CHAPTER III

RESEARCH METHOD

This chapter will discuss about the method used by the researcher in

conducting this research. The discussion will cover research design (a), population,

sample and sampling (b), research variable (c), research instruments (d), validity

and reliability testing (e), Normality Testing (f), data collecting methods (g), and

data analysis (h).

A. Research Design

Research design used in this study is pre-experimental research design

with quantitative approach. Creswell (2008:6) states that quantitative research

is a type of educational research in which the researcher decides what to study,

asks specific, narrow question, collects quantifiable data from participants,

analyze these numbers using statistics, and conducts the inquiry in an unbiased,

objective manner.

Quantitative is research that is used statistical technique to understand

and explain phenomena. Ary et al (2010 : 26) states that experimental research

involves a study of the effect of the systematic manipulation of one variable on

another variable. The manipulated variable is called the experimental treatment

or the independent variable.

The goal of experimental research is to test hypothesis to establish

cause and effect relationships. The overarching purpose of experimental

25
 26

research is to determine whether a particular approach or way of doing

something is “better” than the “older” or more traditional approach that has

served as the standard practice (Lodico et al, 2006 : 12).

There are many kinds of the experimental research design, such as pre-

experimental design, true experimental, factorial design and quasi-experimental

designs (Ary et al ,2010 : 302). This research used pre-experimental design

because they provide little or no control of extraneous variables in the form of

one-group pretest-posttest design. This research used pre-test and post-test to

see the result of the treatment.

The one-group pretest and posttest design usually involves three steps :

(1) administering a pretest measuring the dependent variable, (2) applying the

experimental treatment X to the subjects, and (3) administering a post test,

again measuring the dependent variable. Differences attributed to application

of the experimental treatment are then evaluated by comparing the pretest and

post-test scores (Ary et al, 2010 : 303).

Table 3.1 A Diagram One Group Pretest-Posttest Design

Pre-test Independent Variable Post-test

Y1 X Y2
 27

The procedures of the pre-experimental research with one-group

pretest-posttest design in this research were described as follows:

1. Administering a pretest (Y1) which proposed to measure students’

mastery in vocabulary before given a treatment.

2. Applying an experimental that was using song (X) to teach

vocabulary.

3. Administering a posttest (Y2) which proposed to measure students’

mastery in vocabulary after given a treatment.

Applying one group pretest-posttest, the researcher wanted to find out

whether there is any significant different of students’ mastery before and after

being taught using song at MTsN Tunggangri in academic year 2015/2016.

B. Population, Sample and Sampling

1. Population

Before the sample was collected, the researcher had to determine

the population. Arikunto (2006: 130) stated that a Population is formulated

as the whole groups of people or object that have been formulated clearly.

Based on Creswell (2008: 151) a population is a group of individuals who

have the same characteristics. It means population has at least one

characteristic that differentiates it from other groups.

The population of this study was taken from MTsN Tunggangri

Kalidawir Tulungagung. The population of this research is all of the class

of eighth grade students in MTsN Tunggangri that is consist of ten classes

 28

in academic year 2015/2016 with total of students 392 students, 169 male,

and 223 female. The quantity of students in each class of the population as

follows :

Table 3.2 Population of the Research

Class X (Total of Students)

VIII – A 28

VIII – B 42

VIII – C 42

VIII – D 42

VIII – E 42

VIII – F 41

VIII – G 39

VIII – H 37

VIII – I 38

VIII – J 41

∑X 393

2. Sample

Sample is part of population. A sample must be representative to a

population (Arikunto, 2006: 131). It means that good sample must be

representative of the entire as possible, so that the generalization of the

sample of this research. Based on Creswell (2008: 152) a sample is a

 29

subgroup of the target population that the researcher plans to study for

generalizing about the target population.

According to Sugiyono, (2007 : 73) states that the sample is part of

the total and characteristics possessed by the population. It means if the

population is very large, it is unlikely that the researcher learn all that exists

in the population. Therefore, researcher can use the sample taken from the

population. The sample of this research was the VIII E grade students at

MTsN Tunggangri academic year 2015/2016 consists of 42 students, 18

male, and 24 female. The researcher decides to choose VIII E because the

researcher believes that VIII E has average proficiency in English language

subject. Based on the researcher consideration this class is appropriate one

for doing the research.

3. Sampling

Sampling technique is the process of selecting a number of

individuals for a study in such a way the individual represent the large

group from they were selected. Sampling is the process of selecting a

number of individuals for a study in such a way that the individuals

represent the larger group from which they were selected (Gay, 1992:123).

The purpose of sampling is to gain information about a population; rarely is

a study conducted that includes the total population of interest as subject.

According to Sugiyono (2007: 73) stated that sampling is technique

to take sample. Thus, in selecting the sample in this research, the researcher
 30

chooses one kinds of non probability sampling. For this research, the

researcher used purposive sampling technique. Purposive sampling also

referred to as judgment sampling-sample elements judged to be typical, or

representative, are chosen from the population (Ary et al, 2010: 156).

The researcher used purposive sampling to choose the sample.

Purposive sampling is technique to chosen the sample with specific purpose

(Sukardi, 2012: 64). By using purposive sampling, the expected criteria for

sample obtained completely in accordance with the research to be

conducted. Choosing the sample is based on purposive sampling depends

on what criteria are used. So first determined what criteria sample taken

researcher took sample of class VIII. The researcher chooses the class VIII-

E as the sample because in applying the experimental design, the sample

must not be too “good” and too “bad” in their English achievement. It is

intended to reduce the extraneous variable may appears since the design is

pre-experimental research without control group. And also, the researcher

chooses class VIII-E as the sample because among other classes the

students of the VIII-E class had average capability based on purposive

sampling depends on what criteria are used.

C. Research Variables

Variable is research object or something that becomes view point of

research. Variable is a characteristics or attribute of an individual or an

organization that researchers can measure or observe and varies among

 31

individuals or organizations studied (Creswell, 2008: 123). This study has two

variables; they are independent variable and dependent variable.

a. Independent variable

Independent variable is the variable which influence dependent variable, in

the other words independent variable is causes variable. Independent

variable is an attribute or characteristic that is dependent on or influenced

by the independent variable (Creswell, 2008: 126). In this research, the

independent variable is song.

b. Dependent variable

Dependent variable is the variable which is influenced by independent

variable. According to Creswell (2008: 126) dependent variable is an

attribute or characteristics that influences or affects an outcome or

dependent variable. The dependent variable in this study is the students’

vocabulary mastery.

D. Research Instrument

The researcher must use instrument in order to get the better

data. The instrument of the research is a tool or facility that is used by

researcher for collecting data in order to get better result or in other

words, it can be occur complete and systematic. The instrument is used by

the researcher is test. The instrument of the test in this research is objective

test. Objective test is frequently criticized on the grounds that they are simpler
 32

to answer than subjective test. This vocabulary testing used by the researcher to

know the students mastery in vocabulary before and after taught by using Song.

Tests are valuable measuring instruments for educational research. A

test is not necessarily a written set of questions to which an individual responds

in order to determine whether he or she “passes” (Gay, 1992: 154). According

to Ary et al (2010; 201) stated that a test is a set of stimuli presented to an

individual in order to elicit responses on the basis of which a numerical score

can be assigned. Arikunto (2010:193) explain that a test is a set of question or

exercise or by any means which is used to measure the skill and the knowledge,

intelligence, ability or talent proposed by individual or a group of people.

The researcher gave twice of test to the students. The first test was

pretest where it was given to the students before giving treatment by teaching

vocabulary using song. It aims to know students mastery of vocabulary

achievement before taught by using song. The second test is given to the

students after giving treatment by teaching vocabulary using song. It aims to

know students mastery achievement after taught by using song.

The test items in the pretest are exactly same as those in the posttest.

The researcher gave the test that contains 25 items and consists of 2 kinds. The

first kind is multiple choices that contain 20 items. Second is changing the verb

from verb 1 to verb 2. It was done to know the final score and to know the

students’ difference competence before and after they getting treatment. Thus,

in measuring the students’ comprehension achievement the teat was considered

quit representative to topic in the class.

 33

E. Validity and Reliability Testing

The quantitative research always depends on measurement. These

measurements are validity and reliability through instrument of research. The

instrument of this research was test. In this research, vocabulary test used to

measure the students’ achievement in mastering vocabulary after they have

been taught by using song. Validity and reliability testing of this test was very

important to know scores derived from instrument used in the research.

1. Validity

Validity is concerned with how accurate the test measure and the

appropriate of the test for the subjects. It is one characteristics of a good

test. According to Ary et al (2010:225) validity was defined as the extent to

which an instrument measured what it claimed to measure. Heaton (1988;

159) states that the validity of test is the extent to which it measures what it

is supposed to measures and nothing else. There are four types of validity,

they are content validity, criterion-related validity, construct validity, and

face validity. In this research, the researcher used content validity and

construct validity to measure whether the test has a good validity or not.

The description of both types of validity is described below.

a. Content Validity

A test is said to have content validity if its content constitutes a

representative sample of the language skills, structures, etc with which

it is meant to be concerned. The test would have content validity only

if it includes a proper sample of the relevant structures (Hughes,

 34

1989:22). The researcher create questions of the test based on the

components of vocabulary testing that is matched by syllabus so that is

was not out of contents. The researcher made this test based on the

course objectives in the syllabus of eighth grade in MTsN

Tungganggri. The content validity in this research can be showed as

follow:

Table 3.3 Content Validity of the Objective of Syllabus

Standard Competence 12. Mengungkapkan makna dalam teks

tulis fungsional dan esai pendek
berbentuk recount text dan narrative
untuk berinteraksi dengan lingkungan
sekitar

Basic Competence 12.1 Mengungkapkan makna dan

langkah retorika dalam esai pendek
sederhana dengan menggunakan ragam
bahasa tulis akurat, lancar dan berterima
untuk berinteraksi dengan lingkungan
sekitar berbentuk recount dan narrative

Continued
 35

Continuation

Indicator  Mengidentifikasi cirri kebahasaan

teks fungsional pendek berbentuk
recount text / narrative
 Menulis teks pendek dan sederhana
dalam bentuk recount text / narrative
dengan langkah retorika yang benar
 Mengidentifikasi berbagai makna
recount text / narrative
Technique Writing

Instrument of Test Pre-test

Post-test

b. Construct Validity

A test is said have construct validity if it can be demonstrated

that it measures just the ability which is supposed to measure. Based

on Heaton (1989 : 161) if a test has construct validity, it is capable of

measuring certain specific characteristics in accordance with a theory

of language behavior and learning. Creswell (2008: 173) stated that

construct validity is established by determining if the scores from an

instrument are significant, meaningful, useful, and have a purpose.

 36

2. Reliability

Reliability is a necessary characteristic of any good test for it to be

valid at all and a test must be reliable as measuring instrument. According

to Ary et al (2010: 236) the reliability of a measuring instrument is the

degree of consistency whit which is measures whatever it is measuring.

Creswell (2008: 169) states that a goal of good research is to have measures

or observations that are reliable. This quality is essential in any kind of

measurement. So, the reliability is important part to know the test is good

or not in using the test to measure students master in vocabulary of this

research. A reliable test is consistent and dependable. Thus, if the students

are given the same test on two different occasions, the test should yield

similar result and the more similar the scores are, the more reliable the test

is.

The researcher consider the reliability coefficient use the formula

from Kuder Richardson Reliability or KR-20, which is based on the

proportion of correct and incorrect responses to each of the items on a test

and the variance of the total scores (Ary et al, 2010:245). The researcher

uses KR-20 Formula by calculating the data using Microsoft Excel.

According to Vansickle (2015: 4) the classification of reliability test is:

 37

Table 3.4 The Criteria of Coefficient Correlation

Reliability Test Coefficient Classification

Less than 0.20 Poor

0.20 to 0.40 Fair

0.40 to 0.60 Moderate

0.60 to 0.80 Good

0.80 to 1.00 Very Good

Based on table above the researcher got value of coefficient

reliability was 0.704 for tryout pre-test instrument and the value of

coefficient reliability was 0.783 for tryout post-test instrument.

According to Ary et al (2010: 241) if the reliability coefficient is

near 1.00, the instrument has relatively little error and high reliability. From

this statement, the researcher could be conclude that both instrument in pre-

test and post-test was reliable, so it can be used as appropriate instrument to

measure student’s mastery in vocabulary with needed some any revising.

 38

F. Normality Testing

Normality test are used to determine whether a data set is well-modeled

by a normal distribution or not. Normality test is intended to show that the

sample data come from a normality distribution population. To know the

normality, the researcher used One-Sample Kolmogorov-Smirnov test with

SPSS 16.0. The hypotheses for testing normality are:

a. Ho : Data is in normal distribution

b. Ha : Data is not in normal distribution

c. The hypotheses for normality testing explain that the data is normal

distribution if Ho is accepted and the data is not in normal distribution if

Ha is accepted. The Ho is accepted when the significance value is higher

than 0.05 (α = 5%), while H0 is rejected when the significance value is

lower than 0.05 (α = 5%). The result for normality testing can be seen

as follows;
 39

Table 3.5 One-Sample Kolmogorov-Smirnov Test

One-Sample Kolmogorov-Smirnov Test

Unstandardized
Residual

N 42

Normal Mean .0000000

a
Parameters
Std. Deviation 10.38670264

Most Extreme Absolute .104

Differences
Positive .104

Negative -.098

Kolmogorov-Smirnov Z .673

Asymp. Sig. (2-tailed) .756

a. Test distribution is Normal.

Based on the data above, the result of normality test uses one-sample

kolmogorov-smirnov test shows the subjects is 42 students. The value of

Kolmogorov-Smirnov Z is 0.673 and the Asymp. Sig (2-tailed) is 0.756. It

shows that the data is normal, because the value of Kolomogorov-Smirnov Z is

0.673 more than 0.05 (0.673>0.05) and the Asymp. Sig (2-tailed) is 0.756 more

than 0.05 (0.756>0.05).

 40

G. Data Collecting Method

The data collecting method is the method to obtain the data in the

research. The method in collecting data of this research is administering test.

The researcher gave twice test to the students. The first test is pre-test and the

second post-test while there were teaching activities or treatment between both

tests.

1. Pretest

Pre-test was given before the treatments. Lodico et al (2006: 178) a

pretest is a test given before the experimental treatment. So pretest is done

before treatment process to know the students’ vocabulary mastery before

taught used by song. The researcher gave the test that contains twenty five

items and consists of two kinds. The first kind is multiple choices that

contain twenty items. Second is changing the verb from verb 1 to verb 2.

The numbers of students who took the pre-test there were 42 students.

Then, the result of the test is students’ score. The score obtained were

analyzed to determine between pre-test and post-test.

2. Posttest

Post-test was held after all treatments were conducted. A posttest is

a measure on some attribute or characteristic that is assessed for

participants in an experiment after a treatment (Creswell, 2008: 297). The

test items in the post-test are exactly the same as those in pre-test. The post-

test is done to see final score and to know the different of the students’

score before get the treatment and after they get the treatment. The goal of
 41

this test is to measure students’ vocabulary after being taught using song.

The researcher gave the test that contains twenty five items and consists of

two kinds. The first kind is multiple choices that contain twenty items.

Second is changing the verb from verb 1 to verb 2. The numbers of students

who took the pre-test there were 42 students.

H. Data Analysis

The data obtained from research result students test that were analyzed

quantitatively. Quantitative analysis was done using statistic which is called

statistical analysis. This technique was used to find the significant difference

on students’ vocabulary mastery before after being taught using Song. The data

was collected from students’ score in pretest and posttest. To know the

significant difference on the students’ vocabulary mastery before and after

being taught by using song, the researcher in this research using paired sample

T-test as SPSS 16.0 for windows, because of some reason. There are, first the

data is interval data (see from the score of the student), second the data is

parametric. Parametric is the test for analyzing interval and ratio data. Third,

the data is normal it can be seen from normality testing which show that the

data is normal. Fourth, the researcher use only one class and consists of 42

students, and use the treatment before – after section. The total number of

students which is higher than from 30 students is one requirement for use t-test.