You are on page 1of 50

Chapter 7: Sampling Distributions

Contents
The Airport Problem
This problem begins with binomial calculations in the context of random airport security
checks and guides students through an investigation of the sampling distribution of a
sample proportion using simulation.

Quizzes, Tests, and Solutions

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 297


The Airport Problem

At Guadalajara Airport in Mexico, passengers must claim their luggage and then proceed to
Customs. In the Customs area, each passenger will press a button that activates a modified
stoplight. This light has only red and green bulbs. If the green light shows, the passenger is free
to go. If the light turns red, then Customs agents will inspect the passenger’s luggage. Customs
officials claim that the light has probability 0.30 of showing red on any press of the button.

You are traveling with 6 family members, and you are the seventh person in line.

1. Find the probability that exactly one of the family members in front of you is stopped.

2. Find the probability that more than two of the family members in front of you are stopped.

3. Let X = the number of people in front of you who are stopped. What type of random variable
is X? Be very specific, and justify your answer.

4. Find P ( X  4) .

5. Find the mean and the variance of X.

6. What is the probability that you are the first person stopped?

Now suppose that there are 50 passengers in the queue for the stoplight. Let’s perform a
simulation to discover how many of them are stopped by Customs.

7. On your TI-83/84, press MATH, arrow left to PRB, and choose 7:RandBin. Then type
50,0.3) and press ENTER. How many passengers were stopped?

8. The Customs officials claim that the probability that they will stop an individual passenger is,
in fact, 0.3 and that individuals are stopped independently of one another. Let’s perform a
large number of simulations to see what proportion of passengers are stopped if the Customs
agents’ claim is true. Press MATH, arrow over to PRB, and choose 7:RandBin. Then type
50,0.3) but do not press ENTER yet. We already know that this procedure will give us the
number of passengers (out of 50) who will be stopped. What we want, though, is the
proportion who will be stopped. To accomplish this, continue typing: /50. This will convert
the number of passengers stopped into a proportion. Finally, let’s run this simulation 50
times and examine the proportion of passengers stopped each time. Type
RandBin(50,0.3,50)/50->L1. It may take a few minutes for the calculator to
simulate the trials.

9. Now, let’s analyze the proportions that you simulated in Step 8. As always, we begin by
making a graphical display. Press 2nd, then Y=, and ENTER. Use the arrow keys to toggle
between options in the Statistics Plot menu, and ENTER to select the desired option. Press
ZOOM and select 9:ZoomStat to draw the histogram.

298 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


10. The calculator chooses an “ideal” window when you use ZoomStat. In reality, this window
rarely makes sense. Adjust your window by pressing WINDOW and changing the values to
“logical” numbers.

11. Describe your histogram. What does it tell you about the proportion of passengers stopped
by Customs for baggage inspection?

12. Next, we should compute numerical summary statistics for our 50 simulations. Press STAT,
arrow right to CALC, and choose 1:1-VarStats. Then press 2nd and 1 to type L1.
Record your mean and standard deviation. How does the average proportion stopped in your
50 trials compare to the 0.30 claimed by the Customs officials?

13. What if the number of people in the queue changes? Will that affect the distribution of
proportion values that you obtain from a simulation? Let’s find out. Record your mean and
standard deviation each time.
 First, let there be 100 passengers in the queue for the stoplight. Run a simulation to
discover the proportion stopped in each of 50 trials.
 Repeat the simulation with 150 passengers in line.
 Now try again with 75 passengers.
 Finally, use 25 passengers.

14. Share results with your classmates. Complete the table that follows for the standard
deviation of the distribution of sample proportions for the given sample size.

Sample size Standard deviation


25
50
75
100
150

15. Make a scatterplot of the standard deviation versus sample size. Describe what you see.

p(1  p )
16. Compare your results with the theoretical formula:  pˆ  , where p  0.30 .
n

17. What proportion of a 50-passenger queue would need to be stopped by Customs officials for
you to disbelieve their claim of a 0.30 probability of stopping an individual passenger?

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 299


Answers: The Airport Problem
6
1. P ( X  1)     0.3  0.7   0.302526 OR
1 5
binompdf(6,0.3,1) = 0.3025
1

2.

6 6 6  6 6


P ( X  2)   0.3 0.7    0.3 0.7    0.3 0.7    0.3 0.7    0.3 0.7 
2 4 3 3 4 2 5 1 6 0

 2  3  4  5 6

OR

 6 6 6 4
P ( X  2)  1  P ( X  2)  1    0.3 0.7    0.3 0.7    0.3 0.7    0.25569
0 6 1 5 2

 0 1  2 

OR 1 – binomcdf(6,0.3,2) = 0.2557

3. X is a binomial random variable with parameters n = 6 and p = 0.3, since the scenario
presented meets the 4 criteria for a binomial situation: fixed number of trials, two possible
outcomes on each trial: “success” = red and “failure” = green, same probability of success on
each trial, and independent trials.

4. P ( X  4)  P( X  3)  binomcdf (6, 0.3,3)  0.92953

5.  X  np  6(0.3)  1.8  X  np(1  p)  6(0.3)(0.7)  1.1225

6. This is a geometric situation, since we are waiting on the first “success.” For this geometric
distribution, p = 0.3. Let Y = number of trials up to and including the first person stopped.
P (Y  7)  (0.7)6 (0.3)  0.0353 OR geometpdf(0.3,7)=0.0353

7. In our simulation, randBin(50,0.3) = 22.

8.

300 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


9.

10.

11. The histogram of the sample proportions of people who were stopped seems to be roughly
symmetrical. By tracing, we determined that the histogram is centered at about 0.30.

12.

In our 50 simulations, the mean proportion of passengers who were stopped is 0.3036. The
standard deviation of these 50 sample proportions is 0.571.

13. For instance, RandBin(100,0.3,50)/100L2 will accomplish the 100-passenger simulation.

14. Theoretical answers are given in the table.


Sample size Standard deviation
25 0.09165
50 0.0648
75 0.0529
100 0.0458
150 0.0374

15. As sample size increases, the standard deviation decreases, but not linearly.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 301


16. Answers will vary.

17. We might be suspicious if the proportion stopped was more than 2 standard deviations from
what we expect (0.30). So, reject the Customs officials’ claim if
0.3 1  0.3
pˆ  0.30  2  0.30  0.130  0.170 or
50
0.3 1  0.3
pˆ  0.30  2  0.30  0.130  0.430
50

302 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.1A AP Statistics Name:
1. For each description below, identify each underlined number as a parameter or statistic. Use
appropriate notation to describe each number, e.g., pˆ  0.96 .

(a) A 1993 survey conducted by the Richmond Times-Dispatch one week before election day
asked voters which candidate for the state’s attorney general they would vote for. 37% of
the respondents said they would vote for the Democratic candidate. On election day,
41% actually voted for the Democratic candidate.

(b) The National Center for Health Statistics reports that the mean systolic blood pressure for
males 35 to 44 years of age is 128 and the standard deviation is 15. The medical director
of a large company looks at the medical records of 72 executives in this age group and
finds that the mean systolic blood pressure for these executives is 126.07.

2. Suppose two different statistics—call them Statistic A and Statistic B—can be used to
estimate the same population parameter. Statistics A has lower bias than B, but A also has
high variability compared to B. On the two axes below, draw two parallel dotplots showing
8 values of each statistic that are consistent with these characteristics. Assume that the
parameter value is at the arrow on the axes.

Statistic A

Statistic B

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 303


3. A large pet store that specializes in tropical fish has several thousand guppies. The store
claims that the guppies have a mean length of 5 cm and a standard deviation of 0.5 cm. You
come to the store and buy 10 randomly-selected guppies and find that the mean length of
your 10 guppies is 4.8 cm. This makes you suspect that the mean fish length is not what the
store says it is. To explore this further, you assume that the length of guppies is Normally
distributed and use a computer to simulate 200 samples of 10 guppies from the store’s
claimed population. Below is a dotplot of the means from these 200 samples.

(a) What is the population in this situation, and what population parameters have we been
given?

(b) The distribution of one sample is described in the opening paragraph. What information
have we been given about this sample?

(c) Is the dotplot above a sampling distribution? Explain.

(d) Do you think the store is being honest about the length of its guppies? Justify your
answer.

304 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.1B AP Statistics Name:
1. For each description below, identify each underlined number as a parameter or statistic. Use
appropriate notation to describe each number, e.g., pˆ  0.96 .

(a) Nationwide, 84% of people are living in the same house they were living in one year ago.
The town council of Pleasant Valley surveys 100 residents and find that 75% of them
have not moved in the past year.

(b) The mean birthweight of infants in the United States in 2006 was 3.3 kg with a standard
deviation of 0.57 kg. An obstetrician determines that among her own patients, the mean
birthweight is 3.6 kg.

2. Suppose two different statistics—call them Statistic A and Statistic B—can be used to
estimate the same population parameter. Statistic A has lower bias than B, and A also has
low variability compared to B. On the two axes below, draw two parallel dotplots showing
8 values of each statistic that are consistent with these characteristics. Assume that the
parameter value is at the arrow on the axes.

Statistic A

Statistic B

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 305


3. Inexpensive bathroom scales are not consistently accurate. A manufacturer of bathroom
scales says that when a 150 pound weight is placed on all the scales produced in his factory,
the weight indicated by the scales is Normally distributed with a mean of 150 pounds and a
standard deviation of 2 pounds. A consumer advocacy group acquires a randomly-selected
group of 12 scales from the manufacturer and weighs a 150 weight on each one. They get a
mean weight of 151 pounds, which makes them suspicious about the company’s claim. To
test this, they use a computer to simulate 200 samples of 12 scales from a population with a
mean of 150 pounds and standard deviation 2 pounds. Below is a dotplot of the means from
these 200 samples.

(a) What is the population in this situation, and what population parameters have we been
given?

(b) The distribution of one sample is described in the opening paragraph. What information
have we been given about this sample?

(c) Is the dotplot above a sampling distribution? Explain.

(d) Do you think the manufacturer is being honest about the accuracy of its bathroom scales?
Justify your answer.

306 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.1C AP Statistics Name:
1. Suppose that in a certain community, 40% of the residents would answer “Yes” to the
question, “Do you know the names of at least five other people who live on your block?”
Suppose you plan to take a random sample of 100 people from this community and calculate
the proportion of people in your sample whose response to this question is “Yes”.

(a) What are the parameter and the statistic in this situation?

(b) What does the sampling distribution of this statistic describe?

(c) What does it mean to say that the statistic in this case is a unbiased estimator of the
parameter?

(d) Suppose that in a much larger community, 40% of the residents would also answer “Yes”
to the question. If you took a sample of 100 individuals from this much larger
community, would the sampling distribution of the statistic be different? In what way?

(e) If you took a sample of 50 individuals instead of 100 from the original community, would
the sampling distribution of the statistic change? In what way?

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 307


2. Below are histograms of the values taken by three sample statistics in several hundred
samples from the same population. The true value of the population parameter is marked on
each histogram.

(a) Which statistic has the largest bias among these three? Justify your answer.

(b) Which statistic has the lowest variability among these three? Justify your answer.

(c) Based on the performance of the three statistics in many samples, which is preferred as an
estimate of the parameter? Why?

308 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.2A AP Statistics Name:
1. According to the 2000 U.S. Census, 80% of Americans over the age of 25 have earned a high
school diploma. Suppose we take a random sample of 120 Americans and record the
proportion, pˆ , of individuals in our sample that have a high school diploma.

(a) What are the mean and standard deviation of the sampling distribution of pˆ ?

(b) What is the approximate shape of the sampling distribution? Justify your answer.

(c) Suppose our sample size was 30 instead of 120. Compare the shape, center, and spread
of this sampling distribution to the one in parts (a) and (b).

(d) You live in a small town with only 500 residents over the age of 25. What is the largest
possible sample you can take from your town and still be able to calculate the standard
deviation of sampling distribution of p̂ using the method presented in the textbook?
Explain.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 309


2. George is a big fan of music from the 1960s, and 22% of the songs on his mp3 player are
Beatles songs. Suppose George sets his mp3 player to “shuffle,” so that it selects songs
randomly (assume the shuffle function permits repetition of songs). During a long drive,
George plays 50 randomly-selected songs.

(a) What are the mean and standard deviation of the proportion of the 50 randomly-selected
songs that are Beatles songs?

(b) Calculate the probability that more than 30% of the 50 randomly-selected songs are
Beatles songs.

310 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.2B AP Statistics Name:
1. According to a poll, 22% of high school students in the United Kingdom say that Dobby is
their favorite character in the Harry Potter books. Let’s assume this is the parameter value
for the entire population of high school students in the U.K. You take a sample of 150 high
school students and record the proportion, pˆ , of individuals in your sample who say Dobby is
their favorite character.

(a) What are the mean and standard deviation of the sampling distribution of pˆ ?

(b) What is the approximate shape of the sampling distribution? Justify your answer.

(c) Suppose our sample size was 36 instead of 150. Compare the shape, center, and spread
of this sampling distribution to the one in parts (a) and (b).

(d) A small town in the U.K. has only 600 high school students. What is the largest possible
sample you can take from this town and still be able to calculate the standard deviation of
the sampling distribution of p̂ using the method presented in the textbook? Explain.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 311


2. Power companies severely trim trees growing near their lines to avoid power failures due to
falling limbs in storms. Applying a chemical to slow the growth of the trees is cheaper than
trimming, but the chemical kills some of the trees. Suppose that one such chemical would
kill 20% of sycamore trees. The power company tests the chemical on 250 sycamores.
Consider these an SRS from the population of all sycamore trees.

(a) What are the mean and standard deviation of the proportion of trees that are killed in
samples of 250 trees?

(b) Calculate probability that at least 60 trees (24% of the sample) are killed.

312 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.2C AP Statistics Name:
1. Suppose you are going to roll a fair six-sided die 60 times and record pˆ , the proportion of
times that a 1 or a 2 is showing.

(a) What is the mean of the sampling distribution of pˆ ?

(b) What is the standard deviation of the sampling distribution of pˆ ?

(c) Describe the shape of the sampling distribution of p̂ . Justify your answer.

(d) Suppose that when you actually roll the die 60 times, you get 30 rolls of 1 or 2, for a p̂ of
0.5. Are you suspicious about whether the die is fair? Justify your answer.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 313


2. Are attitudes toward shopping changing? Sample surveys show that fewer people enjoy
shopping than in the past. A recent survey asked a nationwide random sample of 2500 adults
if they agreed or disagreed with the statement, “I like buying new clothes, but shopping is
often frustrating and time-consuming.” In this survey, 1520 agreed. Suppose that in fact
60% of all adult U.S. residents would say “Agree” if asked the same question.

(a) What is the sample proportion of U.S. adults who agreed with the statement?

(b) If, in fact, the proportion of all U.S. adults who would agree with the statement is 0.60,
what is the probability that the proportion in a random sample of 2500 adults is as far
from 0.60—above or below—as the results of this survey? Check that the necessary
conditions are met before calculating this probability.

314 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.3A AP Statistics Name:
1. The customer care manager at a cell phone company keeps track of how long each help-line
caller spends on hold before speaking to a customer service representative. He finds that the
distribution of wait times for all callers has a mean of 12 minutes with a standard deviation of
5 minutes. The distribution is moderately skewed to the right. Suppose the manager takes a
random sample of 10 callers and calculates their mean wait time, x .

(a) What is the mean of the sampling distribution of x ?

(b) Is it possible to calculate the standard deviation of x ? If it is, do the calculation. If it isn’t,
explain why.

(c) Do you know the approximate shape of the sampling distribution of x ? If so, describe the
shape and justify your answer. If not, explain why not.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 315


2. The weights of Granny Smith apples from a large orchard are Normally distributed with a
mean of 380 gm and a standard deviation of 28 gm.

(a) A single apple is selected at random from this orchard. What is the probability that it
weighs more 400 gm?

(b) Three apples are selected at random from this orchard. What is the probability that their
mean weight is greater than 400 gm.?

(c) Explain why the probabilities in (a) and (b) are not equal.

316 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.3B AP Statistics Name:
1. The service department of a large automobile dealership keeps records of the odometer
readings of cars that it repairs and determines that the distribution of miles driven per year by
all of its customers has a mean of 14,000 miles and a standard deviation of 4000. The
distribution is skewed to the right. Suppose a random sample of 12 cars is taken from the
hundreds of cars for which they have records, and mean number of miles per year, x , is
calculated.

(a) What is the mean of the sampling distribution of x ?

(b) Is it possible to calculate the standard deviation of x ? If it is, do the calculation. If it


isn’t, explain why.

(c) Do you know the approximate shape of the sampling distribution of x ? If so, describe
the shape and justify your answer. If not, explain why not.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 317


2. The weights of newborn children in the United States vary according to the Normal
distribution with mean 7.5 pounds and standard deviation 1.25 pounds. The government
classifies a newborn as having low birth weight if the weight is less than 5.5 pounds.

(a) What is the probability that a baby chosen at random weighs less than 5.5 pounds at birth?

(b) You choose three babies at random and compute their mean weight, x . What is the
probability that their average birth weight is less than 5.5 pounds?

(c) Explain why the probabilities in (a) and (b) are not equal.

318 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Quiz 7.3C AP Statistics Name:
1. The distribution of actual weights of 8-ounce chocolate bars produced by a certain machine is
Normal with mean 8.1 ounces and standard deviation 0.1 ounces. Company managers do not
want the weight of a chocolate bar to fall below 7.85 ounces, for fear that consumers will
complain.

(a) Find the probability that the weight of a randomly selected candy bar is less than 7.85
ounces.

Four candy bars are selected at random and their mean weight, x , is computed.

(b) Describe the center, shape, and spread of the sampling distribution of x .

(c) Find the probability that the mean weight of the four candy bars is less than 7.85 ounces.

(d) Would your answers to (a), (b), or (c) be affected if the weights of chocolate bars produced
by this machine were distinctly non-Normal? Explain.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 319


2. Suppose you roll a six-sided 50 times and calculate the mean roll, x .

(a) We can consider these 50 rolls as a SRS of rolls from the population of all possible rolls of
this die. Using formulas from Chapter 6, we can determine that if the die is fair, the mean of
this population is 3.5 and the standard deviation 1.71. How would you describe the shape of
this population distribution?

(b) Find the shape, center, and spread of the sampling distribution of x .

(c) Suppose that the mean of your 50 rolls is x  3.25. Are you suspicious about the fairness of
the die? Justify your answer.

320 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Test 7A AP Statistics Name:
Part 1: Multiple Choice. Circle the letter corresponding to the best answer.

1. Following a dramatic drop of 500 points in the Dow Jones Industrial Average in September
1998, a poll conducted for the Associated Press found that 92% of those polled said that a year
from now their family financial situation will be as good as it is today or better. Which of the
following terms describes the number 92%?
(a) statistic.
(b) sample.
(c) sample parameter.
(d) population parameter.
(e) population.

2. What is distribution of values taken by a statistic in all possible samples of the same size from
the same population called?
(a) the probability that the statistic is obtained.
(b) the population parameter.
(c) the variance of the values.
(d) the sampling distribution of the statistic.
(e) the distribution of sample data.

3. If a statistic used to estimate a parameter is such that the mean of its sampling distribution is
equal
to the true value of the parameter being estimated, what is the statistic said to be?
(a) random
(b) biased
(c) a proportion
(d) unbiased
(e) non-varying.

4. Suppose you take a random sample of size 25 from a population with mean of 120 and a
standard deviation of 15. Your sample has a mean of 115 and a standard deviation of 13.8.
Which of the following has a mean of 120 and a standard deviation of 3?
(a) the distribution of the population
(b) the distribution of the sample data.
(c) the sampling distribution of the sample mean.
(d) the sampling distribution of the population mean.
(e) No important distribution related to this situation has the given mean and standard
deviation.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 321



5. In order to use the formula  X  to calculate the standard deviation of the sampling
n
distribution of the sample mean, which of the following conditions must be met?

I. n  30
II. The population’s distribution is approximately Normal.
III. The sample size is less than 10% of the population size.

(a) I only
(b) II only
(c) III only
(d) III and either I or II
(e) All three conditions must be met.

6. The central limit theorem refers to which of the following characteristic of the sampling
distribution of the sample mean?
(a) Regardless of the shape of the population’s distribution, the sampling distribution of the
sample mean from sufficiently large samples will be approximately Normally distributed.
(b) Regardless of the shape of the population’s distribution, the standard deviation of the

sampling distribution of the sample mean from sufficiently large samples will be .
n
(c) Regardless of the shape of the population’s distribution, the mean of the sampling
distribution of the sample mean from sufficiently large samples will be equal to the mean
of the population.
(d) As you take larger and larger samples from a Normally distributed population, the
standard deviation of the sampling distribution of the sample mean gets smaller and
smaller.
(e) As you take larger and larger samples from a Normally distributed population, the mean
of the sampling distribution of the sample mean gets closer and closer to the population
mean.

7. A simple random sample of 1000 Americans found that 62% were satisfied with the service
provided by the dealer from which they bought their car. A simple random sample of 1000
Canadians found that 59% were satisfied with the service provided by the dealer from which
they bought their car. The sampling variability associated with these statistics is
(a) exactly the same.
(b) not exactly the same, but very close.
(c) much smaller for the sample of Canadians because the population of Canada is much
smaller than that of the United States, hence the sample is a larger proportion of the
population.
(d) smaller for the sample of Canadians because the percent satisfied was smaller than that
for the Americans.
(e) larger for the Canadians because Canadian citizens are more widely dispersed throughout
the country than are citizens in the United States, hence they have more variable views.

322 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


8. A student investigating study habits asks a simple random sample of 16 students at her
school how many minutes they spent on their English homework the previous night.
Suppose the actual parameter values for this variable are   45 minutes and   15 minutes.
Which of the following best describes what we know about the sampling distribution of
means for the student’s sample?
(a)  x  45;  x unknown; shape of distribution unknown
(b)  x  45;  x  15; distribution approximately Normal
(c)  x  45;  x  15; shape of distribution unknown
(d)  x  45;  x  3.75; distribution approximately Normal
(e)  x  45;  x  3.75; shape of distribution unknown

9. Olive weights are classified according to a unique set of adjectives implying great size. For
example, the mean weight of olives classified as “Colossal” is 7.7 gm. Suppose a particular
company’s crop of “Colossal” olives is approximately Normally distributed with a mean of
7.7 gm and a standard deviation of 0.2 gm. Which of the following represents the probability
that the mean weight of a random sample of 3 olives from this population is greater than 8
gm?
     
 8  7.7   7.7  8   8  7.7 
(a) P  z   (b) P  z   (c) P  z  
 0.2   0.2   0.2 
     
 3   3   3 
   
   
 8  7.7   8  7.7 
(d) P z  (e) P z 
  0.2  0.8     0.2  0.8  
  
 3   3 

10. In a large population, 46% of the households own DVD recorders. A simple random sample
of 100 households from this population is to be contacted and the sample proportion
computed. Which of the following expressions represents the probability that more than half
the households sampled will own a DVD recorder?
   
     
 0.46  0.5   0.5  0.46  0.46  0.5
(a) P z  (b) P z  (c) P  z  
  0.5  0.5     0.5  0.5     0.46  0.54 100  
     
 100   100 
   
   
 0.46  0.5   0.5  0.46 
(d) P z  (e) P z 
  0.46  0.54     0.46  0.54  
  
 100   100 

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 323


Part 2: Free Response
Show all your work. Indicate clearly the methods you use, because you will be graded on the
correctness of your methods as well as on the accuracy and completeness of your results and
explanations.

11. A study of college freshmen’s study habits found that the time (in hours) that freshmen use
to study each week follows a distribution with a mean of 7.2 hours and a standard deviation
of 5.3
hours.

(a) Can you calculate the probability that a randomly chosen freshman studies more than 9
hours? If so, do it. If not, explain why not.

(b) What is the shape of the sampling distribution of the mean x for samples of 55 randomly
selected freshmen? Justify your answer.

(c) In all possible samples of size 55, what is the mean and standard deviation of the distribution
of ?

(d) Find the probability that the average number of hours spent studying by an SRS of 55
students is greater than 9 hours. Show your work.

324 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


12. A friend has offered to play a gambling game with you that involves flipping a coin that he
has provided. Since a flip of heads will be to his advantage, you want to test the coin for
fairness before you begin to play. Your friend is willing to let you flip the coin 50 times to
determine if the probability of getting heads is actually 0.50, as it should be if the coin is fair.

(a) Assume for the moment that the coin is fair. If p̂ is the proportion of heads in 50 flips of
the coin, what are the mean and standard deviation of the sampling distribution of p̂ ?

(b) Explain why you can use the formula for the standard deviation of p̂ in this setting.

(c) You flip the coin 50 times and get 30 heads. Do you risk insulting your friend by
refusing to play with his coin? Support your answer with an appropriate probability
calculation.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 325


13. Teenagers send many text messages—recent polls cite medians of more than 50 per day.
Consider a large population of teenagers for whom the distribution of the number of text
messages sent per day is strongly skewed to the right. Here is the five-number summary for
number of texts per day for this population:

Minimum = 0 Quartile 1 = 20 Median = 55 Quartile 3 = 140 Maximum = 250

Suppose we take random samples of size 32 from this population and calculate Q1 for each of
our samples. Below is a dotplot of Q1 from 50 such samples.

(a) Briefly explain what the dot at 35 represents.

(b) Is the sample Q1 an unbiased estimator of the population Q1? Justify your answer.

326 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Test 7B AP Statistics Name:
Part 1: Multiple Choice. Circle the letter corresponding to the best answer.

1. In a study of the effects of acid rain, a random sample of 100 trees from a particular forest is
examined. Forty percent of the trees show some signs of damage. Which of the following
statements is correct?
(a) 40% is a parameter
(b) 40% is a statistic
(c) 40% of all trees in the forest show some signs of damage
(d) More than 40% of the trees in the forest show some signs of damage
(e) Less than 40% of the trees in the forest show some signs of damage

2. The sampling distribution of a statistic is


(a) the probability that we obtain the statistic in repeated random samples.
(b) the mechanism that determines whether randomization was effective.
(c) the distribution of values taken by a statistic in all possible samples of the same sample
size from the same population.
(d) the extent to which the sample results differ systematically from the truth.
(e) the distribution of values in a sample of size n from the population

3. A statistic is said to be unbiased if


(a) the survey used to obtain the statistic was designed so as to avoid even the hint of racial
or sexual prejudice.
(b) the mean of its sampling distribution is equal to the true value of the parameter being
estimated.
(c) both the person who calculated the statistic and the subjects whose responses make up the
statistic were truthful.
(d) the value from any sample is equal to the parameter being estimated.
(e) it is used for honest purposes only.

4. Which of the following distributions has a mean that varies from sample to sample?
I. The population distribution
II. The distribution of sample data
III. The sampling distribution
(a) I only
(b) II only
(c) III only
(d) II and III
(e) all three distributions

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 327


5. You take a sample of size 25 from a very large population in which the true proportion is
p  0.1 , thus violating the condition that np  10 and n 1  p   10. Which statement below
best describes what you know about the sampling distribution of pˆ ?

(a)  pˆ  0.1; pˆ 
 0.1 0.9  ; the distribution is not approximately Normal.
25

(b)  pˆ  0.1; pˆ 
 0.1 0.9  ; the distribution is approximately Normal.
25

(c)  pˆ  0.1; we cannot use the formula  pˆ 


 0.1 0.9  ; the distribution is approximately
25
Normal.

(d)  pˆ  0.1; we cannot use the formula  pˆ 


 0.1 0.9  ; the distribution is not
25
approximately Normal.

(e) We do not know  pˆ ;  pˆ 


 0.1 0.9  ; the distribution is not approximately Normal.
25

6. The number of hours a light bulb burns before failing varies from bulb to bulb. The
distribution of burnout times is strongly skewed to the right. The central limit theorem says
that
(a) as we look at more and more bulbs, their mean burnout time gets ever closer to the
mean  for all bulbs of this type.
(b) the mean burnout time for any number of bulbs has a distribution of the same shape
(strongly skewed) as the distribution for individual bulbs.
(c) the mean burnout time for any number of bulbs has a distribution that is close to Normal.
(d) the mean burnout time for a large number of bulbs has a distribution of the same shape
(strongly skewed) as the distribution for individual bulbs.
(e) the mean burnout time for a large number of bulbs has a distribution that is close to
Normal.

7. You take an SRS of size 500 from the 37,000 students at Purdue University and measure
individual’s heights. You then take an SRS of size 500 from the 4,400,000 adults in the state
of Indiana and measure their heights. Assuming the standard deviation of individual heights
in the two populations is the same, the standard deviation of the sampling distribution of
mean heights for the Indiana sample is
(a) approximately the same as for the Purdue sample because both are samples of size 500.
(b) smaller than for the Purdue sample because the population of Indiana is much larger.
(c) larger than for the Purdue sample because the population of Indiana is much larger.
(d) larger, because the Indiana sample is smaller relative to the population from which it’s
been taken.
(e) either larger or smaller than for the Purdue sample because it varies from sample to
sample.

328 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


8. The chipmunk population in a certain area is known to have a mean weight of 84 gm and a
standard deviation of 18 gm. A wildlife biologist weighs 9 chipmunks that have been caught
in live traps before releasing them. Which of the following best describes what we know
about the sampling distribution of means for the biologist’s sample?
(a)  x  84;  x  18; distribution approximately Normal
(b)  x  84;  x  6; shape of distribution unknown
(c)  x  84;  x  6; distribution approximately Normal
(d)  x  84;  x unknown; distribution approximately Normal
(d)  x  84;  x unknown; shape of distribution unknown

9. Interpupillary distance (IPD) is the distance between the centers of the pupils of a person’s
left and right eyes. In adult males IPD is approximately Normally distributed with a mean of
62.5 mm and a standard deviation of 6 mm. Suppose you randomly select 5 adult males.
What is the probability that their mean IPD is greater than 60 mm?
 
 60  62.5   62.5  60   60  62.5 
(a) P  z   (b) P  z   (c) P  z  
 6   6   6 
 
 5 
   
 60  62.5   62.5  60 
(d) P  z   (e) P  z  
 6   6 
   
 5   5 

10. A survey asks a random sample of 500 adults in Ohio if they support an increase in the state
sales tax from 5% to 6%, with the additional revenue going to education. Let p̂ denote the
proportion in the sample who say they support the increase. Suppose that 53% of all adults
in Ohio support the increase. What is the probability that less than half the sample will say
they support the increase?
 0.5  0.53   0.5  0.53 
(a) P  z   (b) P  z  

  0.5  0.5  500  


  0.53  0.47  500  

   
   
 0.53  0.5   0.5  0.53 
(c) P z  (d) P z 
  0.5  0.5     0.53 047  
  
 500   500 
 
 
 0.5  0.53 
(e) P z 
  0.53 0.47  

 500 

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 329


Part 2: Free Response
Show all your work. Indicate clearly the methods you use, because you will be graded on the
correctness of your methods as well as on the accuracy and completeness of your results and
explanations.

11. The weight of the eggs produced by a certain breed of hen is Normally distributed with mean
65 grams (g) and standard deviation 5 g.

(a) Calculate the probability that a randomly selected egg weighs between 62.5 g and 68.75
g. Show your work.

(b) Think of cartons of such eggs as SRSs of size 12 from the population of all eggs.
Calculate the probability that the mean weight of the eggs in a carton falls between 62.5 g
and 68.75 g. Show your work.

(c) Did you need to know that the population distribution of egg weights was Normal in order
to complete parts (a) or (b)? Justify your answer.

330 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


12. Companies are interested in the demographics of those who listen to the radio programs they
sponsor. A radio station has determined that only 20% of listeners phoning in to a morning
talk program are male. The station management wonders if adding a male host to the
program will increase the proportion of callers who are male. After adding the male host, the
station records the gender of 200 people who phone in to the program during a particular
week. The station is willing to view these 200 callers as an SRS from the population of all
those who call in to this program.

(a) For the moment, assume that the addition of the male host has no effect on the proportion
of callers who are male. If p̂ is the proportion of callers in the sample who are male,
what are the mean and standard deviation of the sampling distribution of p̂ ?

(b) What assumption are you making when you use the formula for the standard deviation of
p̂ in this setting?

(c) In fact, during this particular week, 50 of the 200 callers were male. Does this provide
sufficient evidence to suggest that the proportion of male callers has increased from
20%? Support your answer with an appropriate probability calculation.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 331


13. Buying a year’s worth of textbooks for college can be expensive! Consider a large
population of college students for whom the distribution of the annual cost of textbooks is
slightly skewed to the left. Here is the five-number summary for the annual cost of textbooks
for this population:

Minimum = 80 Quartile 1 = 215 Median = 335 Quartile 3 = 380 Maximum = 440

Suppose we take random samples of size 32 from this population and calculate the
interquartile range (IQR) for each of our samples. Below is a dotplot of the IQR from 50
such samples.

(a) Briefly explain what the dot at 240 represents.

(b) Is the sample IQR is an unbiased estimator of the population IQR? Justify your answer.

332 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Test 7C AP Statistics Name:
Part 1: Multiple Choice. Circle the letter corresponding to the best answer.

1. According to the DuPont 2007 Global Automotive Color Popularity Report, 19% of all cars
manufactured in 2007 were white. In a random sample of 100 cars parked in long-term
parking at Philadelphia International Airport, 22% of the cars were white. Which of the
following statements is true?
(a) 19% and 22% are parameters, 100 is a statistic.
(b) 19% is a parameter, 22% is a statistic.
(c) 19% is a statistic, 22% is a parameter.
(d) 19% and 22% are statistics, 100 is neither a parameter nor a statistic.
(e) 19%, 22%, and 100 are all statistics.

2. The best statistic for estimating a parameter has which of the following characteristics?
(a) Low bias, high variability.
(b) High bias, high variability.
(c) Low bias. Since a statistic should always equal the parameter it is estimating, there
should be no variability.
(d) Low bias, low variability.
(e) Low bias, high variability.

3. Which of the following best describes a sampling distribution?


(a) The distribution of all values of a statistic found in a large number of simulated samples
of
size n.
(b) The set of all values of a variable in a sample of size n.
(c) The set of all values of a variable in a large number of samples of size n.
(d) The distribution of parameter values in all possible samples of size n.
(e) A probability distribution that describes the relative likelihood of all possible values of a
statistic.

4. In a large population of adults, the mean IQ is 112 with a standard deviation of 20. Suppose
200 adults are randomly selected for a market research campaign. The sampling distribution
of the sample mean IQ is
(a) exactly Normal, mean 112, standard deviation 20.
(b) approximately Normal, mean 112, standard deviation 0.1.
(c) approximately Normal, mean 112, standard deviation 1.414.
(d) approximately Normal, mean 112, standard deviation 20.
(e) exactly Normal, mean 112, standard deviation 1.414.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 333


5. In a study of the effects of acid rain, a random sample of 100 trees from a particular forest is
examined. Forty percent of these show some signs of damage. Which of the following
statements is correct?
(a) The sampling distribution of the sample proportion of damaged trees is approximately
Normal.
(b) If we took another random sample of trees, we would find that 40% of these would show
some signs of damage.
(c) If a sample of 1000 trees was examined, the variability of the sampling distribution of the
sample proportion would be larger than in a sample of 100 trees.
(d) The standard deviation of the sampling distribution of the sample proportion of damaged
trees is 100  0.4  0.6  .
(e) None of the above statements is correct.

6. Which of the following statements is/are true when taking an SRS from a large population?
I. The sampling distribution of x has standard deviation  n even if the population is
not Normally distributed.
II. The sampling distribution of x is Normal if the population has a Normal distribution.
III. When n is large, the sampling distribution of x is approximately Normal even if the
population is not Normally distributed.

(a) I and II
(b) I and III
(c) II and III
(d) I, II, and III
(e) None of the above gives the correct set of responses.

7. Suppose you are sampling from a distribution that is strongly skewed left. Which of the
following statements about the sampling distribution of the sample mean is true?
(a) As the sample size increases, the shape of the sampling distribution gets closer and closer
to a Normal distribution.
(b) As the sample size increases, the shape of the sampling distribution gets closer and closer
to the shape of the population distribution.
(c) As the sample size increases, the mean of the sampling distribution gets closer to the
population mean.
(d) Regardless of the sample size, the shape of the sampling distribution is similar to the
shape of the population distribution.
(e) Regardless of the sample size, the standard deviation of the sampling distribution is
approximately equal to the standard deviation of the population.

334 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


8. Suppose we wish to estimate the percentage of students who smoke cigarettes at each of
several colleges and universities. Two of the colleges are Wabash College (enrollment 900)
and Purdue University (enrollment 36,000). What should the relative size of our samples be
from each school if we want the two sampling distributions to have approximately the same
standard deviation?
(a) Because the population sizes are so different, it’s impossible to make the standard
deviations equal by adjusting the two sample sizes.
(b) We should take a larger number of Purdue students, since there are more of them.
(c) We should take a larger number of Wabash students, since there are fewer of them.
(d) We should take the same size samples from each school.
(e) We should take samples that are exactly 10% of each school’s enrollment.

p 1  p 
9. In order to use the formula  pˆ  to calculate the standard deviation of the sampling
n
distribution of the sample proportion, which of the following conditions must be met?

I. np  10 and n 1  p   10
II. The population’s distribution is approximately Normal.
III. The sample size is less than 10% of the population size.

(a) I only
(b) II only
(c) III only
(d) I and III
(e) All three conditions must be met.

10. A census of the labor force in a large metropolitan area found that the time it takes for people
to commute to work has a mean of 20.5 minutes and a standard deviation of 15.4 minutes.
What is the probability that a random sample of 40 people have a mean commute time that is
greater than 25 minutes?
     
 20.5  25   25  20.5   25  20.5 
(a) P  z   (b) P  z   (c) P  z  
 15.4   15.4   15.4 
     
 40   40   39 
 25  20.5   20.5  24 
(d) P  z   (e) P  z  
 15.4   15.4 

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 335


Part 2: Free Response
Show all your work. Indicate clearly the methods you use, because you will be graded on the
correctness of your methods as well as on the accuracy and completeness of your results and
explanations.

11. A certain beverage company is suspected of underfilling its cans of soft drink. The company
advertises that its cans contain, on average, 12 ounces of soda with standard deviation 0.4
ounce.
For the questions that follow, suppose that the company is telling the truth.

(a) Can you calculate the probability that a single randomly selected can contains 11.9 ounces
or less? If so, do it. If not, explain why you cannot.

(b) A quality control inspector measures the contents of an SRS of 50 cans of the company’s
soda and calculates the sample mean x . What are the mean and standard deviation of the
sampling distribution of x for samples of size n = 50?

(c) The inspector in part (b) obtains a sample mean of x  11.9 ounces. Calculate the
probability that a random sample of 50 cans produces a sample mean amount of 11.9 ounces
or less. Be sure to explain why you can use a Normal calculation.

(d) What would you conclude about whether the company is underfilling its cans of soda?
Justify your answer.

336 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


12. An opinion poll asks a sample of 500 adults (an SRS) whether they favor giving parents of
school-age children vouchers that can be exchanged for education at any public or private
school of their choice. Each school would be paid by the government on the basis of how
many vouchers it collected. Suppose that in fact 45% of the population favor this idea.

(a) What is the mean of the sampling distribution of p̂ , the proportion of adults in samples
of 500 who favor giving parents of school-age children these vouchers?

(b) What is the standard deviation of p̂ ?

(c) Check that you can use the Normal approximation for the distribution of p̂ .

(d) What is the probability that more than half of the sample are in favor? Show your work.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 337


13. A small internet mail-order company keeps track of the number of orders it fills per day for
many years and determines that the distribution of the variable “orders filled per day” is
roughly symmetric and has the following five-number summary:

Minimum = 20 Quartile 1 = 32 Median = 46 Quartile 3 = 63 Maximum = 80

Suppose we take random samples of size 40 from this distribution and calculate the range for
each of our samples. Below is a dotplot of the ranges from 50 such samples.

Is the sample range an unbiased estimator of the population range? Use the dotplot above to
justify your answer.

338 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Chapter 7 Solutions

Quiz 7.1A
1. (a) pˆ  0.37 is a statistic; p  0.41 is a parameter. (b)   128 is a parameter;   15 is a
parameter; x  126.07 is a statistic. 2. Answers will vary, but the center of values for Statistic
A should be closer to the arrow than the center of the values for Statistic B, and there should be
more spread in the values for Statistic A. 3. (a) The population is all the guppies in the pet
store. We’ve been given the population mean   5 cm and the population standard deviation
  0.5 cm. (b) The sample mean is x  4.8 cm and the sample size is n  10. (c) No, it’s
merely an approximation of a sampling distribution generated by simulating 200 sample means.
The actual sampling distribution includes the means from all possible samples of size 10 from
the population—many more than 200 values. (d) 21 out of 200, or 10.5% of the sample means
in our simulation are as far or farther below 5.0 as our sample was. Our sample is not
sufficiently unusual to arouse suspicions about the store’s claim.

Quiz 7.1B
1. (a) p  0.84 is a parameter; pˆ  0.75 is a statistic. (b)   3.3 kg is a parameter;   0.57 kg
is a parameter; x  3.6 kg is a statistic. 2. Answers will vary, but the center of values for
Statistic A should be closer to the arrow than the center of the values for Statistic B, and there
should be more spread in the values for Statistic B. 3. (a) The population is all bathroom scales
produced by the manufacturer. We’ve been given the population mean   150 pounds and the
population standard deviation   2 pounds. (b) The sample mean is x  151 pounds and the
sample size is n  12. (c) No, it’s merely an approximation of a sampling distribution generated
by simulating 200 sample means. The actual sampling distribution includes the means form all
possible samples of size 12 from the population—many more than 200 values. (d) Only 8 out of
200, or 4% of the sample means in our simulation are as far or farther above 150 pounds as our
sample was. If the population mean is really 150 pounds, then our sample is unusual, and we
should be somewhat suspicious about the manufacturer’s claim.

Quiz 7.1C
1. (a) The parameter is the proportion of people in the entire community who would answer
“Yes” to the question. It’s equal to 0.40. The statistic is the proportion of people in the sample
of 100 who would answer “Yes” to the question. (b) The sampling distribution describes the
distribution of the proportion of people who would answer “Yes” to this question in all possible
samples of size 100 from this population. (c) The mean of the statistic’s sampling distribution is
equal to the parameter. (d) No. As long as the sample is less than 10% of the population, the
size of the population from which the sample is taken does not influence the sampling
distribution. (e) Yes. The standard deviation of the sampling distribution would be larger if the
sample size were smaller. 2. (a) C has the largest bias: the center of the histogram is clearly to
the left of the parameter value. (b) A has the lowest variability, since most values of the statistic
are close to the parameter value. (c) Distribution A is unbiased and has the lowest variability, so
it should give the best estimate.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 339


Quiz 7.2A
 0.30  .022 
1. (a)  pˆ  0.22;  pˆ  0.0589 (b) P ( pˆ  0.30)  P  z    P( z  1.36)  0.0869 . (c)
 0.0589 
 p̂ would not change,  p̂ would be larger (0.073) and the distribution would be non-Normal,
since n 1  p   30  0.2   6 , which is less than 10.
(d) The largest sample we can take is 50, otherwise the sample would be more than 10% of the
population, and sampling without replacement would require a finite population correction to
calculate standard deviation.
 0.08  0.06 
2. (a)  pˆ  0.06;  pˆ  0.0336 . (b) P  pˆ  0.08   P  z    P  z  0.60   0.2743 .
 0.0336 
Quiz 7.2B
1. (a)  pˆ  0.22;  pˆ  0.034 (b) Since np  150  0.22   33  10 and
n 1  p   150 .78   117  10 , the distribution is approximately normal. (c)  p̂ would not
change,  p̂ would be larger (0.069) and the distribution would be non-Normal, since
np  36  0.22   7.92 , which is less than 10.
(d) The largest sample we can take is 60, otherwise the sample would be more than 10% of the
population, and sampling without replacement would require a finite population correction to
calculate standard deviation.
 
 
 0.24  0.20 
2. (a)  pˆ  0.20;  pˆ  0.025 . (b) P  pˆ  0.24   P z   P  z  1.58   0.0571 .
  0.2  0.8  
 
 250 

Quiz 7.2C
1. (a) Since the probability of rolling a 1 or a 2 is 1/3, the mean of the sampling distribution is
1/3.
 1  2 
  
3 3 1 2
(b)  pˆ      0.061 . (c) Since np   60     20  10 and n 1  p   60    40  10 ,
60 3 3
the sampling distribution is approximately Normal.
 0.5  0.33 
(d) P  pˆ  0.5   P  z    P  z  2.73  0.0032 . The probability of getting a sample
 0.061 
proportion as high or higher than 0.5 if the true proportion of rolling a die and getting a 1 or a 2
1520
is 0.33 is so small that it seems unlikely that the die is fair. 2. (a) pˆ   0.608 . (b) The
2500
sample size (2500) is clearly less than 10% of the population (all adults);
np   2500  0.6   1500  10 and n 1  p   2500  0.4   1000  10 , so we can use the Normal
distribution to calculate this probability. We want to find P  pˆ  0.608  and then double it to

340 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


find the probability that a sample proportion is as far from 0.60 on either side of the Normal
 
 
 0.608  0.6 
curve. P  pˆ  0.608   P z   P  z  0.82   0.2061 times 2 = 0.4122.
  0.6  0.4  
 
 2500 

Quiz 7.3A
1. (a)  x  12 minutes (the same as the population mean). (b) Yes. It seems reasonable to
5
assume that the sample of 10 is less than 10% of the entire population calls.  x   1.58 .
10
(c) No. The population distribution is skewed, and n = 10, which is not large enough for the
central limit theorem to apply.
 400  380 
2. (a) P  x  400   P  z    P  z  0.71  0.2389
 28 
 
 400  380 
(b) P  x  400   P  z   P  z  1.24   0.1075 .
  28  
   
  3 
(c) The mean weight of a random sample of three apples is less variable than the weight of a
single randomly-selected apple, so we are less likely to get a mean weight that is 20 gm above
the mean when we take a sample of three apples.

Quiz 7.3B
1. (a)  x  14000 miles (the same as the population mean). (b) Yes. It seems reasonable to
assume that the sample of 12 is less than 10% of the entire population of customers.
4000
x   1154.70 . (c) No. The population distribution is skewed, and n = 12, which is not
12
large enough for the central limit theorem to apply.

 5.5  7.5 
2. (a) P  x  5.5   P  z    P  z  1.60   0.0548
 1.25 
 
 5.5  7.5 
(b) P  x  5.5   P  z    P  z  2.77   0.0028 .
 1.25 
 
 3 
(c) The mean weight of a random sample of three babies is less variable than the weight of a
single randomly-selected baby, so we are less likely to get a mean weight that is 2 pounds above
the mean when we take a sample of three babies.

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 341


Quiz 7.3C
 7.85  8.1  0.1
1. (a) P  x  7.85   P  z    P  z  2.50   0.0062 (b)  x  8.1 ,  x   0.05 ,
 0.1  4
shape is approximately Normal since the population distribution of weights is Normal.
 7.85  8.1 
(c) P  x  7.85   P  z    P  z  5   0 . (d) If the distribution were distinctly
 0.05 
non-Normal, we would not be able to calculate the probability in part (a). In part (b), the mean
and standard deviation of the sampling distribution would be the same, but the shape would be
non-Normal since n is too small for the central limit theorem to apply. Thus we would also not
be able to calculate the probability in part (c). 2. (a) The population would be a discreet
probability distribution with six possible outcomes—one through six—all of which had a
probability of 1/6. Thus its shape would be a uniform distribution. (b)
1.71
 x  3.5;  x   0.242 . Since n > 30, the means would be approximately Normally
50
distributed.
 3.25  3.5 
(c) P  x  3.25   P  z    P  z  1.03  0.1515 . There is roughly a 15% chance
 0.242 
of getting a mean of 3.25 or lower if the die is fair. (And there is roughly a 30% chance of
getting a mean this far from 3.5 in either direction). This is not a sufficiently unusual outcome
for us to be suspicious about the fairness of this die.

342 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


Test 7A
Part 1
1. a This proportion is the result of a poll—a sample—so it’s a statistic.
2. d This is the definition of a sampling distribution.
3. d Any time the center of a statistic’s sampling distribution is at the parameter value, the
statistic is unbiased.
4. c The mean of the sampling distribution is the same as the mean of the population, and the
 15
standard deviation is   3.
n 25
5. c Even if the sampling distribution is non-Normal, if the 10% condition is satisfied, this is
the appropriate formula for the standard deviation.
6. a See definition of central limit theorem on page 450 in text.

7. b For the U.S.,  pˆ 


 0.62  0.38  0.01535 . For Canada,
1000

 pˆ 
 0.59  0.41  0.01555
1000

8. e  x    45;  x   3.75. Since n is small and we don’t know the shape of the
n
population distribution, the shape of the sampling distribution is unknown.
   
 8  x   8    8  7.7 
9. a P  x  8   P  z    P  z    P  z  
 x      0.2 
   
 n   3 
   
   
 0.5  p   0.5  0.46 
10. e P  pˆ  0.5   P z  P z
  p 1  p     0.46  0.54  
  
 n   100 
Part 2
11. (a) No. We don’t know the shape of the distribution, so we can’t calculate this probability.
(b) Individual coin flips are independent. (Since the population of possible flips of this coin is
infinite, we need not be concerned about the 10% condition for finite populations.)
5.3  9  7.2 
(c)  x  7.2;  x   0.715. (d) P  x  9   P  z    P  z  2.52   0.0059 . 12.
55  0.715 

(a)  pˆ  p  0.5;  pˆ 
 0.5  0.5   0.071. (b) Since the population of possible flips of this
50
coin is infinite, we need not be concerned about the 10% condition for finite populations.
(c) The probability of getting 30 or more heads in 50 flips of a fair coin is
 0.6  0.5 
P  pˆ  0.6   P  z    P  z  1.41  0.0793 . Roughly 2 out of 25 times, we will get
 0.071 
this many or more heads. This is probably not unusual enough to risk your friend’s good will by
accusing him of using an unfair coin. 13. (a) The dot at 35 represents the first quartile of one of
the 50 samples taken from this population. (b) We know from the five-number summary that

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 343


the true population value for Q1 is 20. It appears that the distribution of sample first quartiles is
skewed right, but its median is 20, so it’s safe to say that the center of the distribution is near 20.
Therefore the sample first quartile is an unbiased estimator.

Test 7B
Part 1

1. b This proportion is the result of a sample, so it’s a statistic.


2. c This is the definition of a sampling distribution.
3. b When the center of a statistic’s sampling distribution is at the parameter value, the
statistic is unbiased.
4. b The mean of both the population and the sampling distribution are fixed—only the means
of individual samples vary.
5. a Since the 10% condition is satisfied, this is the appropriate formula for the standard
deviation. The sample is too small for the sampling distribution of p̂ to be Normal.
6. e See definition of central limit theorem on page 450 in text.
7. a The standard deviation of the sampling distribution of means depends on sample size, not
population size.

8. b  x    84;  x   6. Since n is small and we don’t know the shape of the
n
population distribution, the shape of the sampling distribution is unknown.
   
 60   x   60     60  62.5 
9. c P  x  60   P  z    P z     P z  
 x     6 
   
 n   5 
   
   
 0.5  p   0.5  0.53 
10. d P  pˆ  0.5   P z  P z
  p 1  p     0.53 0.47  
  
 n   500 

Part 2
 62.5  65 68.75  65 
11. (a) P  62.5  x  68.75   P  z   P  0.5  z  0.75   0.4649
 5 5 
 
 62.5  65 68.75  65 
(b) P  62.5  x  68.75   P  z   P  1.73  z  2.60   0.9535
 5 5 
 
 12 12 
(c) Yes. The calculations in both (a) and (b) assumed the Normality of the underlying
distribution. In (a) the population was given as Normal. In (b), we would not be able to assume
the Normality of the sampling distribution because the sample size is less than 30.

344 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers


12. (a)  pˆ  p  0.2;  pˆ 
 0.2  0.8  0.028.
(b) We can use this formula if we assume
200
that there are more than 10  200   2000 listeners who call in to the program.
(c) The probability of getting 50 or more males in 200 callers if the true proportion of males is
 0.25  0.2 
still 0.20 is P  pˆ  0.25   P  z    P  z  1.79   0.0367 . Roughly 1 out of 25
 0.028 
times, we will get this many or more male callers. This is probably unusual enough to suggest
that the true proportion of male listeners is higher than 0.20. 13. (a) The dot at 240 represents
the interquartile range of one of the 50 samples taken from this population. (b) We know from
the five-number summary that the true population value for IQR is 380 – 215 = 165. It appears
that the distribution of sample IQR’s is slightly skewed right, but its median is 165, so it’s safe to
say that the center of the distribution is near 165. Therefore the sample first quartile is an
unbiased estimator.

Test 7C
Part 1
1. b 19% is the proportion of white cars in the population (the parameter) and 22% is the
proportion of white cars in the sample (the statistic).
2. d Low bias means the statistic’s sampling distribution is centered around the parameter.
Low variability means the typical value of the statistic is close to the parameter.
(Variability in samples is inevitable, hence c is incorrect).
3. e A sampling distribution is values of a statistic from all possible samples of a given size
from the population. If we think of the statistic as a random variable, this is its
probability distribution.
4. c Since n > 30, the sampling distribution is approximately Normal, with
20
 x  112 and  x   1.414 .
100
5. a Since n > 30, the sampling distribution is approximately Normal. None of the other
statements is correct.
6. d Statement I is true for all infinite populations and if 10% condition is met for finite
populations. Statement II is based on the idea of combining multiple Normal
distributions, Statement III is the central limit theorem.
7. a A restatement of the central limit theorem.
8. d The standard deviation of the sampling distribution depends on the sample size but not on
the population size (as long as the 10% condition is met).
9. c The only condition that is required for using the formula is that samples from a finite
population are less than 10% of the population size.
   
 25   x   25     25  20.5 
10. b P  x  25   P  z    P  z     P  z  15.4 
 x     
   
 n   40 

Part 2

© 2011 BFW Publishers The Practice of Statistics, 4/e- Chapter 7 345


11. (a) No. We don’t know the shape of the distribution, so we can’t calculate this probability.
0.4
(b)  x  12 and  x   0.0566 . (c) Since n = 50, which is greater than 30, we can use the
50
 11.9  12 
Normal probability distribution. P  x  11.9   P  z    P  z  1.77   0.0384
 0.0566 
(d) If the true mean amount of soda in the cans is 12 ounces, there is about a 4% chance of
getting a sample mean as low or lower than 11.9 ounces. This result is unlikely enough to make
us suspicious and lead us to conclude that the company is under-filling its cans of soda!

12. (a)  pˆ  p  0.45. (b)  pˆ 


 0.45 0.55  0.0222 . (c) np  500  0.45   225  10 and
500
n 1  p   500  0.55   275  10 . So the sampling distribution is approximately Normal.
 0.5  0.45 
(d) P  pˆ  0.5   P  z    P  z  2.25   0.0122. 13. Sample range is not an
 0.0222 
unbiased estimator of population range. The population range is 80 – 20 = 60. The range of a
sample will only be this large if the population’s minimum and maximum values in the
distribution are both in the sample. Otherwise, the sample range will be smaller. Thus the center
of the sampling distribution of sample ranges will be somewhere below 60. In this particular
case, the median of the distribution is 57.

346 The Practice of Statistics, 4/e- Chapter 7 © 2011 BFW Publishers