Population and Sample

B. Population and Sample

1. Population According to Gay & Airasian (2000:122) population is the group of interest to the researcher, the group to which she or he would like the result of the study to be generalized. Sugiyono (2013:119) also states that population is generalization region consisting of objects/subjects that have certain qualities and characteristics defined by the researchers to learn and then drawn the conclusions. So population is not just for person, but also everything included in research.

The population of this research was students at X MIPA of Senior High School 1 Bukit Sundi in academic year 2016/2017. They were 125 students spread into four classes. It shows on the table below:

Table 3.2 Population of the Research

No

Class

Total students

1 X MIPA 1

2 X MIPA 2

3 X MIPA 3

4 X MIPA 4

Total

2. Sample The sample is representative of population. Sugiyono (2013:120) states that sample is part of the number and characteristics possessed by population. A sample comprised the individuals, items, or events selected from a large group referred to as a population. It means, when the population is large, so researcher may not learn everything included in population, because of some limitations of the researcher. So the researcher can use sample took from population.

Since the total of population was very large, the researcher used cluster sampling. Cluster sampling random selects group, not individuals. All the member of selected groups has similar characteristics . They have similar teacher, similar material, and similar technique and strategy. By taking two of four lots randomly, the samples that were chosen become class experiment and class control.

To get representative sample of this research, the researcher did some steps:

1. Collecting the midterm test scores of the students at grade X MIPA from English teacher. (See Appendix)

2. Test of normality, test of normality has an objective to know the population is normal or not. The normality was analyzed by using SPSS and was used Kolmogrov Smirnow and Shapiro Wilk. Based on that test the data stated normal if every classes has significance or probability score is bigger than 0.05. it can be seen on the table below:

Table 3.3 Tests of Normality

Kelas a Kolmogorov-Smirnov Shapiro-Wilk Statistic

df Sig.

Statistic

df Sig.

NILAI MIPA 1 * .089 33 .200 .967

a. Lilliefors Significance Correction

*. This is a lower bound of the true significance.

Based on the table of analysis of normality test above, it can be seen that the significance of all these classes were bigger than 0.05.

To see whether the sample normal or not in distribution, researcher also used normal graphic of Q-Q plot, the data was normal if the distribution of data plot be in the surrounding of aslant and athward line. From the normality test, researcher got the output as below:

Chart 1: The Normality of Class X MIPA 1

Chart 2: The Normality of Class X MIPA2

Chart3: The Normality of Class X MIPA3

Chart 4: The Normality of Class X MIPA 4

From the charts of normal Q-Q Plot above, it can be seen that the drops spread around the line. So, it can be concluded that the distribution of all the populations were normal.

3. After doing the normality test, researcher analyzed the homogeneous variation test to know whether the sample homogeny or not. It had been conducted by using SPSS with livened test. If the data has significant more than 0.05, it means the data is homogeneous.

Table 3.4 Test of Homogeneity of Variance

Levene Statistic

df2 Sig. VAR00001 Based on Mean

df1

3 120 .260 Based on Median

3 120 .277 Based on Median and

3 115.751 .278 with adjusted df

Based on trimmed mean

The decision of column test of homogeneity of variance had shown that p-value 0.260 was bigger than 0.05, so it can be concluded that all classes were homogenous.

4. After analyzing the homogeneity test, researcher chosen two classes as the sample of the research as randomly. Because all of the classes were normal in distribution and also homogeny, the researcher found two sample, they are X MIPA 1 and X MIPA 2.

By flipping the coin, the researcher found that X MIPA 1 as control class, and X MIPA 2 as experimental class.

Table 3.5 Sample of the Research

No. Class

Description

1. X MIPA 1

Control Class

2. X MIPA 2

Experimental Class