Analysis of Variance

Chapter 12

McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved.
 GOALS

1. List the characteristics of the F distribution.

2. Conduct a test of hypothesis to determine whether the
variances of two populations are equal.
3. Discuss the general idea of analysis of variance.
4. Organize data into a one-way and a two-way ANOVA table.
5. Conduct a test of hypothesis among three or more treatment
means.
6. Develop confidence intervals for the difference in treatment
means.
7. Conduct a test of hypothesis among treatment means using a
blocking variable.
8. Conduct a two-way ANOVA with interaction.

12-2
 The F Distribution

Uses of the F Distribution:

– Test whether two samples are from populations having equal variances

– To compare several population means simultaneously. The simultaneous

comparison of several population means is called analysis of
variance(ANOVA).

Assumptions:
In both of the uses above:
– Populations must follow a normal distribution, and the

– Data must be at least interval-scale.

12-3
 Characteristics of F-distribution
Characteristics of the F
Distribution
1. There is a “family” of F Distributions.
A particular member of the family is
determined by two parameters: the
degrees of freedom in the numerator
and the degrees of freedom in the
denominator.
2. The F distribution is continuous
3. F cannot be negative.
4. The F distribution is positively
skewed.
5. It is asymptotic. As F →  the curve
approaches the X-axis but never
touches it.

12-4
 Comparing Two Population Variances

The F distribution is used to test the hypothesis that the variance of

one normal population equals the variance of another normal
population.

Examples:
⚫ Two Barth shearing machines are set to produce steel bars of the same length. The
bars, therefore, should have the same mean length. We want to ensure that in addition
to having the same mean length they also have similar variation.

⚫ The mean rate of return on two types of common stock may be the same, but there may
be more variation in the rate of return in one than the other. A sample of 10 technology
and 10 utility stocks shows the same mean rate of return, but there is likely more
variation in the Internet stocks.

⚫ A study by the marketing department for a large newspaper found that men and women
spent about the same amount of time per day reading the paper. However, the same
report indicated there was nearly twice as much variation in time spent per day among
the men than the women.

12-5
 Comparing Two Population Variances

12-6
 Test for Equal Variances - Example
Lammers Limos offers limousine service from the city Step 1: The hypotheses are:
hall in Toledo, Ohio, to Metro Airport in Detroit. H0: σ12 = σ22
Sean Lammers, president of the company, is H1: σ12 ≠ σ22
considering two routes. One is via U.S. 25 and
the other via I-75. He wants to study the time it Step 2: The significance level is .05.
takes to drive to the airport using each route
and then compare the results. He collected Step 3: The test statistic is the F distribution.
the following sample data, which is reported
in minutes. Step 4: State the decision rule.
Reject H0 if F > F/2,v1,v2
F > F.10/2,7-1,8-1
is there a
Using the .10 significance level, F > F.05,6,7
difference in the variation in the F > 3.87
driving times for the two routes?

12-7
 Test for Equal Variances - Example

Step 5: Compute the value of F and make a decision

The decision is to reject the null hypothesis, because the computed F value (4.23) is larger than the critical value
(3.87).

We conclude that there is a difference in the variation of the travel times along the two routes.

12-8
 Comparing Means of Two or More
Populations
The F distribution is also used for testing whether two or more sample means came from the same or equal
populations.

Assumptions:
– The sampled populations follow the normal distribution.
– The populations have equal standard deviations.
– The samples are randomly selected and are independent.

The Null Hypothesis is that the population means are the same. The Alternative Hypothesis is that at least one of the
means is different.

H0: µ1 = µ2 =…= µk
H1: The means are not all equal
Reject H0 if F > F,k-1,n-k

12-9
 One-Way ANOVA
⚫ Sum of Square Treatments=SST=Treatment Variation
– METHOD: Sum of Squared Differences between Each Treatment
Mean and Grand Mean

⚫ Sum of Square Error=SSE=Random Variation

– METHOD: Sum of Squared Differences Between Each
Observation and Treatment Mean

⚫ Total Variation=SS Total

– SS Total=SST+SSE
– METHOD: Sum of Squared Differences Between Each
Observation and Overall (GRAND) Mean
12-10
 FORMULAS: One-Way ANOVA

⚫ Where “k”= number of treatments

⚫ Where “n” is the total number of observations

12-11
 One-Way ANOVA:
F-calculated and F-critical

12-12
 Comparing Means of Two or More
Populations – Example

EXAMPLE Step 1: State the null and alternate hypotheses.

Recently a group of four major carriers joined in
hiring Brunner Marketing Research, Inc., to H0: µE = µA = µT = µO
survey recent passengers regarding their H1: The means are not all equal
level of satisfaction with a recent flight. The Reject H0 if F > F,k-1,n-k
survey included questions on ticketing,
boarding, in-flight service, baggage handling, Step 2: State the level of significance.
pilot communication, and so forth. The .01 significance level is stated in the problem.

Twenty-five questions offered a range of Step 3: Find the appropriate test statistic.
possible answers: excellent, good, fair, or Use the F statistic
poor. A response of excellent was given a
score of 4, good a 3, fair a 2, and poor a 1. Step 4: State the decision rule.
These responses were then totaled, so the Reject H0 if: F > F,k-1,n-k
total score was an indication of the
satisfaction with the flight. Brunner F > F.01,4-1,22-4
Marketing Research, Inc., randomly selected F > F.01,3,18
and surveyed passengers from the four
airlines. F > 5.09

Is there a difference in the mean satisfaction level

among the four airlines?
Use the .01 significance level.

The computed value of F is 8.99, which is greater than the critical

value of 5.09, so the null hypothesis is rejected.
Conclusion: The mean scores are not the same for the four
airlines; at this point we can only conclude there is a difference in
the treatment means. We cannot determine which treatment
groups differ or how many treatment groups differ.
12-13
 ANOVA TEST
⚫ Sum of Square Treatments=SST=Treatment Variation
– METHOD: Sum of Squared Differences between Each Treatment
Mean and Grand Mean

⚫ Sum of Square Error=SSE=Random Variation

– METHOD: Sum of Squared Differences Between Each
Observation and Treatment Mean

⚫ SST+SSE=Total Variation
– METHOD: Sum of Squared Differences Between Each
Observation and Overall (GRAND) Mean

12-14
 FORMULAS

⚫ Where “k”= number of treatments

⚫ Where “n” is the total number of observations

12-15
 Treatment Means and Grand Mean

Northern WTA Poconco Branson

94 75 70 68
90 68 73 70
85 77 76 72
80 83 78 65
88 80 74
68 65
65
TOTAL
Sum of all
Column Total 349 391 510 414 Values 1664
Total
n= 4 5 7 6 Observations= 22
Treatment
Means 87.25 78.2 72.8571 69 GRAND MEAN=75.636
12-17
 Sum of Squares Total (SS TOTAL)

WORKING for Northern

Column ONLY Northern WTA Poconco Branson
=(94-75.636)^2= 337.22 0.40 31.77 58.31
=(90-75.636)^2= 206.31 58.31 6.95 31.77
=(85-75.636)^2= 87.68 1.86 0.13 13.22 75.636 is the GRAND
=(80-75.636)^2= 19.04 54.22 5.59 113.13 MEAN
152.86 19.04 2.68
58.31 113.13
113.13
SS TOTAL
Column Total 650.26 267.66 234.93 332.25 1485.09

12-18
 Sum of Squares Treatment (SST)
Northern WTA Poconco Branson
94 75 70 68
90 68 73 70
85 77 76 72
80 83 78 65
88 80 74
68 65
65

Column Total 349 391 510 414

n= 4 5 7 6
Treatment Means 87.25 78.2 72.85 69 75.636 is the GRAND
MEAN

Northern WTA Poconco Branson SST

539.51 32.86 54.07 264.25 890.7
WORKING
For Values
12-19 given ABOVE =4*(87.25-75.636)^2 =5*(78.2-75.636)^2 =7*(72.857-75.636)^2 =6*(69-75.36)^2
 Sum of Square Errors (SSE)
Northern WTA Poconco Branson
94 75 70 68
90 68 73 70
85 77 76 72
80 83 78 65
88 80 74
68 65
65
Column Total 349 391 510 414
n= 4 5 7 6
Treatment Means 87.25 78.2 72.85714 69
WORKING for Northern
Column Only Northern WTA Poconco Branson
=(94-87.25)^2 45.5625 10.24 8.16 1
=(90-87.25)^2 7.5625 104.04 0.02 1
=(85-87.25)^2 5.0625 1.44 9.88 9
=(80-87.25)^2 52.5625 23.04 26.45 16
96.04 51.02 25
23.59 16
61.73 SSE
12-20
110.75 234.8 180.9 68 594.407
 ANOVA TABLE

k=number of treatments=4, n=number of observations=22

ANOVA

Source of Working for F-

Variation SS df Working MS calculated F-calculated
SST=
Treatments 890.68 k-1= 3.00 =890.68/3 296.89 =296.894/33.02 8.99

Error 594.41 n-k= 18.00 =594.071/18 33.02

SS Total=
Total 1485.09 21.00

12-21
 Comparing Means of Two or More
Populations – Example

EXAMPLE Step 1: State the null and alternate hypotheses.

Recently a group of four major carriers joined in
hiring Brunner Marketing Research, Inc., to H0: µE = µA = µT = µO
survey recent passengers regarding their H1: The means are not all equal
level of satisfaction with a recent flight. The Reject H0 if F > F,k-1,n-k
survey included questions on ticketing,
boarding, in-flight service, baggage handling, Step 2: State the level of significance.
pilot communication, and so forth. The .01 significance level is stated in the problem.

Twenty-five questions offered a range of Step 3: Find the appropriate test statistic.
possible answers: excellent, good, fair, or Use the F statistic
poor. A response of excellent was given a
score of 4, good a 3, fair a 2, and poor a 1. Step 4: State the decision rule.
These responses were then totaled, so the Reject H0 if: F > F,k-1,n-k
total score was an indication of the
satisfaction with the flight. Brunner F > F.01,4-1,22-4
Marketing Research, Inc., randomly selected F > F.01,3,18
and surveyed passengers from the four
airlines. F > 5.09

Is there a difference in the mean satisfaction level

among the four airlines?
Use the .01 significance level.

The computed value of F is 8.99, which is greater than the critical

value of 5.09, so the null hypothesis is rejected.
Conclusion: The mean scores are not the same for the four
airlines; at this point we can only conclude there is a difference in
the treatment means. We cannot determine which treatment
groups differ or how many treatment groups differ.
12-22
12-23
12-24
12-25
12-26
12-27
12-28
12-29
12-30
12-31
12-32
12-33
12-34
12-35
 SST (also called as SS between) and
MST

0mg 50 mg 100 mg
9 7 4
8 6 3
7 6 2
8 7 3
8 8 4
9 7 3
8 6 2
TOTAL
Total 57 47 21 125
Observations 7 7 7 21
Treatment Grand
Means 8.14 6.71 3.00 Mean 5.95
df MST
33.59 4.06 61.02 SST= 98.67 2 49.33
WORKING =7*(8.14-5.95)^2 =7*(6.71-5.95)^2 =7*(3-5.95)^2

12-36
 SS Total, SSE and MSE

5.95 is the GRAND

MEAN
WORKING 0mg 50 mg 100 mg
=(9-5.95)^2 9.29 1.10 3.81
=(8-5.95)^2 4.19 0.00 8.72
=(7-5.95)^2 1.10 0.00 15.62
=(8-5.95)^2 4.19 1.10 8.72
=(8-5.95)^2 4.19 4.19 3.81
=(9-5.95)^2 9.29 1.10 8.72
=(8-5.95)^2 4.19 0.00 15.62
df SS Total
36.44 7.49 65.02 SS Total= 108.95 20 108.95

df MSE
SSE=108.95-
SSE=SSTotal-SST 98.67 10.29 18 0.57
12-37
12-38
12-39
 ANOVA TABLE

ANOVA
Source of
Variation SS df MS F-cal F crit

Treatment 98.67 2 49.33 86.33 3.55

Error 10.29 18 0.57

Total 108.95 20

12-40
12-41
12-42
 Inferences About Pairs of Treatment
Means

⚫ From the working done in the ANOVA table, we can

conclude whether all the means are different or not.

⚫ Now, if the treatment means are different, THEN we want

to know which particular treatment means differ.

⚫ For this purpose, we construct a confidence interval for

the difference of two treatment means.

⚫ For this purpose, we use the t-distribution and the

MSE from the ANOVA table.
12-43
 Confidence Interval for The Difference
in Treatment Means

12-44
 Interpretation of Confidence Interval

⚫ If confidence interval includes “0”, then there

is no difference between treatment means.

⚫ If endpoints have the same sign, then

treatment means differ.

12-45
 PROCESS

⚫ INESTIGATION is a step-by-step process.

⚫ First, conduct ANOVA test.

⚫ If null hypothesis is rejected then we can say

that treatment means are not equal to each
other

⚫ After that, we can conduct an analysis of

12-46 individual treatment means
 EXAMPLE:Sum of Squares Treatment
(SST)

Northern WTA Poconco Branson

94 75 70 68
90 68 73 70
85 77 76 72
80 83 78 65
88 80 74
68 65
65

Column Total 349 391 510 414

n= 4 5 7 6
Treatment Means 87.25 78.2 72.85 69 75.636 is the GRAND
MEAN

Northern WTA Poconco Branson SST

539.51 32.86 54.07 264.25 890.7
WORKING For
Values given
12-47 ABOVE =4*(87.25-75.636)^2 =5*(78.2-75.636)^2 =7*(72.857-75.636)^2 =6*(69-75.36)^2
 Sum of Squares Total (SS TOTAL)

WORKING for Northern

Column ONLY Northern WTA Poconco Branson
=(94-75.636)^2= 337.22 0.40 31.77 58.31
=(90-75.636)^2= 206.31 58.31 6.95 31.77
=(85-75.636)^2= 87.68 1.86 0.13 13.22 75.636 is the GRAND
=(80-75.636)^2= 19.04 54.22 5.59 113.13 MEAN
152.86 19.04 2.68
58.31 113.13
113.13
SS TOTAL
Column Total 650.26 267.66 234.93 332.25 1485.09

12-48
 ANOVA TABLE

k=number of treatments=4, n=number of observations=22

ANOVA

Source of Working for F-

Variation SS df Working MS calculated F-calculated

Treatments 890.68 k-1= 3.00 =890.68/3 296.89 =296.894/33.02 8.99

Error 594.41 n-k= 18.00 =594.071/18 33.02

SS Total=
Total 1485.09 21.00

12-49
 Comparison of Mean Scores of Northern
and Branson

⚫ Null hypothesis that treatment means are the same is

rejected.
⚫ Construct a 95% Confidence Interval to See whether
mean scores of Northern and Branson are the same

⚫ t=2.101 because df=n-k=22-4=18 and confidence

interval is 95%
⚫ MSE is 33.0 from ANOVA table with

12-50
 Interpretation

⚫ Confidence Interval ranges from 10.46 up to

26.04
⚫ (10.46,26.04)
⚫ Both endpoints are of the same sign i.e.
positive
⚫ Hence, we can conclude that the treatment
means differ.
⚫ The mean scores of Northern Airlines and
Branson Airlines is different.
12-51
12-52
 One-Way vs. Two-Way ANOVA

⚫ Previously, in ONE-WAY ANOVA:

⚫ 2 Sources of Variation
– Variation Between Treatments.
– Variation Within Treatments

⚫ Other Sources of Variation Are Also Possible

– Blocking Variable: A second treatment
variable that when included in the ANOVA
analysis reduces the SSE term.

12-53
 Two Sets of Hypothesis

12-54
 Two-Way Analysis of Variance

⚫ For the two-factor ANOVA we test whether

there is a significant difference between
the treatment effect and whether there
is a difference in the blocking effect.

⚫ Let Br be the block totals (r for rows)

⚫ Let SSB represent the sum of squares

for the blocks where:

SSB = k( x b − x G ) 2

12-55
 Sum of Squares Error (SSE) for 2 Way ANOVA

12-56
 2 Way ANOVA TABLE

Source of Variation SS df MS F

Treatments SST k-1 SST/(k-1) MST/MSE

Blocks SSB b-1 SSB/(b-1) MSB/MSE

Error SSE (k-1)*(b-1) SSE/((k-1)(b-1))

Total SSTotal

12-57
 EXAMPLE
⚫ EXAMPLE:
WARTA, the Warren Area Regional Transit Authority, is expanding bus service from the suburb of
Starbrick into the central business district of Warren. There are four routes being considered from
Starbrick to downtown Warren: (1) via U.S. 6, (2) via the West End, (3) via the Hickory Street
Bridge, and (4) via Route 59. WARTA conducted several tests to determine whether there was a
difference in the mean travel times along the four routes. Because there will be many different drivers,
the test was set up so each driver drove along each of the four routes. Next slide shows the travel
time, in minutes, for each driver-route combination.

⚫ At the .05 significance level, is there a difference in the mean travel time along the four routes?

⚫ If we remove the effect of the drivers, is there a difference in the mean travel time?

12-58
 Find Grand Mean, Treatment Means and
Block Means

West Hickory SUM Observations per Block

Driver US6 End St. Rte. 59 BLOCKS Block Mean

Deans 18 17 21 22 78 4 19.5

Snarverly 16 23 23 22 84 4 21

Ormson 21 21 26 22 90 4 22.5

Zollaco 23 22 29 25 99 4 24.75

Filbeck 25 24 28 28 105 4 26.25

SUM Treatments 103 107 127 119 456

Observations Per
Treatment 5 5 5 5 20

Treatment Means 20.6 21.4 25.4 23.8

GRAND MEAN= 22.8

12-59
 SS TOTAL

22.8 is the GRAND

MEAN

West Hickory
WORKING US6 End St. Rte. 59

=(18-22.8)^2 23.04 33.64 3.24 0.64

=(16-22.8)^2 46.24 0.04 0.04 0.64

=(21-22.8)^2 3.24 3.24 10.24 0.64

=(23-22.8)^2 0.04 0.64 38.44 4.84

=(25-22.8)^2 4.84 1.44 27.04 27.04 SS TOTAL

Column Total 77.4 39 79 33.8 229.2

12-60
 SST and SSB
West Hickory SUM Observations Block
Driver US6 End St. Rte. 59 BLOCKS per Block Mean
Deans 18 17 21 22 78 4 19.5
Snarverly 16 23 23 22 84 4 21
Ormson 21 21 26 22 90 4 22.5
Zollaco 23 22 29 25 99 4 24.75
Filbeck 25 24 28 28 105 4 26.25
SUM
Treatments 103 107 127 119 456
Observations
Per Treatment 5 5 5 5 20
Treatment
Means 20.6 21.4 25.4 23.8

GRAND
MEAN= 22.8
SST
SST= 24.2 9.8 33.8 5 72.8
SSB=
SSB= 43.56 12.96 0.36 15.21 47.61 119.7
12-61
 2 WAY- ANOVA TABLE

Source of
Variation SS df MS F P-value F crit
Treatments
(Routes) 72.8 3 24.3 7.9 0.0035 3.5
Blocks( Drivers) 119.7 4 29.9 9.8 0.0009 3.3
Error 36.7 12 3.1
Total 229.2 19

12-62
 Two-Way Analysis of Variance - Example

Step 1: State the null and alternate hypotheses.

H0: µu = µw = µh = µr
H1: Not all treatment means are the same
Reject H0 if F > F,k-1,n-k

Step 2: State the level of significance.

The .05 significance level is stated in the
problem.

Step 3: Find the appropriate test statistic.

Because we are comparing means of more than
two groups, use the F statistic

Step 4: State the decision rule.

Reject H0 if F > F,v1,v2
F > F.05,k-1,(b-1)(k-1)
F > F.05,5-1,(5-1)(4-1)
F > F.05,4,12
F > 3.26
Using Excel to perform the calculations,
we conclude:
(1) The mean time is not the same for all
drivers
(2) The mean times for the routes are not
12-63
all the same
 2 Way ANOVA with Interaction

⚫ Interaction
– The effect of one factor on a response variable
differs depending on the value of another factor.

⚫ Interaction Effect of Diet and Exercise on

Weight.
– General recommended method to control weight
is based on the combined or interaction effect of
diet and exercise.

12-64
 Hypothesis
– Ho: The means of Factor A (treatments or column means) are all
equal
– H1: The means of Factor A(treatments or column means) are not
the same

– Ho: The means of Factor B (blocks or row means) are all equal.
– H1: The means of Factor B(blocks or row means) are not the
same.

– Ho: There is no interaction between Factor A (treatment) and

Factor B (block)
– H1: There is interaction between Factor A (treatment) and Factor B
12-65
(block)
 2 Way ANOVA TABLE with
Interaction
k=number of treatments, b=number of blocks,
n=total number of observations

Source of Variation SS df MS F

Treatments (Columns) SST k-1 MST=SST/(k-1) MST/MSE

Blocks (Rows) SSB b-1 MSB=SSB/(b-1) MSB/MSE

Interaction SSI (k-1)*(b-1) MSI=SSI/((k-1)*(b-1)) MSI/MSE

Error SSE n-(k*b) MSE= SSE/(n-k*b)

12-66 Total SSTotal n-1

 Mean Time of Each Driver Across All
Routes

TABLE 1 TABLE 2
Hickory Route
MEAN TIME OF EACH DRIVER ACROSS ALL THE
US 6 West End St 59
ROUTES
Deans 18 14 20 19
15 17 21 22
21 20 22 25 US 6 West End Hickory St Route 59
Snaverly 19 20 24 24
Deans 18 17 21 22
15 24 23 22
14 25 22 20 Snaverly 16 23 23 22
Ormson 19 23 25 23
Ormson 21 21 26 22
21 21 29 23
23 19 24 20 Zollaco 23 22 29 25
Zollaco 24 20 30 26
20 24 28 25 Filbeck 25 24 28 28
25 22 29 24
Filbeck 27 24 28 28 18 = (18 + 15+ 21)/3
25 24 28 30
16 = (19+15+14)/3
23 24 28 26
12-67
 Interaction Effect

⚫ Is it possible one of the drivers is especially

good at driving one or more of the routes?

⚫ To measure interaction effects, we need to

have at least two observations in each cell.

12-68
 What could suggest Interaction between
Factors?

⚫ Use Interaction Plots, to see if there is any

interaction

⚫ If line segments are not parallel and cross

this suggests some interactions between the
factors.

12-69
 Interaction Plot for Travel Time
from TABLE 2 on previous slide

12-70
 Hypothesis: 2 Way ANOVA with Interaction

12-71
 2 Way ANOVA With Interaction:
SUMMARY TABLES

TABLE 3
US 6 West End Hickory St Route 59 MEAN
Deans 18 14 20 19 Deans 19.5
15 17 21 22
21 20 22 25
Snaverly 19 20 24 24 Snaverly 21
15 24 23 22
14 25 22 20
Ormson 19 23 25 23 Ormson 22.5
21 21 29 23
23 19 24 20
Zollaco 24 20 30 26 Zollaco 24.75
20 24 28 25
25 22 29 24
Filbeck 27 24 28 28 Filbeck 26.25
25 24 28 30
23 24 28 26
Mean All Routes 20.6 21.4 25.4 23.8
GRAND
12-72
MEAN= 22.8
 SST Working (Factor A)

TABLE 3
US 6
West
End
Hickory Route
St 59 MEAN Working for SST
Deans 18 14 20 19 Deans 19.5
15 17 21 22
21 20 22 25
US 6 West End Hickory St Route 59
observation per
Snaverly 19 20 24 24 Snaverly 21 treatment 15 15 15 15
15 24 23 22 mean of each
treatment 20.6 21.4 25.4 23.8
14 25 22 20
Ormson 19 23 25 23 Ormson 22.5
21 21 29 23 Working =15*(20.6-22.8)^2 =15*(21.4-22.8)^2=15*(25.4-22.8)^2 =15*(23.8-22.8)^2 SST
SS Column
23 19 24 20 (Treatment) 72.6 29.4 101.4 15 218.4
Zollaco 24 20 30 26 Zollaco 24.75
20 24 28 25
25 22 29 24
Filbeck 27 24 28 28 Filbeck 26.25
25 24 28 30
23 24 28 26
Mean All
Routes 20.6 21.4 25.4 23.8

12-73 GRAND
MEAN= 22.8
 SSB WORKING (Factor B)

TABLE 3 Working for SSB

Route
US 6 West End Hickory St 59 MEAN
Deans 18 14 20 19 Deans 19.5 Deans Snaverly Ormson Zollaco Filbeck
15 17 21 22
21 20 22 25
Observations
Snaverly 19 20 24 24 Snaverly 21 per block 12 12 12 12 12
15 24 23 22
mean of each
14 25 22 20 block 19.5 21 22.5 24.75 26.25
Ormson 19 23 25 23 Ormson 22.5
21 21 29 23 =12*
23 19 24 20 =12* =12* =12* (24.75- =12*(26.25-
WORKING (19.5-22.8)^2 (21-22.8)^2 (22.5-22.8)^2 22.8)^2 22.8)^2 SSB
Zollaco 24 20 30 26 Zollaco 24.75
20 24 28 25
25 22 29 24 SS ROW
Filbeck 27 24 28 28 Filbeck 26.25 (BLOCK) 130.68 38.88 1.08 45.63 142.83 359.1
25 24 28 30
23 24 28 26
Mean All
Routes 20.6 21.4 25.4 23.8
GRAND
MEAN= 22.8
12-74
 SSTotal Working

TABLE 4: SSTotal Working ∑ (Xi-Xg)^2

SSTOTAL WORKING US 6 West End Hickory St Route 59
Deans 23.04 77.44 7.84 14.44
60.84 33.64 3.24 0.64
3.24 7.84 0.64 4.84
Snaverly 14.44 7.84 1.44 1.44
60.84 1.44 0.04 0.64 SSTOTAL
77.44 4.84 0.64 7.84
Ormson 14.44 0.04 4.84 0.04
859.6
3.24 3.24 38.44 0.04
0.04 14.44 1.44 7.84
Zollaco 1.44 7.84 51.84 10.24
7.84 1.44 27.04 4.84
4.84 0.64 38.44 1.44
Filbeck 17.64 1.44 27.04 27.04
4.84 1.44 27.04 51.84
12-75 0.04 1.44 27.04 10.24
 Mean Time of Each Driver Across All
Routes

TABLE 1 TABLE 2
Hickory Route
MEAN TIME OF EACH DRIVER ACROSS ALL THE
US 6 West End St 59
ROUTES
Deans 18 14 20 19
15 17 21 22
21 20 22 25 US 6 West End Hickory St Route 59
Snaverly 19 20 24 24
Deans 18 17 21 22
15 24 23 22
14 25 22 20 Snaverly 16 23 23 22
Ormson 19 23 25 23
Ormson 21 21 26 22
21 21 29 23
23 19 24 20 Zollaco 23 22 29 25
Zollaco 24 20 30 26
20 24 28 25 Filbeck 25 24 28 28
25 22 29 24
Filbeck 27 24 28 28 18 = (18 + 15+ 21)/3
25 24 28 30
16 = (19+15+14)/3
23 24 28 26
12-76
 SSI Working
TABLE 5 :SSI Working 0.49 =(18-20.6-19.5+22.8)^2
West Hickory Route SSI= 3* (0.49+1.21+1.21+……..+0.5625)
US 6 End St 59
where 3 is the number of observations per cell
Deans 0.49 1.21 1.21 2.25 West Route
Snaverly 7.84 11.56 0.36 0 US 6 End Hickory St 59 MEAN
Ormson 0.49 0.01 0.81 2.25 Deans 18 14 20 19 Deans 19.5
15 17 21 22
Zollaco 0.2025 1.8225 2.7225 0.5625 21 20 22 25
Snaverly 19 20 24 24 Snaverly 21
Filbeck 0.9025 0.7225 0.7225 0.5625 15 24 23 22
14 25 22 20
Ormson 19 23 25 23 Ormson 22.5
Mean of Each Driver Across All The Routes 21 21 29 23
Route 23 19 24 20
US 6 West End Hickory St 59 Zollaco 24 20 30 26 Zollaco 24.75
20 24 28 25
Deans 18 17 21 22 25 22 29 24
Snaverly 16 23 23 22 Filbeck 27 24 28 28 Filbeck 26.25
25 24 28 30
Ormson 21 21 26 22 23 24 28 26
Zollaco 23 22 29 25 Mean All
Filbeck 25 24 28 28 Routes 20.6 21.4 25.4 23.8
GRAND
12-78
MEAN= 22.8
 2 Way ANOVA TABLE with
Interaction
k=number of treatments, b=number of blocks,
n=total number of observations, SSE=SSTotal-SST-SSB-SSI

Source of Variation SS df MS F

Treatments (Columns) SST k-1 MST=SST/(k-1) MST/MSE

Blocks (Rows) SSB b-1 MSB=SSB/(b-1) MSB/MSE

Interaction SSI (k-1)*(b-1) MSI=SSI/((k-1)*(b-1)) MSI/MSE

Error SSE n-(k*b) MSE= SSE/(n-k*b)

12-79 Total SSTotal n-1

 2 Way ANOVA TABLE with
Interaction

MS
SS df (Working) df (Working) MS F (Working) F
Treatment (Route) 218.4 k-1=4-1=3 3 218.4/3 72.8 72.8/4.3 16.93
Blocks (Driver) 359.1 b-1=5-1=4 4 359.1/4 89.775 89.775/4.3 20.88
Interaction 110.1 (k-1)*(b-1)=3*4=12 12 110.1/12 9.175 9.175/4.3 2.13
Error (Within) 172 n-(b*k)=60-(5*4)=40 40 172/40 4.3
Total 859.6

12-80
 Source of
Variation SS df MS F P-value F crit
Treatments 218.4 3 72.8 16.93 0.00 2.84
Blocks 359.1 4 89.775 20.88 0.00 2.61
Interaction 110.1 12 9.175 2.13 0.04 2.00
Within 172 40 4.3

Total 859.6 59

12-81
 Conclusion

⚫ The combination of route and driver has a

significant effect on the response variable of
travel time.

12-82