62
B. Normality and Homogeneity Test
Before analyzing the data using inferential analysis, normality and homogeneity test must be done. The normality test is to know that the sample
is in normal distribution and the homogeneity test is to know t hat the data are homogeneous.
1. Normality Test The sample is in normal distribution if L
o
L-obtained is lower than L
t
L- table at the level of significance
= 0.05. L stands for Lilliefors. Table 10. The Normality Test
No Data
The Number of
Sample L-
obtained L
o
L-table L
t
Alfa
Distribution of Population
1 A
1
B
1
11 0.114
0.249 0.05
Normal 2
A
1
B
2
11 0.112
0.249 0.05
Normal 3
A
2
B
1
11 0.182
0.249 0.05
Normal 4
A
2
B
2
11 0.090
0.249 0.05
Normal 5
A
1
22 0.163
0.190 0.05
Normal 6
A
2
22 0.080
0.190 0.05
Normal 7
B
1
22 0.126
0.190 0.05
Normal 8
B
2
22 0.080
0.190 0.05
Normal
2. Homogeneity Test Homogeneity test is done to know that the data are homogenous. If
77
o 2
is lower than
t 2
0.05
, it can be concluded that the data are homogeneous.
Table 11. The Homogeneity Test Sample
df 1df
s
i 2
log s
i 2
df log s
i 2
1 2
3 4
10 10
10 10
0.1 0.1
0.1 0.1
16.05 27.56
17.36 15.82
1.206 1.440
1.240 1.199
12.06 14.40
12.40 11.99
40 0.4
50.85
63
2
= 2.3026{B – log S
i
x n-1} = 2.3026 51.33
– 50.85 = 1.12
Based on the result of the calculation above, it can be seen that the
o 2
1.12 is lower than
t 2
at the level of significance 5 = 7.81.
o 2
t 2
1.12 7.81, so the data are homogeneous.
C. Hypothesis Test
Hypothesis test can be done after the results of normality and homogeneity test are fulfilled. The test is done by using multifactor analysis
of variance 2 x 2. H
o
is rejected if F
o
F
t
. It means that there is a significant difference and there is an interaction effect. If H
o
is rejected, the analysis is continued to know which group is better using Tukey test. The multifactor
analysis of variance 2 x 2 and Tukey test are described as the following:
1. Summary of a 2 x 2 M ultifactor Analysis of Variance Table 12. M ultifactor Analysis of Variance
Source of Variance SS
df M S
F
o
F
t 0.05
F
t 0.01
Between columns 209.455
1 209.455
10.91 4.08
7.31 Between rows
704.000 1
704.000 36.67
4.08 7.31
Columns by rows interaction
1443.273 1 1443.273 75.17
4.08 7.31
Between groups 2356.727 3
785.576 -
- -
Within groups 768.000
40 19.200
- -
- Total
3124.727 43 -
- -
-
64
The table shows that: a. Because F
o
between columns 10.91 is higher than F
t
at the level of significance α = 0.05 4.08 and F
t
at the level of significance α = 0.01 7.31, the difference between columns is significant. It can be concluded
that teaching methods differ significantly from one another in their effect on the performance of the subjects in the exp eriment.
b. Because F
o
between rows 36.67 is higher than F
t
at the level of significance α = 0.05 4.08 and F
t
at the level of significance α = 0.01 7.31, the difference between rows is significant. It can be concluded
that students having high self-esteem and those having low self-esteem are significantly different in their reading skill.
c. Because F
o
interaction 75.17 is higher than F
t
at the level of significance α = 0.05 4.08 and F
t
at the level of significance α = 0.01 7.31, there is an interaction effect between teaching methods and the
degree of self-esteem toward students’ reading skill. It means that the
effect of teaching methods on reading skill depends on the degree of self- esteem.
2. Summary of Tukey Test The finding of q is found by dividing the difference between the
means by the square root of the ratio of the within group variation and the sample size.
Table 13. Summary of Tukey Test Between
Groups q
o
q
t 0.05
q
t 0.01
Significance M eaning
A
1
– A
2
4.67 2.95
4.02 Significant
A
1
A
2
A
1
B
1
– A
2
B
1
11.97 3.11
4.39 Significant
A
1
B
1
A
2
B
1
A
2
B
2
– A
1
B
2
5.37 3.11
4.39 Significant
A
2
B
2
A
1
B
2
B
1
– B
2
8.56 2.95
4.02 Significant
B
1
B
2
65
a. Because q
o
between A
1
and A
2
4.67 is higher than q
t
at the level of significance
α = 0.05 2.95 and q
t
at the level of significance α = 0.01 4.02, Teams-Games-Tournament differs significantly from the
lecture method for teaching reading. The mean score of students who are taught by using Teams-Games-Tournament 69.91 is higher than
that of those who are taught by using lecture 65.55, so Teams- Games-Tournament is more effective than the lecture method for
teaching reading. b. Because q
o
between A
1
B
1
and A
2
B
1
11.97 is higher than q
t
at the level of significance α = 0.05 3.11 and q
t
at the level of significance α = 0.01 4.39, Teams-Games-Tournament differs significantly from the
lecture method to teach reading for students having high self-esteem. The mean score of students having high self-esteem who are taught by
using Teams-Games-Tournament 79.64 is higher than that of those who are taught by using lecture 63.82, so Teams-Games-Tournament
is more effective than lecture method to teach reading for students having high self-esteem.
c. Because q
o
between A
1
B
2
and A
2
B
2
5.37 is higher than q
t
at the level of significance α = 0.05 3.11 and q
t
at the level of significance α = 0.01 4.39, lecture method differs significantly from Teams-Games-
Tournament to teach reading for students having low self-esteem. The mean score of students having low self-esteem who are taught by using
lecture 67.27 is higher than that of those who are taught by using Teams-Games-Tournament 60.18, so lecture is more effective than
Teams-Games-Tournament to teach reading for students having low self-esteem.
d. Because q
o
between B
1
and B
2
8.56 is higher than q
t
at the level of significance
= 0.05 2.95 and q
t
at the level of significance =
0.01 4.02, students having high self-esteem differ significantly from those having low self-esteem in their reading test. The mean score of
students having high self-esteem 71.73 is higher than that of those
66
having low self-esteem 63.73, so students having high self-esteem
have better reading skill than those having low self-esteem.
D. Discussion of the Result of the S tudy