Validity Reliability Test Analysis

30 LD = RU + RL N Where, LD = Level of Difficulty RU = the number of students in the upper group who got the item right RL = the number of students in the lower group who got the item right N = the total number of subjects n is 40 Standardize: Very Difficult 0, 30 – 0, 70 Very Easy

3.6.1.2 Test Analysis

3.6.1.2.1 Validity

Before giving a test to students, the writer must know first whether the questions are valid or not. Tuckman stated that the validity of test represents the extent to which a test measures what is purpose to measure. Without analyzing the validity, a researcher may not use the proper questions which are suitable with the material. The formula that used in this study is stated as follows see appendix 7: where, q p S M M r t t p pbis − = 31 r pbis = the validity of the instrument Mp = mean of the subjects who have correct answer Mt = mean of total score St = standard of deviation p = proportion of subjects answer correctly q = 1 – p Criteria: The item is valid if r xy r table 5 from the number of subjects 38 = 0,321

3.6.1.2.2 Reliability

The reliability of a test is defined as the extent to which the results can be considered consistent or table, Brown, 1988: 98. The reliability shows the consistency of the test from one time to the other. In the other word, the reliability marks whether the test can be accepted or not by the students. The formula is stated as follows see appendix 7: r xy = N. ΣXY – ΣX ΣY . {N. ΣX² – ΣX²} {N. ΣY² – ΣY²} Where, r xy = correlation reliability of the test item N = the number of the students participating in the try out 32 X = scores obtained from the first administration of the test item Y = scores obtained from the second administration of the test item ΣXY = the sum of product of paired X and Y ΣX² = the sum of the squared of x scores ΣY² = the sum of the squared of y scores Criteria: The item is reliable if r xy r table 5 from the number of subjects 38 = 0,321

3.6.2 Test

Test is a kind of instrument that can be given orally, written, or behaviorally. In this research, the test contains vocabulary questions. In this case there were three kinds of test. They were pre-test, formative tests, and post-test. Those tests contain multiple choice items and essays. In order to analyze the Pre- test and post-test, the writer used formula of t-test below: Where, 2 1 n 1 n 1 s x x t 2 1 + − = 2 n n 1 n 1 n s 2 1 2 2 2 2 1 1 − + − + − = s s