Terbaru Dari Ku

CHAPTER III

RESEARCH METHOD

3.1 Research Design

Arikunto (2002, 45) states that research design is a plan or program made by a researcher, as the activity target that to do. So, the researcher also has made a planning for this research that is in this research, the researcher used factorial 2x2 research design. The design can be simplified at the following table:

Table 3.1 The analysis Varian Factorial 2x2 Design

Source: Basrowi (2010, 212)

Notes:

X₁ A₁   : The students of experimental class (using pair discussion) who have high self confidence.

X₁ A₂   : The students of experimental class (using pair discussion) who have low   self confidence.

X₂ A₁   : The students of control class (using repetition drill) who have high self     confidence.

X₂ A₂   : The students of control class (using repetition drill) who have low self      confidence.

This experimental technique deals with two classes: one is an experimental class and another is a control class. Each of classes receives pre-test, treatment and post-test in order to find the progress of students’ speaking ability. The treatment will be conducted for about three times. There are two groups of the research, one as experimental group using Pair Discussion technique and another as the control group using Repetition Drill technique.

A variable is defined by Arikunto (2002, 98) as something that varies from one case to another. In addition, Sudirman (2009, 34) states that variable is something that has variation of score. The dependent variable is variable which one observes and measures to determine the effect of the independent variable. Independent variable (the major variable) is the variable which is selected, manipulated, and measured by the researcher.

In this research, the researcher uses three kinds of variable, they are: independent variable (X), attribute variable and dependent variable (Y). In this research, the independent variable is pair discussion of cooperative learning method (X₁) and repetition drill of audio lingual method (X₂), self confidenceas an attribute veriable, and the dependent variable is speaking ability (Y).

The operational definition of speaking ability is the students’ speaking ability score gotten after following the test. To get the score from the students, he gives the test that consist of the materials as follow: 1) expression love; 2) expression sorrow; 3) expression of showing attention. Whereas the aspect of scoring consist of content, pronouncation, intonation, fluency, and comprehenssion.

Pair discussion is a technique of acting in which the actor recalls emotions and reactions of the problem and it is done by two people (in pair). The reactions come from the past experience and use them in identifying with other people. The technique to give the argument, opinion or statement based on the problem. In other word, discussion is the act of discussing or exchanging reasons; examination by argument; debate; disputation; agitation. The topic for a pair discussion can be anything, from contrasting ways of solving problems, to the reason a science experiment failed to produce the predicted result.

Pair discussion technique is a technique used by the researcher to increase students’ speaking ability in this research by using the following procedures: 1) Give the explanation to the students about the materials are about expression love, sorrow and showing attention. 2) Ask the students to choose their couple. 3) Ask the students to make coversation based on the material by each couple. 4) Ask the students to discuss about the material with their couple.

Repetition drill technique in this research is a technique used by the researcher to increase students’ speaking ability by using the following procedures: 1) Give the example conversation of the material. 2) Give modeling about how to pronounce the words or sentences. 3) Ask the students to repeat quickly. 4) Ask the students to practice by repeating the sentences. 5) Give explanation about the example and material.

Self confidence is someone’s feeling that can make them sure to do everything. It comes from their heart which always positive and gives motivation. In other word, self confidence can be definited as a believable of own capacity and skill to do everything that we can. When someone has self confidence, he always thinks positively and believes that he can do it well.

The operational definition of self confidence is the score of self confidence gotten from the students after answering the questions. The questions of the questionnaire consist of some aspects, as follow: 1) Belief and self prosperity; 2) Optimism; 3) Objective; 4) Responsibility; and 5) Rationalism and realistic.

Arikunto (2002:108) says that population is people or other things discussed in the research. In other words, population is the total number of the subjects of an investigation. In this research, the population is all of the students at the eleventh grade of SMA N 1 Rumbia. The population consists of six classes, and every class, consists of 29 up to 34 students. So, the population of the research is 195 students they are 121 girls and 74 boys.

Arikunto (2002:108) states that sample is a part of population that is observed. Further, Sugiyono (2010:81) says that sample is a part of whole and characteristic from population itself. Base on the definition, the researcher determines the sample of the research which is taken from the population. The sample is taken by using a sampling technique called by cluster random sampling. The sampling technique is done with the steps as follow:

The research instruments are used in this research as follow:

To find out  the result of sudents’ speaking ability by using pair discussion and repetition drill technique, he gives oral test to students. And then, to find out the students’ self confidence, a questionnaire is given to students. The questionnaire consists of 30 items.

Based on the operational definition of students speaking ability, the researcher gives the test by using instrument which has been determined. He gives oral test to students as an instrument of speaking ability. The test includes the material which has been explained. The material is about expressing love, sorrow, and attention.

In this test, the students have to perform in front of the class. The test is done in pair. It means that every student must perform the conversation with their friend. Every couple has 3 up to 5 minutes to make conversation. They have to make conversation about love, sorrow, and attention. Although speaking performance is done in pair, the assessment is done individually. The specification table of instrument is follow:

Table 3.2 The specification table of instrument of students speaking ability

Source: Syllabus for Senor High School

3.4.2 Self confidence’s questionnaire

To know the students’ level of self confidence, the researcher uses questionnaire as an instrument. The questionnaire is given to students before giving treatment. After giving questionnaire, the result of questionnaire is classified into two parts, high and low self confidence.

The instrument of self confidence includes some aspects. They are about Students’ belief and self prosperity, optimism, objective, responsibility, rational and realistic. The total of questions is 30 items. The questions’ form is multiple choices. So, the students must choose the best choice according to themselves. The specification table of instrument is follow:

Table 3.3 The specification table of instrument of students self confidence

Adapted from Lauster (1997)

According to Basrowi (2010:17), a measurement can be said to have validity, when the content of the measurement is eligible to measure an object which should be measured and suitable with the fixed criteria. Validity means an accuracy of a measurement in measuring data. In this research, the researcher uses two kinds of validity test to validate the speaking instrument those are content validity and construct validity.

Based on Basrowi’s opinion (2010:17), he states that a measurement can be said to have validity, when the content of the measurement is eligible to measure an object which should be measured and suitable with the fixed criteria. Validity means an accuracy of a measurement in measuring data. In this research, the researcher uses validity test to know the validity of questionaire instrument of self confidence. The validity test is construct validity. The questionaire of self confidence is check by the expert. And then, it will be validated by the expert if the questionaire has been relevant. The questionnaire of self confidence is validated by two people from guidance and counselor program and another one from English program.

3.4.2.3 Reliability of speaking instrument

ReferringBasrowi (2010:29) reliability is one that produced essential the same result consistently on different occasion when the condition of the test remains the same, reliability is also known as a measurement that has reliable when that measurement is used to measure the data frequently but the result keep in same. In line with that statement, reliability is one that produced essential the same result consistently in different occasion when the condition of the test remains the same.

In gaining the reliability of the test, the researcher conducted try out of speaking on January 2013. The scores of try out will be used to know the reliability of speaking instrument. Then, the researcher uses the formula to know the speaking instrument. The researcher uses reliability of odd even split and the formula of Pearson’s product moment.

The formula is:

 =

Notes :

     : The coefficient correlation between X variable and Y variable

X      : The score of odd score

Y      : The score of even semester

      : The quadrate score of the odd score

    : The quadrate score of the even score

     : The  score of X * Y

(Arikunto,2011,70)

To find reliability of the test, the researcher will use the spearman brown (Split Half).

The formula as follow:

r_i=    2r _b

       1 + r_b

Where:

3.4.2.4 Reliability of self confidence instrument

ReferringBasrowi (2010:29) reliability is one that produced essential the same result consistently on different occasion when the condition of the test remains the same, reliability is also known as a measurement that has reliable when that measurement is used to measure the data frequently but the result keep in same. In line with that statement, reliability is one that produced essential the same result consistently in different occasion when the condition of the test remains the same.

In gaining the reliability of the test, the researcher conducts try out of self confidence instrument on January2013. The scores of try out will be used to know the reliability of self confidence instrument. Then, the researcher uses the formula to know self confidence instrument. The researcher uses reliability of odd even split and the formula of Pearson’s product moment.

The formula is:

 =

Notes :

     : The coefficient correlation between X variable and Y variable

X      : The score of odd score

Y      : The score of even semester

      : The quadrate score of the odd score

    : The quadrate score of the even score

     : The  score of X * Y

(Arikunto,2011,70)

To find reliability of the test, the researcher will use the spearman brown ( Split Half).

The formula as follow:

r_i=    2r_b

       1 + r_b

Where:

In collecting data, the researcher uses the test of each variable of the research. In this research, the researcher gives two kinds of test to the students as a sample. The test consist of oral test (for speaking ability) and questionnaire test of students self confidence. To collect the data, the researcher uses pre-test and post-test.

To collect the data in this pre-test, the researcher does the steps as follow:

First, the pre-test for speaking ability. The researcher gives a conversation to students. And then, they must perform the conversation in front of the class. They perform in front of the class by couple. The test form is oral test and the material is about expressing love, sorrow, and attention.

Second, the pre-test for students’ self confidence. In this test, the researcher gives test to students which formed questionnaire test. The test consist of 30 questions, where the students have to choose one of the choice among 0, 1, 2, 3, or 4. This test is to know the difference level of students’ self confidence in the class.

In the post-test, the researcher gives the test to students about speaking ability. The purpose of the test is to know the result of students’ learning. In this test, the researcher gives a conversation which has some blank parts. The first, the students must complete the conversation by using expression love, sorrow, or attention. Then, they have to understand the conversation. After that, the students have to perform the conversation in front of the class in pair. The couple will be determined randomly before doing the test. This test is done when the end of the instructional learning.

The purpose of normality test is to know the data distributed normally. Some of statistic technique especially parametric statistic requires that the data has to follow normal distribution form.

Normality test can be done by three types that are use parametric statistic test (frequence test) non-parametric statistic test (kolmogrov smirnow test) and usegraph. The normality test which is used by the researcher is based on the kind of experiment which is done. If the research test which determine the specific qualifications about the population parameter which be a sample, the analysis that has to be used is paramitric statistic analysis method. Whereas, if the research without determine the specific qualification about the population parameter which be a sample, so the analysis use non-parametric statistic analysis method.

Normality tests are usually used to determine to know whether a data set is well-modeled by a normal distribution or not, or to calculate how likely an underlying random variable is to be normally distributed. P-value that will be provided by SPSS which is the principal goodness of fit test for normal and uniform data sets will be computed to test the normality.

In order to make the data which will be taken normal, the sample of the research has to be made balance. Because of that, the researcher uses data reduction to make the data of the sample is normal. It is done by omitting some data of students’ self confidence. Some data will be omitted randomly. The data of sample in each class will be made balance base on the classification of self confidence. The result of data reduction will be the sample of the research.

Basrowi (2010:98) states:

Homogeneity is a measurement which can be used to determine data variation. There are so many ways which can be used to measure the homogeneity of a sample, such as by using explore analysis test and analysis test one way ANOVA. In this case the writer uses model of One-Way ANOVA analysis test. The kind of this test is used to determine the mean of two or more groups manifestly.

In this research, the researcher uses the ANOVA test called Univariate: Analysis Varian Factorial Design. The data will be analyzed by using SPSS program to know whether the data homogeny or not. If hypothesis of the research as follow:

Ho       : Four of the models have variance score which not homogeny

H₁        :  Four of the models have variance score which homogeny.

The criterion:

If sig. score < α (0,05), Ho is accepted

If sig. score > α (0,05), Ho is not accepted

The data also can be calculated by using test of homogenity of variances, the formula:

            F =

Where :

F : the homogenity of variance

S : Devisiation standard

Criteria:

If score Sig. < 0,05 so H_o is accepted

If score Sig. > 0,05 so H_o is not accepted

This test is used to know whether the hypotheses proposed by the writer are proved or not. The formula used in this test is Varian Analysis Test (ANOVA). This test is commonly used to test the hypothesis of the mean comparative sample when the data in theform of interval or ratio (Basrowi, 2003:204).

In this research, the researcher uses the ANOVA test called Univariate: Analysis Varian Factorial Design. This design is used to analyze two data clusters by following the controlling variable (Basrowi, 2010:212).

The criterion:

If score of probability (sig.) > 0,05, so Ho is accepted

If score of probability (sig.) < 0,05, so Ho is not accepted

Besides that, hypothesis test will be analyzed by using SPSS program of analyze compares mean.

The criterion:

If score of sig. (2-tailed) > 0,025, so Ho is accepted

If score of sig. (2-tailed) < 0,025, so Ho is not accepted
Baca Juga Bagian Lainnya :
Halaman Judul & Abstrak 
Halaman BAB I 
Halaman BAB II 
Halaman BAB III 
Halaman BAB IV 
Halaman BAB V