TOTAL SCORE = Content + Organization + Vocabulary + Language Use + Mechanics
E. Technique of Data Collecting
To collect the data, the researcher used a written test as the primary instrument. There are two types of tests; pretest and posttest. The pre-test was
given in experimental and control class to know how far the students’ recount
writing ability before receiving treatment. The post-test was given to know their writing ability after the treatment. The pre-test and post-test included in these
processes: 1. The study conducted on August 3
rd
2015. The researcher asked the information from the English teacher to know the circumstances of learning
English and teaching recount writing. 2. The researcher gave pre-test in control class on August 5
th
and in experiment class on August 7
th
2015. 3. On August 10
th
until 26
th
2015, the researcher conducted the certain treatment in the experimental class by giving the explanation about clustering technique,
the characteristics of recount text, and by asking them to write recount text by using clustering technique; whereas in the control class did not use clustering
technique. 4. The researcher administered post-test in control class On August 28
th
2015, and in experimental class on August 31
st
2015. The researcher asked the students to write a recount text with the same topic in pre-
test, ‘Holiday’. 5. After getting the whole data, the researcher calculated the result of the
students’ score in pre-test and post-test by using the some formulations. So, at the end of the study, the researcher could see how far their ability and
confidence were increased in writing recount text through clustering technique. By the time, the students slowly could understand about language and writing
structure so they could start writing correctly because the writing competence cannot be reached at one moment, but it takes a lot of time to process it.
F. Technique of Data Analysis
In analyzing the data, the researcher used t
test
formula through SPSS Special Package of the Social Sciences version 22 software. The t-test is one of a number
of hypothesis tests. Before calculated t-test, the researcher did normality and homogeneity tests first.
1. Normality Test
Normality test is performed to show whether the data from the sample is normal or not, the sample is taken from experimental and controlled group, both
post-test and pre-test group. If the normality of the data is more than the level of significance a 0.05, scores are normally distributed. The normality test is
performed using Kolmogrov Smirnov and Shapiro-Wilk. This is the example of the data using SPSS:
Table 3.2 The example of Normality Test in SPSS 22
Tests of Normality
Kelas Kolmogorov-Smirnov
a
Shapiro-Wilk Statistic
df Sig.
Statistic Df
Sig. Pretest Experiment
.170 20
.134 .907
20 .055
Control .174
20 .115
.908 20
.059 a. Lilliefors Significance Correction
2. Homogeneity Test
Homogeneity test is performed to show whether the data from the two groups, experimental and controlled class, have the same variant in order that the
hypothesis can be tested by t-test or not. Here is the result of homogeneity test of the data:
Table 3.3 The example of Homogeneity Test in SPSS 22
3. Hypothesis Test
After getting the data from pre-test and post-test from experimental and control class, the researcher needs to find out the differences score using
Clustering technique. Here, the two classes are compared to the independent variable, the experimental class is X variable and the controlled class is Y
variable. The researcher used statistical calculation of the t-test with significant degree 5 and 1. The formula of t
test
is expressed as follows:
3
Where: Mx
= mean of variable X My
= mean of variable Y SE
= standard error
But before calculate the data using t-test formula; the researcher analyzed the students’ writing recount text score by using several processes as follows:
1. Determining Mean of Variable X: ∑
3
Anas Sudijono, Pengantar Statistik Pendidikan. Jakarta: PT. Raja Grafindo Persada, 2008, p.324.
Test of Homogeneity of Variances
Pretest Levene Statistic
df1 df2
Sig. .140
1 38
.711
2. Determining Mean of Variable Y: ∑
3. Determining Standard of Deviation Score of Variable X: √
∑
4. Determining Standard of Deviation Score of Variable Y: √
∑
5. Determining Standard Error of Mean of Variable X: √
6. Determining Standard Error of Mean of Variable Y: √
7. Determining Standard Error of Difference of Mean of Variable X and Y: √
The last procedure is determining df degree of freedom with formula: Where:
M = the average of students score
SD = standard deviation
SE = standard errors
X = experimental class
Y = control class
N
x
= number of students of Experiment class
N
y
= number of students of Control class Df
= degree of freedom
G. Statistical Hypothesis
The statistical hypothesis of this study can be seen as:
t
o
t
t
, H
o
is rejected and H
1
is accepted t
o
t
t
, H
o
is accepted and H
1
is rejected And then, the criteria used as follows:
1. If t-test t
o
t-table t
t
in significant degree of 0.05, Ho null hypothesis is rejected. It means that the rates of the mean score of the experimental group
are higher than the controlled group. The using of clustering technique is effective on students’ writing recount text.
2. If t-test t
o
t-table t
t
in significant degree of 0.05, Ho null hypothesis is accepted. It means that the rates of the mean score of the experimental group
are same as or lower than the controlled group. The using of clustering technique is not effective on students’ writing recount text.
37
CHAPTER IV FINDINGS AND DISCUSSIONS
This chapter presents the data description which consists of the score of pre- test and post-test from the experimental class and the control class. Moreover, the
discussion of the research finding is also explained here.
A. Data Description
The research was conducted in SMA Al-Hasra Depok on August 2015. The researcher took two classes that consist of experimental and controlled class.
Those classes were from the first grade and each class consisted of 20 students, so there were 40 students. The material that was taught is recount text. The
researcher gave pre-test in the first meeting and post-test in the last meeting to both classes experimental and controlled classes
to know the students’ achievement after the using of clustering technique on their writing recount text
and without using clustering technique.
1. The Score of Experimental Class’ Pre-test and Post-test
The data of experimental class were collected from the result of the students’
score of pre-test and post-test in class X.2. For the complete score, it can be seen in the APPENDIX 1.
Based on the result of pre-test in experimental class, the highest score and the lowest score in the experimental class those consist of 20 students. In pre-test, the
highest score was 73 obtained only by one student and the lowest score in pre-test was 35 obtained by one student. The mean score of the pretest was 49.75. From
that data , it could be seen that most of the experimental students’ writing ability in
writing recount text was still very low. In post-test, the mean score of post-test was improved and it was 66.
Moreover, the mean of gained score was 16.25. The highest score of post-test was 80 obtained by two students and the lowest score in post-test was 43 obtained only
by one student.
Kategori