Method of the Study Procedure of Inventing
4 Technique of Data Collecting
the writer used the test to collect the data. The pre-test and post-test waas given in both of the two classes. The pre-test was given before the writer
began the teaching and learning process in two classes. Meanwhile, the post- test was given in order to know the students’ improvement in mastery
vocabulary. The pre-test and post-test are multiple choice form and it was consists
of 20 items. The teacher gave 30 minutes in both of the two classes for finishing pre-test and post-test.
5 Technique of Data Analysis
The writer has conducted the test; pre-test and post-test. The data is compared from the mean of the score from pre-test and post-test. After getting
the data from the score, then the data was analyzed and processed by using statistic calculation of T-test formula with significance degree 5 and gained
scores. T-test in this research used to test the average difference count, was there a significant difference or not between the experimental group and the
control group. While the gained score is the difference between pre-test and post-test score of each class of the experimental and the control groups. Gain
scores are used to determine the increase or decrease in scores and to determine the effectiveness of the media used. However, prior to the tests the
hypothesis necessary analysis prerequisite tests first, namely the distribution normality test and homogeneity test.
1. Pre requisites Test Analysis a Normality Test Population
Normality test data and research using chi squared using the following formula by Riduwan
5
�
2
= ∑
�� − ��
2
��
Specification: X
2
= value of chi-squared Fo= frequency obtained based on the data
Fe= the expected frequency frequency theoretical
Normality testing criteria: If X
2
o ≤
X
2
t, so the data distribution
are normal. If X
2
o ≥ X
2
t, so the data distribution are not normal. b Population Variance Homogeneity test
Homogeneity test is conducted to determine whether the both of groups have the same variant or not homogeneous or not.
According to Riduwan
6
test in this research using Barlet test with significance level
α = 0.01, with db1= N1 -1 and db2 = N2 -1.
� = �ℎ� ������� �������
�ℎ� �������� �������
Terms Homogeneous: If Fo
≤ Ft, then Hois accepted homogeneous and Ha rejected. If Fo
≥ Ft, then Ho is rejected not homogeneous and Ha accepted.
5
Riduwan, Belajar Mudah penelitian unttuk Guru, Karyawan, dan Peneliti Pemula, Bandung: ALFABETA,2011, p.121.
6
Ibid. P. 119
c T-test According to Sudjiono that the formula is follows:
7
t ₒ =
M
1
− M
2
SE м
1
− м
2
M 1 : Mean of the Difference of Experiment Class
M 2 : Mean of the Difference of Control Class
SE M 1 : Standard of Error of Experiment Class SE M 2 : Standard of Error of Control Class
In order to get the calculation of T-test, there are several steps to be taken, there are as follows:
1 Determine mean of variable X with formula:
�
1
=
∑ � �
1
2 Determine mean of variable Y with formula:
�
2
=
∑ � �
2
3 Determine Standard deviation variable X with formula:
SD
1 =
�
∑�
2
�
4 Determine Standard deviation of variable Y with formula:
SD
2 =
�
∑ �
2
�
5 Determine Standard error of variable X with formula:
SE M
1 =
��
1
√�−1
6 Determine standard error of variable Y with formula:
SE M
2 =
��
2
√�−1
7
Sudjiono, Anas. Pengantar Statistik Pendidikan., Jakarta: Rajawali Pers, 2010. p.240
7 Determine standard error means of differences mean of
variable X and variable Y, with formula: SE
M1-M2 =
���
�1 2
+ ��
�2 2
8 Determining t
with formula: t
ₒ = M
₁ − M₂ SE
м
1
− м
2
9 Determining t-table in significance level 5 with Degree
of Freedom df: df = N1+N2 – 2
6 The Statistical Hypotheses
A hypothesis is a tentative statement about the relationship between two or more variables. This research is designed to find out whether there is a significant
progress of cooperative learning specifically using songs in teaching vocabulary. In order to get the answer of that hypothesis, the writer proposed Alternative
Hypothesis H
a
and the Null Hypothesis H which is described to the following
statistical hypothesis: a If t
test
t t-
table
t
t
in significant degree of 0,05, the alternative hypothesis H
a
is accepted and the null hypothesis H is rejected
b If t
test
t t-
table
t
t
in significant degree of 0,05, the alternative hypothesis H
a
is rejected and the null hypothesis H is accepted.
Meanwhile, the degree of freedom df = N1+N2-2 = 20+20-2 = 38. It must be calculated with t-table of df. If df is 38, the value of significance level 5
is 1,686.
27