University of Nueva Caceres

School of Graduate Studies

Naga City

MAED 113: INFERENTIAL

STATISTICS ANALYSIS OF
VARIANCE (ANOVA)

Romeo C. Layones, M.A.T.

POSSIBILITIES WHEN TWO GROUPS ARE
COMPARED
 Case I: Case II:
X Y X Y
5 5 5 7
5 5 5 7
5 5 5 7
between groups
No variance
between groups No variance within
groups
No variance within
groups RLayones

Variance exists
POSSIBILITIES WHEN TWO GROUPS ARE
COMPARED
 Case III: Case IV:
X Y X Y
5 5 5 4
6 6 6 7
7 7 7 9
between groups
No variance between
groups Variance exists
within groups
Variance exists within
groups RLayones

Variance exists
COMPARING TWO OR MORE GROUPS
GROUP 1 GROUP 2 GROUP 3 GROUP 4

In comparing 4 groups, 6 t-tests are needed.

Use ANOVA instead!

RLayones
IMPORTANT NOTES ON ANOVA
Description:

Analysis of Variance (ANOVA) describes a

technique whereby the total variation is being
analyzed or divided into meaningful
components.

It is a technique designed to test whether or not

two or more samples (groups) are significantly
different from each other.

RLayones
IMPORTANT NOTES ON ANOVA
Assumptions:

The individuals in the various subgroups should

be selected on the basis of random sampling.

The populations from which the samples are

taken should be normally distributed.

The variance of the subgroups should be

homogeneous.

RLayones
IMPORTANT NOTES ON ANOVA
Basic Types:

➢One-Way ANOVA (Independent

Groups) ➢One-Way ANOVA (Dependent

Groups) ➢Two-Way ANOVA

(Independent Groups)

➢Two-Way ANOVA with Interaction

(Independent Groups)
 RLayones
Decision Model: One-Way ANOVA (Independent Groups)

(1) Ho: There is no significant difference in the means

of the given groups.
H1: There is a significant difference in the means of
the given groups.
(2) Statistical Test: Analysis of Variance
(ANOVA) (3) Significance level: Either α = 0.05
or α = 0.01 (4) Sampling Distribution: F –
distribution
(5) Region of rejection: F ≥ Fα(v1,v2)
where v1 = k – 1 , k is the no. of columns
v2 = N – k , N is the total number of items
 RLayones
DECISION MODEL:
(6) Computation:
a) Compute the sum of squares
a.1) Total Sum of Squares (SST)
=
2 ∑ ∑−
()
SST
X
2

X
N
 where N = total no. of observations
a.2) Between Column Sum of Squares
or Treatment Sum of Squares (SSC)
∑ =
2
2
∑ ∑−
()()
X
j
X
SSC
n
j
N
where n = no. of items in each group
a.3) Within Column Sum of Squares
or Error Sum of Squares (SSE)
SSE = SST – SSC
 RLayones
DECISION MODEL:
(6) Computation:
b) Compute the degrees of freedom (df) for the
sources of variation.
b.1) dfT = N – 1
b.2) dfC = k – 1
b.3) dfE = N – k
c) Form the ANOVA table.
Source Sum of Degrees Mean Compute
of Squares of Squar dF
Variation SSC Freedom e F-value
Column dfc MSS
C
 Error SSE dfe MSSE
Total SST dft

MSSC
F=
d) Calculate F. MSSE
RLayones
DECISION MODEL:

(7) Decision:
➢ If the computed F value is greater than the tabular F
value, Hois rejected. Otherwise, Hois accepted. (8)
Conclusion:
 RLayones
USING THE F-DISTRIBUTION TABLE:
➢ Find the intersection of the degrees of freedom (df) for
the numerator (v1), located at the first row, and the degrees
of freedom for the denominator (v2), located at the first
column. Use the first table if the significance level is 0.05,
and the second table if it the significance level is 0.01.

Example 1:
1) What is the critical region of rejection if the v1 = 2 and
v2 = 6 and the significance level is α = 0.05?
 ANSWER: F ≥ 5.14RLayones
USING THE F-DISTRIBUTION TABLE:

Example 2:
1) What is the critical region of rejection if the v1 = 2 and
v2 = 6 and the significance level is α = 0.01?
ANSWER: F ≥ 10.92

RLayones
Example 1: Six students were randomly selected from
each of three sections in grade 11. They were all given
the same test in mathematics. The following table lists
the scores of these students in that test.
Section A Section B Section C
89 82 81
92 87 73
84 76 64
86 77 79
91 82 81
 95 85 78

Using the 5% significance level, test the hypothesis that

there is no significant difference in the mean scores in
mathematics of the students from the three sections.
RLayones
Solution:
(1)Ho: There is no significant difference in the mean
scores in mathematics of the students from the three
sections.
H1: There is a significant difference in the mean
scores in mathematics of the students from the three
sections.
(2) Test:
Statistical F-test
(3) Significance Level:
α = 0.05
(4) Sampling F-distribution F ≥
Distribution: (5) 3.68
Region of rejection:
Find v1: v1 = k – 1 = 2
=3–1
Find v2: – 3 = 15 2
v2 = N – k = 18
15 3.68

RLayones
Solution:
(6) Computation:
X1 X2 X3 X12 X22 X32
89 82 81 7921 6724 6561
 92 87 73 8464 7569 5329
84 76 64 7056 5776 4096
86 77 79 7396 5929 6241
91 82 81 8281 6724 6561
95 85 78 9025 7225 6084
ΣX1=537 ΣX2=489 ΣX3=456

2
ΣX=1482 ΣX =122962
89.5 x1 = 81.5 x2 = 76 x3 =
Solution:
a) Find the sum of
squares: 2 ∑
 = −
∑
()
2
a.1) SST
XX
N

()
1482
1229622
=−
18
= 122962 −122018
 ∑ =
SST = 944
()() ∑ ∑ −X
2
2

a.2) SSC j X
n
N
( ) ( ) ( ) ( ) 537 2 2 2 2 489
456
1482
=++−6
6
6
18
= 48061.5 + 39853.5 + 34656
 −122018SSC = 553
a.3) 944 − 553
SSE = SST − SSC =
SSE = 391 RLayones
Solution:
b) Compute the degrees of freedom:
b.1) dfT = N – = 18 – 1
1 = 17
b.2) dfC = = 3 – 1 = 2
k–1
b.3) dfE = N – = 15
k
= 18 – 3

c) Form the ANOVA table:

Source Sum of Degrees Mean Compute
of Squares of Squar dF
Variation Freedom
553 e 10.61
Column 2
276.
5

Error 391 15 26.07

Total 944 17

MSSC
F =26.07
276.5
= = 10.61
MSSE
RLayones
Solution:
7) Decision: Since the computed F-value (10.61) is greater
than the tabular F-value(3.68), Hois rejected.

8) Conclusion: There is a significant difference in the mean

scores in mathematics of the grade 11 students from the
three sections.

RLayones
ONE-WAY ANOVA USING MS EXCEL
 1. Encode data on MS Excel.

RLayones
ONE-WAY ANOVA USING MS EXCEL
 2. Click Data > Data Analysis.
RLayones
ONE-WAY ANOVA USING MS EXCEL
3. The Data Analysis box will appear. Choose
 Anova: Single Factor. Click OK.

RLayones
ONE-WAY ANOVA USING MS EXCEL
4. The Anova: Single Factor box will appear. In "Input
Range" highlight the entire range of data. Specify your
critical level . Click on OK.
RLayones
ONE-WAY ANOVA USING MS EXCEL
 5. This output will be displayed.
RLayones
2) A large marketing firm uses many photocopy machines, several of
each of four different models. During the last six months, the officer
manager has tabulated for each machine the average number of
minutes per week that it is out of service due to repairs, resulting in
the following data:
G H K L
56 74 55 78
61 77 36 72
68 82 59 89
72 63 56 62
72 63 56 62
82 84 74 61
70 58
88

Test at the 0.01 level of significance the hypothesis that there is no

significant difference in the mean number of minutes per week the four
machine models are out of service.RLayones
3) The following data represent the number of packages of 5 popular
brands of cigarettes sold by a supermarket on 7 randomly selected
days:
A B C D E
21 35 45 32 45
35 12 60 53 29
32 27 33 29 31
28 41 36 42 22
14 19 31 40 36
47 23 40 23 29
25 31 43 35 42

Perform an analysis of variance, at the 0.05 level of significance,

and determine whether or not the 5 brands sell, on the average, the
same number of cigarettes at this supermarket.
RLayones