Business Statistics - A Decision-Making Approach

Chapter 8 Student Lecture Notes 8-1
Business Statistics:
A Decision-Making Approach
7th Edition
Chapter 8
Estimating Single
Population Parameters
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-1
Chapter Goals
After completing this chapter, you should be
able to:
 Distinguish between a point estimate and a confidence
interval estimate
 Construct and interpret a confidence interval estimate for a
single population mean using both the z and t distributions
 Determine the required sample size to estimate a single
population mean within a specified margin of error
 Form and interpret a confidence interval estimate for a
single population proportion
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Confidence Intervals
Content of this chapter

 Confidence Intervals for the Population
Mean, μ
 when Population Standard Deviation σ is Known
 when Population Standard Deviation σ is Unknown
 Determining the Required Sample Size
 Confidence Intervals for the Population
Proportion, p
Point and Interval Estimates
 A point estimate is a single number, used to

estimate an unknown population parameter
 a confidence interval provides additional
information about variability
Lower Upper
Confidence Confidence
Point Estimate Limit
Limit
Width of
confidence interval

Point Estimates
We can estimate a with a Sample

Population Parameter … Statistic
(a Point Estimate)
Mean μ x
Proportion π p
 How much uncertainty is associated with a

point estimate of a population parameter?
 An interval estimate provides more

information about a population characteristic
than does a point estimate
 Such interval estimates are called confidence

intervals

Confidence Interval Estimate
 An interval gives a range of values:

 Takes into consideration variation in sample
statistics from sample to sample
 Based on observation from 1 sample
 Gives information about closeness to
unknown population parameters
 Stated in terms of level of confidence
 Never 100% sure
Estimation Process
Random Sample I am 95%

confident that
μ is between
Population Mean 40 & 60.
(mean, μ, is x = 50
unknown)
Sample

General Formula
 The general formula for all

confidence intervals is:
Point Estimate  (Critical Value)(Standard Error)
Confidence Level
 Confidence Level
 Confidence in which the interval
will contain the unknown
population parameter
 A percentage (less than 100%)

Confidence Level, (1-)

(continued)
 Suppose confidence level = 95%
 Also written (1 - ) = .95
 A relative frequency interpretation:
 In the long run, 95% of all the confidence
intervals that can be constructed will contain the
unknown true parameter
 A specific interval either will contain or will
not contain the true parameter
 No probability involved in a specific interval
Confidence
Intervals
Population Population
Mean Proportion
σ Known σ Unknown

Confidence Interval for μ

(σ Known)
 Assumptions
 Population standard deviation σ is known
 Population is normally distributed
 If population is not normal, use large sample
 Confidence interval estimate
σ
xz
n
Finding the Critical Value

 Consider a 95% z  1.96
confidence interval:
1   .95
α α
 .025  .025
2 2
z units: -z = -1.96 0 z = 1.96

Lower Upper
x units: Confidence Point Estimate Confidence
Limit Limit
σ x σ
xz xz
n n

Common Levels of Confidence
 Commonly used confidence levels are

90%, 95%, and 99%
Confidence
Confidence Critical
Coefficient,
Level value, z
1 
80% .80 1.28
90% .90 1.645
95% .95 1.96
98% .98 2.33
99% .99 2.58
99.8% .998 3.08
99.9% .999 3.27
Interval and Level of Confidence

Sampling Distribution of the Mean
/2 1  /2
x
Intervals μx  μ
extend from x1
σ x2 100(1-)%
xz of intervals
n
to constructed
σ contain μ;
xz
n 100% do not.

Margin of Error
 Margin of Error (e): the amount added and

subtracted to the point estimate to form the
confidence interval
Example: Margin of error for estimating μ, σ known:
σ σ
xz ez
n n
Factors Affecting Margin of Error
σ
ez
n
 Data variation, σ : e as σ
 Sample size, n : e as n
 Level of confidence, 1 -  : e if 1 - 

Example
 A sample of 11 circuits from a large normal

population has a mean resistance of 2.20
ohms. We know from past testing that the
population standard deviation is 0.35 ohms.
 Determine a 95% confidence interval for the

true mean resistance of the population.
Example
(continued)
 A sample of 11 circuits from a large normal
population has a mean resistance of 2.20
ohms. We know from past testing that the
population standard deviation is 0.35 ohms.
Solution: σ
 x z
n
 2.20  1.96 (0.35/ 11)
 2.20  .2068
1.9932 ............... 2.4068

Interpretation
 We are 95% confident that the true mean

resistance is between 1.9932 and 2.4068 ohms
 Although the true mean may or may not be in this
interval, 95% of intervals formed in this manner
will contain the true mean
 An incorrect interpretation is that there is 95% probability that this

interval contains the true population mean.
(This interval either does or does not contain the true mean, there is
no probability for a single interval)
Confidence
Intervals
Mean Proportion
σ Known σ Unknown


(σ Unknown)
 If the population standard deviation σ is

unknown, we can substitute the sample
standard deviation, s
 This introduces extra uncertainty, since s
is variable from sample to sample
 So we use the t distribution instead of the
normal distribution

(σ Unknown)
(continued)
 Assumptions
 Population standard deviation is unknown
 Population is normally distributed
 If population is not normal, use large sample
 Use Student‟s t Distribution
 Confidence Interval Estimate
s
xt
n

Student‟s t Distribution
 The t is a family of distributions

 The t value depends on degrees of
freedom (d.f.)
 Number of observations that are free to vary after
sample mean has been calculated
d.f. = n - 1
Degrees of Freedom (df)

Idea: Number of observations that are free to vary
after sample mean has been calculated
Example: Suppose the mean of 3 numbers is 8.0
Let x1 = 7
If the mean of these three
Let x2 = 8
values is 8.0,
What is x3? then x3 must be 9
(i.e., x3 is not free to vary)
Here, n = 3, so degrees of freedom = n -1 = 3 – 1 = 2
(2 values can be any numbers, but the third is not free to vary
for a given mean)

Student‟s t Distribution
Note: t z as n increases
Standard
Normal
(t with df = )
t (df = 13)
t-distributions are bell-
shaped and symmetric, but
have „fatter‟ tails than the t (df = 5)
normal
0 t
Student‟s t Table
Upper Tail Area

Let: n = 3
df .25 .10 .05 df = n - 1 = 2
 = .10
1 1.000 3.078 6.314 /2 =.05
2 0.817 1.886 2.920

3 0.765 1.638 2.353 /2 = .05
The body of the table

contains t values, not 0 2.920 t
probabilities

t distribution values
With comparison to the z value
Confidence t t t z
Level (10 d.f.) (20 d.f.) (30 d.f.) ____
.80 1.372 1.325 1.310 1.28

.90 1.812 1.725 1.697 1.64
.95 2.228 2.086 2.042 1.96
.99 3.169 2.845 2.750 2.58
Note: t z as n increases
Example
A random sample of n = 25 has x = 50 and
s = 8. Form a 95% confidence interval for μ
 d.f. = n – 1 = 24, so t /2 , n1  t.025,24  2.0639
The confidence interval is

s 8
xt  50  (2.0639)
n 25
46.698 …………….. 53.302

Approximation for Large Samples

 Since t approaches z as the sample size increases,
an approximation is sometimes used when n is very
large
 The text t-table provides t values up to 500 degrees
of freedom
 Computer software will provide the correct t-value for
any degrees of freedom
Correct formula, Approximation

σ unknown for very large n
s s
xt xz
n n
Determining Sample Size
 The required sample size can be found to

reach a desired margin of error (e) and
level of confidence (1 - )
 Required sample size, σ known:

2
z σ z 2σ 2
n     2
 e  e

Required Sample Size Example
If  = 45, what sample size is needed to be

90% confident of being correct within ± 5?
z2σ 2 1.645 2 (45) 2

n 2   219.19
e 52
So the required sample size is n = 220
(Always round up)

If σ is unknown
 If unknown, σ can be estimated when

using the required sample size formula
 Use a value for σ that is expected to be at least
as large as the true σ
 Select a pilot sample and estimate σ with the
sample standard deviation, s
 Use the range R to estimate the standard deviation
using σ = R/6 (or R/4 for a more conservative
estimate, producing a larger sample size)

Confidence
Intervals
Mean Proportion
σ Known σ Unknown
Confidence Intervals for the

Population Proportion, π
 An interval estimate for the population

proportion ( π ) can be calculated by
adding an allowance for uncertainty to
the sample proportion ( p )

Confidence Intervals for the

Population Proportion, π
(continued)
 Recall that the distribution of the sample

proportion is approximately normal if the
sample size is large, with standard deviation
π(1 π)
σπ 
n
 We will estimate this with sample data:
p(1 p)
sp 
n
Confidence interval endpoints
 Upper and lower confidence limits for the

population proportion are calculated with the
formula
p(1 p)
pz
n
 where
 z is the standard normal value for the level of confidence desired
 p is the sample proportion
 n is the sample size

Example
 A random sample of 100 people

shows that 25 are left-handed.
 Form a 95% confidence interval for
the true proportion of left-handers
Example
(continued)
 A random sample of 100 people shows
that 25 are left-handed. Form a 95%
confidence interval for the true proportion
of left-handers.
1. p  25/100  .25
2. Sp  p(1 p)/n  .25(.75)/1 00  .0433
3. .25  1.96 (.0433)

0.1651 . . . . . 0.3349

Interpretation
 We are 95% confident that the true

percentage of left-handers in the population
is between
16.51% and 33.49%.
 Although this range may or may not contain

the true proportion, 95% of intervals formed
from samples of size 100 in this manner will
contain the true proportion.
Changing the sample size
 Increases in the sample size reduce

the width of the confidence interval.
Example:
 If the sample size in the above example is
doubled to 200, and if 50 are left-handed in the
sample, then the interval is still centered at .25,
but the width shrinks to
.19 …… .31

Finding the Required Sample Size

for proportion problems
Define the π(1 π)

margin of error: ez
n
z 2 π (1 π)
Solve for n: n
e2
π can be estimated with a pilot sample, if
necessary (or conservatively use π = .50)
What sample size...?
 How large a sample would be necessary

to estimate the true proportion defective in
a large population within 3%, with 95%
confidence?
(Assume a pilot sample yields p = .12)

What sample size...?

(continued)
Solution:
For 95% confidence, use Z = 1.96
E = .03
p = .12, so use this to estimate π
z2 π (1 π) (1.96) 2 (.12)(1 .12)

n   450.74
e2 (.03)2
So use n = 451
Chapter Summary
 Illustrated estimation process
 Discussed point estimates
 Introduced interval estimates
 Discussed confidence interval estimation for
the mean (σ known)
 Addressed determining sample size
the mean (σ unknown)
the proportion

7th Edition
Chapter 9
Introduction to
Hypothesis Testing
Chapter Goals

able to:
 Formulate null and alternative hypotheses for
applications involving a single population mean or
proportion
 Formulate a decision rule for testing a hypothesis
 Know how to use the test statistic, critical value, and
p-value approaches to test the null hypothesis
 Know what Type I and Type II errors are
 Compute the probability of a Type II error

What is a Hypothesis?
 A hypothesis is a claim
(assumption) about a
population parameter:
 population mean
Example: The mean monthly cell phone bill
of this city is µ = $42
 population proportion
Example: The proportion of adults in this
city with cell phones is π = .68
The Null Hypothesis, H0
 States the assumption (numerical) to be

tested
Example: The average number of TV sets in
U.S. Homes is at least three ( H0 : μ  3 )
 Is always about a population parameter,

not about a sample statistic
H0 : μ  3 H0 : x  3

The Null Hypothesis, H0

(continued)
 Begin with the assumption that the null

hypothesis is true
 Similar to the notion of innocent until
proven guilty
 Refers to the status quo
 Always contains “=” , “≤” or “” sign
 May or may not be rejected
The Alternative Hypothesis, HA
 Is the opposite of the null hypothesis

 e.g.: The average number of TV sets in U.S.
homes is less than 3 ( HA: µ < 3 )
 Challenges the status quo
 Never contains the “=” , “≤” or “” sign
 May or may not be accepted
 Is generally the hypothesis that is believed
(or needs to be supported) by the
researcher – a research hypothesis

Formulating Hypotheses
 Example 1: Ford motor company has

worked to reduce road noise inside the cab
of the redesigned F150 pickup truck. It
would like to report in its advertising that
the truck is quieter. The average of the
prior design was 68 decibels at 60 mph.
 What is the appropriate hypothesis test?
 Example 1: Ford motor company has worked to reduce road noise inside
the cab of the redesigned F150 pickup truck. It would like to report in its
advertising that the truck is quieter. The average of the prior design was
68 decibels at 60 mph.
 What is the appropriate test?
H0: µ ≥ 68 (the truck is not quieter) status quo

HA: µ < 68 (the truck is quieter) wants to support
 If the null hypothesis is rejected, Ford has sufficient

evidence to support that the truck is now quieter.

 Example 2: The average annual income of

buyers of Ford F150 pickup trucks is
claimed to be $65,000 per year. An
industry analyst would like to test this
claim.
 What is the appropriate hypothesis test?
 Example 1: The average annual income of buyers of Ford F150
pickup trucks is claimed to be $65,000 per year. An industry
analyst would like to test this claim.
 What is the appropriate test?
H0: µ = 65,000 (income is as claimed) status quo

HA: µ ≠ 65,000 (income is different than claimed)
 The analyst will believe the claim unless

sufficient evidence is found to discredit it.

Hypothesis Testing Process
Claim: the population mean age is 50.
Null Hypothesis: H0: µ = 50

Population
Now select a random sample:
Sample
Suppose the sample Is x = 20 If not likely,
mean age is 20: likely if REJECT
x = 20 µ = 50? Null Hypothesis
Reason for Rejecting H0

Sampling Distribution of x
x
20 μ = 50
If H0 is true
... then we
If it is unlikely that
reject the null
we would get a
... if in fact this were hypothesis that
sample mean of
the population mean… μ = 50.
this value ...

Errors in Making Decisions
 Type I Error
 Reject a true null hypothesis
 Considered a serious type of error
The probability of Type I Error is 

 Called level of significance of the test
 Set by researcher in advance
Errors in Making Decisions

(continued)
 Type II Error
 Fail to reject a false null hypothesis
The probability of Type II Error is β
 β is a calculated value, the formula is

discussed later in the chapter

Outcomes and Probabilities
Possible Hypothesis Test Outcomes
State of Nature
Decision H0 True H0 False
Do Not
No error Type II Error
Reject
Key: (1 -  ) (β)
Outcome H0
(Probability) Reject Type I Error No Error
H0 () (1-β)
Type I & II Error Relationship
 Type I and Type II errors cannot happen at

the same time
 Type I error can only occur if H0 is true
 Type II error can only occur if H0 is false
If Type I error probability (  ) , then

Type II error probability ( β )

Factors Affecting Type II Error

 All else equal,
 β when the difference between
hypothesized parameter and its true value
 β when 
 β when σ
The formula used to
compute the value of β
 β when n is discussed later in the
chapter
Level of Significance, 
 Defines unlikely values of sample statistic if

null hypothesis is true
 Defines rejection region of the sampling
distribution
 Is designated by  , (level of significance)
 Typical values are .01, .05, or .10
 Is selected by the researcher at the beginning
 Provides the critical value(s) of the test

Hypothesis Tests for the Mean
Hypothesis
Tests for 
σ Known σ Unknown
 Assume first that the population

standard deviation σ is known
Process of Hypothesis Testing
 1. Specify population parameter of interest

 2. Formulate the null and alternative hypotheses
 3. Specify the desired significance level, α
 4. Define the rejection region
 5. Take a random sample and determine
whether or not the sample result is in the
rejection region
 6. Reach a decision and draw a conclusion

Level of Significance
and the Rejection Region
Level of significance = 
Lower tail test Upper tail test Two tailed test

Example: Example: Example:
H 0: μ ≥ 3 H 0: μ ≤ 3 H 0: μ = 3
HA: μ < 3 HA: μ > 3 HA: μ ≠ 3
  /2 /2
-zα 0 0 zα -zα/2 0 zα/2
Do not Do not Do not

Reject H0 Reject H0 Reject H0 Reject H0
reject H0 reject H0 reject H0
Critical Value for Lower Tail Test

H0: μ ≥ 3
 The cutoff value, -zα or xα , HA: μ < 3
is called a critical value
Reject H0 Do not reject H0

-zα 0
xα µ=3
σ
x   μ  z
n

Critical Value for Upper Tail Test
 The cutoff value, zα or xα , H0: μ ≤ 3

HA: μ > 3
is called a critical value
Do not reject H0 Reject H0

0 zα
µ=3 xα
σ
x   μ  z
n
Critical Values for

Two Tailed Tests
 There are two cutoff H0: μ = 3
values (critical values): HA: μ  3
zα/2
or /2 /2
xα/2
Lower
Reject H0 Do not reject H0 Reject H0
xα/2 -zα/2 0 zα/2
Upper
xα/2 µ=3 xα/2
Lower Upper
σ
x /2  μ  z /2
n

The Rejection Region

Lower tail test Upper tail test Two tailed test
Example: Example: Example:
H 0: μ ≥ 3 H 0: μ ≤ 3 H 0: μ = 3
HA: μ < 3 HA: μ > 3 HA: μ ≠ 3
  /2 /2
-zα 0 0 zα -zα/2 0 zα/2

xα xα x α/2(L) x α/2(U)
Do not Do not Do not
Reject H0 Reject H0 Reject H0 Reject H0
reject H0 reject H0 reject H0
Reject H0 if z < -zα Reject H0 if z > zα Reject H0 if z < -zα/2 or z > zα/2
i.e., if x < xα i.e., if x > xα i.e., if x < xα/2(L) or x > xα/2(U)

Two Equivalent Approaches

to Hypothesis Testing
 z-units:
 For given , find the critical z value(s):
 -zα , zα ,or zα/2
x μ
 Convert the sample mean x to a z test statistic: z 
σ
 Reject H0 if z is in the rejection region, n
otherwise do not reject H0
 x units:
 Given , calculate the critical value(s)
 xα , or xα/2(L) and xα/2(U)
 The sample mean is the test statistic. Reject H0 if x is in the

rejection region, otherwise do not reject H0

Hypothesis Testing Example

Test the claim that the true mean # of TV
sets in US homes is at least 3.
(Assume σ = 0.8)
 1. Specify the population value of interest

 The mean number of TVs in US homes
 2. Formulate the appropriate null and alternative

hypotheses
 H0: μ  3 HA: μ < 3 (This is a lower tail test)
 3. Specify the desired level of significance
 Suppose that  = .05 is chosen for this test

(continued)
 4. Determine the rejection region
 = .05
-zα= -1.645 0
This is a one-tailed test with  = .05.

Since σ is known, the cutoff value is a z value:
Reject H0 if z < z = -1.645 ; otherwise do not reject H0


 5. Obtain sample evidence and compute the
test statistic
Suppose a sample is taken with the following
results: n = 100, x = 2.84 ( = 0.8 is assumed known)
 Then the test statistic is:
x μ 2.84  3  .16
z     2.0
σ 0.8 .08
n 100

(continued)
 6. Reach a decision and interpret the result
 = .05
z
-1.645 0
-2.0
Since z = -2.0 < -1.645, we reject the null

hypothesis that the mean number of TVs in US
homes is at least 3. There is sufficient evidence
that the mean is less than 3.


(continued)
 An alternate way of constructing rejection region:
Now
expressed
 = .05 in x, not z
units
x
2.8684 3
2.84 σ 0.8
x α  μ  zα  3  1.645  2.8684
Since x = 2.84 < 2.8684, n 100
we reject the null
hypothesis
p-Value Approach to Testing
 Convert Sample Statistic ( x ) to Test Statistic

(a z value, if σ is known)
 Determine the p-value from a table or
computer
 Compare the p-value with 
 If p-value <  , reject H0
 If p-value   , do not reject H0

p-Value Approach to Testing

(continued)
 p-value: Probability of obtaining a test

statistic more extreme ( ≤ or  ) than the
observed sample value given H0 is true
 Also called observed level of significance
 Smallest value of  for which H0 can be

rejected
p-value example
 Example: How likely is it to see a sample mean
of 2.84 (or something further below the mean) if
the true mean is  = 3.0?
 = .05
P( x  2.84 | μ  3.0)
p-value =.0228
 
 2.84  3.0 
 P z  
 0.8 
 100 
2.8684 3 x
 P(z  2.0)  .0228
2.84
-1.645 0 z
-2.0

p-value example
(continued)
 Compare the p-value with 

 If p-value <  , reject H0
 If p-value   , do not reject H0

 = .05
Here: p-value = .0228 p-value =.0228
 = .05
Since .0228 < .05, we reject
the null hypothesis
2.8684 3
2.84
Example: Upper Tail z Test

for Mean ( Known)
A phone industry manager thinks that
customer monthly cell phone bill have
increased, and now average over $52 per
month. The company wishes to test this
claim. (Assume  = 10 is known)
Form hypothesis test:

H0: μ ≤ 52 the average is not over $52 per month
HA: μ > 52 the average is greater than $52 per month
(i.e., sufficient evidence exists to support the
manager‟s claim)

Example: Find Rejection Region

(continued)
 Suppose that  = .10 is chosen for this test
Find the rejection region: Reject H0
 = .10

0 zα=1.28
Reject H0 if z > 1.28

Review:
Finding Critical Value - One Tail
Standard Normal
What is z given  = 0.10? Distribution Table (Portion)
.90 .10
Z .07 .08 .09
 = .10
1.1 .3790 .3810 .3830
.50 .40
1.2 .3980 .3997 .4015
z 0 1.28
1.3 .4147 .4162 .4177
Critical Value
= 1.28

Example: Test Statistic

(continued)
Obtain sample evidence and compute the test

statistic
Suppose a sample is taken with the following
results: n = 64, x = 53.1 (=10 was assumed known)
 Then the test statistic is:
x μ 53.1  52
z    0.88
σ 10
n 64
Example: Decision
(continued)
Reach a decision and interpret the result:
Reject H0
 = .10

1.28
0
z = .88
Do not reject H0 since z = 0.88 ≤ 1.28
i.e.: there is not sufficient evidence that the
mean bill is over $52

p -Value Solution
(continued)
Calculate the p-value and compare to 

p-value = .1894
P(x  53.1 | μ  52.0)

Reject H0
 = .10  
 53.1  52.0 
 P z  
 10 
0  64 
1.28  P(z  0.88)  .5  .3106
z = .88  .1894
Do not reject H0 since p-value = .1894 >  = .10

Critical Value
Approach to Testing
 When σ is known, convert sample statistic ( x ) to
a z test statistic
Hypothesis
Tests for 
 Known  Unknown
The test statistic is:
x μ
z 
σ
n

Critical Value
Approach to Testing
 When σ is unknown, convert sample statistic ( x )
to a t test statistic
Hypothesis
Tests for 
 Known  Unknown
x μ
t n1 
s
(The population must be
approximately normal) n
Hypothesis Tests for μ,

σ Unknown
 1. Specify the population value of interest
 2. Formulate the appropriate null and
alternative hypotheses
 3. Specify the desired level of significance
 4. Determine the rejection region (critical values
are from the t-distribution with n-1 d.f.)
 5. Obtain sample evidence and compute the
t test statistic
 6. Reach a decision and interpret the result

Example: Two-Tail Test

( Unknown)
The average cost of a

hotel room in New York
is said to be $168 per
night. A random sample
of 25 hotels resulted in
x = $172.50 and H0: μ = 168
s = $15.40. Test at the HA: μ  168
 = 0.05 level.
(Assume the population distribution is normal)
Example Solution: Two-Tail Test
H0: μ = 168
/2=.025 /2=.025
HA: μ  168
 = 0.05
 n = 25 Reject H0
-tα/2
tα/2
0
 Critical Values: -2.0639 1.46 2.0639
t24 = ± 2.0639
x μ 172.50  168
  is unknown, so t n1    1.46
s 15.40
use a t statistic n 25
Do not reject H0: not sufficient evidence that

true mean cost is different than $168

Hypothesis Tests for Proportions
 Involves categorical values

 Two possible outcomes
 “Success” (possesses a certain characteristic)
 “Failure” (does not possesses that characteristic)
 Fraction or proportion of population in the

“success” category is denoted by π
Proportions
(continued)
 The sample proportion of successes is denoted

by p :
x number of successes in sample

p 
n sample size
 When both nπ and n(1- π) are at least 5, p

is approximately normally distributed with mean
and standard deviation

μp  π π(1 π)
σp 
n

Hypothesis Tests for Proportions
 The sampling
distribution of p is Hypothesis
normal, so the test Tests for π
statistic is a z
value:
nπ  5 nπ < 5
pπ and or
z n(1-π)  5 n(1-π) < 5
π(1 π)
Not discussed
n in this chapter
Example: z Test for Proportion
A marketing company
claims that it receives
8% responses from its
mailing. To test this
claim, a random sample
of 500 were surveyed
Check:
with 25 responses. Test
at the  = .05 n π = (500)(.08) = 40 
significance level. n(1-π) = (500)(.92) = 460

Z Test for Proportion: Solution
H0: π = .08 Test Statistic:

HA: π  .08 pπ .05  .08
z   2.47
π(1 π) .08(1  .08)
 = .05
n 500
n = 500, p = .05
Critical Values: ± 1.96 Decision:
Reject Reject Reject H0 at  = .05
Conclusion:
.025 .025
There is sufficient
-1.96 0 1.96 z evidence to reject the
-2.47 company‟s claim of 8%
response rate. Chap 9-97
p -Value Solution
(continued)
Calculate the p-value and compare to 
(For a two sided test the p-value is always two sided)
Do not reject H0
Reject H0 Reject H0 p-value = .0136:
/2 = .025 /2 = .025
P(z  2.47)  P(x  2.47)
.0068 .0068  2(.5  .4932)
 2(.0068)  0.0136
-1.96 0 1.96
z = -2.47 z = 2.47
Reject H0 since p-value = .0136 <  = .05


Type II Error
 Type II error is the probability of
failing to reject a false H0
Suppose we fail to reject H0: μ  52
when in fact the true mean is μ = 50
50 52
Reject Do not reject
H0: μ  52 H0 : μ  52
Type II Error
(continued)
 Suppose we do not reject H0:   52 when in fact
the true mean is  = 50
This is the range of x where

This is the true H0 is not rejected
distribution of x if  = 50
50 52
H0:   52 H0 :   52

Type II Error
(continued)
 Suppose we do not reject H0: μ  52 when
in fact the true mean is μ = 50
Here, β = P( x  cutoff ) if μ = 50
 β
50 52
H0: μ  52 H0 : μ  52
Calculating β
 Suppose n = 64 , σ = 6 , and  = .05
σ 6
cutoff  x   μ  z  52  1.645  50.766
(for H0 : μ  52) n 64
So β = P( x  50.766 ) if μ = 50
50 50.766 52
H0: μ  52 H0 : μ  52

Calculating β
(continued)
 Suppose n = 64 , σ = 6 , and  = .05
 
 50.766  50 
P( x  50.766 | μ  50)  P z    P(z  1.02)  .5  .3461  .1539
 6 
 64 
Probability of
type II error:
 β = .1539
50 52
H0: μ  52 H0 : μ  52
Chapter Summary
 Addressed hypothesis testing methodology

 Performed z Test for the mean (σ known)
 Discussed p–value approach to
hypothesis testing
 Performed one-tail and two-tail tests . . .

Chapter Summary
(continued)
 Performed t test for the mean (σ

unknown)
 Performed z test for the proportion
 Discussed Type II error and computed its
probability
7th Edition
Chapter 10
Estimation and Hypothesis Testing
for Two Population Parameters

Chapter Goals

able to:
 Test hypotheses or form interval estimates for
 two independent population means
 Standard deviations known
 Standard deviations unknown
 two means from paired samples
 the difference between two population
proportions
Estimation for Two Populations

Estimating two
population values
Population
means, Paired Population
independent samples proportions
samples
Examples:
Group 1 vs. Same group Proportion 1 vs.
independent before vs. after Proportion 2
Group 2 treatment

Difference Between Two Means
Population means, Goal: Form a confidence

independent
samples
* interval for the difference
between two population
means, μ1 – μ2
σ1 and σ2 known
The point estimate for the

σ1 and σ2 unknown difference is
but assumed equal
x1 – x2
σ1 and σ2 unknown,
not assumed equal
Independent Samples
 Different data sources

Population means,
independent
samples
*  Unrelated
 Independent
 Sample selected from
σ1 and σ2 known one population has no

effect on the sample
selected from the other
σ1 and σ2 unknown population
but assumed equal
 Use the difference between
2 sample means
not assumed equal

σ1 and σ2 known
Population means, Assumptions:

independent
samples  Samples are randomly and
independently drawn
σ1 and σ2 known *  population distributions are
normal or both sample sizes
σ1 and σ2 unknown
but assumed equal are  30
 Population standard
not assumed equal deviations are known
σ1 and σ2 known
(continued)
When σ1 and σ2 are known and

Population means,
both populations are normal or
independent
both sample sizes are at least 30,
samples
the test statistic is a z value…
σ1 and σ2 known * …and the standard error of

x1 – x2 is
σ1 and σ2 unknown
but assumed equal 2 2
σ1 σ 2
σ x1  x 2  
not assumed equal
n1 n2

σ1 and σ2 known
(continued)
Population means,
independent The confidence interval for
samples μ1 – μ2 is:
σ1 and σ2 known *
x 1 
 x 2  z/2
σ12 σ 22

n1 n2
σ1 and σ2 unknown
but assumed equal
not assumed equal
large samples

independent
 Samples are randomly and
samples
independently drawn
σ1 and σ2 known  Population standard

deviations are unknown
σ1 and σ2 unknown
but assumed equal
*  The two standard deviations
are equal
not assumed equal

large samples
(continued)
Population means,
independent
Forming interval estimates:
samples
 The population standard
deviations are assumed equal,
σ1 and σ2 known so use the two sample
standard deviations and pool
σ1 and σ2 unknown
but assumed equal
* them to estimate σ
 the test statistic is a t value

σ1 and σ2 unknown, with (n1 + n2 – 2) degrees
not assumed equal of freedom
large samples
(continued)
Population means, The pooled standard

independent deviation is
samples
σ1 and σ2 known
sp 
n1  1s12  n2  1s22
σ1 and σ2 unknown
but assumed equal
* n1  n2  2
not assumed equal

large samples
(continued)
Population means, The confidence interval for

independent μ1 – μ2 is:
samples
σ1 and σ2 known x 1 
 x 2  t /2 sp
1 1

n1 n2
σ1 and σ2 unknown
but assumed equal
* Where t/2 has (n1 + n2 – 2) d.f.,
and
sp 
n1  1s12  n2  1s22
not assumed equal n1  n2  2
small samples

independent
samples  populations are normally
distributed
σ1 and σ2 known  there is a reason to believe

that the populations do not
σ1 and σ2 unknown have equal variances
but assumed equal
 samples are independent
not assumed equal
*

small samples
(continued)
Forming interval
Population means, estimates:
independent
samples  The population variances
are not assumed equal, so
σ1 and σ2 known we do not pool them
 the test statistic is a t value

σ1 and σ2 unknown with degrees of freedom
but assumed equal given by:
(s12 /n1  s22 /n2 )2
not assumed equal
* df 
  
 s2 /n 2 s2 /n 2 
 1 1  2 2  
 n1  1 n2  1 
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.  Chap 10-119
small samples
(continued)
Population means, The confidence interval for

samples
σ1 and σ2 known x 1 
 x 2  t α/2
s12 s22

n1 n2
σ1 and σ2 unknown
but assumed equal Where t/2 has d.f. given by
(s12 /n1  s22 /n2 )2

not assumed equal
* df 
  
 s2 /n 2 s2 /n 2 
 1 1  2 2  
 n1  1 n2  1 

Hypothesis Tests for the

Difference Between Two Means
 Testing Hypotheses about μ1 – μ2
 Use the same situations discussed already:

 Standard deviations known
 Standard deviations unknown
 Assumed equal
 Assumed not equal
Hypothesis Tests for

Two Population Proportions
Two Population Means, Independent Samples
Lower tail test: Upper tail test: Two-tailed test:
H0: μ1  μ2 H0: μ1 ≤ μ2 H0: μ1 = μ2

HA: μ1 < μ2 HA: μ1 > μ2 HA: μ1 ≠ μ2
i.e., i.e., i.e.,
H0: μ1 – μ2  0 H0: μ1 – μ2 ≤ 0 H0: μ1 – μ2 = 0
HA: μ1 – μ2 < 0 HA: μ1 – μ2 > 0 HA: μ1 – μ2 ≠ 0

Hypothesis tests for μ1 – μ2
Population means, independent samples
σ1 and σ2 known Use a z test statistic
Use sp to estimate unknown

σ1 and σ2 unknown σ , use a t test statistic with
but assumed equal n1 + n2 – 2 d.f.
σ1 and σ2 unknown, Use s1 and s2 to estimate

not assumed equal unknown σ1 and σ2 , use a t
test statistic and calculate the
required degrees of freedom
σ1 and σ2 known
Population means,
independent The test statistic for
samples μ1 – μ2 is:
σ1 and σ2 known * z
x 1 
 x 2   μ1  μ2 
2 2
σ1 and σ2 unknown σ1 σ 2
but assumed equal 
n1 n2
not assumed equal

large samples
The test statistic for
Population means,
x 
samples
 x 2   μ1  μ2 
t
1
σ1 and σ2 known
1 1
sp 
σ1 and σ2 unknown
but assumed equal
* n1 n2
Where t has (n1 + n2 – 2) d.f.,
σ1 and σ2 unknown, and
not assumed equal sp 
n1  1s12  n2  1s2 2
n1  n2  2
small samples
The test statistic for
Population means,
μ1 – μ2 is:
independent
samples
t
x 1 
 x 2   μ1  μ2 
2 2
σ1 and σ2 known s1 s2

n1 n2
σ1 and σ2 unknown
but assumed equal Where t has d.f. given by
(s12 /n1  s22 /n2 )2
not assumed equal
* df 
  
 s2 /n 2 s2 /n 2 
 1 1  2 2  
 n1  1 n2  1 

Hypothesis tests for μ1 – μ2

Two Population Means, Independent Samples
H0: μ1 – μ2  0 H0: μ1 – μ2 ≤ 0 H0: μ1 – μ2 = 0
HA: μ1 – μ2 < 0 HA: μ1 – μ2 > 0 HA: μ1 – μ2 ≠ 0
Example: σ1 and σ2 known:
  /2 /2
-z z -z/2 z/2

Reject H0 if z < -z Reject H0 if z > z Reject H0 if z < -z/2
or z > z/2
Pooled sp t Test Example

σ1 and σ2 unknown, assumed equal
You‟re a financial analyst for a brokerage firm. Is there a
difference in dividend yield between stocks listed on the
NYSE & NASDAQ? You collect the following data:
NYSE NASDAQ
Number 21 25
Sample mean 3.27 2.53
Sample std dev 1.30 1.16
Assuming equal variances, is

there a difference in average
yield ( = 0.05)?

Calculating the Test Statistic

t
x 1 
 x 2  μ1  μ2  3.27  2.53   0
  2.040
1 1 1 1
sp  1.2256 
n1 n2 21 25
Where:
sp 
n1  1s12  n2  1s2 2 
21  11.30 2  25  11.16 2  1.2256
n1  n2  2 21  25  2
Solution
Reject H0 Reject H0
H0: μ1 - μ2 = 0 i.e. (μ1 = μ2)
HA: μ1 - μ2 ≠ 0 i.e. (μ1 ≠ μ2)
.025 .025
 = 0.05
df = 21 + 25 - 2 = 44 -2.0154 0 2.0154 t
Critical Values: t = 2.0154
2.040
Test Statistic: Decision:
3.27  2.53 Reject H0 at  = 0.05
t  2.040
1 1
1.2256  Conclusion:
21 25 There is evidence that
the means are different.

Paired Samples
Tests Means of 2 Related Populations

Paired  Paired or matched samples
samples  Repeated measures (before/after)
 Use difference between paired values:
d = x 1 - x2
 Eliminates Variation Among Subjects

 Assumptions:
 Both Populations Are Normally Distributed
 Or, if Not Normal, use large samples
Paired Differences
The ith paired difference is di , where
Paired di = x1i - x2i
samples
n
The point estimate for
the population mean
d i
d i 1
paired difference is d : n
n
The sample standard
deviation is  (d  d) i
2
sd  i1
n 1
n is the number of pairs in the paired sample

Paired Differences
(continued)
Paired The confidence interval for d is

samples
sd
dt
n
n
Where t has n - 1 d.f. and sd is:  (d  d)

i
2
sd  i1
n 1
n is the number of pairs in the paired sample
Hypothesis Testing for

Paired Samples
The test statistic for d is
Paired
samples
d  μd
t
sd
n
n is the
number n
of pairs
in the
Where t has n - 1 d.f.  (d i  d)2
paired and sd is: sd  i 1
sample n 1

Hypothesis Testing for

Paired Samples
(continued)
Paired Samples
H0: μd  0 H0: μd ≤ 0 H0: μd = 0

HA: μd < 0 HA: μd > 0 HA: μd ≠ 0
  /2 /2
-t t -t/2 t/2

Reject H0 if t < -t Reject H0 if t > t Reject H0 if t < -t/2
or t > t/2
Where t has n - 1 d.f.
Paired Samples Example

 Assume you send your salespeople to a “customer
service” training workshop. Is the training effective?
You collect the following data:
Number of Complaints: (2) - (1)  di

Salesperson Before (1) After (2) Difference, di d = n
C.B. 6 4 - 2
= -4.2
T.F. 20 6 -14
M.H. 3 2 - 1
R.K.
M.O.
0
4
0
0
0
- 4 sd 
 (d  d)
i
2
-21 n 1
 5.67

Paired Samples: Solution

 Has the training made a difference in the number of
complaints (at the 0.05 level)?
Reject Reject
H0: μd = 0
HA: μd  0
/2 /2
 = .05 d = - 4.2 - 2.7765 2.7765
- 1.66
Critical Value = 2.7765
d.f. = n - 1 = 4
Decision: Do not reject H0
(t stat is not in the reject region)
Test Statistic:
Conclusion: There is not a
d  μd  4.2  0
t   1.66 significant change in the
sd / n 5.67/ 5 number of complaints.
Goal: Form a confidence interval for

Population or test a hypothesis about the
proportions difference between two population
proportions, π1 – π2
Assumptions:
n1π1  5 , n1(1-π1)  5
n2π2  5 , n2(1-π2)  5
The point estimate for

the difference is p1 – p2

Confidence Interval for

Population The confidence interval for

proportions
π1 – π2 is:
p1(1 p1 ) p2 (1 p2 )
 p1  p2   z 
n1 n2

Population proportions
H0: π1  π2 H0: π1 ≤ π2 H0: π1 = π2

HA: π1 < π2 HA: π1 > π2 HA: π1 ≠ π2
i.e., i.e., i.e.,
H0: π1 – π2  0 H0: π1 – π2 ≤ 0 H0: π1 – π2 = 0
HA: π1 – π2 < 0 HA: π1 – π2 > 0 HA: π1 – π2 ≠ 0


Since we begin by assuming the null
hypothesis is true, we assume π1 = π2
Population
and pool the two p estimates
proportions
The pooled estimate for the
overall proportion is:
n1p1  n2p2 x1  x 2
p 
n1  n2 n1  n2
where x1 and x2 are the numbers from
samples 1 and 2 with the characteristic of interest

(continued)
Population The test statistic for

proportions π1 – π2 is:
z
 p1  p2    π1  π 2 
 1 1 
p (1 p )   
 n1 n2 


Population proportions
H0: π1 – π2  0 H0: π1 – π2 ≤ 0 H0: π1 – π2 = 0
HA: π1 – π2 < 0 HA: π1 – π2 > 0 HA: π1 – π2 ≠ 0
  /2 /2
-z z -z/2 z/2

Reject H0 if z < -z Reject H0 if z > z Reject H0 if z < -z/2
or z > z/2
Example:
Two population Proportions
Is there a significant difference between the
proportion of men and the proportion of
women who will vote Yes on Proposition A?
 In a random sample, 36 of 72 men and 31 of

50 women indicated they would vote Yes
 Test at the .05 level of significance

Example:
(continued)
 The hypothesis test is:
H0: π1 – π2 = 0 (the two proportions are equal)
HA: π1 – π2 ≠ 0 (there is a significant difference between proportions)
 The sample proportions are:

 Men: p1 = 36/72 = .50
 Women: p2 = 31/50 = .62
 The pooled estimate for the overall proportion is:

x1  x 2 36  31 67
p    .549
n1  n2 72  50 122
Example:
(continued)
Reject H0 Reject H0
The test statistic for π1 – π2 is:
.025 .025
z
 p1  p 2    π1  π 2 
 1 1 
p (1 p )    -1.96 1.96
 1
n n 2 
-1.31

 .50  .62    0    1.31
 1 1  Decision: Do not reject H0
.549 (1 .549)   
 72 50 
Conclusion: There is not
significant evidence of a
Critical Values = 1.96
For  = .05 difference in the proportion
who will vote yes between
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. men and women. Chap 10-146

Chapter Summary
 Compared two independent samples
 Formed confidence intervals for the differences between two
means
 Performed z test for the differences in two means
 Performed t test for the differences in two means
 Compared two related samples (paired samples)
 Formed confidence intervals for the paired difference
 Performed paired sample t tests for the mean difference
 Compared two population proportions
 Formed confidence intervals for the difference between two
population proportions
 Performed z test for two population proportions
7th Edition
Chapter 11
Hypothesis Tests for One and Two
Population Variances

Chapter Goals

able to:
 Formulate and complete hypothesis tests for a single
population variance
 Find critical chi-square distribution values from the
chi-square table
 Formulate and complete hypothesis tests for the
difference between two population variances
 Use the F table to find critical F values
Hypothesis Tests for Variances
Hypothesis Tests
for Variances
Tests for a Single Tests for Two

Population Variance Population Variances
Chi-Square test statistic F test statistic

Single Population
Tests for a Single * H0: σ2 = σ02

HA: σ2 ≠ σ02
Two tailed test
Population Variance
H0: σ2  σ02
Lower tail test
HA: σ2 < σ02
Chi-Square test statistic
H0: σ2 ≤ σ02
Upper tail test
HA: σ2 > σ02
Chi-Square Test Statistic
The chi-squared test statistic for

a Single Population Variance is:
Tests for a Single
Population Variance (n  1)s2
 2
σ2
Chi-Square test statistic * where
2 = standardized chi-square variable
n = sample size
s2 = sample variance
σ2 = hypothesized variance

The Chi-square Distribution

 The chi-square distribution is a family of
distributions, depending on degrees of freedom:
 d.f. = n - 1
0 4 8 12 16 20 24 28 2 0 4 8 12 16 20 24 28 2 0 4 8 12 16 20 24 28 2
d.f. = 1 d.f. = 5 d.f. = 15

The critical value,   , is found from the

2

chi-square table
Upper tail test:
H0: σ2 ≤ σ02
HA: σ2 > σ02
2
 2

Example
 A commercial freezer must hold the selected
temperature with little variation. Specifications call
for a standard deviation of no more than 4 degrees
(or variance of 16 degrees2). A sample of 16
freezers is tested and
yields a sample variance
of s2 = 24. Test to see
whether the standard
deviation specification
is exceeded. Use
 = .05

 Use the chi-square table to find the critical value:
2 = 24.9958 ( = .05 and 16 – 1 = 15 d.f.)
(n  1)s2 (16  1)24

 
2
  22.5
σ2 16
Since 22.5 < 24.9958,
do not reject H0  = .05
There is not significant

evidence at the  = .05 level 2
that the standard deviation Do not reject H0 Reject H0
specification is exceeded  2
Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. = 24.9958 Chap 11-156

Lower Tail or Two Tailed

Chi-square Tests
Lower tail test: Two tail test:
H0: σ2  σ02 H0: σ2 = σ02
HA: σ2 < σ02 HA: σ2 ≠ σ02
 /2
/2
2 2
Reject Do not reject H0 Reject Do not Reject
21- reject H0
21-/2 2/2
(2L) 2
( U)
Confidence Interval Estimate

for σ2
 The confidence interval estimate for σ2 is
(n  1)s2 (n  1)s2
/2 σ 
2
/2
χU2 χL2
21-/2 2/2 2
(2L) (2U) Where 2L and 2U are from the
2 distribution with n -1 degrees
of freedom

Example
 A sample of 16 freezers yields a sample
variance of s2 = 24.
 Form a 95% confidence interval for the
population variance.
Example
(continued)
 Use the chi-square table to find 2 L and 2 U :
( = .05 and 16 – 1 = 15 d.f.)
/2=.025 (n  1)s 2 (n  1)s 2
 σ 2

/2=.025 χU2 χL2
(16  1)24 (16  1)24

2.975 2.025  σ2 
27.4884 6.2621
(2L) (2U)
6.2621 27.4884 13.096  σ 2  57.489
We are 95% confident that the population variance is between 13.096
and 57.489 degrees2. (Taking the square root, we are 95% confident that
the population standard deviation is between 3.619 and 7.582 degrees.)

F Test for Difference in Two

H0: σ12 = σ22

Two tailed test
* Tests for Two
HA: σ12 ≠ σ22 Population Variances
H0: σ12  σ22 Lower tail test

HA: σ12 < σ22 F test statistic
H0: σ12 ≤ σ22 Upper tail test

HA: σ12 > σ22
F Test for Difference in Two

The F test statistic is:
s12 Tests for Two

F 2 Where F has D1
numerator and D2
s2 denominator
degrees of freedom
s12 = Variance of Sample 1

* F test statistic
D1 = n1 - 1 = numerator degrees of freedom
2
s 2 = Variance of Sample 2
D2 = n2 - 1 = denominator degrees of freedom

The F Distribution
 The F critical value is found from the F table

 The are two appropriate degrees of freedom:
D1 (numerator) and D2 (denominator)
s12
F 2 where D1 = n1 – 1 ; D2 = n2 – 1
s2
 In the F table,
 numerator degrees of freedom determine the row
 denominator degrees of freedom determine the column
Formulating the F Ratio
s12
F 2 where D1 = n1 – 1 ; D2 = n2 – 1
s2
 For a two-tailed test, always place the larger

sample variance in the numerator
 For a one-tailed test, consider the alternative
hypothesis: place in the numerator the sample
variance for the population that is predicted
(based on HA) to have the larger variance


H0: σ12  σ22 H0: σ12 = σ22
HA: σ12 < σ22 HA: σ12 ≠ σ22
H0: σ12 ≤ σ22
HA: σ12 > σ22
 /2
0 F 0 F
Do not Reject H0 Do not Reject H0
reject H0 F reject H0 F/2
 rejection region  rejection region for
for a one-tail test is a two-tailed test is
s12 s12
F  2  F F  2  F / 2
s2 s2
(where the larger sample variance in the numerator)
F Test: An Example
You are a financial analyst for a brokerage firm. You

want to compare dividend yields between stocks listed
on the NYSE & NASDAQ. You collect the following data:
NYSE NASDAQ
Number 21 25
Mean 3.27 2.53
Std dev 1.30 1.16
Is there a difference in the

variances between the NYSE
& NASDAQ at the  = 0.05 level?

F Test: Example Solution

 Form the hypothesis test:
H0: σ21 = σ22 (there is no difference between variances)
HA: σ21 ≠ σ22 (there is a difference between variances)
 Find the F critical value for  = .05:

 Numerator:
 D1 = n1 – 1 = 21 – 1 = 20
 Denominator:
 D2 = n2 – 1 = 25 – 1 = 24
F.05/2, 20, 24 = 2.327
F Test: Example Solution

(continued)
 The test statistic is: H0: σ12 = σ22

HA: σ12 ≠ σ22
s12 1.30 2
F 2   1.256
s2 1.16 2
/2 = .025
0
 F = 1.256 is not greater than Do not Reject H0
reject H0 F/2
the critical F value of 2.327, so
=2.327
we do not reject H0
 Conclusion: There is no evidence of a

difference in variances at  = .05

Using EXCEL and PHStat
EXCEL
 F test for two variances:
 Data | Data Analysis | F-test: Two Sample for Variances
PHStat
 Chi-square test for the variance:
 PHStat | One-sample Tests | Chi-square Test for the Variance
 F test for two variances:
 PHStat | Two-sample Tests | F Test for Differences in Two
Variances
Chapter Summary
 Performed chi-square tests for the variance

 Used the chi-square table to find chi-square
critical values
 Performed F tests for the difference between two
population variances
 Used the F table to find F critical values

7th Edition
Chapter 12
Analysis of Variance
Chapter Goals
After completing this chapter, you should be able

to:
 Recognize situations in which to use analysis of variance
 Understand different analysis of variance designs
 Perform a single-factor hypothesis test and interpret results
 Conduct and interpret post-analysis of variance pairwise
comparisons procedures
 Set up and perform randomized blocks analysis
 Analyze two-factor analysis of variance test with replications
results

Chapter Overview
Analysis of Variance (ANOVA)
One-Way Randomized Two-factor

ANOVA Complete ANOVA
Block ANOVA with replication
F-test
F-test
Tukey-
Kramer Fisher‟s Least
test Significant
Difference test
General ANOVA Setting
 Investigator controls one or more independent

variables
 Called factors (or treatment variables)
 Each factor contains two or more levels (or
categories/classifications)
 Observe effects on dependent variable
 Response to levels of independent variable
 Experimental design: the plan used to test
hypothesis

One-Way Analysis of Variance
 Evaluate the difference among the means of

three or more populations
Examples: Accident rates for 1st, 2nd, and 3rd shift
Expected mileage for five brands of tires
 Assumptions
 Populations are normally distributed
 Populations have equal variances
 Samples are randomly and independently

drawn
Completely Randomized Design
 Experimental units (subjects) are assigned

randomly to treatments
 Only one factor or independent variable
 With two or more treatment levels
 Analyzed by
 One-factor analysis of variance (one-way ANOVA)
 Called a Balanced Design if all factor levels
have equal sample size

Hypotheses of One-Way
ANOVA
 H0 : μ1  μ2  μ3    μk
 All population means are equal
 i.e., no treatment effect (no variation in means among
groups)

HA : Not all of the population means are the same
 At least one population mean is different
 i.e., there is a treatment effect
 Does not mean that all population means are different
(some pairs may be the same)
One-Factor ANOVA
H0 : μ1  μ2  μ3    μk
HA : Not all μi are the same
All Means are the same:

The Null Hypothesis is True
(No Treatment Effect)
μ1  μ2  μ3

One-Factor ANOVA
(continue
H0 : μ1  μ2  μ3    μk d)
HA : Not all μi are the same
At least one mean is different:
The Null Hypothesis is NOT true
(Treatment Effect is present)
or
μ1  μ2  μ3 μ1  μ2  μ3
Partitioning the Variation

 Total variation can be split into two parts:
SST = SSB + SSW
SST = Total Sum of Squares

SSB = Sum of Squares Between
SSW = Sum of Squares Within


(continue
d)
SST = SSB + SSW
Total Variation (SST) = the aggregate dispersion of the

individual data values across the various factor levels
Between-Sample Variation (SSB) = dispersion among the

factor sample means
Within-Sample Variation (SSW) = dispersion that exists

among the data values within a particular factor level
Partition of Total Variation
Total Variation (SST)
Variation Due to Variation Due to Random

= Factor (SSB) + Sampling (SSW)
Commonly referred to as: Commonly referred to as:

 Sum of Squares Between  Sum of Squares Within
 Sum of Squares Among  Sum of Squares Error
 Sum of Squares Explained  Sum of Squares Unexplained
 Among Groups Variation  Within Groups Variation

Total Sum of Squares
SST = SSB + SSW

k ni
SST   ( x ij  x )2
i1 j1
Where:
SST = Total sum of squares
k = number of populations (levels or treatments)
ni = sample size from population i
xij = jth measurement from population i
x = grand mean (mean of all data values)
Total Variation
(continued)
SST  ( x11  x )2  ( x12  x )2  ...  ( x kn k  x )2
Response, X
Group 1 Group 2 Group 3

Sum of Squares Between
SST = SSB + SSW

k
SSB   ni ( x i  x )2
i1
Where:
SSB = Sum of squares between
k = number of populations
xi = sample mean from population i
Between-Group Variation
k
SSB   ni ( x i  x )2
i1
SSB
Variation Due to MSB 
Differences
GroupsAmong k 1
Mean Square
Between =
SSB/degrees of
i j
freedom

Between-Group Variation
(continued)
SSB  n1 ( x1  x )2  n2 ( x 2  x )2  ...  nk ( x k  x )2
Response, X
x3
x2
x
x1

Sum of Squares Within
SST = SSB + SSW

k nj
SSW    ( x ij  x i )2
i1 j1
Where:
SSW = Sum of squares within
xi = sample mean from population i
xij = jth measurement from population i

Within-Group Variation
k nj
SSW    ( x ij  x i )2
i1 j1
SSW
Summing the variation MSW 
within each group and nT  k
then adding over all
groups Mean Square Within =
SSW/degrees of
freedom
i
Within-Group Variation
(continued)
SSW  ( x11  x1 )2  ( x12  x 2 )2  ...  ( x kn k  x k )2
Response, X
x3
x2
x1


One-Way ANOVA Table
Source of SS df MS F ratio
Variation
Between SSB MSB
SSB k-1 MSB =
Samples k - 1 F = MSW
Within SSW
SSW nT - k MSW =
Samples nT - k
SST =
Total nT - 1
SSB+SSW
nT = sum of the sample sizes from all populations
df = degrees of freedom
One-Factor ANOVA
F Test Statistic
H0: μ1= μ2 = … = μ k
HA: At least two population means are different
 Test statistic MSB

F
MSW
MSB is mean squares between variances
MSW is mean squares within variances
 Degrees of freedom
 df1 = k – 1 (k = number of populations)
 df2 = nT – k (nT = sum of sample sizes from all populations)

Interpreting One-Factor ANOVA

F Statistic
 The F statistic is the ratio of the between
estimate of variance and the within estimate
of variance
 The ratio must always be positive
 df1 = k -1 will typically be small
 df2 = nT - k will typically be large
The ratio should be close to 1 if

H0: μ1= μ2 = … = μk is true
The ratio will be larger than 1 if

H0: μ1= μ2 = … = μk is false
One-Factor ANOVA
F Test Example
You want to see if three Club 1 Club 2 Club 3

different golf clubs yield 254 234 200
different distances. You 263 218 222
randomly select five 241 235 197
measurements from trials on 237 227 206
an automated driving 251 216 204
machine for each club. At the
.05 significance level, is there
a difference in mean
distance?

One-Factor ANOVA Example:

Scatter Diagram
Distan
Club 1 Club 2 Club 3 270
ce
254 234 200 •
260 ••
263 218 222 x1
241 235 197 250
•
237 227 206 240 • ••
251 216 204
230
•x
2 •
x
••
220
x1  249.2 x 2  226.0 x 3  205.8
•• x 3
210 ••
x  227.0
200

190 1 2 3Chap 12-195
Club
One-Factor ANOVA Example

Computations
Club 1 Club 2 Club 3 x1 = n1 = 5
254 234 200 249.2
n2 = 5
263 218 222
x2 =
241 235 197 n3 = 5
226.0
237 227 206
nT = 15
251 216 204 x3 =
205.8 k=3
SSB = 5 [ (249.2 – 227)2 + (226 – 227)2 + (205.8 – 227)2 ]
SSW = (254 – 249.2)= 2 + 4716.4
(263 – 249.2)2 +…+ (204 –
x = 227.0
205.8)2 = 1119.6
MSB = 4716.4 / (3-1) = 2358.2
F  25.275
MSW = 1119.6 2358.2/ (15-3) = 93.3
93.3

One-Factor ANOVA Example

Solution
H0: μ1 = μ2 = μ3 Test Statistic:
HA: μi not all equal
MSB 2358.2
 = .05 F   25.275
df1= 2 df2 = 12 MSW 93.3
Decision:
Critical
Value: Reject H0 at  = 0.05
Conclusion:
F =
3.885 =
.05 There is evidence that
0 Do not Reject H0 at least one μi differs
reject H0 F= from the rest
F.05 =
Business Statistics: A Decision-Making Approach, 7e © 200825.275
Prentice-Hall, Inc. Chap 12-197
3.885
ANOVA -- Single Factor:

Excel Output
EXCEL: tools | data analysis | ANOVA: single factor
SUMMARY
Groups Count Sum Average Variance
Club 1 5 1246 249.2 108.2
Club 2 5 1130 226 77.5
Club 3 5 1029 205.8 94.2
ANOVA
Source of
SS df MS F P-value F crit
Variation
Between
4716.4 2 2358.2 25.275 4.99E-05 3.885
Groups
Within
1119.6 12 93.3
Groups
Total 5836.0 14

The Tukey-Kramer Procedure

 Tells which population means are significantly
different
 e.g.: μ1 = μ2  μ3
 Done after rejection of equal means in ANOVA
 Allows pair-wise comparisons
 Compare absolute mean differences with critical
range
μ1= μ2 μ3 x
Tukey-Kramer Critical Range
MSW 1 1
Critical Range  q   
2 n n 
 i j 
where:
q = Value from standardized range table
with k and nT - k degrees of
freedom for the desired level of 
MSW = Mean Square Within
Business Statistics: An
i and nj = Sample sizes from populations (levels)
Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 12-200
i and j

The Tukey-Kramer Procedure:

Example
1. Compute absolute mean
Club 1 Club 2 Club 3 differences:
254 234 200
263 218 222 x1  x 2  249.2  226.0  23.2
241 235 197 x1  x 3  249.2  205.8  43.4
237 227 206
251 216 204 x 2  x 3  226.0  205.8  20.2
2. Find the q value from the table in appendix J with

k and nT - k degrees of freedom for
the desired level of 
qα  3.77
The Tukey-Kramer Procedure:

Example
3. Compute Critical Range:
1 1
Critical Range  qα
MSW     3.77 93.3  1  1   16.285
2 n n  2 5 5
 i j 
4. Compare:
5. All of the absolute mean differences x1  x 2  23.2
are greater than critical range.
Therefore there is a significant x1  x 3  43.4
difference between each pair of
means at 5% level of significance. x 2  x 3  20.2

Randomized Complete Block

ANOVA
 Like One-Way ANOVA, we test for equal population
means (for different factor levels, for example)...
 ...but we want to control for possible variation from a

second factor (with two or more levels)
 Used when more than one factor may influence the

value of the dependent variable, but only one is of key
interest
 Levels of the secondary factor are called blocks

 Total variation can now be split into three parts:
SST = SSB + SSBL + SSW
SST = Total sum of squares

SSB = Sum of squares between factor levels
SSBL = Sum of squares between blocks
SSW = Sum of squares within levels

Sum of Squares for Blocking
b
SSBL   k( x j  x )2
j1
Where:
k = number of levels for this factor
b = number of blocks
xj = sample mean from the jth block

 Total variation can now be split into three parts:
SST and SSB are SSW = SST – (SSB + SSBL)

computed as they were
in One-Way ANOVA

Mean Squares
SSBL
MSBL  Mean square blocking 
b 1
SSB
MSB  Mean square between 
k 1
SSW
MSW  Mean square within 
(k  1)(b  1)
Randomized Block ANOVA

Table
Source of SS df MS F ratio
Variation
Between MSBL
SSBL b-1 MSBL
Blocks MSW
Between MSB
SSB k-1 MSB
Samples MSW
Within
SSW (k–1)(b-1) MSW
Samples
Total SST nT - 1
k = number of populations nT = sum of the sample sizes from all populations
b = number of blocks df = degrees of freedom

Blocking Test
H0 : μb1  μb2  μb3  ...
HA : Not all block means are equal
MSBL  Blocking test: df1 = b - 1

F=
MSW df2 = (k – 1)(b – 1)
Reject H0 if F > F
Main Factor Test

H0 : μ1  μ2  μ3  ...  μk
HA : Not all population means are equal
MSB  Main Factor test: df1 = k - 1

F=
MSW df2 = (k – 1)(b – 1)
Reject H0 if F > F

Fisher‟s
Least Significant Difference Test
 To test which population means are significantly
different
 e.g.: μ1 = μ2 ≠ μ3
 Done after rejection of equal means in randomized
block ANOVA design
 Allows pair-wise comparisons
 Compare absolute mean differences with critical
range
1= 2 3 x
Fisher‟s Least Significant

Difference (LSD) Test
2
LSD  t /2 MSW
b
where:
t/2 = Upper-tailed value from Student‟s t-
distribution
for /2 and (k - 1)(b - 1) degrees of freedom
MSW = Mean square within from ANOVA table
b = number of blocks
k = number of levels of the main factor

Fisher‟s Least Significant

Difference (LSD) Test
(continued)
2
LSD  t /2 MSW
b
Compare:
Is xi  x j  LSD ? x1  x 2
If the absolute mean difference x1  x 3

is greater than LSD then there
is a significant difference x2  x3
between that pair of means at
the chosen level of significance. etc ...
Two-Factor ANOVA
 Examines the effect of

 Two or more factors of interest on the
dependent variable
 e.g.: Percent carbonation and line speed on
soft drink bottling process

 Interaction between the different levels of these
two factors
 e.g.: Does the effect of one particular
percentage of carbonation depend on which

level the line speed is set?

Two-Factor ANOVA
(continued)
 Assumptions
 Populations are normally distributed

 Populations have equal variances
 Independent random samples are
drawn
Two-Way ANOVA
Sources of Variation
Two Factors of interest: A and B

a = number of levels of factor A
b = number of levels of factor B
nT = total number of observations in all cells

Two-Way ANOVA
Sources of Variation
(continued)
SST = SSA + SSB + SSAB + SSE Degrees
of
SSA Freedom
Variation due to factor A a–
:
1
SSB
SST b–
Variation due to factor B
Total Variation 1
SSAB
Variation due to interaction (a – 1)(b –
between A and B 1)
nT -
1 SSE nT –
Inherent variation (Error) ab
Two Factor ANOVA Equations
Total Sum of Squares: a b n

SST   ( x ijk  x )2
i1 j1 k 1
Sum of Squares Factor A: a

SS A  bn ( x i  x )2
i1
Sum of Squares Factor B: b

SS B  an ( x j  x )2
j1


(continued)
Sum of Squares
Interaction Between a b
A and B: SS AB  n ( x ij  x i  x j  x )2
i1 j1
Sum of Squares Error:

a b n
SSE   ( x ijk  x ij )2
i1 j1 k 1

(continued)
a b n
where:  x
i1 j1 k 1
ijk
x  Grand Mean
b n
abn
 x
j1 k 1
ijk
xi   Mean of each level of factor A

bn
a n
 x ijk
xj  i1 k 1
 Mean of each level of factor B
an
n x ijk
x ij  
a = number of levels of factor A
 Mean of each cell
k 1 n
b = number of levels of factor B
n‟ = number of replications in each cell

Mean Square Calculations

SS A
MS A  Mean square factor A 
a 1
SSB
MS B  Mean square factor B 
b 1
SS AB
MS AB  Mean square interactio n 
(a  1)(b  1)
SSE
MSE  Mean square error 
nT  ab
Two-Way ANOVA:
The F Test Statistic
F Test for Factor A Main
H0: μA1 = μA2 = μA3 =
MS AEffect Reject
••• F
MSE H0 if F >
HA: Not all μAi are
F
equal F Test for Factor B Main
H0: μB1 = μB2 = μB3 =
MS Effect
••• F B Reject
MSE H0 if F >
HA: Not all μBi are
F
equal
F Test for Interaction
H0: factors A and B do not
MSEffect
interact F AB Reject
to affect the mean
response MSE H0 if F >
HA: factors A and B do FChap 12-222
interact

Two-Way ANOVA
Summary Table
Source of Sum of Degrees of Mean F
Variation Squares Freedom Squares Statistic
MSA MSA
Factor A SSA a–1
= SSA /(a – 1) MSE
MSB MSB
Factor B SSB b–1
= SSB /(b – 1) MSE
AB MSAB MSAB
SSAB (a – 1)(b – 1)
(Interaction) = SSAB / [(a – 1)(b – 1)] MSE
MSE =
Error SSE nT – ab
SSE/(nT – ab)
Total SST nT – 1
Features of Two-Way ANOVA

F Test
 Degrees of freedom always add up
 nT - 1 = (nT - ab) + (a - 1) + (b - 1) + (a - 1)(b - 1)
 Total = error + factor A + factor B + interaction
 The denominator of the F Test is always the

same but the numerator is different
 The sums of squares always add up
 SST = SSE + SSA + SSB + SSAB
 Total = error + factor A + factor B + interaction

Examples:
Interaction vs. No Interaction
 Interaction is
 No interaction:
present:
Factor B Level 1
Mean Response
Mean Response
Factor B Level 1
Factor B Level 3
Factor B Level 2
Factor B Level 2
Factor B Level 3
Factor A Levels 1 Factor A Levels

1 2 2
Chapter Summary
 Described one-way analysis of variance
 The logic of ANOVA
 ANOVA assumptions
 F test for difference in k means
 The Tukey-Kramer procedure for multiple comparisons
 Described randomized complete block designs
 F test
 Fisher‟s least significant difference test for multiple
comparisons
 Described two-way analysis of variance
 Examined effects of multiple factors and interaction

Business Statistics - A Decision-Making Approach

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Business Statistics - A Decision-Making Approach

Uploaded by

Copyright:

Available Formats

Chapter 8 Student Lecture Notes 8-1

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-1

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-2

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Content of this chapter

Point and Interval Estimates

 A point estimate is a single number, used to

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

We can estimate a with a Sample

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-5

 How much uncertainty is associated with a

 An interval estimate provides more

 Such interval estimates are called confidence

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-6

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Confidence Interval Estimate

 An interval gives a range of values:

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-7

Random Sample I am 95%

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-8

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

 The general formula for all

Point Estimate  (Critical Value)(Standard Error)

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-9

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-10

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Confidence Level, (1-)

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-11

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-12

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Confidence Interval for μ

 Population is normally distributed

 If population is not normal, use large sample

 Confidence interval estimate

Finding the Critical Value

z units: -z = -1.96 0 z = 1.96

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Common Levels of Confidence

 Commonly used confidence levels are

Interval and Level of Confidence

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

 Margin of Error (e): the amount added and

Example: Margin of error for estimating μ, σ known:

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-17

Factors Affecting Margin of Error

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-18

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

 A sample of 11 circuits from a large normal

 Determine a 95% confidence interval for the

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-19

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

 We are 95% confident that the true mean

 An incorrect interpretation is that there is 95% probability that this

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-21

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-22

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

Confidence Interval for μ

 If the population standard deviation σ is

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-23

Confidence Interval for μ

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc.

 The t is a family of distributions

Business Statistics: A Decision-Making Approach, 7e © 2008 Prentice-Hall, Inc. Chap 8-25