 1

e) Prob(x=1 λ = 5.4)=

(5.41)(e-5.4) = (5.4)(.0045) = .0244

1! 1

f) Prob(4 < x < 8 λ = 4.4) =

Prob(x=5 λ = 4.4) + Prob(x=6 λ = 4.4) + Prob(x=7 λ = 4.4)=

(4.45)(e-4.4) + (4.46)(e-4.4) + (4.47)(e-4.4) =

5! 6! 7!

(1649.162)(.012277) + (7256.314)(.012277) + (31927.781)(.012277)

120 720 5040

= .1687 + .1237 + .0778 = .3702

5.16 a) Prob(x=6 λ = 3.8) = .0936

b) Prob(x>7 λ = 2.9):

x Prob
8 .0068
9 .0022
10 .0006
11 .0002
12 .0000

x > 7 .0098
 Chapter 5: Discrete Distributions 15

c) Prob(3 < x < 9 λ = 4.2)=

x Prob
3 .1852
4 .1944
5 .1633
6 .1143
7 .0686
8 .0360
9 .0168
3<x<9 .7786

d) Prob(x=0 λ = 1.9) = .1496

e) Prob(x < 6 λ = 2.9)=

x Prob
0 .0050
1 .1596
2 .2314
3 .2237
4 .1622
5 .0940
6 .0455
x< 6 .9214

f) Prob(5 < x < 8 λ = 5.7) =

x Prob
6 .1594
7 .1298
8 .0925
5<x < 8 .3817

5.17 a) λ = 6.3 mean = 6.3 Standard deviation = 6.3 = 2.51

x Prob
0 .0018
1 .0116
2 .0364
3 .0765
4 .1205
5 .1519
6 .1595
 Chapter 5: Discrete Distributions 16

7 .1435
8 .1130
9 .0791
10 .0498
11 .0285
12 .0150
13 .0073
14 .0033
15 .0014
16 .0005
17 .0002
18 .0001
19 .0000

b) λ = 1.3 mean = 1.3 standard deviation = 1.3 = 1.14

x Prob
0 .2725
1 .3542
2 .2303
3 .0998
4 .0324
5 .0084
6 .0018
7 .0003
8 .0001
9 .0000
 Chapter 5: Discrete Distributions 17

c) λ = 8.9 mean = 8.9 standard deviation = 8.9 = 2.98

x Prob
0 .0001
1 .0012
2 .0054
3 .0160
4 .0357
5 .0635
6 .0941
7 .1197
8 .1332
9 .1317
10 .1172
11 .0948
12 .0703
13 .0481
14 .0306
15 .0182
16 .0101
17 .0053
18 .0026
19 .0012
20 .0005
21 .0002
22 .0001
 Chapter 5: Discrete Distributions 18

d) λ = 0.6 mean = 0.6 standard deviation = 0.6 = .775

x Prob
0 .5488
1 .3293
2 .0988
3 .0198
4 .0030
5 .0004
6 .0000

5.18 λ = 2.84 minutes

a) Prob(x=6 λ = 2.8)

from Table A.3 .0407

 Chapter 5: Discrete Distributions 19

b) Prob(x=0 λ = 2.8) =

from Table A.3 .0608

c) Unable to meet demand if x > 44 minutes:

x Prob.
5 .0872
6 .0407
7 .0163
8 .0057
9 .0018
10 .0005
11 .0001
x>4 .1523

.1523 probability of being unable to meet the demand.

Probability of meeting the demand = 1 - (.1523) = .8477

15.23% of the time a second window will need to be opened.

d) λ = 2.8 arrivals4 minutes

Prob(x=3) arrivals2 minutes = ??

Lambda must be changed to the same interval (½ the size)

New lambda=1.4 arrivals2 minutes

Prob(x=3)  λ=1.4) = from Table A.3 = .1128

Prob(x > 5 8 minutes) = ??

Lambda must be changed to the same interval(twice the size):

New lambda= 5.6 arrivals8 minutes

Prob(x > 5  λ = 5.6):

 Chapter 5: Discrete Distributions 20

From Table A.3: x Prob.

5 .1697
6 .1584
7 .1267
8 .0887
9 .0552
10 .0309
11 .0157
12 .0073
13 .0032
14 .0013
15 .0005
16 .0002
17 .0001
x> 5 .6579

5.19 λ = Σx/n = 126/36 = 3.5

Using Table A.3

a) P(x = 0) = .0302

b) P(x > 6) = P(x = 6) + P(x = 7) + . . . =

.0771 + .0385 + .0169 + .0066 + .0023 +

.0007 + .0002 + .0001 = .1424

c) P(x < 4 10 minutes)

λ = 7.0 10 minutes

P(x < 4) = P(x = 0) + P(x = 1) + P(x = 2) + P(x = 3) =

.0009 + .0064 + .0223 + .0521 = .0817

d) P(3 < x < 6 10 minutes)

λ = 7.0  10 minutes

P(3 < x < 6) = P(x = 3) + P(x = 4) + P(x = 5) + P(x = 6)

= .0521 + .0912 + .1277 + .1490 = .42

 Chapter 5: Discrete Distributions 21

e) P(x = 8  15 minutes)

λ = 10.5  15 minutes

λX ⋅ e − λ (10.58 )(e −10.5 )

P(x = 8  15 minutes) = = = .1009
X! 8!

5.20 λ = 5.6 days3 weeks

a) Prob(x=0  λ = 5.6):

from Table A.3 = .0037

b) Prob(x=6  λ = 5.6):

from Table A.3 = .1584

c) Prob(x > 15  λ = 5.6):

x Prob.
15 .0005
16 .0002
17 .0001
x > 15 .0008

Because this probability is so low, if it actually occurred, the researcher would

actually have to question the Lambda value as too low for this period.

5.21 λ = 0.6 trips 1 year

a) Prob(x=0  λ = 0.6):

from Table A.3 = .5488

b) Prob(x=1  λ = 0.6):

from Table A.3 = .3293

c) Prob(x > 2  λ = 0.6):

 Chapter 5: Discrete Distributions 22

from Table A.3 x Prob.

2 .0988
3 .0198
4 .0030
5 .0004
6 .0000

x > 2 .1220

d) Prob(x < 3  3 year period):

The interval length has been increased (3 times)

New Lambda = λ = 1.8 trips3 years

Prob(x < 3  λ = 1.8):

from Table A.3 x Prob.

0 .1653
1 .2975
2 .2678
3 .1607

x < 3 .8913

e) Prob(x=4  6 years):

The interval has been increased (6 times)

New Lambda = λ = 3.6 trips6 years

Prob(x=4  λ = 3.6):

from Table A.3 = .1912

5.22 λ = 1.2 collisions4 months

a) Prob(x=0  λ = 1.2):

from Table A.3 = .3012

b) Prob(x=2 2 months):
 Chapter 5: Discrete Distributions 23

The interval has been decreased (by ½)

New Lambda = λ = 0.6 collisions2 months

Prob(x=2  λ = 0.6):

from Table A.3 = .0988

c) Prob (x < 1 collision6 months):

The interval length has been increased (by 1.5)

New Lambda = λ = 1.8 collisions6 months

Prob(x < 1 λ = 1.8):

from Table A.3 x Prob.

0 .1653
1 .2975

x< 1 .4628

The result is likely to happen almost half the time (46.26%). Ship channel and
weather conditions are about normal for this period. Safety awareness is
about normal for this period. There is no compelling reason to reject the
lambda value of 0.6 collisions per 4 months based on an outcome of 0 or 1
collisions per 6 months.

5.23 λ = 1.2 penscarton

a) Prob(x=0  λ = 1.2):

from Table A.3 = .3012

b) Prob(x > 8  λ = 1.2):

from Table A.3 = .0000

c) Prob(x > 3  λ = 1.2):

from Table A.3 x Prob.

4 .0260
5 .0062
 Chapter 5: Discrete Distributions 24

6 .0012
7 .0002
8 .0000

x>3 .0336

5.24 n = 100,000 p = .00004

Prob(x > 7 n = 100,000 p = .00004):

λ = µ = n⋅p = 100,000(.00004) = 4.0

Since n > 20 and n⋅p < 7, the Poisson approximation to this binomial problem is
close enough.

Prob(x > 7  λ = 4):

Using Table A.3 x Prob.

7 .0595
8 .0298
9 .0132
10 .0053
11 .0019
12 .0006
13 .0002
14 .0001

x > 7 .1106

Prob(x>10  λ = 4):

Using Table A.3 x Prob.

11 .0019
12 .0006
13 .0002
14 .0001

x > 10 .0028

Since getting more than 10 is a rare occurrence, this particular geographic region
appears to have a higher average rate than other regions. An investigation of
particular characteristics of this region might be warranted.
 Chapter 5: Discrete Distributions 25

5.25 p = .009 n = 200

Use the Poisson Distribution:

λ = n⋅p = 200(.009) = 1.8

P(x > 6) from Table A.3 =

P(x = 6) + P(x = 7) + P(x = 8) + P(x = 9) + . . . =

.0078 + .0020 + .0005 + .0001 = .0104

P(x > 10) = .0000

P(x = 0) = .1653

P(x < 5) = P(x = 0) + P(x = 1) + P(x = 2) + P( x = 3) + P(x = 4) =

.1653 + .2975 + .2678 + .1607 + .0723 = .9636

5.26 n = 300, p = .01, λ = n(p) = 300(.01) = 3

a) Prob(x = 5):

Using λ = 3 and Table A.3 = .1008

b) Prob (x < 4) = Prob.(x = 0) + Prob.(x = 1) + Prob.(x = 2) + Prob.(x = 3) =

.0498 + .1494 + .2240 + .2240 = .6472

c) The expected number = µ = λ = 3

5.27 a) Prob(x = 3 N = 11, A = 8, n = 4)

C3 ⋅3 C1 (56)(3)
8
= = .5091
11 C 4 330

b) Prob(x < 2)N = 15, A = 5, n = 6)

Prob(x = 1) + Prob (x = 0) =

C1 ⋅10 C 5 C 0 ⋅10 C 6 (5)(252) (1)(210)

5
+ 5
= +
15 C 6 15 C 6 5005 5005
 Chapter 5: Discrete Distributions 26

.2517 + .0420 = .2937

c) Prob(x=0 N = 9, A = 2, n = 3)

C 0 ⋅7 C 3 (1)(35)
2
= = .4167
9 C3 84

d) Prob(x > 4 N = 20, A = 5, n = 7) =

Prob(x = 5) + Prob(x = 6) + Prob(x = 7) =

5 C 5 ⋅15 C 2 5 C 6 ⋅15 C1 5 C 7 ⋅15 C 0

+ + =
20 C 7 20 C 7 20 C 7

(1)(105)
+ 5C6 (impossible) + 5C7(impossible) = .0014
77520

5.28 N = 19 n = 6

a) P(x = 1 private) A = 11

C1 ⋅8 C 5 (11)(56)
11
= = .0227
19 C 6 27,132

b) P(x = 4 private)

C 4 ⋅8 C 2 (330)(28)
11
= = .3406
19 C 6 27,132

c) P(x = 6 private)

C 6 ⋅8 C 0 (462)(1)
11
= = .0170
19 C 6 27,132

d) P(x = 0 private)

C 0 ⋅8 C 6 (1)(28)
11
= = .0010
19 C 6 27,132
 Chapter 5: Discrete Distributions 27

5.29 N = 17 A=8 n=4

8 C 0 ⋅9 C 4 (1)(126)
a) P(x = 0) = = = .0529
17 C 4 2380

8 C 4 ⋅9 C 0 (70)(1)
b) P(x = 4) = = = .0294
17 C 4 2380

9 C 2 ⋅8 C 2 (36)(28)
c) P(x = 2 non computer) = = = .4235
17 C 4 2380

5.30 N = 20 A = 16 white N - A = 4 red n=5

16 C 4 ⋅ 4 C1 (1820)(4)
a) Prob(x = 4 white) = = = .4696
20 C 5 15504

4 C 4 ⋅16 C1 (1)(16)
b) Prob(x = 4 red) = = = .0010
20 C 5 15504

C 5 ⋅16 C 0
4
c) Prob(x = 5 red) = = .0000 because 4C5 is impossible to determine
20 C 5
The participant cannot draw 5 red beads if there are only 4 to draw from.

5.31 N = 10 n=4

a) A = 3 x = 2

C 2 ⋅7 C 2 (3)(21)
3
= = .30
10 C 4 210

b) A = 5 x = 0

C 0 ⋅5 C 4 (1)(5)
5
= = .0238
10 C 4 210

c) A = 5 x = 3

C 3 ⋅5 C1 (10)(5)
5
= = .2381
10 C 4 210
 Chapter 5: Discrete Distributions 28

5.32 N = 16 A = 4 defective n=3

C 0 ⋅12 C3 (1)(220)
a) Prob(x = 0) = 4
= = .3929
16 C 3 560

C 3 ⋅12 C 0 (4)(1)
b) Prob(x = 3) = 4
= = .0071
16 C 3 560
4 C 2 ⋅12 C1
c) Prob(x > 2) = Prob(x=2) + Prob(x=3) = + .0071 (from part b.) =
16 C 3

(6)(12) + .0071 = .1286 + .0071 = .1357

560

d) Prob(x < 1) = Prob(x=1) + Prob(x=0) =

4 C1 ⋅12 C 2 (4)(66)
+ .3929 (from part a.) = + .3929 = .4714 + .3929 = .8643
16 C 3 560

5.33 N = 18 A = 11 Hispanic n=5

Prob(x < 1) = Prob(1) + Prob(0) =

C1 ⋅7 C 4 C 0 ⋅7 C5 (11)(35) (1)(21)
11
+ 11
= + = .0449 + .0025 = .0474
18 C 5 18 C 5 8568 8568

It is fairly unlikely that these results occur by chance. A researcher might

investigate further the causes of this result. Were officers selected based on
leadership, years of service, dedication, prejudice, or what?

5.34 a) Prob(x=4 n = 11 and p = .23)

4 7
11C4(.23) (.77) = 330(.0028)(.1605) = .1482

b) Prob(x > 1n = 6 and p = .50) =

1 - Prob(x < 1) = 1 - Prob(x = 0) =

1-6C0(.50)0(.50)6 = 1-(1)(1)(.0156) = .9844

 Chapter 5: Discrete Distributions 29

c) Prob(x > 7 n = 9 and p = .85) = Prob(x = 8) + Prob(x = 9) =

8 1
9C8(.85) (.15) + 9C9(.85)9(.15)0 =

(9)(.2725)(.15) + (1)(.2316)(1) = .3679 + .2316 = .5995

d) Prob(x < 3 n = 14 and p = .70) =

Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x = 0) =

3 11
14C3(.70) (.30) + 14C2(.70)2(.30)12 +
1 13
14C1(.70) (.30) + 14C0(.70)0(.30)14 =

(364)(.3430)(.00000177) + (91)(.49)(.000000047)=

(14)(.70)(.00000016) + (1)(1)(.000000047) =

.0002 + .0000 + .0000 + .0000 = .0002

5.35 a) Prob(x = 14 n = 20 and p = .60) = .124

b) Prob(x < 5 n = 10 and p =.30) =

Prob(x = 4) + Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x=0) =

x Prob.
0 .028
1 .121
2 .233
3 .267
4 .200

x<5 .849

c) Prob(x > 12 n = 15 and p = .60) =

Prob(x = 12) + Prob(x = 13) + Prob(x = 14) + Prob(x = 15)

x Prob.
12 .063
13 .022
 Chapter 5: Discrete Distributions 30

14 .005
15 .000

x > 12 .090

d) Prob(x > 20 n = 25 and p = .40) = Prob(x = 21) + Prob(x = 22) +

Prob(x = 23) + Prob(x = 24) + Prob(x=25) =

x Prob.
21 .000
22 .000
23 .000
24 .000
25 .000

x > 20 .000

5.36
a) Prob(x=4  λ = 1.25)

(1.254)(e-1.25) = (2.4414)(.2865) = .0291

4! 24

b) Prob(x < 1 λ = 6.37) = Prob(x = 1) + Prob(x = 0) =

(6.37)1(e-6.37) + (6.37)0(e-6.37) = (6.37)(.0017) + (1)(.0017) =

1! 0! 1 1

.0109 + .0017 = .0126

c) Prob(x > 5 λ = 2.4) = Prob(x = 6) + Prob(x = 7) + ... =

(2.4)6(e-2.4) + (2.4)7(e-2.4) + (2.4)8(e-2.4) + (2.4)9(e-2.4) + (2.4)10(e-2.4) + ...

6! 7! 8! 9! 10!

.0241 + .0083 + .0025 + .0007 + .0002 = .0358

for values x > 11 the probabilities are each .0000 when rounded off to 4
decimal places.
 Chapter 5: Discrete Distributions 31

5.37 a) Prob(x = 3  λ = 1.8) = .1607

b) Prob(x < 5 λ = 3.3) =

Prob(x = 4) + Prob(x = 3) + Prob(x = 2) + Prob(x = 1) + Prob(x = 0) =

x Prob.
0 .0369
1 .1217
2 .2008
3 .2209
4 .1823

x<5 .7626

c) Prob(x > 3 λ = 2.1) =

x Prob.
3 .1890
4 .0992
5 .0417
6 .0146
7 .0044
8 .0011
9 .0003
10 .0001
11 .0000

x> 5 .3504

d) Prob(2 < x < 5 λ = 4.2) = Prob(x=3) + Prob(x=4) + Prob(x=5) =

x Prob.
3 .1852
4 .1944
5 .1633

2<x <5 .5429

5.38 a) Prob(x = 3 N = 6, n = 4, A = 5) =

C 3 ⋅1 C1 (10)(1)
5
= = .6667
6 C4 15
 Chapter 5: Discrete Distributions 32

b) Prob(x < 1 N = 10, n = 3, A = 5) = Prob(x = 1) + Prob(x = 0) =

C1 ⋅5 C 2 C 0 ⋅5 C 3 (5)(10) (1)(10)
5
+ 5
= +
10 C 3 10 C 3 120 120

= .4167 + .0833 = .5000

c) Prob(x > 2  N = 13, n = 5, A = 3) =

Prob(x=2) + Prob(x=3) Note: only 3 x's in population

C 2 ⋅10 C 3 C 3 ⋅10 C 2 (3)(120) (1)(45)

3
+ 3
= +
13 C 5 13 C 5 1287 1287

= .2797 + .0350 = .3147

5.39 n = 25 p = .20 retired

from Table A.2: P(x = 7) = .111

P(x > 10): P(x = 10) + P(x = 11) + . . . + P(x = 25) =

.012 + .004 + .001 = .017

Expected Value = µ = n⋅p = 25(.20) = 5

n = 20 p = .40 mutual funds

P(x = 8) = .180

P(x < 6) = P(x = 0) + P(x = 1) + . . . + P(x = 5) =

.000 + .000 + .003 +.012 + .035 + .075 = .125

P(x = 0) = .000
P(x > 12) = P(x = 12) + P(x = 13) + . . . + P(x = 20) =

.035 + .015 + .005 + .001 = .056

x=8

Expected Number = µ = n p = 20(.40) = 8

 Chapter 5: Discrete Distributions 33

5.40 λ = 3.2 cars2 hours

a) Prob(x=3) cars per 1 hour) = ??

The interval has been decreased by ½.

The new λ = 1.6 cars1 hour.

Prob(x = 3 λ = 1.6) = (from Table A.3) .1378

b) Prob(x = 0 cars per ½ hour) = ??

The interval has been decreased by ¼ the original amount.

The new λ = 0.8 cars½ hour.

Prob(x = 0 λ = 0.8) = (from Table A.3) .4493

c) Prob(x > 5 λ = 1.6) = (from Table A.3)

x Prob.
5 .0176
6 .0047
7 .0011
8 .0002

.0236

Either a rare event occurred or perhaps the long-run average, λ, has changed
(increased).

5.41 N = 32 A = 10 n = 12

10 C 3 ⋅ 22 C 9 (120)(497,420)
a) P(x = 3) = = = .2644
32 C12 225,792,840

10 C 6 ⋅ 22 C 6 (210)(74,613)
b) P(x = 6) = = = .0694
32 C12 225,792,840

10 C 0 ⋅ 22 C12 (1)(646,646)
c) P(x = 0) = = = .0029
32 C12 225,792,840
 Chapter 5: Discrete Distributions 34

d) A = 22

22 C 7 ⋅10 C5 22 C8 ⋅10 C 4 22 C9 ⋅10 C 3

P(7 < x < 9) = + +
32 C12 32 C12 32 C12

(170,544)(252) (319,770)(210) (497,420)(120)

= + +
225,792,840 225,792,840 225,792,840

= .1903 + .2974 + .2644 = .7521

5.42 λ = 1.4 defects 1 lot If x > 3, buyer rejects If x < 3, buyer accepts

Prob(x < 3 λ = 1.4) = (from Table A.3)

x Prob.
0 .2466
1 .3452
2 .2417
3 .1128

x<3 .9463

5.43 a) n = 20 and p = .25

The expected number = µ = n⋅p = (20)(.25) = 5.00

b) Prob(x < 1 n = 20 and p = .25) =

Prob(x = 1) + Prob(x = 0) = 20C1(.25)1(.75)19 + 20C0(.25)0(.75)20

= (20)(.25)(.00423) + (1)(1)(.0032) = .0212 +. 0032 = .0244

Since the probability is so low, the population of your state may have a lower
percentage of chronic heart conditions than those of other states.

5.44 a) Prob(x > 7 n = 10 and p = .70) = (from Table A.2):

x Prob.
8 .233
9 .121
10 .028

x>7 .382
 Chapter 5: Discrete Distributions 35

Expected number = µ = n⋅p = 10(.70) = 7

b) Expected number = µ = n⋅p = 15 (1/3) = 5

Prob(x=0⋅ n = 15 and p = 1/3) =

0 15
15C0(
) () = .0023

c) Prob(x = 7 n = 7 and p = .53) = 7C7(.53)7(.47)0 = .0117

Probably the 53% figure is too low for this population since the probability of
this occurrence is so low (.0117).

5.45 n = 12

a.) Prob(x = 0 long hours):

0 12
p = .20 12C0(.20) (.80) = .0687

b.) Prob(x > 6) long hours):

p = .20

Using Table A.2: .016 + .003 + .001 = .020

c) Prob(x = 5 good financing):

5 7
p = .25, 12C5(.25) (.75) = .1032

d.) p = .19 (good plan), expected number = µ = n(p) = 12(.19) = 2.28

5.46 n = 100,000 p = .000014

Worked as a Poisson: λ = n⋅p = 100,000(.000014) = 1.4

a) P(x = 5):

from Table A.3 = .0111

b) P(x = 0):
 Chapter 5: Discrete Distributions 36

from Table A.3 = .2466

c) P(x > 6): (from Table A.3)

x Prob
7 .0005
8 .0001
.0006

5.47 Prob(x < 3) n = 8 and p = .60) = (from Table A.2)

x Prob.
0 .001
1 .008
2 .041
3 .124

x< 3 .174

17.4% of the time in a sample of eight, three or fewer customers are walk-ins by
chance. Other reasons for such a low number of walk-ins might be that she is
retaining more old customers than before or perhaps a new competitor is
attracting walk-ins away from her.

5.48 n = 25 p = .20

a) Prob(x = 8 n = 25 and p = .20) = (from Table A.2) .062

b) Prob(x > 10) n=25 and p = .20) = (from Table A.2)

x Prob.
11 .004
12 .001
13 .000
x > 10 .005

c) Since such a result would only occur 0.5% of the time by chance, it is likely
that the analyst's list was not representative of the entire state of Idaho or the
20% figure for the Idaho census is not correct.

5.49 λ = 0.6 flats2000 miles

Prob(x = 0 λ = 0.6) = (from Table A.3) .5488

 Chapter 5: Discrete Distributions 37

Prob(x > 3 λ = 0.6) = (from Table A.3)

x Prob.
3 .0198
4 .0030
5 .0004

x>3 .0232

One trip is independent of the other.

Let F = flat tire and NF = no flat tire

P(NF1 _ NF2) = P(NF1) ⋅ P(NF2)

P(NF) = .5488
P(NF1 _ NF2) = (.5488)(.5488) = .3012

5.50 N = 25 n=8

a) P(x = 1 in NY) A=4

4 C1 ⋅ 21 C 7 (4)(116,280)
= = .4300
25 C 8 1,081,575

b) P(x = 4 in top 10) A = 10

C 4 ⋅15 C 4 (210(1365)
10
= = .2650
25 C 8 1,081,575

c) P(x = 0 in California) A=4

C 0 ⋅ 21 C8 (1)(203,490)
4
= = .1881
25 C 8 1,081,575

d) P(x = 3 with M) A=3

C 3 ⋅ 22 C 5 (1)(26,334)
3
= = .0243
25 C 8 1,081,575

5.51 N = 24 n=6 A=8

 Chapter 5: Discrete Distributions 38

C 6 ⋅16 C 0 (28)(1)
a) P(x = 6) = 8
= = .0002
24 C 6 134,596

C 0 ⋅16 C 6 (1)(8008)
b) P(x = 0) = 8
= = .0595
24 C 6 134,596

d) A = 16 East Side

C 3 ⋅8 C 3 (560)(56)
P(x = 3) = 16
= = .2330
24 C 6 134,596

5.52 n = 25 p = .20

Expected Value = µ = n⋅p = 25(.20) = 5

µ = 25(.20) = 5

σ = n ⋅ p ⋅ q = 25(.20)(.80) = 2
P(x > 12) = (from Table A.2)

x Prob
13 .0000

The values for x > 12 are so far away from the expected value that they are very
unlikely to occur.

P(x = 14) = 25C14(.20)14(.80)11 = .000063 which is very unlikely.

If this value (x = 14) actually occurred, one would doubt the validity of the
p = .20 figure or one would have experienced a very rare event.
 Chapter 5: Discrete Distributions 39

5.53 λ = 2.4 calls1 minute

a) Prob(x = 0 λ = 2.4) = (from Table A.3) .0907

b) Can handle x < 5 calls Cannot handle x > 5 calls

Prob(x > 5 λ = 2.4) = (from Table A.3)

x Prob.
6 .0241
7 .0083
8 .0025
9 .0007
10 .0002
11 .0000

x>5 .0358

c) Prob(x = 3 calls 2 minutes)

The interval has been increased 2 times.

New Lambda = λ = 4.8 calls2 minutes.

from Table A.3: .1517

d) Prob(x < 1 calls15 seconds):

The interval has been decreased by ¼.

New Lambda = λ = 0.6 calls15 seconds.

Prob(x < 1 λ = 0.6) = (from Table A.3)

Prob(x = 1) = .3293
Prob(x = 0) = .5488
Prob(x < 1) = .8781
 Chapter 5: Discrete Distributions 40

5.54 n = 160 p = .01

Working this problem as a Poisson problem:

a) Expected number = µ = n(p) = 160(.01) = 1.6

b) P(x > 8):

Using Table A.3: x Prob.

8 .0002
9 .0000
.0002

c) P(2 < x < 6):

Using Table A.3: x Prob.

2 .2584
3 .1378
4 .0551
5 .0176
6 .0047
P(2 < x < 6) .4736

5.55 p = .005 n = 1,000

λ = n⋅p = (1,000)(.005) = 5

a) P(x < 4) = P(x = 0) + P(x = 1) + P(x = 2) + P(x = 3) =

.0067 + .0337 + .0842 + .1404 = .265

b) P(x > 10) = P(x = 11) + P(x = 12) + . . . =

.0082 + .0034 + .0013 + .0005 + .0002 = .0136

c) P(x = 0) = .0067
 Chapter 5: Discrete Distributions 41

5.56 n=8 P = .36 x = 0 Women

0 8
8C0(.36) (.64) = (1)(1)(.0281475) = .0281

It is unlikely that a company would randomly hire 8 physicians from the U.S. pool
and none of them would be female. If this actually happened, the figures might
be used as evidence in a lawsuit.

5.57 N = 32

a) n = 5 x=3 A = 10

C 3 ⋅ 22 C 2 (120)(231)
10
= = .1377
32 C 5 201,376

b) n = 8 x < 2 A=6

6 C 0 ⋅ 26 C8 6 C1 ⋅ 26 C 7 6 C 2 ⋅ 26 C 6
+ + =
32 C 8 32 C 8 32 C 8

(1)(1,562,275) (6)(657,800) (15)(38,760)

+ + = .1485 + .3752 + .0553 = .5790
10,518,300 10,518,300 10,518,300

c) n = 5 x = 2 p = 3/26 = .1154
2 3
5C2(.1154) (.8846) = (10)(.013317)(.692215) = .0922

5.58 N = 14 n = 4

a) P(x = 4 N = 14, n = 4, A = 10 Northside)

C 4 ⋅ 4 C 0 (210((1)
10
= = .2098
14 C 4 1001

b) P(x = 4 N = 14, n = 4, A = 4 West)

C 4 ⋅10 C 0 (1)(1)
4
= = .0010
14 C 4 1001

c) P(x = 2 N = 14, n = 4, A = 4 West)

 Chapter 5: Discrete Distributions 42

C 2 ⋅10 C 2 (6)(45)
4
= = .2697
14 C 4 1001

5.59 a) λ = 3.841,000

3.840 ⋅ e −3.84
P(x = 0) = = .0215
0!

b) λ = 7.682,000

7.686 ⋅ e −7.68 (205,195.258)(.000461975)

P(x = 6) = = = .1317
6! 720

c) λ = 1.61,000 and λ = 4.83,000

from Table A.3:

P(x < 7) = P(x = 0) + P(x = 1) + . . . + P(x = 6) =

.0082 + .0395 + .0948 + .1517 + .1820 + .1747 + .1398 = .7907

5.60

This is a binomial distribution with n = 15 and p = .36.

µ = n⋅p = 15(.36) = 5.4

σ = 15(.36)(.64) = 1.86

The most likely values are near the mean, 5.4. Note from the printout that the
most probable values are at x = 5 and x = 6 which are near the mean.

5.61 This printout contains the probabilities for various values of x from zero to eleven from a
Poisson distribution with λ = 2.78. Note that the highest probabilities are at x = 2 and x
= 3 which are near the mean. The probability is slightly higher at x = 2 than at x = 3 even
though x = 3 is nearer to the mean because of the “piling up” effect of x = 0.

5.62 This is a binomial distribution with n = 22 and p = .64.

The mean is n⋅p = 22(.64) = 14.08 and the standard deviation is:

σ = n ⋅ p ⋅ q = 22(.64)(.36) = 2.25
 Chapter 5: Discrete Distributions 43

The x value with the highest peak on the graph is at x = 14 followed by x = 15

and x = 13 which are nearest to the mean.

5.63 This is the graph of a Poisson Distribution with λ = 1.784. Note the high
probabilities at x = 1 and x = 2 which are nearest to the mean. Note also that the
probabilities for values of x > 8 are near to zero because they are so far away
from the mean or expected value.