Statistical Design METHODS OF INVESTIGATION

38 Where, D : discriminating power, BA : number of students in the upper group, who answered the item correctly, BB : number of students in the lower group, who answered the item correctly, JA : number of all students in the upper group, and JB : number of students in the lower group. The criteria are: D ≤ 0,20 is poor 0,20 D ≤ 0,40 is fair 0,40 D ≤ 0,70 is good 0,70 D ≤ 1 is very good

3.8 Statistical Design

Statistics is calculations made to describe a sample. I analyzed the significant difference between teaching descriptive text using process approach and teaching descriptive text with conventional method using t-test formula to see the differences between the two means. Here it is, the t-test formula: 2 1 2 1 1 1 n n s x x t + − = 2 1 1 2 1 2 2 2 2 1 1 − + − + − = n n s n s n s 39 Where: 1 x : mean score of the experimental group, 2 x : mean score of the control group, s : standard deviation, 1 n : the number of students in experimental group, and 2 n : the number of students in control group. 40

CHAPTER IV FINDING AND DISCUSSION

This chapter dealt with the result of the experiment that has been conducted. Because the explanation of the t-test has been presented in the previous chapter, in this part the writer will only discuss the detailed calculation and the result of the t-test value.

4.1 Research Finding.

In the previous chapter, it has been stated that the method used in the data collecting was writing test that is writing descriptive texts. Therefore, the instruments used in this study were in the form of writing descriptive texts. The text was administered in the both control group and the experimental group. Also, with regard to the research design that has been chosen, two types of the test was administered, there are pre-test and post-test. Two groups firstly were given pre-test. Than the two groups were given different treatments and then finally they were post tested. The control group was taught by using the conventional method that is writing descriptive text without process approach, while the experimental group was taught by using process approach in the writing descriptive text. A post-test was then administered in both groups. The data were obtained from the students’ achievement of the writing descriptive text. They were pre-test and pos test scores from the control group and the experimental group. The total scores for the control group were 2840 for the pre-test and 3112 for the post-test. So the difference score between the pre-test