chapter3

CHAPTER III

RESEARCH METHODOLOGY

3.1 Research Design

According to Nazir (2005:84) design of study is all process that is needed in research planning and implementation. In limited meaning, design of study can be said as data collecting and analysis process. There are some techniques of conducting research which guide an enable the researcher to gather and analyze data. The present study is a descriptive correlation research because it deals with the description and interpretation whether there is or not a correlation between variables. Based on Nazir (2005:123) variable is concept that has many values. In this case the variables are students’ mathematics achievement that is taken from students’ report score (raport) and students’ English grammar achievement on second year students of MTs Nurul Islam Slatri Brebes.
This correlation research tries to describe the relationship between the variables by using coefficient of correlation, because such the writer has explained in chapter one, mathematic can influence English grammar based on their characteristics. In this research mathematics' achievement as independent (X) variable, while English grammar's achievement as dependent (Y) variable. The data of students' achievements in mathematics and English grammar are taken by using questionnaire.


3.2 Populations and Sample
One of research steps that also important is choosing the source of the data. According to Arikunto (2006:129) data source is the subject where the data is obtained. This data source is including population and sample that will be explained below.
3.2.1 Population
Such is stated in http://infinity.cos.edu//faculty/ A population is a collection of data whose properties are analyzed. The population is the complete collection to be studied; it contains all subjects of interest.
According to Nazir (2005:271) population is collection of individual with quality and characteristics that have been determined. The population of the study is the second year of MTs Nurul Islam Slatri Brebes. There are three classes namely; VIII A, VIII B and VIII C. The whole number is 100 students which consist of; 51 students of VIII A, 49 students of VIII B.
3.2.2 Sample
A sample is a part of the population of interest, a sub-collection selected from a population (http://infinity.cos.edu//faculty/). According to Nazir (2005:271) sample is part of population. So sample is taken from the population that has been determined before. In choosing the sample, the writer is using random sample. Because the writer chooses the sample is not based on personal consideration, but based on probability application.
Because the number of population is 100 students, so to minimize the sampling error and get precise result, the writer choose 50 students from 100 students of second year students in MTs Nurul Islam. That consists of 26 students from A class and 24 students of B class.
However, for interest only, there is a list of the recommended sample sizes for different populations. The table below, from a book on surveys, indicates some recommended sample sizes:

Recommended Sample Sizes for Different Populations and Permissible Sampling Errors 

Sampling Error Allowed
Population Size



500
1000
10,000
100,000
1 million


±10
83
93
99
100
100


±5
222
286
385
398
400


±4
250
385
588
621
625


±3
250
500
1000
1099
1111


±2
250
500
2000
2439
2500


±1
250
500
5000
9091
10,000


3.3 Instruments
In order to collect the data for this study, the instruments that are used are documentation and achievement test.




a.Documentation
The first instrument is documentation; according to Arikunto (2006:158) documentation is written things, it means that data that is taken from book or report that, or in the wider meaning documentation is not only in writing shape but also can be inheritance things, such as prasasti or archaeological symbols. This instrument is to get data about students’ achievement on mathematics. This data is taken from students’ semester report of second year students at MTs Nurul Islam Slatri Kab. Brebes . This mathematics’ score is as independent variable (X).
b.Achievement test
According to Arikunto (2006:151) Achievement test is a test that is used to measure someone achievement after learning something. In this case this achievement test is intended to get the information how far the students’ comprehensions and English grammar are. The way to get it is by taking English grammar test on second year students of MTs Nurul Islam Slatri Kab. Brebes.
The achievement tests consist of Questions of English grammar; this test is held to know how far students’ comprehension on English grammar is or to know the progress towards the instructional objectives of study. The English grammar score is as a dependent variable (Y).



3.4 Procedure of Data Collection

To organize the data obtained, we need certain steps. It needs valid data to be analyzed in order to get the calculation at the exact condition. That is way, in this research, the data collection have to be done first. The data that can be analyzed in this research should have the following qualification. There are two data in this study. They are mathematics’ score and English grammar score. Both score are taken from the result of the tests that that are held by the writer on second year students of MTs Nurul Islam Slatri Kab. Brebes.
Having been collected the data. The data were listed on two groups. This has purpose to make data analysis easier.
3.5. Data Analysis

This data is an attempt to find out whether there is or not a correlation between students’ achievement in mathematics and their English grammar achievement on the second year students of MTs Nurul Islam Slatri, Kab. Brebes. To compute the two variables the Pearson product moment is used. Because Pearson Product Moment Correlation (Bivariate Statistic) used to describe the strength of relationship between two variables. Beside that its formula is most stable technique that has smallest standard error. Formula used in this technique is:





r : correlation coefficient
N : number of sample
X : mathematics score
Y : English grammar score
The result of above computation will show the significance of the correlation.

4.1.1 Validity Test
Before we try to find out the correlation between mathematics and English grammar score, first of all we have to find the validity of English grammar. This is because the writer uses test questions that are made by him, so to know that the questions are qualified, the writer should find the validity of the test.
According to Arikunto (2006:168) validity is a parameter that shows levels of validity or truth of an instrument. A valid instrument has a high validity; otherwise, a less valid instrument means has low validity. The way to measure the validity of test is by giving two tests that consist of same questions or materials to the sample that has been chosen, and the test time is held in different time.
The computation of the coefficient of correlation using Pearson Product Moment correlation formula resulted in the coefficient of correlation is indicated in the following table, which consists of English grammar scores and its analysis. Score from the first test of English grammar is as independent variable (X), while English grammar score from the second test (Y).
Data are known as follows:

From the data above we can find the validity of the test by using Pearson’s Correlation product moment.

r : Coefficient correlation
N : Number of sample
X : fisrt English grammar score
Y : second English grammar score

After we get the result, the next step is to compare the result to the value of r in the table. In the table we find that: (1) on the significance level 5%: 0, 297 (2) on the significance level 1%: 0, 361. We find that r value above is bigger than r value in the table (0, 297 (0, 7606) 0, 361), it means that there is significance correlation between X variable and Y variable and it significance because right after coma there is no other nil. And the correlation index is big because the value is close to 1, 000.
Conclusion: because there is positive significance correlation, so formative test that is being tested the comparison validity can be said as valid test.