46

CHAPTER III
RESEARCH METHODOLOGY

This chapter presents the research methodology used in this study. The
research methodology consists of place and time of research, approach and types
of research, the source of data, method of collecting data, procedure of the
research, and technique of analyzing data. The complete information about
research methodology use by researcher in this research will be presented as
follows:
A. Place and Time of Research
1.

Place of Research
This research takes place at SMPN 4 Madiun that is located on
Abdurrahman Saleh 3rd street, Madiun City. The reason for selecting this
school as the place of research is that there have not been any researches
at this school yet, particularly concerning with the application of films or
songs as learning media in teaching writing.

2.

Time of Research
This study has been conducted for 5 months; from August to
December 2013 The research activities that the researcher does during
the time are presented as follows:

47

Table 3.1. Time of Research

No.
1.

2.
3.
4.

Research
Activities
Pre-Research
Arranging
for
research design
Observing
related literature
Preparing
research
proposal
Collecting data
by pretest and
post test
Analyzing data
Reporting
the
result of study

Aug.

Sept.

2013
Oct.

Nov.

Dec.

B. The Method and Design of the Research

1. Types of Research
This was a quantitative research based on computation and
measurement, operational variables and statistics. The design of the
research used experiment design. It concerned primarily with discovering
the effectiveness between or among interrelationship of two variables at
the same time. Best (1977: 102) stated that experiment design is the
blueprint of the procedures that enables the researcher to test his
hypothesis by reaching valid conclusions about relationships between
independent and dependent variables.
2. Design of Research
This reserach is a kind of true experiment-quasi quantitative in the
form of comparative study. According to Aquino (2006:5), true experiment

48

method properly seeks answers to question by investigate possible cause

and effect relationships by exposing one or more treatment conditions and
comparing the results to one or more control groups not receiving the
treatment, random assignment being essential. The term quasi has a
purpose to approximate the conditions of the true experiment in a setting
which does not allow the control or manipulation of all relevant variables.
Gay (1987:250) says that the basic causal comparative research
involving selecting two groups differing on some independent variables
and comparing them on the some dependent variable. The following is the
design of the research.
Table 3.2. Research Design
Group
Group: E1
Films/movies
Group; E2
Songs

Pre test

Treatment

Post test

O1

X1

O2

O1

X2

O2

Note:
GE1: Films/movies

: Experiment group 1 using films/movies media

GE2: songs

: Experiment group 2 using songs

X1

: Treatment using films or movies

X2

: Treatment using songs

O1

: Pre test

O2

: Post test
(Arikunto, 2006: 86)
In this design, the observation was done twice, before and after the

experiment. The observation that is done before the experiment (O 1) is

49

called pre test, and the observation that is done after the experiment (O 2) is
called post-test. The difference between O1 and O2 is assumed as the effect
of the treatment/experiment.
C. Population, Sample, and Sampling
1. Population
A population is a set (or collection) of all elements processing one
or more attributes of interest (Arikunto, 2006: 130). However,
Population is number of individual or thing that will make an object
ofresearch (Hadi, 1995:72). From the statement above, the researcher
conclude that the population is a number of people that have similar
characteristics and become the object of the research. The population of
this research was the eighth grade of SMPN 4 Madiun in the academic
year of 2013/2014.
There are seven classes, namely VIII-A, VIII-B, VIII-C, VIII-D,
VIII-E, VIII-F, VIII-G, VIII-H, and VIII-I. The total number of the
students is 308 students. All of them were given English materials in
teaching and learning process. Moreover, they were chosen as the
population of the research based on some reasons, those are:
a. The students were all in the same grade.
b. The students were relatively in the same age.
c. The students had been studying English for the same period.
2. Sample

50

According to Arikunto (2010: 130), Sample is a part that can

represent all the population observed. It is called sample research when
people want to generalize the sample research result. In this research, not
all data and information will be processed and not all people but simply
studied using a representative sample. In this study, the writer took class
VIII A VIII G as the sample.
c. Sampling
In this research, it is more convenient to study subject in naturally
occurring groups or clusters. To gain the sample for this research, the
researcher use a cluster random sampling. This means that the researcher
pick up the sample from the available population by random. According to
Setyosari (2010: 191-192),
teknik kelompok atau rumpun (cluster) digunakan apabila populasi
atau sampel yang tersedia adalah berupa unit-unir rumpun dalam
populasi Penelitian eksperimental tentang pengaruh metode
mengajar biasanya menggunakan kelas-kelas atau kelompokkelompok, dan tidak mungkin mengambil secara acak setiap
individual anak dari setiap kelas. Sampel sering diambil dari
kelompok-kelompok yang telah ditentukan atau tersedia.
It means that cluster technique use if population or sample is in
cluster units. Experimental research is impossible to pick up the individual
from class by random. Sample usually pick from groups.
According to this theory, in this study the researcher use cluster
random sampling. By random, the researcher decide the sample by lottery.
The first lottery was to determine the experimental group. From seven
class, the class is selected as experimental group is VIII A which consists

51

of 33 students. Experimental group is taught using movies. While class is

selected as control group is VIII G which consists of 31 students. Control
group is taught using songs.
D. The Technique of Collecting the Data
There are some methods of collecting data which can be used for
collecting the data in a research. The researcher uses some techniques as
follows:
1. Observation
Observation has purpose to get the information from the object of
observation. According to Murray (2003:60) Observation is gathering
information involves watching and listening to events then recording what
occurred. It means that observation is the way to research directly by
watching and listening the object the record what happened in the events.
In this research, the researcher uses the direct observation so the researcher
involved directly in teaching process. It is used to find the teaching writing
conditions in eighth grades of SMPN 4 Madiun in the academic year of
2013/2014 as pre-research.
2. Documentation
This research also uses documentation to collect the data. Lincoln
and Guba (in Gall et al. 2007: 166) stated, Documentation is as written
communication that are prepared for personal rather than official reasons.
While, Grabe and Stoller (2002: 166) say Document gathering is
collection of sets of documents (interpreted broadly) that are relevant to

52

the research question (e.g. in-class exam; lesson plans; pre-and post-test,
software, students exercises, worksheets, writing assignment or projects,
student record; textbooks).
Based on the statement above, it can be said that documentation in
collecting data means collecting all written documents which are relevant
to the research question. From the documentation technique, the researcher
gets some data of the students writing achievement and any document
used in teaching writing process in SMPN 4 Madiun. The data from the
documentation technique are then interpreted and evaluated as the
information to support this research.
3. Test Method
The research instrument is a tool used for capturing data in the study.
In this study, the research instrument used is a test technique. Brown
(1994: 254) defines a test as a method of measuring a persons ability or
knowledge in a given area. A test is used to collect the data of students
writing skill.
E. Procedure and Method of Data Collection
There were some steps in gathering the data in this study, they are
scoring technique, try out, pre-test, and post-test.

1. Scoring Technique

53

Heaton (1974:135) points out three types of method in grading

writing score; they are: the impression methods, analytic method and the
mechanical accuracy.
In facilitating to analyze the students test result in writing report of
the two groups of this study, the writer attempted to apply analytic method
which consists of an attempt to separate the various features of a
composition for scoring purposes (Heaton, 1974:136). The writer only
limited to the items as follows:
a. Grammar refers to appropriate tenses;
b. Vocabulary refers to diction;
c. Mechanics refers to punctuation and spelling;
d. Fluency refers to style and ease of communication; and
e. Relevance refers to the content in relation to the task demanded of the
students (Heaton, 1974:137)
According to Heaton (1974:136), analytical method is ideally suited
to the classroom situation: because certain features have been graded
separately, each student is able to see how his particular grade has been
obtained. Here, underlying such reason, the writer attempted to use
Heatons Grid and Categories as the scoring system of this study.

Table 3.3. Scoring Guidance Taken from Heatons Grid and Categories

54

Criteria
of
Mastery

Score

Criterion of Scoring

Fluency

Excellent: flow style, very easy to understand, both

complex and simple sentences, very effective.
Good: Quite flowing in style, mostly ease to
understand, and a few complex sentences, very
effective.
Fair: style reasonably smooth, not too hard to
understand mostly (but not all), simple sentence, fairly
effective.
Inadequate: bad style, an effort to understand and
enjoy, complex sentences, confusing, mostly simple
sentences or compound sentences.
Unacceptable: very hard to understand, cannot enjoy
reading, almost all simple sentences, complex
sentences, confusing, excessive use of and.
Excellent: mastery of grammar taught on course, only
1 or 2 minor mistakes.
Good: a few mistakes only (preposition, articles, etc).
Fair: only 1 or 2 major mistakes but a few minor
ones.
Inadequate: major mistakes, which lead to difficult to
understand, lack of mastery sentence construction.
Unacceptable: numerous serious mistakes, no
mastery of sentence construction, almost intelligible.
Excellent: use of wide range of vocabulary taught
previously.
Good: good use of new words acquired, fairly
appropriate synonym, and circumlocutions.
Fair: attempt to use words acquired, fairly appropriate
vocabulary on the whole but sometimes restricted, has
to resort to use of synonyms, circumlocution, etc on a
few creation
Inadequate: restricted vocabulary, use of synonyms
(but always inappropriate), imprecise and vague,
affects meaning
Unacceptable:
very
restricted
vocabulary,
inappropriate use of synonyms seriously hinders
communication.
Excellent: all sentences support the topic, highly
organized, clear progression of ideas well linked, like
educated native speaker.
Good: ideas well organized, links could occasionally
be clearer but communication not impaired.
Fair: some lack of organization, re-reading required
for clarifying ideas.

4
3
2
1
Grammar

5
4
3
2
1

Vocabulary

5
4
3

2
1
Content

5
4
3
2

Inadequate: little or no attempt at connectivity,

though reader can deduce some organization,

55

Criteria
of
Mastery

Score

1
Spelling

5
4
3
2
1

Criterion of Scoring
individual ideas, may be clearer but very difficult to
deduce
Unacceptable: lack of organization to serve that
communication is seriously imparted.
Excellent: no error.
Good: 1 or 2 minor errors only.
Fair: several errors, not too hard to understand.
Inadequate: several errors, some interfere with
communication, some words very hard to recognize.
Unacceptable: numerous errors, hard to recognize,
several words communication made very difficult.

Based on the scoring guidance above, the maximum score is 25 for

there are five aspects and each is scored five. The scoring was based on the
analytical method. To get the score of each student, the scoring considered
the five components above. Each student should be able to write 10-15
sentences related to the topic and construct them into a report text and the
students score were multiplied by 4, so the maximum score was 100.
In classifying the score, the writer used the measurement of students
achievement suggested by Harris (1969:134) that is placing the score into
some categories as follows:
Table 3.4. Scoring Guidance by Harris
Grade
A
B
C
D
E
Less than 50

2. Try Out

Criteria of Mastery
91 100
81 90
71 80
61 70
51 60

Level
Excellent
Very good
Good
Fair
Poor
Very poor

56

The instrument to collect the data will clearly affect to the quality of
the data obtained. A good instrument is to fulfill two important
qualifications, they are valid and reliable. Therefore, to measure the
validity and reliability of the instrument it should be tried out first to the
students in another class. Try out is also important since try out is a kind
of pre-testing, which provides opportunities for the test-maker to try out
the test directions and to check the estimated time required to work the
items of the test. If the directions are not clear to the subjects, this should
certainly be noted at the time of pre-testing, so that the instructions can be
clarified in the final form (Harris, 1969:104). Thus, through try out, it
could be identified whether the test instrument is valid and reliable or not.
This try out was also to identify the appropriateness of the scoring system
applied on the instrument.
3. Pre-Test
Pre-test was given before doing the experiment in starting the data
collection to identify the writing report achievement. The test was in the
form of writing test. After having been explained about what they were
going to do, the students were given the instrument. They were to write the
text in 40 minutes.
4. Post-Test
Post-test was given after conducting all the activities during
treatment. It was to measure the students achievement in writing report
after the treatment. The test was also as the same as the pre-test, i.e.

57

writing report text with the same. The experimental group was shown a
movie before starting to do their post test. The control group, on the other
hand, was not given the similar method with the experimental one. The
students of the control group were songs to help them in doing post-test.
F. Research Instrument
Instrument is important thing in an experiment since it will affect on the
data obtained. As Kerlinger (1965: 118) states an instrument plays an
important role in a study in the sense that the reliability of data obtained.
Arikunto (2010: 203) also defines that research instrument is Alat atau
fasilitas yang digunakan oleh peneliti dalam pengumpulan data agar
pekerjaannya lebih mudah dan hasilnya lebih baik, dalam arti lebih cermat,
lengkap, dan sistematis sehingga lebih mudah diolah. (a device used by the
researcher while collecting data to make his work easier and get a better,
complete and systematic result in order to make the data easy to be processed).
It is important to use instrument in doing experiment. In this study, the
test was used as the instrument to obtain the data that is the students ability in
writing. Test can be defined as a set of techniques, procedures, or items of
measuring a person ability, knowledge, or performance in a given domain
(Brown, 2004:3).
The writer carried out the writing test in this research since the aim of
this study is to investigate the students achievement in writing after having
been given the treatment. The writer particularly chose writing composition
test to be implemented on this study. The writing of a composition is a task

58

which involves the students in manipulating words in grammatically correct

sentences and in linking those sentences to form a piece of continuous writing
which successfully communicate the writers thoughts and ideas on a certain
topics (Heaton, 1974:127). Having gathered the data, then the students result
was scored based on a scoring guidance.
Harris (1969: 13) states that all good tests possessed three qualities, i.e.
validity, reliability, and practicality. The test that the researcher use has to be
appropriate in terms of four objectives, dependable in the evidence it provides,
and applicable to our particular situation. Those characteristics of a good test
will be explained further below.
1. Validity of the Test
According to Harris (1969: 19-21) states that validity is distinguished
into three kinds. They are content validity, empirical validity, and face
validity. In this study, the writer concerns to content validity because
content validity is particularly important to achievement test. In content
validity, a valid instrument is an instrument which has high validity. On
the contrary, an invalid instrument is an instrument which has low validity.
The formula used in measuring the validity of an instrument
according to Arikunto (2007: 69) is by using Product Moment Formula
below:
rxy

N xy x y

N x x N y y

Where:

59

rxy : validity of the item

N

: total number of the respondent

: the sum of total item

: the sum of item score

xy : the sum of item score multiplying the total score

x2 : the sum of the square of the total item
y2 : the sum of the square of the item score
Then, the result of rXY is consulted to rtabel product moment with = 5%.
When rXY > rtabel the test item is considered to be valid and when rXY < rtabel,
it is invalid.
2. Reliability of the Test
Human error, subjectivity and bias may enter into scoring process, so
two raters are needed to score the test. According to Brown (2004: 21),
rater-reliability issues are not limited to context where two or more scores
are involved. According to Madsen (in Galuh 2009: 40), a reliable test is
one that produces essentially when the consistently on different occasions
when the conditions of the test remain the same. The reliability of the
variables, which follows:

r11

k
b2
=
1 2
k 1
t

60

In which:
r11

= instrument reliability

= number of test

b2 = question of test might show the reliability of test, used K-R.20

formula as variance
t2 = number of test variance.

The result of r11 is then compared with rtabel of product moment with =
0,05. The item is reliable if r11 is higher than rtabel or r11 < rtabel.
3. Item Analysis
According to Tinambunan (1988: 137), the function of item analysis
is to reexamine each item to find out its strength and weakness. This item
analysis involves analysis of index of item difficulty and index of
discriminating power.
a. Item Difficulty
Heaton (1975: 172) says, The index of difficulty of an item
simply shows how easy or difficult the particular item proved in the
test. It is needed to identify whether an item is difficult or easy one.
According to Arikunto (2007: 207) to calculate the index of difficulty
of an item, we can use the following formula which is found by
Dubois:
P=

B
JS

In which:

61

= index of difficulty

= The number of students who answered correctly

JS

= The total number of students

The index of difficulty can be classified as follow:

P < 0,30

: difficult

0,30 0,70 : medium

P > 0,70

: easy

b. Discriminating Power
The index of discriminating power is needed to identify whether
or not the item can be used to differentiate the more able students from
the less able students. Heaton (1975: 178) states the discrimination
index of an item indicates the extent to which the item discriminates
between the testees, separating the more able testees from the less
able.
According to Arikunto (2007: 213) the calculation of the
discrimination power can be formulated as follows:

D=

BA BB

J A JB

Note:

62

= Discrimination index

BA

= The number of students in the upper group who answered the

item correctly

BB

= The number of students in the lower group who answered the

item correctly

JA

= The number of students in upper group

JB

= The number of students in lower group

The classification of the discrimination index is presented below:

D < 0,0

: very poor

0,0 < D 0,20 : poor

0,20 < D 0,40 : fair
0,40 < D 0,70 : good
0,70 < D 1,00 : excellent
F. The Technique of Analyzing Data
As a requirement for the t-test, firstly the data have to be tested using
balanced, normality, and homogeneity tests. The discussion of normality test
and homogeneity test are as follows.

1. Balanced Test

63

Based on Winner in Budiono (2004: 176), before the experiment is

conducted by using t-test at the significance level of = 0,05 to find out
that experiment group 1 and experiment group 2 have the same or
balanced initial ability.
According to Budiono (2004: 185) the procedures of t-test are as
follows.
a. Determine the hypothesis
Ho : 1 = 2 = 3 = n

(all of population have same t-average

count)

H1 : 1 2 3 = n

(all of population not have same t-average

count)

b. Level of significancy : = 0,05

c. Statistical test

X2

S12 S 22

n1 n2

t=

fi ( xi X ) 2
S
n 1
2

(Sudjana, 2001: 95)

Note:
t

= the significant score of group

= average score of experiment group 1 test

= average score of experiment group 2 test

s12

= the variance of experiment group 1

s22

= the variance of experiment group 2

64

n1

= total number of the member of experiment group 1

n2

= total number of the member of experiment group 2

s2

= combined variance

= combined standard deviation

d. Critical area:
Df = {t|t < - t/2; n1 + n2 2 atau t > t/2; n1 + n2 2}
e. Decision : H0 rejected if t Df
f. Summary
Two populations is balanced if H0 accepted
(Budiyono, 2003: 151)
2. Normality Test
Based on Budiono (2004: 170), normality test is used to know
whether the data have normal distribution or not. In order to test the
normality, the researcher uses Lilliefors formula. The procedures are as
follows.
a. Having an observation of X1, X2, X3, Xn. The result X1, X2, X3,
Xn become standard number Z1, Z2, Z3, Zn by using the formula

Zi

Xi X
= ( X is the average while S is standard deviation of the
S

sample).
b. Using standard number distribution list for each of the standard
number, then count the probability F(Zi )= P(Z Zt).

65

c. Counting of proportion of Z1, Z2, Z3, Zn which less than or equal to

Zi. If the proportion stated by using S(Zi), so S(Zo) is the number of
Z1, Z2, Z3, Zn which is less than or equal to Zi divided by n.
d. Counting the result of F(Zi ) - S (Zi) then state the absolute value.
e. Taking the highest value among the absolute values from that result.
The maximum result of F(Zt ) - S (Zi) is Lo.
Criteria:
a. Lo (L obtained) > Lt (L table) = data

do

not

have

normal

distribution Lo
b. (L obtained) Lt (L table)

= data have normal distribution.

3. Homogeneity Test
Based on Winner in Budiono (2004: 176), homogeneity test is used
to know whether the data is homogeneous or not. In order to test the
homogeneity, the researcher use steps below:
a.

Computing the S of each variance (sample groups: experimental and

(X i ) 2
X
control group) with formula: 2
n
Si
n 1
2
1

b.

Computing the Smerge from the computing step 1 with formula:

(ni 1) S i2
S
(ni 1)
2

c.

Find the log S2

66

d.

Computing the B from step c with formula: B = (log s2) (ni-1)

e.

Computing the o2 with formula: 2 = (ln 10) {B (ni 1)log s i2 }

Where:
2

= the homogeneity test

= the total samples that is used

ni

= the total sample in each class.

= the total variance

Criteria:
1) o2 < table = the data are homogeneous.
2
2) o > table = the data are not homogeneous

4. Hypothesis test
The researcher uses t test to analyze the data of writing skill achieved
by the eighth grade students writing skill of SMPN 4 Madiun in the
academic year of 2013/2014 who have been taught using films and taught
using songs. The t test to analyze the data, with the following formula:
x1 x 2

t0
S

1
1

n1 n2

(Sudjana, 2001: 239)

Where:
t0 = significant figures the value of group
x1 = average of students writing score test who have been taught using
films (experiment group 1)

67

x2 = average of students writing score test who have been taught using
songs (experiment group 2)
S = standard deviation
n1 = the number of the experiment group 1
n2 = the number of the experiment group 2
with the determination criteria specified hypothesis, namely:
-

Hypotheses is rejected if t0 t (1 ; df)

Hypotheses is accepted if t0 > t (1 ; df)