33

CHAPTER III

RESEARCH METHOD

A. The Research Design

In conducting the research, the researcher uses Quantitative Research

Method with the general type is experimental method. According to Kothari

(2004:3), “ Quantitative research is based on the measurement of quantity or

amount. It is applicable to phenomena that can be express in term of quantity”.

Regarding to the definition, the researcher use quasi experimental design because

it is more flexible and use previously in teaching learning process. Anderson (

2005:263) says, “ Quasi experimental is a form of research that examines

differences between research groups based on some natural characteristic using

treatments or interventions, but not randomization”.

In applying quasi experimental design, the researcher will divide the

sample into two groups. They are experimental group and control group. Pretest is

conducted for both of them. The researcher treats Time Token Strategy in

experimental group while for control group the researcher use conventional

teaching method. After that, the researcher conducts posttest for both groups to

find the effect of Time Token Strategy on the students’ ability in speaking. For

more understanding, the research design can be shown in the table on the next

page.

33
 34

Table 5

RESEARCH DESIGN

Class Pretest Treatment Posttest

Experimental Group T1(e) X(e) T2(e)

Control Group T1(c) X(c) T2(c)

In which:

T1 (e) = Giving the pretest to the experimental group

T1 (c) = Giving the pretest to the control group
X (e) = Teaching by using Time Token Strategy
X (c) = Teaching by using Conventional Teaching Method
T2 (e) = Giving the posttest to experimental group
T2 (c) = Giving the posttest to control group.

B. The Research Variables

There are two variables in this research. This variables are Time Token

Strategy as independent variable (X) and the students’ ability in speaking as the

dependent variable (Y). The two variables are use by the researcher in conducting

the research to know the effect of Time Token Strategy on the students’ ability in

speaking at the eighth grade of SMP Negeri 1 Sogae’adu in 2018/2019.

C. Population and Sample

1. Population

In conducting the research, the researcher needs the population.

According toWirodikromo (2001:3) defines the population as all objects of a

defined group that study for collecting the information on data driven decision. In

addition, Anggoro, et al., (2011:53) assert that the population is the total numbers

of individuals which their characteristics would be searched. Furthermore,

 35

Herhyanto, et al, (2014:1.10) confirm that population is covering all the members

of the defined group as the target of the research. Briefly, the researcher has to

determine the population based on the researcher’s observation.

The population of the research is the eighth grade of SMP Negeri 1

Sogae’adu in 2018/2019 that consists of50 students. Clearly, the total of the

research population can be seen in the following table.

Table 6

THE POPULATION OF THE EIGHTH GRADE STUDENTS OF SMP NEGERI

1 SOGAE’ADU IN 2018/2019

No Grade Total
1 VIII-A 25
2 VIII-B 25
Total 50 Students
Source: Office Administration of SMP Negeri 1Sogae’adu in 2018/2019

2. Sample

Sample is a part of the population that will represent the characteristic of

population. According to Herhyanto, et al., (2014:1.10), sample as amount of

partly members of population. The existence of sample in research takes a very

important role because basically it determines the quality of the researcher itself.

Furthermore, Wirodikromo (2001:3) defines sample is part of population. It

means that, sample is part of population that will describe and denote the students’

characteristics.

In choosing sample of the research, the researcher use the cluster random

sampling to decide the sample. As Anggoro, et al. (2011:4.7) say that on cluster

random sampling, the sample is not selected individually directly, but it is select
 36

through random groups. The procedures to select the research sample through

cluster random sampling as follows.

a) Identify the research population.

The research population is 50 students of the eighth grade SMP Negeri 1

Sogae’adu.

b) Identify the desired sample size.

The desired sample size is 50 students which considered as fulfillment of a

condition of the chosen of statistic non parametric form. So the size of each

sample should be even (≤30).

c) Identify and define the logical cluster.

The logical cluster of the research is classroom.

d) List the entire cluster (or obtain list) that make up the population of clusters.

The list of cluster can be seen on Table 6. The clusters consist of twogroups.

e) Estimate the average number of the population of clusters.

𝑇ℎ𝑒 𝑎𝑣𝑒𝑟𝑎𝑔𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑚𝑒𝑚𝑏𝑒𝑟 𝑝𝑒𝑟 𝑐𝑙𝑢𝑠𝑡𝑒𝑟

𝑇ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑟𝑒𝑠𝑒𝑎𝑟𝑐ℎ 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛

=
𝑇ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑙𝑢𝑠𝑡𝑒𝑟

f) Determine the number of cluster needed by dividing the sample size by the

estimated size of a cluster.

𝑇ℎ𝑒 𝑠𝑎𝑚𝑝𝑙𝑒 𝑠𝑖𝑧𝑒

𝑇ℎ𝑒 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑐𝑙𝑢𝑠𝑡𝑒𝑟 𝑛𝑒𝑒𝑑𝑒𝑑 =
𝑇ℎ𝑒 𝑠𝑖𝑧𝑒 𝑜𝑓 𝑐𝑙𝑢𝑠𝑡𝑒𝑟

g) Select the numbers of cluster randomly.

Therefore, the researcher will use lottery to random for selecting control

and experimental group. The researcher makes the lottery as many the classroom

with makes number 1 to 2, then asks the representation from each classroom to
 37

take one lottery. So, the classroom which gets number 1 as a control group while

the class which gets number 2 as an experimental group.

D. Data and Research Instrument

In this research, the kind of the data use is the primary data. Herrhyanto, et

al., (2014:1.4) define that the primary data is a data obtained directly by the

certain organizations, includes the researcher. So, in the research, the primary data

as the sources has takes from the students at the eighth grade of SMP Negeri 1

Sogae’adu directly.

The instrument that use to get the data by the researcher in conducting the

research is by using Handycam. So, to know the students’ ability in speaking with

the material about asking and giving opinion, the researcher performs test that

consists of two kinds, namely pretest and posttest. Pretest will give to the students

before they get the treatment while posttest will give to the students after they get

treatment.

E. Procedures of Collecting the Data

In conducting the data collection, the researcher determines a control

group and an experimental group. Then, the researcher follows some procedures

of collecting data, as follows:

1. Finding the location. The location of the research is SMP Negeri 1

Sogae’adu.

2. Selecting the population. The population of the research is the eighth

grade of SMP Negeri 1 Sogae’adu.

 38

3. Selecting sample. The sample in the research are control group and

experimental group.

4. Holding the pretest on experimental group and control group.

5. Holding the treatment by using different strategy in which Time Token

Strategy will be used to the experimental group and conventional

teaching method or traditional method will be use to control group.

6. Holding the post test on experimental group and control group.

7. After doing it, the researcher analyzes the result by using the test.

8. Analyzing the data, getting finding and drawing the conclusion.

F. The Techniques of Analyzing the Data

1. Testing of Instruments

a. Validity

Validity means that a tool measures what it sets out to measure. The

instrument use by the researcher in conducting the research by using sounds

handycam and the test is oral test.

The step that is done by the researcher to know the validity of the

instrument is conducting the content validity of the instrument.

Kothari (2004:14) states,

Content validity is extent to which a measuring instrument can be

determinedby using a panel of persons who shall judge how well the
measuringinstrument meets the standards, but there is no numerical way
to express it. Kind of validity in this research is internal validity.The
internal validity of a research design isits ability to measure what it aims
to measure. The instrument is validated by three validators, one person of
the lecturer of the English Department, two person of the English
teachers.
 39

b. Reliability

Reliability means the results of the measurement which has been done by

using the test continuous to the same subject and lastly, it always shows the same

or stability result.

Sugiyono (2011:122) says that a reliable instrument is not definitely valid,

but a valid instrument is trusted reliable. So, reliable measuring instrument does

contribute to validity, but a reliable instrument needs not be a valid instrument,

the researcher just conducts content validity to the items, as means the test is

absolutely reliable.

2. The Data Analysis

a. The Students’ Ability in Speaking

In analyzing the students’ ability in speaking, the researcher firstly makes

the phonetic transcription to find out whether the students pronounce the words

correctly or not.

The researcher follows the aspect of scoring of Hughes (2003) state that

there are five criteria levels should be pay attention, they are: accent, grammar,

vocabulary, fluency, and comprehension (more elaboration can be seen in

Chapter II).

Score: Accent + Grammar + Vocabulary + Fluency + Comprehension =

𝑜𝑏𝑡𝑎𝑖𝑛𝑒𝑑 𝑠𝑐𝑜𝑟𝑒
𝑣𝑎𝑙𝑢𝑒 = 𝑥100
𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑠𝑐𝑜𝑟𝑒

After that the researcher describes the level of the students’ ability in

speakingas Cartier in Nurgiyantoro (2010:363)classifiesthat 0-39 is fail, 40-59 is

less, 0-74 is enough, 75-84 is good, 85-00 is very good.

 40

b) Mean Score

In getting the mean score and standard deviation of the students’ test result

even in experimental group and control group, the researcher seek sit by using the

formula from Sudjana (2005:109) as follows.

𝑋
̅
X=
𝑁

In which:

̅ = Mean
X
X = Thesum of all scores
N = Thenumbers of sample

Based on formula above, the researcher use formula of mean score in pretest

and posttest.

c) Standard Deviation

Besides, to get the standard deviation, the researcher seek sit by using the

formula from Sudjana (2005:114):

√∑(𝑋−𝑋)2
S=
N−1

In which:
S = Standard deviation of the sample
X = The score of item
̅ =
X The mean of the data
N = The number of the sample

Based on formula above, the researcher uses formula of deviation in pre

test and post test.

 41

d) Variance

Variant is the quadrate from standard deviation. To determine the variant,

the researcher uses the formula suggested by Djiwandono (2008:217):

2
2
∑(X − ̅
X)
S =
N

Which:

S2 = Variance
X = The score of item
X̅ = The mean for data
N = The number of the sample

Based on formula above, the researcher uses formula of variance in pre

test andpost test.

e) Normality of the Data

In examining the normality in functions to know whether the sample

comes from population which has the normal distribution or not, the researcher

seeksit by using Liliefors’ formula as Herhyanto, et al., (2014:8.17) suggest as

follows.

(1) Make the work table as give next.

(2) Input the data or score in the table.

(3) Find the Z score by using the formula:

Xi− X
Zi=
S X

In Which.

Zi = the standard score

Xi = the score of i
S = the standard deviation
 42

X = the mean score

(4) Determine the F (zi) through Table Standard Normal based on the Z score.

𝐹 𝑐𝑢𝑚𝑚𝑢𝑙𝑎𝑡𝑖𝑣𝑒
(5) Determine the S (z) by using the formula: S(z) =
𝑁

(6) Count the quarrel between F(z) and S(z) by using the formula.

Lc = │F(z) – S(z)│(decide the highest one as Liliefors count (L0).

(7) Determine the score of Liliefors table (Lt) with 𝛼 = 0,05.

(8) Compare Liliefors count and Lilieforstable, then take the conclusion like as.

(a) If L0< Lt, so the data is normal and the population has the normal

distribution.

(b) If L0> Lt, so the data is not normal and the population has not normal

distribution.

Table 7

WORK TABLE FOR LILIEFORS FORMULA

fi xi zi F (Zi) S (zi) F(zi)-S(zi) │F(zi)-S(zi)│

Lcount
Ltable
Conclusion

f) Homogeneity of the Sample

To find the homogeneity of the sample, the researcher seeks it by using

Harley formula as suggested by Irianto, et al., (2007:276) through several steps as

follows.

(1) Assure that population has normal distribution.

 43

(2) Count the mean score of the experimental group and control group.

(3) Count the variance of the both groups through formula like as.

∑(𝑋 − 𝑋̅) 2
𝑆2 =
𝑁−1

(4) Determine Fcountwith formula.

𝐻𝑖𝑔ℎ𝑒𝑠𝑡 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒
𝐹𝑐𝑜𝑢𝑛𝑡 = ,
𝐿𝑜𝑤𝑒𝑠𝑡 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒

(5) Determine Ftable through formula.

𝐹 𝑡𝑎𝑏𝑙𝑒 = 𝐹𝛼 (𝑛 ℎ𝑖𝑔ℎ𝑒𝑠𝑡 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 − 1, 𝑛 𝑙𝑜𝑤𝑒𝑠𝑡 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 − 1),

then confirm it on the table of F distribution critics value.

(6) Compare Fcount and Ftableand get decision on it as follows.

If Fcount≤ Ftable, so the both samples are homogeny.

If Fcount>Ftable, so the both samples are not homogeny.

g) Hypotheses Testing

The result of the both groups (experimental and control group) is used by

the researcher to tests the hypotheses such as:

Ha : There is a significant effect of Time Token Strategyonthe students’

ability in speaking at the eighth grade of SMP Negeri 1 Sogae’adu in

2018/2019.

H0 : There is no any significant effect of Time Token Strategy on thestudents’

ability in speaking at the eighth grade of SMP Negeri 1 Sogae’adu in

2018/2019.

If the result of the post test data shows the normal distribution and

homogeny, the researcher uses Chi Square (X2) test many samples as the kind of
 44

non parametric statistic because the number of sample in each group in this

research is less than 30(≤ 30).The researcher uses the following formula suggested

by Sugiyono (2006:104):

(𝑓0 −𝑓ℎ )2
𝑋2 = ∑ ; dk= k-1
𝑓ℎ

Which :

𝑓 𝑓 ×𝑓
ℎ= 𝑘𝑛 𝑏
𝑡

In Which:

X2 = Chi Square test

fo = The frequency that observed
fh = The frequency that hoped
fk = The number of frequency based on the column
fb = The number of frequency based on the row
nt = The total number of sample group

In the overall activity, the researcher confirms the Chi Square (X2) which

X2tablevalue of X2 distribution of significance dk=k – 1. Determining the criteria of

testing which Ho is accepted and Ha is rejected if X2α(dk)<X2<X2α(dk)<and Ho is

rejected and Ha is accepted in the other condition.

If the result is not homogeneous, so the researcher will use Chi Square

one sample formula, and it means Ho is refused and Ha is accepted.

 45

BIBLIOGRAPHY

Aguilesssra,LuciaCarelandFilologia,LienciadaEn,2012.ProductiveLanguageskills

learningandteaching:SpeakingandWriting,PublicacionesDidacticas,

Ingles

Alshatti, (2013).The Effectiveness of Using Time Token Strategy to Improve

Reading Comprehension for the Eighth Grade Students in Communication

Skills Book. Jordan

Andrade. 2000. Language Level.Department of Education and Training,

Melbourne, Australia.

Bashir, dkk, 2011, Factor Effecting Students’ English Speaking Skills, Vol. 2, No.

1, 34-50, University of Education, UK.

Brown, 2004, Language Assessment: Principles and Classroom Practices,

Pearson Education, Inc, NY.

DebbieCole,ChristineEllis,BarbaraMason,JohnMeed,DeborahRecord,Anna

Rossetti,GrahamWillcocks,2007.TeachingSpeakingandListening:a

Toolkitforparticioner,PortishedPress,Bristol,England
 46

De Bono, Edward. 1995. Six Thinking Hat.The International

BestSeller.GranicaEditions.Mica Management Inc. Resources. England

De Bono, E. 1999. Six Thinking Hats.MICA ,Management Resources ,Inc

Ferrance, Eileen. 2010. Action Research. Themes In Education. Northeast and

Islands Regional Educational Laboratory At Brown University. New

England, New York, Puerto Rico, and the Virgin Islands.

Harmer, J. 2011.The Practice of English Language Teaching. Edinburgh: Pearson

Education limited.

Hughes, Artur, 2003, Testing for Language Teachers, Cambridge University

Press, New York.

Illinois, 1994, New Ways in Teaching Speaking. Pantagraph Printing,

Bloomington. USA

Kasbolah, Kasihani & Wayan Sukarnyana. 2006. Classroom Action Research.

University. Malang.

Koşar & Bedirre. 2014. Speaking Skill. Working the American Way: How to

Communicate Successfully with Americans At Work, UK.

 47

Manktelow,J . 2010.The Mind Tools E-book. Excellent skills for an excellent

career. UK.

McAleer, F. 2006. Put On Six Thinking Hats. Pennsylvania Association for

Supervision and Curriculum Development. Pennsylvania, U.S.A.

Nurgiantoro, Burhan. 2001. Assessing. Yogyakarta: BPFE.

Porto. 2012. Six Thinking Hats for Electricity Markets Simulation. Adapting

Meeting Tools to Agent Decision.Knowledge-based and Decision Support

Technologies.Isep

Professional Development Services for Teacher .2011.Active

LearningMethodologies (Revised Edition).

Proctor, Tony, 2014, Creative Problem Solving for Managers. Developing Skills

For Decision Making and Innovation. Fourth Edition. Florence Production

Ltd, Stoodleigh, Devon, UK

Richard,JackC,2008.TeachingListeningandSpeaking:FromtheorytoPractice,

CambridgeUniversityPress,USA

Rivers, W. 2009.Speaking Skill. Modern Language Journal 50(4): 196-204.

 48

Salem, (2010).“The Effect of Using the Six Thinking Hats Method on the

Development of Eleventh Grade Students' Writing Skill in Southern Al-

Mazar”.Southern

Sanjani, (2015). “Improving the Students’ Ability in Speaking by Using Time

Token Strategy at the eighth grade of SMP N. Karangmojo in 2014/2015”.

Karangmojo

Singh, Yoseph Kumar. 2006. Fundamental of Research Methodology and

Statistic. New Age International.

TimPenulis,2008.PedomanPenulisanSkripsi(PPS),IKIPGunungsitoli,

Gunungsitoli