C. Data Analysis
1. Result of Normality Test
The normality test is used to measure whether the data in both control class and experimental class are normally distributed or not. In this case, the writer used
liliefors
formula to test the normality of the data. The hypothesis for normality test formulated as follows:
H = the data have normal distribution
H
a
= the data do not have normal distribution The test criteria:
H is accepted if L
observed
is lower than L
critical
it means that the distribution of the data is normal.
H
a
is rejected if L
observed
is higher than L
critical
it means that the distribution of the data is not normal.
L
critical
was gotten by looking at table of vritical values for Lilifors test for normality Appendix 36. Because each class had the same number of sample, and the number
of sample was 31, and also this research was education or social research, the researcher used 0.05 significant level. Then, it was gotten pattern 0.886
� . Finally,
it was gotten that 0.886 31
was 0.159.
Table 4.3 Normality of the Control and Experimental Class
Class Pre-test
Post-test L
observed
L
critical
L
observed
L
critical
Control 0.0969
0.159 Normal
0.1438 0.159
Normal Experimental
0.0922 0.159
Normal 0.1357
0.159 Normal
Based on the table above, it can be seen that in control and experimental class, shown
Ho was accepted because L
observed
L
critical.
The researcher concluded that the data in both classes were normality distributed. see Appendix 17-20
2. Result of Homogeneity Test
The hypothesis for the homogeneity test formulated as follows: H
: Data have the homogenous variances H
a :
Data have not the homogenous variances
The criteria for homogeneity test are as follows: a.
H is accepted if F
observed
F
critical,
b. H
a
is rejected if F
observed
F
critical,
From the data gained the homogeneity test for pre-test was 1.13. From the result of homogeneity was consulted to
fisher
table, it was found that F
critical
of 0.05 30,30 = 1.84 see Appendix 37. The data come from homogeneous data provided F
observed
is lower than F
critical
F
observed
F
critical
it means that the variance of the data of post-test in both control class and experimental class is homogenous.
From the data gained the homogeneity test for post-test was
1.51
. From the result of homogeneity was consulted to
fisher
table, it was found that F
critical
of 0.05 30,30 = 1.84. The data come from homogeneous data provided F
observed
is lower than F
critical
F
observed
F
critical
it means that the variance of the data of post-test in both control class and experimental class is homogenous. see Appendix 21-23
3. Result of Hypothetical Test
The hypotheses are: H
o
= There is no influence of using audio visual media towards students‟
pronunciation mastery of the eighth grade at the second semester of SMPN 01 Rebang Tangkas Way Kanan in academic year of 20152016.
H
a
= There is influence of using audio visual media towards students‟ pronunciation
mastery of the eighth grade at the second semester of SMPN 01 Rebang
Tangkas Way Kanan in academic year of 20152016. The criteria of the test as follows:
H
a
is accepted if t
observed
is higher than t
critical
, or t
observed
t
critical
H
o
is rejected if t
observed
is lower than t
critical
, or t
observed
t
critical
In this case, the writer used the level of significant α = 0.05.
Based on the calculation was obtained t-test is 7.695, and the result of level of significant 0.05 is 1.671see appendix 24-26 and 38. From this it can be seen that the