• In a random sample of the length (in mins) of songs on CDs produced by

hard rock bands, mean is 3.35 and s.d is 0.5.

• Find the approximate proportion of observations between 2.35 and 4.35

Ans: at least 1-(1/4)=3/4 of the observations

• Approximately what proportion of songs last more than 5 mins.

Ans: at most 1-0.91=0.09 (or 9%) of the observations are outside of this
interval, either less than 1.7 or greater than 5. Since we can not assume that
distribution is symmetric, we do not know what part of the 9% is less than 1.7
and what part is more than 5. To be conservative, the best we can say is at
most 9% of the observations are more than 5 mins long.

HINT: Use Chebyshev’s empirical rule

• First Horizon Pharmaceutical has just developed a new
medicine for treatment of routine aches and pains. The
company claims the distribution of pain-relief times (in hours)
is approximately symmetric with mean =8 and s.d=0.2. A
patient with a typical muscle ache is randomly selected and
the medicine is given. The patient reports pain relief for only 7
hours. Is there any evidence to refute the manufacturer’s
claim?
• Most college career counselors agree that starting salary is
associated with academic major. Even, if a person’s first job is
not directly related to their course of study, their salary may
still be related to their academic major. A recent survey of
academic major, and starting salary of graduates showed the
following information:

Mean S.D
English $32,300 $1,175
CS $ 42,500 $ 2,375

A CS major who responded to the survey received a starting

salary of $45,000, and an English major received an offer of
$35,000. Which salary is better in terms of statistics?
• Marketing research by Coffee Beanery in Detroit, Michigan,
indicates 70% of all customers put sugar in their coffee, 35%
add milk, and 25% use both. Suppose a Coffee Beanery
customer is selected at random.

• What is the probability the customer uses at least one of the

items
Ans:0.80
• What is the probability the customer uses neither
Ans: 0.20
• What is the probability the customer uses just sugar
Ans: 0.45
• What is the probability the customer uses just one of these
two items
Ans: 0.55
• Over the past 15 years, there has been a decline in traditional
benefits packages offered to workers (for example, lump sum
payments upon retirement) and an increase in the number of
defined contribution plans. According to the Department of
Labor, 43% of all workers in the private sector have a defined
contribution plan and 14% have both defined-contribution
plan and defined benefits package. Suppose a worker in the
private sector is selected at random. If the worker has a
defined-contribution plan, what is the probability the worker
also has a defined-benefits plan?

Ans: Use conditional probability

it is 0.325
• Suppose that 18% of employees of a given corporation engage in
physical exercise during the lunch hour. Moreover, assume that 57%
of all employees are male, and 12% of all employees are males who
engage in physical exercise during the lunch hour.
• If we choose an employee at random from this corporation then
what is the probability that this person is a female who engages in
physical exercise during the lunch hour?
• If we choose an employee at random from this corporation then
what is the probability that this person is a female who does not
engage in physical exercise during the lunch hour?
• What proportion of female employees engage in physical exercise
during the lunch hour?
• What proportion of employees who engage in physical exercise
during the lunch hour are male?

HINT:n see next slide!!!

Provided probabilities in the table ..

Exercise No Exercise All

Male 12% 45% 57%
Female 6% 37% 43%
All 18% 82% 100%

One issue with this question is how probability questions can be worded. One might ask:
“What proportion of employees are female?” The answer is 43%. I could also ask “if I
choose an employee at random, what is the probability that they are female?” The answer
is again 43%.

(a) The proportion of employees who are female and exercise is 6%.
(b) The proportion of employees who are female and do not exercise during the
lunch hour is 37%?
(c) This question restricts attention to females only. This is a conditional probability.
Out of the 43% of female employees, 6% exercise so the (conditional) probability is
6/43=14%. In terms of a formula

(d) This question restricts attention to 18% of employees who exercise. The
proportion of these who are male is 12% so the (conditional) probability of being male is
67%. In terms of a formula
• The probability a randomly selected family belongs to the AAA
automobile club is 0.25. If a family belongs to AAA, the
probability they have more than one car is 0.45. Suppose a
family is randomly selected. What is the probability they have
more than one car and belong to AAA?

Ans:
Use joint probability (not conditional!!)
It is 0.1125
• Better Bedding in East Hartford, Connecticut, claims 99.4% of
all mattress deliveries are on time. Suppose two mattress
deliveries are selected at random.

• What is the probability both mattresses will be delivered on

time
Ans: 0.994*0.994=0.988
• What is the probability both mattresses will be delivered late
Ans: 0.006*0.006=0.000036
• What is the probability exactly one mattress will be delivered
on time
Ans: 0.0119
• A picnic is arranged to be held on a particular day. The
weather forecast suggests that there is a 70% chance of rain
on that day. If it rains, probability of good picnic is 0.2, while if
it does not rain, probability of is 0.9

• What is the probability that picnic will be good

Ans: 0.41
• If the picnic was good, what is the probability that it did not
rain on that day
Ans: 0.65
• Suppose 20% of clerical staff in an office smoke cigarrettes.
Research shows that 60% of smokers and 15% of nonsmokers
suffer a breathing illness by age 65.
• (a) Do these percentages indicate that smoking and breathing
illness are independent?
• (b) what's the probability that a randomly selected 65 year old
employee who has breathing illness smokes.
• In a manufacturing plant, there are three machines producing
50%, 30% and 20% respectively of the total output. Out of the
items produced by 1st machine, 4% are defective.
Corresponding percentages for 2nd and 3rd machines are 5 and
3 respectively. An item is drawn at random from the
production line.
• Find the probability that this item is defective .
• Given that this item defective, what is the conditional
probability that it has been produced by machine 2?
• You are trying to assess the effectiveness of an
advertising campaign for a product which is bought by
20% of people. A survey shows that 40% of people
recognise your advertisement while 12% of people
both recognise your advertisement and buy your
product.

• Does seeing the advertisement increase the probability

of purchasing?

• For another advertisement, 30% of people recognise it

while 10% both recognise it and buy the product.

• Which of the two ads seems to be the most effective?

• A manufacturing process produces computer chips in large
numbers. Over the long-run the fraction of bad chips
produced by the process is around 20%. Thorough testing is
expensive but there is a cheap test. All good chips pass the
cheap test but so do 10% of bad chips.
• Given that a chip passes the test what is the probability that it
is a good chip?
• If a company sold all chips that passed the test what fraction
of those they sold would be bad?
Each chip can be bad or good. So the two possible unknown states of the world are B or
G. The data described in the first part is “passes the test”.
a. According to the information given Pr(D|G)=1, Pr(D|B)=0.1. The prior probabilities are
Pr(G)=0.8, Pr(B)=0.2. The Bayes calculation is set out below.
State Prior Data-L’hood Product Posterior
G 0.8 1 0.8 97.56%
B 0.2 0.1 0.02 2.44%
Total 100.0% 0.82 100.0%
b. The second question is really the same question, but expressed in real world terms.
For each chip tested and passed, we used Bayes to calculate Pr(B|D)=2.44%. So in
the long run this is the fraction of bad chips we would expect to sell.
• Hartnett car company has 12% market share. They are trying to
assess the value of their long-standing advertising campaign. A
standard way of doing this is to ask purchasers whether they recall
the advertising campaign at the point of purchase. It turns out that
75% can recall the main details of the campaign. They also have
general market research data which indicates that amongst those
people who do not own a Hartnett, only 21% recall the major
details of the campaign. On average, Hartnett makes a profit of
k$12,200 on each customer.
• What proportion of people overall recall the major details of the
campaign?
• Given that someone recalls the major details of the campaign, what
is the probability that they will buy a Hartnett?
• On average, Hartnett makes a profit of k$12,200 on 12% of the car
buying population so their average profit per member of this target
population is 12% of k$12,200 which is $1,460. Given that someone
recalls the major details of the campaign, what is the average profit
per member of the target population?
• Set up a Bayes table. The two states of the world are that someone is a buyer
or a non-buyer. The data is that the major details of the advertising campaign
are recalled.

• State Prior DL Product Posterior Value

Buyer 0.12 0.75 0.090 0.328 $12.2
Non-Buyer 0.88 0.21 0.185 0.672 $0.0
$1.464 0.275 1.000 $3.996

• The total of the product column gives the probability of the data – here the
probability that someone recalls the campaign is 27.5%.

• The posterior distribution gives the probability of the states given the data.
Conditional on recalling the campaign, the probability of being a buyer is
32.8% - much higher than the overrall 12%.

• Just calculate 32.8% of k$12,200 which is $3,996. The average value of a target
customer to the company increases by $2,532 if they can just recall the details
of the campaign.
 Bayes in Market Research
• You are considering launching a new product- male aftershave fragrance. It is very
difficult for new products in this market. On the basis of your experience, expert
judgment and (to some extent) historical records, you “guestimate” that there is
around 10% chance your product will be a success. It is much more likely to be a
flop. To try and reduce your risk, you are considering retaining a marketing
company to do some test marketing based on product testing and focus groups.

• The company claims the following accuracy rates: 99% of ultimately

successful products will test well and return and overall positive
recommendation. On the other hand, 89% of products that ultimately flop
will test poorly and return an overall negative recommendation.

• Q-1: How much does the company’s service help you to reduce the
uncertainty in the product’s outcome? Does it actually reduce uncertainty
at all?
• Q-2: How much would the marketing company’s service be worth to you?
For the sake of argument, suppose that a success makes you INR
1,000,000 and a flop loses you INR 100,00
Example: Disease detection

19
• A game is being played between two players A and B
according to following rule: Each of them rolls a die with faces
1,2,..,6. If the total score is a perfect square, then A receives
as many rupees as total score, otherwise B receives from A as
many rupees as total score.

• Find the probability distribution of the gain of A

• Find the mean gain for A
• The Hard Rock Café in Dallas carefully monitors customers
orders and has found that 70% of all customers ask for some
kind of coffee (C) while the remainder order a specialized tea
(T ). Suppose four customers are selected at random. Let the
random variable X be the number of customers who order
coffee

• Find the probability distribution of X

• Find the probability more than two customers order coffee
• Suppose at least two customers order coffee. What is the
probability all four customers order coffee?
• A red and a black die are thrown, and X denote the sum of the
two dice. What is the X’s expected value, variance and
standard deviation? What fraction of the probability mass lies
within one standard deviation of the expected value?
• Suppose the discrete random variable X, the age of a
randomly selected child at the UniKids Academy Day Care
Center has the probability distribution given in the following:

x: 1 2 3 4 5 6 7
f(x): 0.05 0.10 0.15 0.25 0.20 0.15 0.10

Find the expected value and variance of X

Find the probability the random variable X takes on a value

within one standard deviation of the mean
An insurance salesman meets with 5 prospective customers each week. From
historical data, the proportion of these customers who take out a policy is
18%.
• What is the probability that the salesman will sell 3 or more policies in a
given week?
• What is the probability that the salesman will sell 12 or more policies in a
given month (assuming a month is exactly 4 weeks)?
• What is the mean and standard deviation of the number of policies he
sells in a week, a month, a year? What do you notice about these three
distributions?
• Using the binomial distribution with n=8 and p=0.18, the probability of 3
or more is 0.161.

• In 4 weeks the number of trials (i.e. customers) is 32. Using the binomial
distribution with n=32 and p=0.18, the probability of 12 or more is 0.007.

• In 52 weeks the number of trials (i.e. customers) is 416. Using the

binomial distribution with n=416 and p=0.18, the probability of 156 or
more is zero to 8 significant figures. While it is not at all impossible to
make three sales in a week (see part (a)), it is virtually impossible for him
to average 3 sales per week (by selling 156 in 52 weeks).

• Over a week the mean is 1.4 and the standard deviation is 1.1. Over a
month, the mean is 5.8 and the standard deviation is 2.17. Over a year, the
mean is 74.9 and the standard deviation is 7.84. Obviously the mean
number of sales increases in proportion to the number of weeks. But the
standard deviation does not. The standard deviation of 7.84 over a year is
a much smaller proportion of the mean of 74.8 during a year, than it is for
a week. The yearly sales are much more predictable than the weekly sales.
• Can a Binomially distributed r.v. have expectation 10 and
variance 3?
• Three contractors A, B, and C are bidding for n contracts.
Suppose A has exactly half the chance as B; B has 4/5th as
likely chance as C to win a contract. Results for different
contracts are independent

• What is the probability that A will win at least one contract

• Find the expected difference in numbers of contracts to be
won by A and B
• Researchers conducting a study of buying habits interviewed
thousands of customers and claim that 60% of all
supermarket purchases are completely unplanned, or impulse
purchase. Suppose 100 customers are selected at random

• Find the mean, variance and std deviation of the number of

customers who make an impulse purchase

• Suppose 55 of the 100 customers make an impulse purchase.

Is there any evidence to suggest the study’s claim is false?
• You survey 500 customers on whether or not they recognize
your brand. On the basis of historical norms you expect 35%
or 175 to recognize your brand. But you have been spending
more money than usual on advertising. How many more than
175 would start to convince you that the extra money had
been worth the expense? Is it feasible?
• The number who recognize your brand is
binomial with n=500 and, if nothing has
changed, p=0.35. The mean is 175 and the
standard deviation is 10.7. You would start to
think things have improved if your results are
say two standard deviations above the mean –
say around 175+22=197. The chance of 197 or
more is 0.023 which is unlikely enough that
you would think the campaign has worked.
Even with 190 you would be pretty convinced
since the chance of doing this well by luck is
0.088.
XYZ mutual fund has outperformed the index in 37 of the past 52
weeks. On this basis they are claiming that they can
systematically outperform the market.
• What do you think of this claim?
• Assuming that XYZ are not outperforming the market, what is
the chance they would get to make this claim during a 10 year
period?
• XYZ mutual fund has outperformed the index in 37 of the past 52 weeks?
Think of each week as a trial. You either outperform or you do not. Each
week is independent because financial performances in each week are
independent (a theorem of finance). What is the chance XYZ outperform
the market in a given week? If they really are systematic outperformers
then the probability would be high. But if they are just guessing then the
chance of outperforming will not be high. Let us assume that there is a
50% chance of someone outperforming the market by sheer luck. We can
vary this slightly if we need to. With these assumptions, the number of
weeks where XYZ outperform the market is binomial with n=52 and p=0.5.

• The chance of 37 or more success is 1-binomdist(36,52,.5,true) which is

0.0182 which is pretty low. If we change the chance of outperforming the
market by luck to 555 then this changes to 0.085 which is not so low.

• In each year they get to make the claim or not. The years are now trials
and the chance of success is 0.0182. The chance of at least one successful
year in 5 equals 1 minus the chance of zero successful years. The answer is
0.088 from typing 1-binomdist(0,5,.0182,true). With 55% chance of
outperforming this is 0.359.
You are determined that the underlying proportion of customers
who give your products the lowest possible rating for meeting
expectations on a 5 point scale is to be no higher than 10%.
Suppose it is exactly 10%.

From a survey of 100 customers describe in words the likely

number who will give your product the lowest possible rating.

What is the chance that, by bad luck, 15 or more out of the

sample report the lowest level of satisfaction?
Emergency patients arrive at a large hospital at the rate of 0.33
per minute. On average, 22% of emergency patients are triaged
into the most serious category.

• What is the probability of 6 or more arrivals during the next

30 minutes?
• What is the probability of 6 or more arrivals during the next
30 minutes that are into the most serious category?
• If there are 25 patients that arrive in the next hour, what is
that chance that more than half of them will be into the most
serious category?
• A company has 2 machines A and B. On
average there are 0.8 breakdowns per week
on machine A and 1.2 breakdowns on
machine B. What is the probability of there
being a total of 2 breakdowns on these two
machines in a given week?
• The number of calls coming per minute into a hotels
reservation center is poisson r.v. with mean 3.

• 1) Find the probability that no calls in a given 1 minute period

• 2) Assume that the number of calls arriving in two different

minutes are independent. Find the probability that at least
two calls will arrive in a given two minute period.
• The number of people arriving at a fast food drive thru (or
through) in any given 2 minute interval obeys a Poisson
process with mean 1. Suppose that the waiters can only
process 3 orders in any given 4 minute interval (also, assume
the waiters can process an order instantaneously, but are
limited in how many they can process in given time intervals,
as indicated).

• What is the expected number of people that leave the drive

thru with their orders filled in any given 4 minute interval?
• An aptitude test, when applied to the general population,
gives scores that are normally distributed with mean 570 and
standard deviation 60.

• What proportion of people sitting the test will score less than
500?

• What is the probability that a person scores between 570 and

600?
• Checking the level of cholesterol in the blood is important
because high cholesterol may increase the risk of heart
disease. The distribution of blood cholesterol levels in a large
population of people of the same age and sex is roughly
normal. For 14-year old boys, the mean is 170 mg/dl and
s.d=30 mg/dl. Levels above 240 may require medical
attention.

• What percent of 14-year old boys have more than 240 mg/dl
of cholesterol?
• Scores on the SAT test in 2002 followed approximately
normal distribution N(504, 111) .

• How high must a student score in order to place in the top

10% of all students taking the SAT?
• In an exam, scores (out of 100 total marks) are distributed
normally. It is known that 10% of students get distinction, i.e.,
80 or more. While 20% of them fail, i.e., score below 40

• Find the mean and standard deviation of the distribution of

scores

• Find the percentage of students who obtained 90 or more

• Suppose a r.v X follows a normal distribution with mean zero
and variance unity. What is the probability distribution of
Y=X2
• An industrialist finds that the measured quality characteristics of 20% of
his product fall short of the specified minimum quality level and thus fail
to be accepted. The quality characteristics is normally distributed with a
coefficient of variation of 30%. The manufacturer wants to ensure that
95% of articles manufactured exceed the specified minimum quality level
and thus get accepted, which he can attain either by increasing the mean
level of quality or by reducing the variation in the quality. Find which way
will be profitable to him if the cost in increasing the mean level of 1% is
twice the cost of reducing the variation by 1%
• There are six hundred economics students in the
postgraduate classes of a university, and the probability for
any student to need a copy of a particular book from the
university library on any day is 0.05. How many copies of the
book should be kept in the university library so that the
probability may be greater than 0.9 that none of the students
needing a copy from the library is disappointed?
• Are attitudes towards shopping changing? Sample surveys
show that fewer people enjoy shopping than in the past. A
survey asked a nationwide random sample of 2500 adults if
they agreed or disagreed that “ I like buying new clothes, but
shopping is often frustrating and time-consuming”. Suppose
60% of all adult residents of India say “Agree” if asked the
same question.

• What is the probability that 1520 or more of the sample

agree?
• Two independent r.v. X and Y are both normally
distributed with means 1 and 2 and s.d. 3 and 4
respectively. If Z=X-Y, what is the p.d.f of Z?
Find Prob {Z+1<0}
• An industrialist finds that the measured quality
characteristic of 20% of his product fall short of
the specified minimum quality level and thus fail
to be accepted. The quality characteristic is
normally distributed with a coefficient of
variation of 30%. The manufacturer wants to
ensure that 95% of articles manufactured exceed
the specified minimum quality level and thus get
accepted, which can attain either by increasing
the mean level of quality or by reducing the
variation in the quality. Find which way will be
profitable to him if the cost in increasing the
mean level 1% is twice the cost of reducing the
variation by 1%
• Some people complain about the length of time it takes for a
pharmacist to fill a prescription. A careful audit at Religare has
shown that the time it takes to fill a randomly selected
prescription has a uniform distribution between 5 and 25
mins. Suppose a prescription is selected at random.

• Find the probability it takes at most 10 mins to fill the

prescription

• Find the probability it takes between 10 and 20 mins to fill the

prescription

• Find the mean time it takes to fill a prescription.

• Female bald eagles lay only two eggs per year. The eaglets
usually hatch a few days apart and normally only one eaglet
will survive. Suppose the weight of a randomly selected eaglet
at birth, X, has a uniform distribution between 100 and 125
grams. Find the weight w, such that 75% of all eaglets weigh
at most w.
• Two companies are competing for control of a limited amount
of a certain resource. Let X=the proportion of the resource
controlled by company 1 and suppose X has a uniform
distribution on [0,1]. Compute the expected value and
variance of the amount controlled by the company having
majority control.
• The waiting time in minutes at a bus stop follows a uniform
distribution over the range [0,5]. Find the probability that an
individual has to wait for more than 4 minutes on at least 3
days out of 5 randomly chosen days
• Novartis Pharma claims that drug A relives many symptoms
due to the common cold. After the prescribed dose is taken,
suppose the length of time (in hrs) until symptoms return is a
random variable, X, that has an exponential distribution with
parameter 0.1.

• What is the probability the length of time until symptoms

return is less than mean?

• What is the probability the length of time until symptoms

return is at least 12 hrs

• What is the probability the length of time until symptoms

return is between 8 and 16 hrs?
• In a factory, there are 3 machines producing respectively 50%,
20% and 30% of the total output. The life distribution of a
typical item produced by the machines are exponential with
mean 8 (in hrs), 10 and 12 for first, second and third machine
respectively. An item is drawn at random from the production
line; it is seen to survive beyond 15 hrs.

• what is the probability that it came from machine 1?

• A carrier XH heat pump, designed to heat a home in the
winter and cool the home in summer, lasts, on average , 16
years. The life time (in yrs) of this system can be modeled by
an exponential distribution with parameter 0.0625. suppose a
heat pump is selected at random

• What is the probability the heat pump will last for at least 5
yrs

• Suppose the heat pump lasts for five years. What is the
probability it will last for at least another 5 years.
• Distribution of life in hours of a certain kind of electric bubs is
known to be exponential. It is also known that a typical bulb
of this kind survives for at least 8 hrs with probability 0.64.
Find the probability that out of 5 randomly chosen bulbs of
this kind, at least 4 will survive for 12 hours or more.
• In upstate New York, milk tanker trucks follow a daily routine,
stopping at the same dairy farms every day. Farm output
varies due to weather, time of year, number of cows, and
other factors. From years of recorded data, the mean amount
of milk delivered for processing by a truck is 7750 liters, with a
standard deviation of 150 liters. Suppose 36 trucks are
randomly selected.

• Find the probability the sample mean amount of milk

delivered for the 36 trucks I more than 7800 liters

• Find the value m such that the probability the sample mean is
less than m is 0.1
• In 2003, a major vendor of antispam software claimed that
the proportion of email consisting of unsolicited spam was
0.40. Suppose 120 email messages are selected at random

• Find the distribution of the sample proportion

• What is the probability the sample proportion is greater than
0.5

• Find the probability the sample proportion will be between

0.32 to 0.37
• A quality control plan for an assembly line involves sampling n
finished items per day and counting Y, the number of
defectives. If P is the probability that a finished item is
defective on a given day, then Y has Y~Bin(n,P) on that day,
assuming that the number of items produced per day is very
large. Suppose P varies from day to day and a has a uniform
distribution on (0,θ). Find the expected value and variance of
Y.
• Five friends have purchased tickets to a certain concert. If the
tickets are for seats 1-5 in a particular row and the tickets are
randomly distributed among the five, what is the expected
number of seats separating any particular two of the five?
• The total weight of a filled tire can dramatically affect the
performance and safety of an automobile. Some
transportation officials argue that mechanics should check the
tire weights of every vehicle as part of the annual inspection.
Suppose the weight of a filled tire is normally distributed with
standard deviation 1.25 pounds. In a random sample of 15
filled tires, the sample mean weight was 18.75 pounds. Find a
95% confidence interval for the true mean weight of the tires.
• Lighthouses are constructed to guide ships travelling in rocky
waters and to allow sailing at night. There are many ways to
report the size of a lighthouse. However, the height is usually
measured from the base of the tower to the top of the
ventilator ball. A random sample of 18 lighthouses in France
and England was obtained, and the height of each was
recorded. The sample mean was 33.75 meters. Assume the
distribution of lighthouse heights is normal and population
s.d. is 5.4 meters.
• Find 95% CI for the mean height
• How large a sample is necessary to ensure that the width of
the resulting 95% CI is 2 meters?
• Distillation is a process for separating and collecting
substances according to their reaction to heat. When heat is
applied to a mixture, the substance that evaporates and is
collected as it cools is distillate. The unevaporated portion of
the mixture is the residue. Oil obtained from orange blossoms
through distillation is used in perfume. Suppose the oil yield
is normally distributed. In a random sample of eleven
distillations, the sample mean oil yield was 980.2 grams with
standard deviation 27.6 grams. Find 95% confidence interval
for the true mean oil yield per batch.
• High blood pressure, or hypertension, occurs when the force
of blood against your artery walls is too strong. This added
pressure can cause a stroke. In a recent survey, 1100 adult are
randomly selected and examined for high blood pressure. The
number of patients classified with hypertension is 319.

• Find 95% CI for the true proportion of adult with hypertension

• The department of Health and Human Services has a target of

16% for the hypertension prevalence by the year 2010. Is
there any evidence to suggest the prevalence of hypertension
is different from the target value?
• Each year a regional Bell telephone company inspects
telephone poles and replaces those that are defective. In
order to effectively allocate resources, the company plans to
estimate the proportion of defective poles. A 95% confidence
interval for p with bound on the error of estimation 0.02 is
needed. How large a sample size is necessary in each of the
following cases

• Prior experience suggest p=0.1

• There is no prior information regarding the proportion of

defective poles.
• Earthenware dishes are made from clay and are fired, or
exposed to heat, in a large kiln. Large fluctuations in the kiln
temperature can cause cracks, bumps or other flaws. With kiln
set at 800 degree, a random sample of 19 temperature
measurements was obtained. The sample variance was 17.55

• Find the 95% CI for the true population variance

• Millions of Indians suffer from back pain, and some
experience disc problems so sever that simple tasks, such as
driving, sitting in a chair, or even sleeping, are painful. The
traditional remedy for a damaged disc is surgery; spinal
fusion. Historical records indicate that 65% of all patients who
endure this costly, complicated surgery, actually experience
reduced pain and greater mobility. A news treatment (IDET,
intradiscal eletrothermal annuloplasty) has been developed
and researchers claim this procedure is more effective,
cheaper, and less painful. An experiment is conducted to
determine whether IDET is more effective than spinal fusion.
What null and alternative hypothesis should be used?
• Although the divorce rate has stabilized over the past few
years, the institutions of marriage has changed dramatically. A
divorce procedure can be very messy, time consuming,
contentious, and costly. Historical records indicate that the
mean cost of an uncontested divorce processed through
attorneys is $1200. A new on-line service offers an
uncontested divorce service with a cost structure that may
save couples money. A observational study is conducted to
determine whether the mean cost of an on-line divorce is less
than that of a lawyer-processed divorce. State the appropriate
null and alternative hypotheses.
• The thickness (measured in inches) of the recycled printer is
important, because sheets that are too thick will clog in
machine, and paper that is too thin will rip and bleed toner.
The variance in thickness for 20-lb printer paper at
manufacturing plant is known to be 0.007. A new process
developed using more recycled fiber, and an experiment is
conducted to detect any difference in variance in paper
thickness. State the appropriate null and alternative
hypothesis.
• Hospital emergency rooms across the country are
experiencing shortages of doctors and nurses, and have too
few beds. These constraints make it difficult to treat patients
in a timely manner. Breach candy hospital, bombay, which
treats approximately 50,000 patients in the emergency room
each year, decided to address this issue by moving into the
waiting room to treat patients. Prior to this experiment, the
mean time to treat very ill patients in the emergency room
was 20 minutes (with s.d=5 mins). During the waiting room
experiment, a random sample of 36 very ill patients was
selected and the time to treatment for each was recorded.
The sample mean was 16.1 mins. Conduct a hypothesis test to
determine whether there is any evidence to suggest the
waiting room experiment has reduced the mean time to
treatment for very ill patients.
• U.S military personnel must relocate to different bases often,
due to orders or a request for transfer. These frequent moves
create family hardships and deter some from entering, or
remaining in, the military. The mean time for military
personnel to stay in any one location is known to be 1.95
years (with s.d=0.6 yrs). Recent changes in salary and benefits
have been designed to provide military personnel greater
satisfaction and to lengthen the time they are sanctioned at
any location. Several years following implementation of these
changes, a random sample of 40 relocated military personnel
was obtained. The mean time they stayed at their previous
location was 2.36 yrs. Is there any evidence to suggest that
the new pay scale and benefit package has increased the
mean time before relocation?
• A manufacturer of cereal wants to test the hypothesis that a
filling machine is set to load boxes with a mean load of 12
ounces per box. If the mean load either exceeds 12 ounces
per box or is less than this amount, the manufacturer wants
to detect the situation. Suppose that 100 boxes of cereal are
filled and weighted and that sample mean and sample
standard deviation were calculated to be 11.85 ounces and
0.5 ounce.

• Test to see whether the data indicate that the population

mean filling weight differs from 12 ounces. Use 5% as
significance level
• Test to see whether the data indicate that the machine is
under-filling the boxes or not. Use 5% significance level
• Consider a coin which shows a head with probability θ.
Consider null hypothesis θ=1/2 against the alternative θ≠1/2.
Coin is tossed thrice. For a test rule which rejects null if and
only if all the 3 tosses yield identical results, find the
probability of type I error and also the power of the test.
• Consider a population with proportion p. Let
X be the number of success in a random sample
of size 100 with model X~Binomial (100,p).
Consider the hypotheses NULL: p=0.3 vs.
ALT: p<0.3
The researcher reject the null hypothesis if X≤22

Find the type I error and the power of the test if

p=0.2
• A certain daily delivery route for Hostess breads and snack cakes
includes eight grocery stores and four convenience stores. The
historical mean time to complete these deliveries (to the 12 stores)
and return to the distribution center is 6.5 hrs. A new driver has
been assigned to this route, and a random sample of his route
completion time in hrs was obtained. The data given are;
6.61, 6.25, 6.40, 6.57, 6.35, 5.95, 6.53, 6.29

Assume the underlying population is normal.

Is there any evidence to suggest the new driver has been able to
shorten the route completion time. Use α=0.01

Find the bounds on the p-values associated with this test.

• In 2003, records indicated that 35.4 million Americans, or
approximately 19% of all credit cardholders carried an
American Express (AE) card. An advertising campaign was
developed and implemented in order to entice more credit
cardholders to carry an AE card. Following the promotion, a
random sample of 2500 card holders was obtained and 496
were identified as AE cardholders. Is there any evidence the
proportion of cardholders who carry an AE card has
increased? Use α=0.01
• The Us department of agriculture is concerned about graying
of American farms. It has been reported that 25% of all US
farmers are over the age of 65 and that fewer children of
farmers are choosing their parents profession. A random
sample of 350 American farers was obtained and 77 were
over 65.

• Is there any evidence to suggest the proportion of farmers

over 65 is different from 0.25? Use α=0.05

• Find the p value associated with this hypothesis test

• The mean weight of US Fancy Grade watermelons is 22
pounds. Watermelons that are too light are dry and are less
flavorful. When a large grocery chain receives a shipment of
watermelons for sale, the store supervisor randomly selects
36 watermelons and weighs each. A hypothesis test is
conducted to determine whether there is any evidence that
the mean weight of the watermelons is less than 22 pounds.
Let =0.025, and assume population s.d=3 and find the
probability of a type II error if the true mean weight of the
watermelon is 20.5 pounds.
• One measure of permafrost is the volumetric liquid water
content (VWC), a unitless quantity. This measurement is used
to study climate changes in the Arctic ice cap, greenhouse
gases, and vegetation. Along the northern part of the Trans-
Alaskan Pipeline, variation in permafrost was an important
consideration in planning and construction. Suppose a
random sample of 15 locations was obtained, and the VWC
was measured at each location during the winter months. The
sample mean was 0.225 and the sample variance was 0.0025.
Is there any evidence to suggest the population variance is
greater that 0.002? Use α=0.05
• Reports completed by independent truck drivers indicate that
the mean amount of disel fuel purchased per week is 346
gallons with a standard deviation of 23.6 gallons. Truck-stop
managers use this information to plan work schedules and to
order gasoline supplies. A random sample of independent
truck drivers was obtained, and their weekly gas purchases (in
gallons) are given below:

339 348 324 343 365 355 347 343 343 358
328 326 343 358 340 331 326 355 342 349

Is there any evidence to suggest the true population variance in

disel fuel purchased per week is different from 556.96 gallons,
Use α=0.01
Find bonds on the p-value associated with this hypothesis test
• Most casts formed to help heal a broken limb are made from
plaster of Paris and cotton wool. Another common type of
cast is fabricated from polyurethane and fibreglass. This
combination of materials is quick-setting and light. The most
important aspect of any cast, however, is he crush resistance
or crush strength. Independent random samples of each type
of cast were obtained, and the crush resistance was measured
for each cast one week post-fabrication. The summary
statistics and known variances are given below:

• Sample-1: n=16, mean=210.5, pop. Var=416.16

• Sample-2: n=25, mean=225.1, pop. Var=96.04
• Is there any evidence to suggest that the mean crush strength
of plaster of Paris casts is less than the mean crush strength
of polyurethane casts? Use α=0.05
• Pizza stones designed for home use help cooks produce baked
goods with brick-oven qualities. However, pizza stones can be
very heavy and can also take up a lot of space in a traditional
residential oven. Independent random samples of two similar
types of round pizza stones were obtained, and the weight (In
pounds) of each was recorded. The summary statistics and
known variances are given below:

• Sample-1: n=35, mean =6.21, pop. Var=2.1

• Sample-2: n=31, mean=7.08, pop. Var=3.5

• Find the 95% C.I. for the difference in population mean pizza-
stone weights.
• Although smoking is banned in many public places in Europe,
a high proportion of European adults (ages 15 and older) still
smoke daily. In a random sample of 250 adults in Bulgaria, 92
were daily smokers, and in a random sample of 300 adults in
Greece, 114 were daily smokers. Is there any evidence to
suggest the true proportion of daily smokers is different in
Bulgaria and in Greece? Use level of significance=0.05