Anova

Slides by
JOHN
LOUCKS
& Updated by
SPIROS
VELIANITIS
© 2008 Thomson South-Western. All Rights Reserved Slide

1
Chapter 13
Experimental Design and Analysis of Variance
 Introduction to Experimental Design
and Analysis of Variance
 Analysis of Variance
and the Completely Randomized Design
 Multiple Comparison Procedures

2
An Introduction to Experimental Design
 Statistical studies can be classified as being either
experimental or observational.
 In an experimental study, one or more factors are
controlled so that data can be obtained about how the
factors influence the variables of interest.
 In an observational study, no attempt is made to
control the factors.
 Cause-and-effect relationships are easier to establish
in experimental studies than in observational studies.
 Analysis of variance (ANOVA) can be used to analyze
the data obtained from experimental or observational
studies.

3
 In this chapter three types of experimental designs
are introduced.
• a completely randomized design
• a randomized block design
• a factorial experiment

4
 A factor is a variable that the experimenter has
selected for investigation (the independent variable).
 A treatment is a level of a factor.
 Experimental units are the objects of interest in the
experiment.
 A completely randomized design is an experimental
design in which the treatments are randomly
assigned to the experimental units.

5
Analysis of Variance: A Conceptual Overview
Analysis
Analysis of
of Variance
Variance (ANOVA)
(ANOVA) cancan be
be used
used to
to test
test
for
for the
the equality
equality of
of three
three or
or more
more population
population means.
means.
Data
Data obtained
obtained from
from observational
observational or
or experimental
experimental
studies
studies can
can be
be used
used for
for the
the analysis.
analysis.
We
We want
want to
to use
use the
the sample
sample results
results to
to test
test the
the
following
following hypotheses:
hypotheses:
H0: 1=2=3=. . .=

k
Ha: Not all population means are equal

6
H0: 1=2=3=. . .=

k
If H00 is
If H is rejected,
rejected, we
we cannot
cannot conclude
conclude that
that all
all
population
population means
means are
are different.
different.
Rejecting H00 means

Rejecting H means that
that at
at least
least two
two population
population
means
means have
have different
different values.
values.

7
 Assumptions for Analysis of Variance
For
For each
each population,
population, the
the response
response (dependent)
(dependent)
variable
variable is
is normally
normally distributed.
distributed.
The
The variance
variance of
of the
the response
response variable, denoted 
variable, denoted 22,,
is
is the
the same
same for
for all
all of
of the
the populations.
populations.
The
The observations
observations must
must be
be independent.
independent.

8
 Sampling Distribution of x Given H0 is True
Sample means are close together

because there is only
one sampling distribution
when H0 is true.
2
 x2 
n
x2  x1 x3

9
 Sampling Distribution of x Given H0 is False
Sample means come from

different sampling distributions
and are not as close together
when H0 is false.
x3 3 x1 1 2 x2

10
Analysis of Variance
 Between-Treatments Estimate of Population Variance

 Within-Treatments Estimate of Population Variance
 Comparing the Variance Estimates: The F Test
 ANOVA Table

11
Between-Treatments Estimate
of Population Variance 2
 The estimate of 2 based on the variation of the
sample means is called the mean square due to
treatments and is denoted by MSTR.
k
 j j
n (
j 1
x  x ) 2
MSTR 
k1
Denominator is the Numerator is called

degrees of freedom the sum of squares due
associated with SSTR to treatments (SSTR)

12
Within-Treatments Estimate
of Population Variance 2
 The estimate of 2 based on the variation of the
sample observations within each sample is called the
mean square error and is denoted by MSE.
k
2
 (n j  1) s 2j
j1
MSE 
nT  k
Numerator is called
Denominator is the the sum of squares
degrees of freedom due to error (SSE)
associated with SSE

13
Comparing the Variance Estimates: The F Test
 If the null hypothesis is true and the ANOVA

assumptions are valid, the sampling distribution of
MSTR/MSE is an F distribution with MSTR d.f.
equal to k - 1 and MSE d.f. equal to nT - k.
 If the means of the k populations are not equal, the
value of MSTR/MSE will be inflated because MSTR
overestimates 2.
 Hence, we will reject H0 if the resulting value of
MSTR/MSE appears to be too large to have been
selected at random from the appropriate F
distribution.

14
Comparing the Variance Estimates: The F Test
 Sampling Distribution of MSTR/MSE
Sampling Distribution
of MSTR/MSE
Reject H0
Do Not Reject H0 
MSTR/MSE
F
Critical Value

15
Test for the Equality of k Population Means
 Hypotheses
H0: 1=2=3=. . .=

k 
 Test Statistic
F = MSTR/MSE

16
Test for the Equality of k Population Means
 Rejection Rule
p-value Approach: Reject H0 if p-value < 
Critical Value Approach: Reject H0 if F > F
where the value of F is based on an

F distribution with k - 1 numerator d.f.
and nT - k denominator d.f.

17
ANOVA Table
 The ANOVA table decomposes the variance into two
components (sources): a between-group component
(treatments) and a within-group component (error). The F-
ratio, which in this case equals 15.6234, is a ratio of the
between-group estimate to the within-group estimate. Since
the P-value of the F-test is less than 0.05, there is a statistically
significant difference between the means from one level to
another at the 95.0% confidence level.

18
Testing for the Equality of k Population Means:
A Completely Randomized Experimental Design
 Example: AutoShine, Inc.
AutoShine, Inc. is considering marketing a long-
lasting car wax. Three different waxes (Type 1, Type 2,
and Type 3) have been developed.
In order to test the durability
of these waxes, 5 new cars were
waxed with Type 1, 5 with Type
2, and 5 with Type 3. Each car was then
repeatedly run through an automatic carwash until the
wax coating showed signs of deterioration.

19
 Example: AutoShine, Inc.
The number of times each car went through the
carwash before its wax deteriorated is shown on the
next slide. AutoShine, Inc. must decide which wax
to market. Are the three waxes
equally effective?
Factor . . . Car wax

Treatments . . . Type I, Type 2, Type 3
Experimental units . . . Cars
Response variable . . . Number of washes

20
Wax Wax Wax

Observation Type 1 Type 2 Type 3
1 27 33 29
2 30 28 28
3 29 31 30
4 28 30 32
5 31 30 31
Sample Mean 29.0 30.4 30.0

Sample Variance 2.5 3.3 2.5

21
 Hypotheses
H0: 1=2=3
Ha: Not all the means are equal
where:
1 = mean number of washes using Type 1 wax

22
 Test Statistic
F = MSTR/MSE
 Conclusion
The p-value of 0.42 is greater than 0.10, where F = 2.81.
Therefore, we cannot reject H0.

There is insufficient evidence to conclude that
the mean number of washes for the three wax
types are not all the same (waxes are the same).

23
 ANOVA Table
Source of Sum of Degrees of Mean

Variation Squares Freedom Squares F p-Value
Treatments 5.2 2 2.60 .939 .42

Error 33.2 12 2.77
Total 38.4 14

24
An Observational Study
 Example: Reed Manufacturing
Janet Reed would like to know if
there is any significant difference in
the mean number of hours worked per
week for the department managers
at her three manufacturing plants
(in Buffalo, Pittsburgh, and Detroit).
An F test will be conducted using
 = .05.

25
A simple random sample of five
managers from each of the three plants
was taken and the number of hours
worked by each manager in the
previous week is shown on the next
slide.
Factor . . . Manufacturing plant
Treatments . . . Buffalo, Pittsburgh, Detroit
Experimental units . . . Managers
Response variable . . . Number of hours worked

26
Plant 1 Plant 2 Plant 3

Observation Buffalo Pittsburgh Detroit
1 48 73 51
2 54 63 63
3 57 66 61
4 54 64 54
5 62 74 56
Sample Mean 55 68 57
Sample Variance 26.0 26.5 24.5

27
 p -Value and Critical Value Approaches
Develop the hypotheses.

H0: 1=2=3
Ha: Not all the means are equal
where:
1 = mean number of hours worked per
week by the managers at Plant 1
2 = mean number of hours worked per
3 = mean number of hours worked per

28
 ANOVA Table

Variation Squares Freedom Square F p-Value
Treatment 490 2 245 9.55 .0033
Error 308 12 25.667
Total 798 14
The p-value < .05, so we reject H0.
We have sufficient evidence to conclude that the
mean number of hours worked per week by
department managers is not the same at all 3 plant.
29
Multiple Comparison Procedures
 Suppose that analysis of variance has provided

statistical evidence to reject the null hypothesis of
equal population means.
 Fisher’s least significant difference (LSD) procedure
can be used to determine where the differences
occur.

30
Fisher’s LSD Procedure
 Hypotheses
H 0 : i   j
H a : i   j
 Test Statistic
xi  x j
t
MSE( 1 n  1 n )
i j

31
 Rejection Rule
p-value Approach:
Reject H0 if p-value < 
Critical Value Approach:
Reject H0 if t < -ta/2 or t > ta/2
where the value of ta/2 is based on a

t distribution with nT - k degrees of freedom.

32
Based on the Test Statistic xi - xj
Recall that Janet Reed wants to know
if there is any significant difference in
the mean number of hours worked per
week for the department managers
at her three manufacturing plants.
Analysis of variance has provided
statistical evidence to reject the null
hypothesis of equal population means.
Fisher’s least significant difference (LSD) procedure
can be used to determine where the differences occur.

33
 LSD for Plants 1 and 2
• Hypotheses (A) H 0 : 1   2
H a : 1   2
• Rejection Rule
Reject H0 if x1  x2 > 6.98
• Test Statistic
x1  x2 = |55  68| = 13
• Conclusion
The mean number of hours worked at Plant 1 is
not equal to the mean number worked at Plant 2.

34
• Hypotheses (B) H 0 : 1   3
H a : 1   3
• Rejection Rule
• Test Statistic
x1  x3 = |55  57| = 2
• Conclusion
There is no significant difference between the mean
number of hours worked at Plant 1 and the mean
number of hours worked at Plant 3.

35
• Hypotheses (C) H 0 : 2  3
H a : 2  3
• Rejection Rule
• Test Statistic
x2  x3 = |68  57| = 11
• Conclusion
The mean number of hours worked at Plant 2 is
not equal to the mean number worked at Plant 3.

36
Type I Error Rates
 The comparison-wise Type I error rate  indicates

the level of significance associated with a single
pairwise comparison.
 The experiment-wise Type I error rate EW is the
probability of making a Type I error on at least one of
the (k – 1)! pairwise comparisons.
EW = 1 – (1 – )(k – 1)!
 The experiment-wise Type I error rate gets larger for

problems with more populations (larger k).

37
Chapter 13, Part B
 Randomized Block Design

 Factorial Experiments

38
Randomized Block Design
 Experimental units are the objects of interest in the

experiment.
 A completely randomized design is an experimental
design in which the treatments are randomly
assigned to the experimental units.
 If the experimental units are heterogeneous, blocking
can be used to form homogeneous groups, resulting
in a randomized block design (single sample, each
block(element) is used in all treatments).

39
 ANOVA Procedure
• For a randomized block design the sum of squares
total (SST) is partitioned into three groups: sum of
squares due to treatments, sum of squares due to
blocks, and sum of squares due to error.
SST = SSTR + SSBL + SSE
• The total degrees of freedom, nT - 1, are partitioned

such that k - 1 degrees of freedom go to treatments,
b - 1 go to blocks, and (k - 1)(b - 1) go to the error term.

40
 ANOVA Table
Source of Sum of Degrees of Mean p-

Variation Squares Freedom Square F Value
SSTR MSTR
Treatments SSTR k-1 MSTR 
k-1 MSE
SSBL
Blocks SSBL b-1 MSBL 
b -1
SSE
Error SSE (k – 1)(b – 1) MSE 
( k  1)(b  1)
Total SST nT - 1

41
 Example: Crescent Oil Co.

Crescent Oil has developed three
new blends of gasoline and must
decide which blend or blends to
produce and distribute. A study
of the miles per gallon ratings of the
three blends is being conducted to determine if the
mean ratings are the same for the three blends.

42
 Example: Crescent Oil Co.

Five automobiles have been
tested using each of the three
gasoline blends and the miles
per gallon ratings are shown on
the next slide.
Factor . . . Gasoline blend

Treatments . . . Blend X, Blend Y, Blend Z
Blocks . . . Automobiles
Response variable . . . Miles per gallon

43
Type of Gasoline (Treatment)

Automobile Block
(Block) Blend X Blend Y Blend Z Means
1 31 30 30 30.333
2 30 29 29 29.333
3 29 29 28 28.667
4 33 31 29 31.000
5 26 25 26 25.667
Treatment
Means 29.8 28.8 28.4

44
 ANOVA Table

Variation Squares Freedom Square F p-Value
Treatments 5.20 2 2.60 3.82 .07

Blocks 51.33 4 12.80
Error 5.47 8 .68
Total 62.00 14

45
Factorial Experiments
 In some experiments we want to draw conclusions
about more than one variable or factor.
 Factorial experiments and their corresponding
ANOVA computations are valuable designs when
simultaneous conclusions about two or more factors
are required.
 The term factorial is used because the experimental
conditions include all possible combinations of the
factors.
 For example, for a levels of factor A and b levels of
factor B, the experiment will involve collecting data
on ab treatment combinations.

46
Chapter 13
 Introduction to Experimental Design
 Analysis of Variance
and the Completely Randomized Design
 Multiple Comparison Procedures

47

Anova

Enviado por

Dados do documento

Descrição original:

Título original

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Anova

Enviado por

Direitos autorais:

Formatos disponíveis

Slides by

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

H0: 1=2=3=. . .=

Ha: Not all population means are equal

© 2008 Thomson South-Western. All Rights Reserved Slide

H0: 1=2=3=. . .=

Ha: Not all population means are equal

Rejecting H00 means

© 2008 Thomson South-Western. All Rights Reserved Slide

 Assumptions for Analysis of Variance

© 2008 Thomson South-Western. All Rights Reserved Slide

 Sampling Distribution of x Given H0 is True

Sample means are close together

© 2008 Thomson South-Western. All Rights Reserved Slide

 Sampling Distribution of x Given H0 is False

Sample means come from

© 2008 Thomson South-Western. All Rights Reserved Slide

 Between-Treatments Estimate of Population Variance

© 2008 Thomson South-Western. All Rights Reserved Slide

Denominator is the Numerator is called

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

 If the null hypothesis is true and the ANOVA

© 2008 Thomson South-Western. All Rights Reserved Slide

 Sampling Distribution of MSTR/MSE

© 2008 Thomson South-Western. All Rights Reserved Slide

H0: 1=2=3=. . .=

© 2008 Thomson South-Western. All Rights Reserved Slide

p-value Approach: Reject H0 if p-value < 

Critical Value Approach: Reject H0 if F > F

where the value of F is based on an

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

Factor . . . Car wax

© 2008 Thomson South-Western. All Rights Reserved Slide

Wax Wax Wax

Sample Mean 29.0 30.4 30.0

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

Therefore, we cannot reject H0.

© 2008 Thomson South-Western. All Rights Reserved Slide

Source of Sum of Degrees of Mean

Treatments 5.2 2 2.60 .939 .42

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

Plant 1 Plant 2 Plant 3

© 2008 Thomson South-Western. All Rights Reserved Slide

Develop the hypotheses.

© 2008 Thomson South-Western. All Rights Reserved Slide

Source of Sum of Degrees of Mean

 Suppose that analysis of variance has provided

© 2008 Thomson South-Western. All Rights Reserved Slide

© 2008 Thomson South-Western. All Rights Reserved Slide

Critical Value Approach:

Reject H0 if t < -ta/2 or t > ta/2

where the value of ta/2 is based on a