Score of Reading Comprehension by using Conventional Method
Tests the null hypothesis that the error variance of the dependent variable is equal across groups.
a. Design: Intercept + A + B + A B
The hypotheses for homogeneity test were set as follows: H
: Data comes from homogenous population H
1
: Data comes from non-homogenous population The criteria were set as follows:
If the Sig value Levene’s test 0.05 means that H
is accepted and H
1
is automatically rejected. On the contrary, the Sig value Leven e‟s test
0.05 means that H
1
is accepted and H is automatically rejected.
Refer to table 4.3 above, it can be seen that the
Sig p value
for reading comprehension was 0.026. It means that
p value
is smaller than 0.05. It means that H
1
is accepted and H is automatically rejected, which implies
that data comes from non-homogenous population. Even though came from non-homogeny variance, the data can be processed by using contrast test in
Anova Multi-factor t- test that can be treated for both homogeny and non- homogeny variance Agung, 2004, p.19.
According to both normality test and homogeneity test revealed above, it can be concluded that the prerequisite test which are needed before
processing the data by using ANOVA test are already fulfilled. 3.
The Testing of Hypotheses Hypothesis testing was intended to determine the proposed null
hypotheses H tested at a certain significance level. Two way ANOVA
analysis was performed and, because in this study to be obtained was how much influence that occurs between the two independent variables and the
dependent variable. Hypothesis testing was done consecutively, starting from the first hypothesis, The Directed Reading Thinking Activity was more
effective than Conventional method toward reading comprehension, the second hypothesis was the students who have high reading interest have
better reading comprehension than those who have low reading interest, the third hypothesis was there was interactional effect between teaching method
and reading interest toward reading comprehension.
The analysis of reading comprehension variable is performed by using two tailed ANOVA test, with the assistance of SPSS version 20 for windows.
The result of ANOVA test then continued to extended test to find out the level of significance among groups significantly simple effect. In other
words, the extended test was performed to find out which group contributes
more to be students‟ reading comprehension according to the teaching method and the level of reading interest.
The computation of data analysis by using ANOVA test can be seen on the Table 4.12 below:
Table 4.12 ANOVA Test 2 x 2
Tests of Between-Subjects Effects
Dependent Variable: Reading Comprehension Y Source
Type III Sum of Squares
df Mean Square
F Sig.
Corrected Model 178.047
a
3 59.349
2.656 .056
Intercept 359850.016
1 359850.016
16101.422 .000
A 31.641
1 31.641
1.416 .239
B 17.016
1 17.016
.761 .386
A B 129.391
1 129.391
5.790 .019
Error 1340.938
60 22.349
Total 361369.000
64 Corrected Total
1518.984 63
a. R Squared = .117 Adjusted R Squared = .073
Further, contrast test with t-test statistic was presented below:
Table 4.13. Contrast Tests
Contrast Value of
Contrast Std.
Error t
df Sig. 2-
tailed
Y Assume equal
variances 1
4.25 1.671 2.543
60 .014
2 1.44
1.671 .860
60 .393
Does not assume equal variances
1 4.25
2.055 2.068 29.823 .047
2 1.44
1.168 1.231 21.763 .232