Anova

Testing Equality of Several Means
ONE WAY ANALYSIS OF VARIANCE

For comparing the cleansing action of three
detergents, the following whiteness readings
were recorded on 15 swatches of white cloth,
which were first soiled with India ink and then
washed in an agitator-type machine with the
respective detergents.
The means of three samples are 77, 68 and 80.
Test (at 1% significance level), whether the
differences among them are significant or they
can be attributed to chance.
Detergent A
77
81
71
76
80
Detergent B
72
58
74
66
70
Detergent C
76
85
82
80
77
Basic Calculations
Yij2 = 85041 ;
Grand Total = Yij = 1125
n = 15
Correction Factor (C. F.)
= (Grand Total)2/n
= 84375
Total Sum of Squares = Yij2 C.F. = 666
Detergent S.S. = (Detergent Total)2 C.F.
= (3852 + 3402 + 4002)/5 - C.F.
= 390
Error S.S. = Total S.S. Detergent S.S. =
= 666 390 = 276
ANOVA for comparing the detergents

Source
of
Variation
d.f.
Sum of
Squares
(S.S.)
Mean
S.S.
Detergent
390
195
Error
12
276
23
TOTAL
14
666
F ratio F0.01(2,12)
8.48
6.93
Pairwise Comparisons
Critical Difference for Detergent:
1 1
t /2;12 d.f . * M.S.E. * ( )
n1 n 2
= 3.0545 {23 *(2/5)}
= 9.265
Detergent Means:
(1) 77a,b
(2) 68a
(3) 80b
FIGURES WITH SAME ALPHABET DO NOT

INDICATE SIGNIFICANT DIFFERENCE.
SPSS Output _ One-way ANOVA

ANOVA
Index
Between Groups
Within Groups
Total
Sum of
Squares
390.000
276.000
666.000
df
2
12
14
Mean Square
195.000
23.000
F
8.478
Sig.
.005
SPSS Output _ One-way ANOVA
Post Hoc Tests

Multiple Comparisons
Dependent Variable: detergen
LSD
(I) group
1
2
3
(J) group
2
3
1
3
1
2
Mean
Difference
(I-J)
Std. Error
9.000
3.033
-3.000
3.033
-9.000
3.033
-12.000*
3.033
3.000
3.033
12.000*
3.033
*. The mean difference is significant at the .01 level.
Sig.
.012
.342
.012
.002
.342
.002
99% Confidence Interval

Lower Bound Upper Bound
-.26
18.26
-12.26
6.26
-18.26
.26
-21.26
-2.74
-6.26
12.26
2.74
21.26
Two-way ANOVA with single observation

per cell
An experiment was designed to study the
performance of four different detergents. The
following whiteness readings were obtained
with specially designed equipments for 12 loads
of washing distributed over three different
models of washing machines.
Test (at 1% significance level) whether there
exists differences in the detergents as well as
the washing machines.
Machine 1
Machine 2
Machine 3
Detergent A
45
43
51
Detergent B
47
46
52
Detergent C
48
50
55
Detergent D
42
37
49
ANOVA for Two-way classified data with

single observation per cell
Source of
Variation
d.f.
Sum of
Squares
Mean
Sum of
Squares
F ratio F (1%)
Detergents
111
37.0
11.6
9.78
Machines
135
67.5
21.1
10.9
Error
19
3.2
TOTAL
11
265
Conclusions :
- There are significant differences in the
effectiveness of the four detergents.
- Performance of three models of washing
machines is significantly different.
Pairwise Comparisons - Detergents

Critical Difference for Detergent:
t /2; 6 d.f.
1
1
M.S.E. ( ) ; here, n1 n2 3
n1 n 2
=3.70743 * {3.2 (2/3)}
= 5.415
Detergent Means:
A(46.33)a,b
B(48.33)a
C(51.0)a
D(42.67)b
FIGURES WITH SAME ALPHABET DO NOT INDICATE

SIGNIFICANT DIFFERENCE.
Pairwise Comparisons - Machines

Critical Difference for Machines:
t /2; 6 d.f.
1 1
M.S.E. ( ) ; here n1 n 2 4
n1 n 2
= 3.70743 * {3.2 (2/4)

=4.69
Machine Means:
1.
45.5a
2.
44.0a
3.
51.75
SPSS Output _ Two-way ANOVA

(Without Replication)
1. Detergent
Dependent Variable: Index
Detergent
1
2
3
4
Mean
46.333
48.333
51.000
42.667
Std. Error
1.023
1.023
1.023
1.023

Lower Bound
Upper Bound
42.541
50.126
44.541
52.126
47.208
54.792
38.874
46.459
2. Machine
Machine
1
2
3
Mean
45.500
44.000
51.750
Std. Error
.886
.886
.886

Lower Bound
Upper Bound
42.216
48.784
40.716
47.284
48.466
55.034

(Without Replication)
Tests of Between-Subjects Effects
Source
Corrected Model
Intercept
Detergent
Machine
Error
Total
Corrected Total
Type III Sum

of Squares
246.083a
26602.083
110.917
135.167
18.833
26867.000
264.917
df
5
1
3
2
6
12
11
Mean Square
49.217
26602.083
36.972
67.583
3.139
a. R Squared = .929 (Adjusted R Squared = .870)
F
15.680
8475.000
11.779
21.531
Sig.
.002
.000
.006
.002

(Without Replication)_ Post Hoc Tests
Detergent
LSD
(I) Detergent
1
(J) Detergent
2
3
4
1
3
4
1
2
4
1
2
3
Mean
Difference
(I-J)
-2.00
-4.67
3.67
2.00
-2.67
5.67*
4.67
2.67
8.33*
-3.67
-5.67*
-8.33*
Std. Error
1.447
1.447
1.447
1.447
1.447
1.447
1.447
1.447
1.447
1.447
1.447
1.447
Based on observed means.

Sig.
.216
.018
.044
.216
.115
.008
.018
.115
.001
.044
.008
.001

Lower Bound
Upper Bound
-7.36
3.36
-10.03
.70
-1.70
9.03
-3.36
7.36
-8.03
2.70
.30
11.03
-.70
10.03
-2.70
8.03
2.97
13.70
-9.03
1.70
-11.03
-.30
-13.70
-2.97

(Without Replication)_ Post Hoc Tests
Machine
LSD
(I) Machine
1
2
3
(J) Machine
2
3
1
3
1
2
Mean
Difference
(I-J)
Std. Error
1.50
1.253
-6.25*
1.253
-1.50
1.253
-7.75*
1.253
6.25*
1.253
7.75*
1.253
Based on observed means.

Sig.
.276
.002
.276
.001
.002
.001

Lower Bound Upper Bound
-3.14
6.14
-10.89
-1.61
-6.14
3.14
-12.39
-3.11
1.61
10.89
3.11
12.39

Anova

Uploaded by

Copyright:

Available Formats

Anova

Uploaded by

Document Information

Original Description:

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

Anova

Uploaded by

Copyright:

Available Formats

Testing Equality of Several Means

ONE WAY ANALYSIS OF VARIANCE

ANOVA for comparing the detergents

FIGURES WITH SAME ALPHABET DO NOT

SPSS Output _ One-way ANOVA

SPSS Output _ One-way ANOVA

Post Hoc Tests

*. The mean difference is significant at the .01 level.

99% Confidence Interval

Two-way ANOVA with single observation

ANOVA for Two-way classified data with

Pairwise Comparisons - Detergents

FIGURES WITH SAME ALPHABET DO NOT INDICATE

Pairwise Comparisons - Machines

= 3.70743 * {3.2 (2/4)

SPSS Output _ Two-way ANOVA

99% Confidence Interval

99% Confidence Interval

SPSS Output _ Two-way ANOVA

Type III Sum

a. R Squared = .929 (Adjusted R Squared = .870)

SPSS Output _ Two-way ANOVA

Based on observed means.

99% Confidence Interval

SPSS Output _ Two-way ANOVA

Based on observed means.

99% Confidence Interval

You might also like