Section 6.

1 Discrete and Continuous Random Variables 359

6.1 T echnology
Corner
TI-Nspire instructions in Appendix B; HP Prime instructions on the book’s Web site.

11. Analyzing random variables on the calculator page 354

Section 6.1 Exercises

1. Toss 4 times Suppose you toss a fair coin 4 times. Let (a) Write the event “plays with at most two toys” in terms
X = the number of heads you get. of X. What is the probability of this event?
(a) Find the probability distribution of X. (b) Describe the event X > 3 in words. What is its
­probability? What is the probability that X ≥ 3?
(b) Make a histogram of the probability distribution.
­Describe what you see. 5. Benford’s law Faked numbers in tax returns, invoices,
pg 349 or expense account claims often display patterns that
(c) Find P(X ≤ 3) and interpret the result.
aren’t present in legitimate records. Some patterns,
2. Pair-a-dice Suppose you roll a pair of fair, six-sided dice. like too many round numbers, are obvious and easily
Let T = the sum of the spots showing on the up-faces. avoided by a clever crook. Others are more subtle. It is
a striking fact that the first digits of numbers in legiti-
(a) Find the probability distribution of T.
mate records often follow a model known as Benford’s
(b) Make a histogram of the probability distribution. law.7 Call the first digit of a randomly chosen record X
­Describe what you see. for short. Benford’s law gives this probability model for
X (note that a first digit can’t be 0):
(c) Find P(T ≥ 5) and interpret the result.
3. Spell-checking Spell-checking software catches First digit: 1 2 3 4 5 6 7 8 9
“nonword errors,” which result in a string of letters that Probability: 0.301 0.176 0.125 0.097 0.079 0.067 0.058 0.051 0.046
is not a word, as when “the” is typed as “teh.” When
undergraduates are asked to write a 250-word essay (a) Show that this is a legitimate probability distribution.
(without spell-checking), the number X of nonword (b) Make a histogram of the probability distribution.
errors has the following distribution: ­Describe what you see.
(c) Describe the event X ≥ 6 in words. What is P(X ≥ 6)?
Value: 0 1 2 3 4
Probability: 0.1 0.2 0.3 0.3 0.1 (d) Express the event “first digit is at most 5” in terms of
X. What is the probability of this event?
(a) Write the event “at least one nonword error” in terms 6. Working out Choose a person aged 19 to 25 years at
of X. What is the probability of this event? ­random and ask, “In the past seven days, how many
times did you go to an exercise or fitness center or
(b) Describe the event X ≤ 2 in words. What is its
work out?” Call the response Y for short. Based on a
­probability? What is the probability that X < 2?
large sample survey, here is a probability model for
4. Kids and toys In an experiment on the behavior of the answer you will get:8
young children, each subject is placed in an area with
five toys. Past experiments have shown that the prob- Days: 0 1 2 3 4 5 6 7
ability distribution of the number X of toys played Probability: 0.68 0.05 0.07 0.08 0.05 0.04 0.01 0.02
with by a randomly selected subject is as follows:
(a) Show that this is a legitimate probability distribution.
Number of toys xi: 0 1 2 3 4 5 (b) Make a histogram of the probability distribution.
Probability pi: 0.03 0.16 0.30 0.23 0.17 0.11 ­Describe what you see.

Starnes-Yates5e_c06_344-419hr2.indd 359 11/13/13 2:34 PM

 360 CHAPTER 6 R a n d o m Va r i a b l e s

(c) Describe the event Y < 7 in words. What is P(Y < 7)? 12. Kids and toys Refer to Exercise 4. Calculate the
mean of the random variable X and interpret this
(d) Express the event “worked out at least once” in terms result in context.
of Y. What is the probability of this event?
13. Benford’s law and fraud A not-so-clever employee
7. Benford’s law Refer to Exercise 5. The first digit of decided to fake his monthly expense report. He
a randomly chosen expense account claim follows believed that the first digits of his expense amounts
Benford’s law. Consider the events A = first digit is 7 should be equally likely to be any of the numbers
or greater and B = first digit is odd. from 1 to 9. In that case, the first digit Y of a ran-
(a) What outcomes make up the event A? What is P(A)? domly selected expense amount would have the
probability distribution shown in the histogram.
(b) What outcomes make up the event B? What is P(B)?
0.4
(c) What outcomes make up the event “A or B”? What
0.3

Probability
is P(A or B)? Why is this probability not equal to
P(A) + P(B)? 0.2

8. Working out Refer to Exercise 6. Consider the 0.1

events A = works out at least once and B = works 0.0

out less than 5 times per week. 0 1 2 3 4 5 6 7 8 9
Outcomes
(a) What outcomes make up the event A? What is P(A)?
(b) What outcomes make up the event B? What is P(B)? (a) Explain why the mean of the random variable Y is
located at the solid red line in the figure.
(c) What outcomes make up the event “A and B”? What Starnes/Yates/Moore: The Practice of Statistics, 4E
(b) New
TheFig.:
first06_UN08
digits of randomly
Perm. Fig.:selected
6019 expense amounts
is P(A and B)? Why is this probability not equal to
actually follow Benford’s law (Exercise 5). According
P(A) # P(B)? First Pass: 2010-04-16
to Benford’s
2nd law, what’s the expected value of the first
Pass: 2010-05-04
9. Keno Keno is a favorite game in casinos, and simi- digit? Explain how this information could be used to
pg 352 lar games are popular with the states that operate detect a fake expense report.
lotteries. Balls numbered 1 to 80 are tumbled in a (c) What’s P(Y > 6) in the above distribution? Ac-
machine as the bets are placed, then 20 of the balls cording to Benford’s law, what proportion of first
are chosen at random. Players select numbers by digits in the employee’s expense amounts should be
marking a card. The simplest of the many wagers greater than 6? How could this information be used
available is “Mark 1 Number.” Your payoff is $3 on to detect a fake expense report?
a $1 bet if the number you select is one of those
chosen. Because 20 of 80 numbers are chosen, your 14. Life insurance A life insurance company sells a
probability of winning is 20/80, or 0.25. Let X = the term insurance policy to a 21-year-old male that pays
net amount you gain on a single play of the game. $100,000 if the insured dies within the next 5 years.
The probability that a randomly chosen male will
(a) Make a table that shows the probability distribution of X. die each year can be found in mortality tables. The
(b) Compute the expected value of X. Explain what this company collects a premium of $250 each year as
result means for the player. payment for the insurance. The amount Y that the
company earns on this policy is $250 per year, less
10. Fire insurance Suppose a homeowner spends the $100,000 that it must pay if the insured dies.
$300 for a home insurance policy that will pay out Here is a partially completed table that shows infor-
$200,000 if the home is destroyed by fire. Let Y = mation about risk of mortality and the values of
the profit made by the company on a single policy. Y = profit earned by the company:
From previous data, the probability that a home in
this area will be destroyed by fire is 0.0002. Age at death: 21 22 23 24 25 26 or more
Profit: −$99,750 −$99,500 −$99,250 −$99,000 −$98,750 $1250
(a) Make a table that shows the probability distribution
Probability: 0.00183 0.00186 0.00189 0.00191 0.00193
of Y.
(b) Compute the expected value of Y. Explain what this (a) Explain why the ­company suffers a loss of $98,750
result means for the insurance company. on such a policy if a client dies at age 25.

11. Spell-checking Refer to Exercise 3. Calculate the (b) Find the missing probability. Show your work.
mean of the random variable X and interpret this (c) Calculate the mean mY. Interpret this value in
result in ­context. context.

Starnes-Yates5e_c06_344-419hr2.indd 360 11/13/13 2:34 PM

 Section 6.1 Discrete and Continuous Random Variables 361

15. Spell-checking Refer to Exercise 3. Calculate and (b) Find the mean number of rooms for both types of
pg 353 interpret the standard deviation of the random vari- housing unit. Explain why this difference makes
able X. Show your work. sense.
16. Kids and toys Refer to Exercise 4. Calculate and in- (c) Find and interpret the standard deviations of both
terpret the standard deviation of the random variable X and Y.
X. Show your work.
20. Size of American households In government data, a
17. Benford’s law and fraud Refer to Exercise 13. It household consists of all occupants of a dwelling unit,
might also be possible to detect an employee’s fake while a family consists of two or more persons who
expense records by looking at the variability in the live together and are related by blood or marriage. So
first digits of those expense amounts. all families form households, but some households
are not families. Here are the distributions of house-
(a) Calculate the standard deviation sY. This gives us
hold size and family size in the United States:
an idea of how much variation we’d expect in the
employee’s expense records if he assumed that first Number of Persons
digits from 1 to 9 were equally likely. 1 2 3 4 5 6 7
(b) Now calculate the standard deviation of first digits that Household probability 0.25 0.32 0.17 0.15 0.07 0.03 0.01
follow Benford’s law (Exercise 5). Would using stan- Family probability 0 0.42 0.23 0.21 0.09 0.03 0.02
dard deviations be a good way to detect fraud? Explain.
Let X = the number of people in a randomly selected
18. Life insurance U.S. household and Y = the number of people in a
(a) It would be quite risky for you to insure the life of a randomly chosen U.S. family.
21-year-old friend under the terms of Exercise 14. (a) Make histograms suitable for comparing the prob-
There is a high probability that your friend would ability distributions of X and Y. Describe any differ-
live and you would gain $1250 in premiums. But if ences that you observe.
he were to die, you would lose almost $100,000. Ex-
(b) Find the mean for each random variable. Explain
plain carefully why selling insurance is not risky for
why this difference makes sense.
an insurance company that insures many thousands
of 21-year-old men. (c) Find and interpret the standard deviations of both
X and Y.
(b) The risk of an investment is often measured by the
standard deviation of the return on the investment. 21. Random numbers Let X be a number between
The more variable the return is, the riskier the invest- 0 and 1 produced by a random number generator.
ment. We can measure the great risk of insuring a Assuming that the random variable X has a uniform
single person’s life in Exercise 14 by computing the distribution, find the following probabilities:
standard deviation of the income Y that the insurer (a) P(X > 0.49)
will receive. Find sY using the distribution and mean
found in Exercise 14. (b) P(X ≥ 0.49)
19. Housing in San Jose How do rented housing units (c) P(0.19 ≤ X < 0.37 or 0.84 < X ≤ 1.27)
differ from units occupied by their owners? Here are 22. Random numbers Let Y be a number between 0
the distributions of the number of rooms for owner- and 1 produced by a random number generator.
occupied units and renter-occupied units in San Assuming that the random variable Y has a uniform
Jose, California:9 distribution, find the following probabilities:
Number of Rooms (a) P(Y ≤ 0.4)
1 2 3 4 5 6 7 8 9 10
(b) P(Y < 0.4)
Owned 0.003 0.002 0.023 0.104 0.210 0.224 0.197 0.149 0.053 0.035
Rented 0.008 0.027 0.287 0.363 0.164 0.093 0.039 0.013 0.003 0.003 (c) P(0.1 < Y ≤ 0.15 or 0.77 ≤ Y < 0.88)

Let X = the number of rooms in a randomly selected 23. Running a mile A study of 12,000 able-bodied male
owner-occupied unit and Y = the number of rooms in a pg 357 students at the University of Illinois found that their
randomly chosen renter-occupied unit. times for the mile run were approximately Normal
with mean 7.11 minutes and standard deviation 0.74
(a) Make histograms suitable for comparing the prob- minute.10 Choose a student at random from this
ability distributions of X and Y. Describe any differ- group and call his time for the mile Y. Find P(Y < 6)
ences that you observe. and interpret the result.

Starnes-Yates5e_c06_344-419hr2.indd 361 11/13/13 2:34 PM

 362 CHAPTER 6 R a n d o m Va r i a b l e s

24. ITBS scores The Normal distribution with mean (d) The number of cars would be within 1.08 from the
m = 6.8 and standard deviation s = 1.6 is a good mean about 68% of the time.
description of the Iowa Test of Basic Skills (ITBS)
(e) The mean number of cars would be about 1.08 from
vocabulary scores of seventh-grade students in Gary,
the expected value.
Indiana. Call the score of a randomly chosen student
X for short. Find P(X ≥ 9) and interpret the result. 29. About what percentage of households have a number
of cars within 2 standard deviations of the mean?
25. Ace! Professional tennis player Rafael Nadal hits the
ball extremely hard. His first-serve speeds follow a (a) 68% (b) 71% (c) 93% (d) 95% (e) 98%
Normal distribution with mean 115 miles per hour
30. A deck of cards contains 52 cards, of which 4 are
(mph) and standard deviation 6 mph. Choose one
aces. You are offered the following wager: Draw one
of Nadal’s first serves at random. Let Y = its speed,
card at random from the deck. You win $10 if the
measured in miles per hour.
card drawn is an ace. Otherwise, you lose $1. If you
(a) Find P(Y > 120) and interpret the result. make this wager very many times, what will be the
mean amount you win?
(b) What is P(Y ≥ 120)? Explain.
(a) About −$1, because you will lose most of the time.
(c) Find the value of c such that P(Y ≤ c) = 0.15. Show
your work. (b) About $9, because you win $10 but lose only $1.
26. Pregnancy length The length of human pregnan- (c) About −$0.15; that is, on average you lose about 15 cents.
cies from conception to birth follows a Normal dis-
(d) About $0.77; that is, on average you win about 77 cents.
tribution with mean 266 days and standard deviation
16 days. Choose a pregnant woman at random. Let X (e) About $0, because the random draw gives you a fair bet.
= the length of her pregnancy.
Exercises 31 to 34 refer to the following setting. Many chess
(a) Find P(X ≥ 240) and interpret the result. masters and chess advocates believe that chess play devel-
ops general intelligence, analytical skill, and the ability to
(b) What is P(X > 240)? Explain.
concentrate. According to such beliefs, improved reading
(c) Find the value of c such that P(X ≥ c) = 0.20. Show skills should result from study to improve chess-playing
your work. skills. To investigate this belief, researchers conducted
a study. All of the subjects in the study participated in a
Multiple choice: Select the best answer for Exercises comprehensive chess program, and their reading perfor-
27 to 30. mances were measured before and after the program. The
Exercises 27 to 29 refer to the following setting. Choose graphs and numerical summaries below provide informa-
an American household at random and let the random tion on the subjects’ pretest scores, posttest scores, and the
variable X be the number of cars (including SUVs and difference (post – pre) between these two scores.
light trucks) they own. Here is the probability model if we
ignore the few households that own more than 5 cars: Descriptive Statistics: Pretest, Posttest, Post – pre
Variable N Mean Median StDev Min Max Q1 Q3
Number of cars X: 0 1 2 3 4 5 Pretest 53 57.70 58.00 17.84 23.00 99.00 44.50 70.50
Probability: 0.09 0.36 0.35 0.13 0.05 0.02 Posttest 53 63.08 64.00 18.70 28.00 99.00 48.00 76.00
Post-pre 53 5.38 3.00 13.02 −19.00 42.00 −3.50 14.00
27. What’s the expected number of cars in a randomly
selected American household?
100
(a) 1.00 (b) 1.75 (c) 1.84 (d) 2.00 (e) 2.50
28. The standard deviation of X is sX = 1.08. If many
Reading performance

households were selected at random, which of the

following would be the best interpretation of the 50
value 1.08? *

(a) The mean number of cars would be about 1.08.

(b) The number of cars would typically be about 1.08 0
from the mean.
(c) The number of cars would be at most 1.08 from the
Pretest Posttest Post – pre
mean.

Starnes-Yates5e_c06_344-419hr2.indd 362 11/13/13 2:34 PM

 Section 6.2 Transforming and Combining Random Variables 363

31. Better readers? (1.3) Did students have higher read-

ing scores after participating in the chess program?
Give appropriate statistical evidence to support your
answer.
32. Chess and reading (4.3) If the study found a statisti-
cally significant improvement in reading scores,
could you conclude that playing chess causes an
increase in reading skills? Justify your answer.

Some graphical and numerical information about the

relationship between pretest and posttest scores is provided
below.
Regression Analysis: Posttest versus Pretest
Predictor Coef SE Coef T P
Constant 17.897 5.889 3.04 0.004
Pretest 0.78301 0.09758 8.02 0.000
S = 12.55 R-Sq = 55.8% R-Sq(adj) = 54.9%

33. Predicting posttest scores (3.2) What is the equa-

tion of the linear regression model relating posttest
and pretest scores? Define any variables used.
34. How well does it fit? (3.2) Discuss what s, r2, and
the residual plot tell you about this linear regression
model.

6.2 Transforming and Combining

Random Variables
What You Will Learn By the end of the section, you should be able to:
• Describe the effects of transforming a random variable • Find probabilities involving the sum or difference of
by adding or subtracting a constant and multiplying or independent Normal random variables.
dividing by a constant.
• Find the mean and standard deviation of the sum or
difference of independent random variables.

In Section 6.1, we looked at several examples of random variables and their proba-
bility distributions. We also saw that the mean mX and standard deviation sX give us
important information about a random variable. For instance, for X = the amount
gained on a single $1 bet on red in a game of roulette, we already showed that
mX = −$0.05. You can verify that the standard deviation is sX = $1.00. That is, a
player can expect to lose an average of 5 cents per $1 bet if he plays many games.
But if he plays only a few games, his actual gain could be much better or worse
than this expected value.

Starnes-Yates5e_c06_344-419hr2.indd 363 11/13/13 2:34 PM