CHAPTER III
RESEARCH METHOD
This chapter presents the research methodology used in this study. It
comprises research design, population, sampling, and sample, research
instrument, validity and reliability testing, data collecting method, and data
analysis.
A. Research design
In this research, the researcher uses quantitative approach. This research
statistics had an important role as a tool to analyze the data in a research, it is
importance to think the design of the research. According to Muijs, (2004: 1)
quantitative research is eexplaining phenomena by collecting numerical data that
are analyze using mathematically based methods (in particular statistic). It means
that quantitative research methods dealing with numbers and anything that is
measurable in systematic way of investigation of phenomena and their
relationships.
In this research, the researcher uses the correlative research. Latief (2011:
111) states that correlative research designs are used to measure the relationship
between two or more continuous variables. It means that, in correlation research,
the researcher looking for variables that seem to interact with one other, so that
when we can see one changing, we have an idea of how the other will change
when we can see one changing
B. Population, Sampling, and Sample
1. Population
As well as deciding what information the researcher need, the researcher
also need to decide exactly what the population going to be. According to (Muijs,
2004: 37) population is the group that the researchers want to generalize finding
to. It means that, population is the totally of a person or group that it able to help
the researcher generalize of finding. According to Arikunto, (1998: 115-117)
state that population is the whole or research subject. In other word population is a
group of individuals or items that share one more characteristics from which data
can be gathered and analyzed. The writer will take tenth grade student of MAN
Kunir in academic 2014/2015 as the populations they are X C consist of 30
students, XB 30 student. The total population is 60 students.
2. Sampling
According to Muij (2004: 38) sampling is sampling technique which based
itself that every member of the population has an equal chance to be selected as
sample. From this devinition above, sampling is the technique a researcher
chooses his/her sample. It involves simple random sampling, systematic random
sampling, stratified random sampling, random sampling. In this study, the writer
used random sampling. Random sampling is a technique or tool or procedure
essentially a maniversion of the initial population (Voegtle, 2006: 143). From this
definition, random sampling is the technique or procedure that used to selecting
sample that every person. The procedure of taking sample begins from invite the
student to write the name. Then, the entire name put inside of bottle, and the entire
name are put in a bowl, and they are shacked thoroughly. Those numbers or
names coming out are considered to be samples.
3. Sample
Sample is a part of population. The researcher sample was the tenth grade
students at MAN Kunir Wonodadi Blitar that are 20 students’ from XB, 20
students’ from XC. The total are 40 students’.
A. Research Instrument
1. Questionnaire
Questionnaire is listed of question or statement about student’s learning
style that is given for them at tenth grade to find the information of learning style
such as of reading comprehension. The process of constructing questionnaire in
this research, the writer takes it from the book quantum learning (Depoter, 1992:
125). The researcher reads some theories in the book especially in part of learning
style theory then deciding the points of learning styles characteristics to made as
Questionnaire which related with the student’s or learner’s learning activity based
on the theory, and it has a function to know the students preference in their
learning style. This method contains some questions about the issue which shown
agreement or disagreement or certain conviction. The simplest way to describe the
idea is to provide member of alternative answer for each question. For the analysis
is this item, an alternative answer could be strongly agree, agree, Undecided
disagree, strongly disagree.
In this research, the question was written in Indonesian. It was done to
make easier the respondent answer the question. The questionnaire consisted of
fifteen (15) multiple choice. In this part, the researcher asked to the students to
choose one option, it means that the answers most reflect their personality or their
real life. Each question consists of five (5) options of the answer, strongly agree,
agree, undecided, disagree, strongly disagree. This qquestionnaire will be given to
the students. The researcher asked to fill the questions which they are listed. The
scores of each student would be marked as: Score 5 for item strongly agree, score
4 for item agree, score 3 for item undecided, score 2 for item disagree, score 1 for
item strongly disagree
2. Test
According to Arikunto, (2008: 33) test is means of collecting information,
but when compared to other tool this test is more formally because it is full
restrictions. According to him, has the dual function of a test that are measured the
students and to measure the success of the teaching program.
Here, the researcher used test to measure students’ comprehension in
reading. Researcher applied and instruction to choose the best answer in reading
comprehension test consist of (25) multiple choice. The students’ must notice the
sentence or the story before they are answer the question.
This test has been conducted try out in another class and the result that the
instruction of this test not to confuse to the student. And they can be done well. It
can be assumed that the instruction was good and clear, so it was not necessary to
revised. The essential requirement of test is valid.
D. Validity and Reliability
1. Validity
Muijh, (2004: 67) state that validity is probably the single most important
aspect of the design of any measurement of the instrument educational research. It
can be conclude that validity is the process that refers to how wel the test
measures what it is purported to measure. However good our research design
statistical analyses, the results will be meaningless if the researchers aren’t
actually measuring what the researcher is purposing to measure. According to
latif, (2011: 223) valid means correct. It means that when the researcher claim that
the result of students writing assessment is valid, the researchers convinced that
the writing assessment result correctly reflects the students’ writing skill as
distinct from their reading skill, or from their knowledge of vocabulary. Validity
is very important because one of main characteristic of test, without having this
characteristic a test is not important.
a. Content Validity
Content validity is the process of establishing the representativeness of
items with respect to the domain of skills, task, knowledge and so forth of
whatever is being measured. Content validity refers to whether or not the content
of the manifest variables. Validating the instrument, the researcher also consults
the instruments related to the test with the lecturer and teachers who expert in
teaching English language. The test in this research had fulfilled the content
validity because the materials of this test were taken from the topic in the English
book of the tenth grade student.
b. Construct validity
Construct validity is the capacity for the measuring certain specific
characteristics in accordance with the theory of language behavior and learning, in
this research, reading comprehension test is tested aim to see their ability of
student in student comprehension in reading text. Construct validity a slightly
more complex issue relating to the internal structure of an instrument and the
concept it is measuring (Muijh, 2004: 68). A test is said to have construct validity
if it can be demonstrated that it measures just the ability which supposed to
measure. Reading comprehension test can be assessed in the form multiple
choices. Meanwhile, learning style is used questionaire,
2. Reliability
While validity refers to degree of correctness of the writing skill
assessment results in representing the writing skill being. Reliability of the result
of language skill assessment refers to the preciseness of the language skill
assessment result in representing the actual level of the skill of the examinees’
(latief, 2011: 212). It means that, reliability of instrument is needed to make sure
that the instrument can be consistent if used in other time. Therefore, the
instrument as the test is reliable.
In this case, before the researcher conduct research in this class, the
researcher made a test consist of 25 question of multiple choice, and then the
researcher asked two experts to give correction about this test, such as style, lay
out the test, grammar, vocabulary and content. After that, the researcher revises
the this test. And the researcher tried it out to 9 students to know how far the
reability of the instrument. The researcher used SPSS 16.0 version to account the
data collected. So the researcher know whether this test have reability or not. The
result of computing can be seen below:
Table Reabiliti 3.1
Based on the table above, it showed the reliability of croncach’s alpha is
0.754. the value conbach’s alpha can be be interpreted as follow:
Table 3.2 Cronbach’s Alpha Interpretation
Cronbach’s Alpha Interpretation
0,00 – 0,20 Less Reliable
0,21 – 0,40 Rather Reliable
0,41- 0, 60 Quite Reliable
0,61 – 0,80 Reliable
0,81 – 1,00 Very Reliable
Based on the table above, it can be conclude that the instrument of this
research was in the category of reliable because 0.61<0.754<0.80
E. The Technique of the Data Collection
The technique of the data collection of learning style is used non-test
technique. There were two instruments of non-test technique applied; they were
questionnaire. In answering the questionnaire, the students are asked to choose
one option by giving mark (X) or checklist (√). To identify the
students’preference in learning style, the writer counted the mean score of each
learning style type (visual, auditory, and kinesthetic). The students’ highest mean
score of learning style determined in which type they were. The data of the
students’ comprehension in reading was collected by the test.
F. Data analysis
Data analysis method is technique to analyze the obtained data. Analyze
data is useful for the research, because the raw data taken from the field are
meaningless if they are not analyzed.
Creswell, (2012: 338) state that, correlation is a statistical test to determine
the tendency or pattern for two (or more) variables or two sets of data to vary
consistently. In the case of only two variables, this means that two variables share
common variance, or they co-vary together. Because the research deals with
quantitative data with consist of two variable that is learning style and reading
comprehension, the researcher applies statistic method, therefore, used correlative
research to analyze the data.
In this research, all the data are gained from the result of the questionnaire
and test in the form of number. All the data therefore analyzed quantitative. Since
this research is to know the correlation between two variables. The research is to
know the correlation between two variables. The researcher applied correlation
analysis by using Spearman Rank Correlation technique. The researcher used it
because the data obtained from the two variables was exposed in term in
quantitative score, and the data was using in ordinal data. Spearman rank
correlatio is a non parametric test that is used to measure th degree of association
between two variables.
Here the researcher used the simple formula of Spearman Rank technique.
It could be gotten the correlation coefficient value (r) of two variables. Those
variables X was students’ score in the Questionnaire and variable y was reading
score in.
Table 3.3 The interpretation Coefficient Correlation (r)
Interval coefficient Interpretation
0,00– 0,20 The correlation is very low
0,20 - 0, 40 The correlation is low
0,40 - 0, 70 The correlation is moderate
0,70- 0,90 The correlation is high
0,90– 1,00 The correlation is very high
` If the correlation coefficient (r) obtained is smaller than or equal to the “r”,
Ha (There is significant correlation between student’s learning style and reading
comprehension of tenth grade students at MAN Kunir Wonodadi Blitar Academic
Year 2014/2015) is rejected. It means that, there is no significant correlation
between student’s visual learning style preference and reading comprehension of
tenth grade students at MAN Kunir Wonodadi Blitar Academic Year 2014/2015).
Conversely, if the correlation coefficient (r) obtained is greater than or equal to
the “r” product moment table value, Ha is accepted or Ho is rejected. It means
that, there is significant correlation between student’s visual learning style and
reading comprehension of Tenth grade students at MAN Kunir Wonodadi Blitar
in academic Year 2014/2015).
All related to data analyze is explained more in chapter IV. Following is
brief description of them.
1. Editing
Editing process is done after the researcher finishes collected data from the
field. This process started with giving identifies the qquestionnaire and Testing of
reading comprehension. Score transcript and then checking the instrument,
pointing and answering one by one.
2. Coding
Instead of reading all the qquestionnaires and looking at scores of
samples’ reading test, coding system of data was very useful. It made data
readable in more efficient way and time. Coding system for each variable will be
displayed in Chapter IV together with analysis.
3. Tabulating
In tabulating, the researcher entered the data in the tables to describe the
data so the researcher was easier in understanding the structure of the data.
4. Analyzing data using SPSS
After Each Variable was coded and put into tables, then the data were
analyzed using SPSS 19.0 (Statistical Package For Social Science). Using
Spearman Correlation Method, this thesis outcome was analyzed after calculating
the entire variable.
G. The Statistical Hypothesis
The statistical hypothesis with consists of Ha (Alternative Hypothesis) and
H0 (Null Hypothesis) with the significance level α =0.05 in this research was
formulated as follow:
H0 : χ2 = 0
Ha : χ2 ≠ 0
H0 : There was no significant relationship between students’visual learning style
in their comprehension in reading
Ha : There was a significant relationship between students’ visual learning style in
their comprehension in reading
The hypothesis criterion was according to Chi-Square analysis “Rejecting
H0 if counted χ2 > χ2 table or χ2 > χ2(k-1)(r-1) and accepting H0 if counted χ2
<χ2 Table or χ2 < χ2 (k-1)(r-1).”