KARYATULISILMIAH, ANGKATAN X, (2016)

The Influence of Think-Talk-Write Strategy Towards Reading

Comprehension

Banowo Thohar Ichsani1 ;Aulia Putri2 ;Siswanto Ismadi Megah3

English Education Department

University of Riau Kepulauan
banowothohar15@gmail.com 1, auliaputri1983@yahoo.com 2, megah76@yahoo.co.id 3

Abstract

Dealing with learning English, Students are always faced by English reading text a
school, but they have some problems within. The students got difficulties if teacher let
them read, for instance, they got difficulties to convey the gist of the text, and the teacher
does not build their background knowledge by using good method in reading activity. To
acquire the students reading comprehension, it is needed to stimulate the students mind
and corrective feedback. Thus, researcher applied think talk write strategy to make the
reading activity in the classroom more effective and could improve students reading
comprehension. Then, the researcher conducted this teaching strategy in process of
reading descriptive text at class VIII-A at SMP N 34 Batam with 40 samples. To prove
that the alternative hypothesis (Ha) is accepted, the researcher used t-test formula
calculate the data. The data gained Ttest< t table (4.93>1.667). It also can be seen from
both classes post-test mean value; control class mean was 62.13 and experimental class
was 67.13. So, it can be concluded that the use of think talk write strategy give significant
influence in reading comprehension than the use discussion method.
Keywords : Students Reading Comprehension, Think Talk Write Strategy.

INTRODUCTION
Language is a communication for submit some information or to share about
ideas. Usually people use language for communication with someone. English as the
foreign language which is learned by students in the schools. It has proved with English
become subject of the study. So that, English becomes an important language to the
students. Every students must master and learn this language. Based on the background
of the study mentioned above According to Grabe and Stoller (2002) in Nurhadi
(2014:32) reading is the ability to draw meaning from the printed page and to interpret
this information appropriately.

1
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

It means that reading is a result of the interaction between the perceptions of

graphics symbols that represent language and readers language skill, cognitive skill,
knowledge of world and so on. According to Panel (2003:22) reading is meaning-
making process that involves a great deal of thinking, problem solving, and decision
making by both the teacher and the child. Furthermore definition of Reading
comprehension According to Wolley (2011:15-16) is the process of making meaning
from text. The aim is to gain an overall understanding of what is described in the text to
get the meaning of a word or phrase isolated. In the text of the information reading
comprehension, children can develop mental models, or a representation of the meaning
of the idea of the text during the reading process.

Some problems that faced by students in reading comprehension: first, The

students still have difficulties in understanding reading texts. When they are require to
identify the detail information about the text,they often make a mistake even they could
not answer the question related to the text given before. Second, The students
motivation in learning English is low; the students do not get more motivation which
make them feel interest in reading. When they get difficulties in studying they do not
want to ask the teacher or students more passive than active in the class.

Because of many problems that faced by the students in reading. The writer need
to limit this problem. This study was limited to the influence of Think Talk Write
(TTW) in students reading comprehension of Descriptive Text eight grade at SMPN 34
BATAM in academic year 2015/2016. The objective of the research is the aim of this
research is to see whether the use of Think Talk Write give significant influence in
developing students reading comprehension eight grade at SMPN 34 BATAM in
academic year 2015/2016.
There were two significance of the study, the study is use to for student to
encourage the students motivation in learning English especially in improve their
reading comprehension and Give the students an interesting technique of presenting the
material in older to improve their ability to comprehend and master reading. And then
for teacher english teacher can use this method (Think Talk Write) in the class, so the
students can be more active and interest in learning and TTW can help the English

2
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

Teacher to teach English in various method and for improving students interest in
reading.

METHODOLOGY
Research could be divided into two broad categories: Quantitative research and
Qualitative research. According to Gay and Airasian (2003:8) quantitative is methods
are based on the collection and analysis of numerical data, usually obtained from
questionnaires, tests, checklists, and other formal paper-and-pencil instruments. The
kind of this research was an experimental research It was the type of research which
focused on establish the cause-effect relationships.

According to Gay and Airasian (2003:355), experimental is the only one type of
research that can test hypotheses to establish cause-and-effect relationship. In
experimental study, the researcher manipulated at least one independent variable,
controled other relevant variables, and observed the effect on one or more dependent
variables. In this research, the one variable and control the other relevant variable wasthink-
talk-write strategy and this variable observate the influenced or effected to one dependent
variable that was students reading comprehension.

There are four designed of experimental research by Sugiyono (2012:110)

A. Pre-Experimental Designs
B. True-Experimental Designs
C. Quasi-Experimental Designs
D. Factorial Designs
This research belong to the quasi-experimental design. This research was
development from true experimental design which has control group. This design was
used because of the fact was very difficult to get control group which used in the
research (Sugiyono, 2012:116). In this research, there were two groups. Experimental
group and control group. Experimental group was treated by using think-talk-write
strategy. Control group was treated by using discussion strategy.

3
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

Population, Sample and Sampling Technique

Population
According to Gay and Airasian (2003: 102) state that the population is the
group of interest to the researcher, the group to which the result of study will ideally
generalize. Furthermore, According to Best (2006: 13) a population is any group of
individuals that has one or more characteristics in common and that are of interest to the
researcher. A population was defined as a group of individuals with at least one
common characteristic which distinguished that group from other individuals.
So the population was not only people, but also objects and objects of other
nature. Population was not just the amount that was the object / subject being studied,
but covered all the characteristics / properties owned by the subject / object. The
population in this proposal was the eighth grade students of SMPN 34Batam in the
academic year of 2016/2017. It has six classes, there were VIII.A, VIII.B, VIII.C,
VIII.D, VIII.E, and VIII.F. The specification of the population could be seen on the
table below :
Table 1. The List of Population
No. Student of Class VIII Total of Students
1. VIII.A 40 Students
2. VIII.B 40 Students
3. VIII.C 40 Students
4. VIII.D 40 Students
5. VIII.E 40 Students
6. VIII.F 40 Students
Total 240 Students

Sample
According to Fraenkel (2012: 91) the sample is of the larger group. One of the
most important steps in the research process was the selection of the sample of
individuals who will participate (be observed or questioned). Sampling refer to the
process of selecting these individuals.

4
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

According to Best (2006: 13) sample is a small proportion of the population

that is selected for observation and analysis. That mean sample was a part of
population that choosen to be research and analysis. In this proposal as quasi experiment
research, there were two sample, experimental class and control class. Experiment class
was VIII.Aand control class was VIII.E.

Sampling Technique
Sugiyono (2009:62-63) stated that sampling technique defines as the technique
that researcher will use to choose the research sample. There were two kinds of
sampling technique. First, Probability sampling consisted of Simple Random Sampling,
proportionate stratified random sampling, disproportionate sampling and Area (Cluster)
sampling. Second, Non-Probability Sampling which consisted of systematic sampling,
Quota Sampling, Insidental sampling, Purposive sampling, sampling jenuh, Snowball
sampling.
Researcher used cluster random sampling. According to Fraenkel (2012:96)
state that cluster random sampling is the selection of groups, or clusters of subjects
rather than individuals is known as cluster random sampling. Just as simple random
sampling was more effective with larger numbers of individuals, cluster random
sampling was more effective with larger numbers of cluster.
Therefore, the researcher used this technique in selecting sample, because all of
the class was taught by the same English teacher, so all of them were taught by using
the same syllabus, lesson plan and material lesson too. So by using cluster random
sampling technique was more practical to get the samples.

Technique of collecting data

Instrument
The instrumentation in this research was reading test. Arikunto (2006:150)
stated that test is an instrument which is arranged specially tome a sure ones skill,
knowledge, intelligence, andability of individual orgroup. The test was aimed to
know so far students ability in reading comprehension.

5
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

In this test, the writer prepared the test in the form of multiple-choice
questions. The researcher focused on genre of descriptive text. The data needed in this
research were collected through:

Reading Comprehension Test

Reading comprehension test was given to make evaluation. It included of various
items which were answered by the students and then the result of the test was about how
effective the think-talk-write strategy to students reading comprehension. The test was
taken by using written test. The test namely:
Pre-test
The function of pre-test was to get the data about students reading
comprehension before applied think-talk-write strategy. Before applied think-talk-write
strategy to the students, teacher gave the test to know how far the students ability in
reading. In this pre-test the writer used reading method to know how far students
reading comprehension progress.
Post-test
The function of post-test was to get the data about to know the students progress
after applied think-talk-write strategy. After gave the treatment, teacher gave the test to
know the progress of students reading comprehension after applied the strategy.

Testing Instrument
Validity Test

In this research, the validity of the test was item validity. Because, the instrument
of this research wass a test which consisted of 50 items of multiple choice questions.The
writer used product moment formula to find out the validity of the test designed by
Arikunto (2009:72)

rxy
XY - X Y
n

n X X n Y Y

2 2 2 2

6
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

Where:
rxy : validity coefficient
X : item score
Y : total score
N : total of subject

Reliability
Arikunto (2006:178) states that reliabillity is the instrument that used for the tool
of collecting data because that instrument has been good. It mean if the instrument was
tested twice, the result still constant, reliable or consistent.
To measure the reliability of the test, Kuder-Richardson Reliability Formula was
used. To get Kuder-Richardson reliability, it required test administration only once. The
formula was KR-20 (Arikunto, 2009:100) as follows:

2
11 = ( )( )
1 2

Where:
r11 = Reliability of instrument
P = porpotion of subject with correct item
q = porpotion of subject with wrong item (q=1-p)
pq = Sum of multiplication p and q
n = total item
s = deviation standard from the test

Technique of analyzing data

Data analyzing was the last step in the procedure of research. In analyzing the data
from the pre-test and post-test, the researcher used the statistical calculation they were:

Normality
Normality test was used to determine whether students reading comprehension in
a class of normal distribution or not. Sugiyono (2009:107) state that Chi Kuadrat (X2)
was aimed of this test to find out whether the research population was normally

7
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

distribution or not, the normal distribution mean that the sample was drew from the
same population. Here the formula (Riduwan, 2011:187):

2
( )2
=

=1

Where:

X2 : Chi quadrate

Fo : Frequency of observation

Fe : Frequency in hope

Homogeneity
This test was used to determine whether the sample has the same variance or not.
To test the homogeneity the writer used formula by Riduwan (2011:186). The formula
:

=

Hypothesis Test

Arikunto (2006:311) states that Ttest statistical method is a technique used to

compare theme an of the scores from two independent samples to know the influence of
think-talk-write strategy.
Sugiyono (2013: 122), to answer the formulation of the study and hypothesis
above the hypothesis need to be the best and the testing that will be use T-test. The
formula designed by Sugiyono (2013: 122):

T-test
1 2
t=
2 2
1 + 2 2( 1 )( 2 )
1 2 1 2

8
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

Where:
X1 : Average score sample 1
X2 : Average score sample 2
S1 : Deviation sample 1
S2 : Deviation sample 2
S11 : Variant sample 1
S22 : Variant sample 2
r : Correlation between two sample

Result
Validity Test
comprehension.In order to saw the validity of the test, the writer has to use a
formula. In this case, the writer used Product Moment formula by Arikunto (2009: 72).
The formula can be seen in formula 1 page 26. Validity of the questions was done,
where the writer gave the test to the class Eight F outside the experimental and control
class. Before the writer give the pre-test to the experimental and control class, the writer
gave the try out test to verify the validation of the tests. The total number of the
respondents are 40 with the significance level = 5% so the rtable = 0,312. If rxy> rtable,
so the data is valid vice versa if rxy< rtable, so the data is not valid.

Reliability Test
The writer used Kuder Ricahrdson formula or called KR-20 formula to measure
the reliability of instrument. The reliability data collected from the valid questions (see
appendix 8 page 158) there were 23 valid questions. The writer counted the total
variance of the test before find the reliability with KR-20 formula.

Normality Test
Normality testing was analyzed toward two of the data groups, they were reading
comprehension narrative text data of pre-test and reading comprehension narrative text
data post-test. The writer used Chi Squared formula by Riduwan (2011: 187). The
criteria of normality test is if the X2observed < X2table, so the instrument is in normal
distribution.

9
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

Homogeneity Test
The homogeneity test used to know whether each group had the same variance or
not. It was tested by using F test.
Table 2. Summary of Homogeneity Test
No Variable Fobserved Ftable Conclusion
1 Pre-Test 1.020 1.757 Homogeneous
2 Post-test 1.051 1.757 Homogeneous

Fobserved was Homogeneous when it is lower than Ftable (Fobserved < Ftable), Ftable in
this data was 1,757 with the total of students in experimental class was 40 and Ftable
1,757 with the total of students in control class was 40. Based on the data above, it
could be summarized that Fobserved for students pre-testing was 1.020 (1.020 < 1.77) and
Fobserved for students post-testing was 1.051 (1.051 < 1.76). Both of the Fobserved was
lower than Ftable was 1.757 and 1.757. It means that both of the variances were
homogeneous. The process of homogenity testing could be seen in appendix 10 page
161.

Testing Hypothesis
In order to see the influence of Think- Talk- Write strategy to students reading
comprehension, the data was analyzed by using T-test formula. Before the data calculate
by using t-test formula, the writer have to found out the correlation between post-test in
reading comprehension and Story Impressions strategy.
1. Between reading comprehension and Think- Talk- Write strategy by using product
moment formula:
Product Moment

=
2 2
(Sugiyono, 2013: 228)

Where:
X : The score of experimental class post-test
Y : The score of control class post-test

10
 Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

The process to find the correlation can be seen as follow:

=
2 2
6672225
=
(7209225)(6175225)
6672225
=
44518586450625
6672225
=
6672225
=1

2. T-test

After the writer found the correlation of the two variables the writer would to find
the t-test to know the effect of the strategy towards students reading comprehension by
using t-test formula. The t-table with 78 of level of freedom and level of significance is
0.05, the t-table was 1,664. Result of tcount (4,93) was bigger than ttable (1,667). Based on
calculation tcount, it means there were significant influence of Think talk - write
strategy. The alternative hypothesis was accepted and the null hypothesis was rejected.
It can be concluded that Think talk - write strategy have significant influence towards
students reading comprehension.

DISCUSSION
Based on the statistical analysis of the hypothesis testing, it was concluded that
thik talk - write Strategy gave significant influence on students reading
comprehension. Mean score of reading comprehension of experimental class and
control class was significantly different.

The mean score of reading comprehension of experimental class respectively were

67.13. In contrast, the mean score of reading comprehension of control class
respectively 62.13. The findings showed that the students mean score of reading
comprehension at post-test was better than students mean score of reading
comprehension at pre-test. It has supported from explanation in the second chapter

11
Banowo Thohar Ichsani., The Influence of Think-Talk-Write Strategy...(1-11)

where the writer conclude think talk - write Strategy was more effective learning
environment. Based on discussion above, it can be concluded that think talk - write
strategy gave better result on students reading comprehension.

REFERENCES
Arikunto, S. (2006). Prosedur Penelitian Suatu Pendekatan Praktik. Jakarta: PT Asdi
Mahasatya
Gay. L, R. And Airasian, Peter. (2003). Educational Research. New Jersey: Pearson
Education, Inc
Jack R, Fraenkel. (2012). How to Design and Evaluate Research in Education. New
York: Mc Graw Hill
John W, Best. (2006). Research in Education Tenth Edition. Boston: Perason Education
Inc
Nurhadi, Kardi. 2014. Story Map and Know, Want to Know and Learn (K-W-L)
Strategies in Teaching Reading Skills of Narrative Texts. Thesis. Indramayu:
Education Faculty Wiralodra Indramayu University
Panel. (2003). Early Reading Strategy. Ontario:
Riduwan. (2011). Dasar-dasar Statistika. Alfabeta. Bandung
Sugiyono. (2009). Statistika Untuk Penelitian. Bandung: Alfabeta
Sugiyono.(2013). StatistikauntukPenelitian. Bandung: Alfabeta
Sugiyono.(2013). StatistikauntukPenelitian. Bandung: Alfabeta
Wolley, G. (2011). Reading Comprehension Assisting Children With Learning
Difficulties. Australia: Springer Science+Business Media

12