COTABATO STATE UNIVERSITY

GRADUATE SCHOOL
Sinsuat Avenue, Cotabato City 9600

Ed 702

Statistical Modelling and Analysis for Decision-Making

Tarhata Piang Solaiman-Guiamalon, Ph.D., Sh. C.

Professor

Omar M. Bulando
Doctor of Philosophy
in Educational Administration
 F-TEST
(ONE-WAY-ANOVA)

What is F-Test?

The F-Test is another parametric test used to compare the means of two

or more groups of independent samples. It is also known as the Analysis

of Variance, (ANOVA).

Three Kinds of Analysis of Variance

1. One-way analysis of variance

2. Two-way analysis of variance

3. Three-way analysis of variance

The F-test is the analysis of variance (ANOVA). This is used in

comparing the means of two or more independent groups.

Two-way ANOVA is used when two variables are involved: the column

and the row variables. The researcher is interested to know if there are

significant difference between and among columns and rows. This is also

used in looking at the interaction effect between the variables being

analyzed.

Why do we use the F-TEST?

Because we want to find out if there is a significant difference

between and among the means of two or more independent groups.

How do we use the F-TEST?

• To get the F-computed value, the following computations should be

done:
 2
(¿)
• Compute the Correction Factor, CF =
n

• TSS is the total sum of squares minus the CF, TSS =∑ X 2 - CF

• BSS is the between sum of squares minus the CF,

2
(∑ X)
BSS = – CF
n

• WSS is the within sum of squares or it is the difference between the

TSS minus the BSS

ANOVA TABLE

Degrees F - Value
Sources of of Sum of Mean
Variance Freedo Squares Squares Computed Tabular
m
See the

Betwee BSS MSB

K-1 BSS F= table
n df MSW
Groups at .05

(N-1)
WSS
Within – (K – WSS
df
Groups
1)

Total N-1 TSS

Example:

A sari-sari store is selling four brands of shampoo. The owner is interested

if there is a significant difference in the average sales for one week. The

following data are recorded:

 Brand

A B C D

7 9 2 4

3 8 3 5

5 8 4 7

6 7 5 8

9 6 6 3

4 9 4 4

3 10 2 5

Perform the analysis of variance and test the hypothesis at .05 level of

significance that the average sales of four brands of shampoo are equal.

SOLVING BY STEPWISE METHOD

I. Problem:

Is there a significant difference in the average sales of the four

brands of shampoo?

II. Hypotheses

H0 : There is no significant difference in the average sales of

the

four brands of shampoo.

H1 : There is a significant difference in the average sales of the

four brands of shampoo.

 III. Level of Significance:

a=.05

dfb = K – 1

= 4 -1 = 3

dfw = (N – 1) – (K – 1)

= (28 – 1) – (4 – 1)

= 27 – 3 = 24

F .05 =

F Table for .05 level of significance

Level of Significance:

a=.05

dfb = K – 1

= 4 -1 = 3

dfw = (N – 1) – (K – 1)

= (28 – 1) – (4 – 1)

= 27 – 3 = 24

F .05 = 3.01
 IV. Statistics:

F-Test One-way Analysis Of Variance

Brand

A B C D

2 2 2 2
X1 X1 X2 X2 X3 X3 X4 X4

7 49 9 81 2 4 4 16

3 9 8 64 3 9 5 25

5 25 8 64 4 16 7 49

6 36 7 49 5 25 8 64

9 81 6 36 6 36 3 9

4 16 9 81 4 16 4 16

3 9 10 100 2 4 5 25

∑ X1 = 37 ∑ X 21 = 225 ∑ X2 = 57 ∑ X 22 = 475 ∑ X3 = 26 ∑ X 23 =110 ∑ X4

= 36 ∑ X 24 = 204

n1 = 7 n2 = 7 n3 = 7

n4 = 7

x̄1 = 5.28 x̄2 = 8.14 x̄3 = 3.71

x̄4 = 5.14

CF = ¿ ¿

CF = ¿ ¿
 CF = ¿ ¿

CF = 869.14

TSS = ∑ X 21+ ∑ X 22+ ∑ X 23+∑ X 24 - CF

2 2 2 2
∑ X 1 =225 ; ∑ X 2 = 475 ; ∑ X 3 = 110 ; ∑ X 4 = 204

CF = 869.14

TSS = 225 + 475 + 110 + 204 - 869.14

TSS = 1014 – 869.14

TSS = 144.86

BSS ¿ ¿ ¿ + ¿ ¿ + ¿ ¿ + ¿ ¿ - CF

∑ X 1 = 37 ; ∑ X 2 = 57 ; ∑ X 3 =26 ; ∑ X 4 = 36

CF = 869.14

BSS = ¿ ¿ + ¿ ¿ + ¿ ¿ + ¿ ¿ - 869.14

BSS = 195.57 + 464.14 + 96.57 + 185.14 -

869.14

BSS = 941.42 – 869.14

BSS = 72.28

Analysis of Variance Table

Source Degrees of Sum of Mean F - Value

of Freedom Squares Squares

Computed Tabular
 Variance
Betwee BSS MSB
= F=
K -1 df MSW
n BSS
4–1 72.28 24.09 3.01
Groups F=
72.28 3 3.02
=3
= 24.09 = 7.98

Within (N-1) – (K – WSS 72.58

=
df 24
Groups 1) WSS
= 3.02
(28-1) – (4-1) 72.58

= 24

N–1

Total 28 – 1 144.86

= 27

IV. Decision Rule:

If the F-computed value is greater than the F-tabular value,

reject

H0.

V. Conclusion:

Since the F-computed value of 7.98 is greater than the F-tabular

value of 3.01 at .05 level of significance with 3 and 24 degrees of

freedom, the null hypothesis is rejected in favor of the research

hypothesis which means that there is a significant difference in the

average sales of the four brands of shampoo.

SCHEFFE’S TEST
 • To find out where the difference lies, another test must be used, the

Scheffe’s test.

• The F-test tell us that there is a significant difference in the average

sales of the four brands of shampoo but as to where the difference

lies, it has to be tested further by another test, the Scheffe’s test.

F’ = ¿ ¿

__________

n 1n2

Where:

F’ = Scheffés test

x̄1 = mean of group 1

x̄2 = mean of group 2

n1 = number samples in group 1

n2 = number samples in group 2

Sw2 = within mean squares

Brand

A B C D

2 2 2 2
X1 X1 X2 X2 X3 X3 X4 X4

7 49 9 81 2 4 4 16
 3 9 8 64 3 9 5 25

5 25 8 64 4 16 7 49

6 36 7 49 5 25 8 64

9 81 6 36 6 36 3 9

4 16 9 81 4 16 4 16

3 9 10 100 2 4 5 25

∑ X1 = 37 ∑ X 21 = 225 ∑ X2 = 57 ∑ X 22 = 475 ∑ X3 = 26 ∑ X 23 =110 ∑ X4

= 36 ∑ X 24 = 204

n1 = 7 n2 = 7 n3 = 7

n4 = 7

x̄1 = 5.28 x̄2 = 8.14 x̄3 = 3.71

x̄4 = 5.14

A vs. B
F´ = ¿ ¿
x̄1= 5.28
__________
x̄2= 8.14
7(7)
n1 = 7
¿¿
n2 = 7
__________
Sw2 = 3.02 49
8.1796
=
0 .86

F´ = 9.51
 A vs. C
x̄1 = 5.28
F´ = ¿ ¿
x̄3 = 3.71
__________
n1 = 7
7(7)
n3 = 7 ¿¿

_____
Sw2 = 3.02
49
2.4649
=
0 .86

F´ = 2.87

A vs. D
x̄1 = 5.28 F´ = ¿ ¿
x̄4 = 5.14 __________
n1 = 7 7(7)

n4 = 7
¿¿
Sw2 = 3.02
_____
49
0.0196
=
0 .86
 B vs. C

x̄2 = 8.14 F´ = ¿ ¿

x̄3 = 3.71 __________

n2 = 7 7(7)
¿¿
n3 = 7
__________
Sw2 = 3.02
49
19.6249
=
0.86

F´ = 22.82

B vs. D
x̄2 = 8.14 F´ = ¿ ¿
x̄4 = 5.14 __________
n2 = 7 7(7)

n4 = 7
¿¿
Sw2 = 3.02
__________
49
9
=
0.86
 C vs. D
F´ = ¿ ¿
x̄3 = 3.71
__________
x̄4 = 5.14
7(7)
n3 = 7
¿¿
n4 = 7
__________
Sw2 = 3.02 49
2.0449
=
0.86

F´ = 2.38

Comparison of the Average Sales of the Four Brands of Shampoo

(F )(K-1)
.05
Between
F´ (3.01) (3) Interpretation
Brand
= 9.03
 A vs. B 9.51 9.03 Significant

Not
A vs. C 2.87 9.03
Significant

Not
A vs. D 0.02 9.03
Significant

B vs. C 22.82 9.03 Significant

B vs D 10.46 9.03 Significant

Not
C vs D 2.38 9.03
Significant

The above table shows that there is a significant difference between

brand A and B, brand B and brand C and also brand B and brand D,

However, brands A and C, A and D and C and D have no significant

differences in their average sales.

This implies that the brand B is more saleable than brands A, C and

D.

Example 2.

The following data represents the operating time in hours of the 3

types of scientific pocket calculators before a recharge is required.

Perform the analysis of variance at .05 level of significance.

Brand

2 2 2
Fx1 X1 Fx2 X2 Fx3 X3

4.9 24.01 6.4 40.96 4.8 23.04

 5.3 28.09 6.8 46.24 5.4 29.16

4.6 21.16 5.6 31.36 6.7 44.89

6.1 37.21 6.5 42.25 7.9 62.41

4.3 18.49 6.3 39.69 6.2 38.44

6.9 47.61 6.7 44.89 5.3 28.09

5.3 28.09 5.9 34.81

4.1 16.81

4.3 18.49

∑ X1 = 32.1 ∑ X 21 = 176.57 ∑ X2 = 52 ∑ X 22 =
308.78

∑ X 3 = 42.2 ∑ X 23 = 260.84

n1 = 6 n2 = 9
n3 = 7

x1 = 5.35 x 2 = 5.78
x3 = 6.03

I. Problem:
 Is there a significant difference in the average operating time

in hours for the 3 types of pocket scientific calculators before a

recharge is required?

II. Hypotheses

H0 : There is no significant difference in the average operating time

in hours for the 3 types of pocket calculators before a

recharge is required.

H1 : There is a significant difference in the average operating

time

in hours for the 3 types of pocket calculators before a

recharge is required.

III. Level of Significance:

df = 2 and 19

a=.05

IV. Statistics:

F-test one-way-analysis of variance

Computation:

CF = ¿ ¿ = ¿ ¿ = 725.08

TSS = ∑ X 21+ ∑ X 22+ ∑ X 23- CF

= 176.57 + 308.78 + 260.84 - 725.08

= 746.19 - 725.08

TSS = 21.11

BSS ¿ ¿ ¿ + ¿ ¿ + ¿ ¿ - CF

= ¿ ¿ + ¿ ¿ + ¿ ¿ - 725.08

= 171.74 + 300.44 + 254.40 - 725.08

 = 726.58 – 725.08

BSS = 1.50

WSS = TSS - BSS

= 21.11 - 1.50

WSS = 19.61

ANOVA TABLE

Degrees F - Value
Sources of of Sum of Mean
Variance Freedo Squares Squares Computed Tabular
m
Betwee 2 1.50 .75 .73 3.53
n
Groups
Within 19 19.61 1.03
Groups
21.1
Total 21
1

V. Decision Rule:

If the F-computed value is greater than the F-tabular value, reject H

VI. Conclusion:

Since the F-computed value of 0.73 is lesser than the F-tabular

value of 3.52 at .05 level of significance, the null hypothesis is accepted.

This means that there is no significant difference in the average operating

time in hours for the 3 types of pocket scientific calculators before a

recharge is required.

Scheffe’s Test
 Brand

2 2 2
Fx1 X1 Fx2 X2 Fx3 X3

4.9 24.01 6.4 40.96 4.8 23.04

5.3 28.09 6.8 46.24 5.4 29.16

4.6 21.16 5.6 31.36 6.7 44.89

6.1 37.21 6.5 42.25 7.9 62.41

4.3 18.49 6.3 39.69 6.2 38.44

6.9 47.61 6.7 44.89 5.3 28.09

5.3 28.09 5.9 34.81

4.1 16.81

4.3 18.49

∑ X1 = 32.1 ∑ X 21 = 176.57 ∑ X2 = 52 ∑ X 22 =
308.78

∑ X 3 = 42.2 ∑ X 23 = 260.84

n1 = 6 n2 = 9
n3 = 7

x1 = 5.35 x 2 = 5.78
x3 = 6.03

x̄1= 5.35 A vs. B

x̄2= 5.78 F´ = ¿ ¿

n1 = 6 __________
 2

A vs. C
x̄1= 5.35
F´ = ¿ ¿
x̄3= 6.03
__________
n1 = 9
6(7)
n2 = 7
¿¿
Sw2 = 1.03 __________
42
0.0625
=
13.39
--------
42
0.0625
=
0.3188
F´ = 0.1960

B vs. C
x̄2= 5.78 F´ = ¿ ¿
x̄3= 6.03 __________
Comparison of the Operating time in hours of the 3 types of

scientific pocket calculators before a recharge is required

(F )(K-1)
.05
Between
F´ (1.03) (2) Interpretation
Brand
= 2.06

Not
A vs. B 0.6462 2.06
Significant

Not
A vs. C 0.1960 2.06
Significant

Not
B vs. C 0.0037 2.06
Significant
 The above table shows that there is no significant difference

between brand A and brand B, brand A and brand C and also brand B and

brand C,

This implies that all the 3 types of scientific pocket calculator have

no significance difference in the operating time in hour before the

recharge is required.