Score of Reading Comprehension by Using DRTA Method with

Scoring the data for reading comprehension was done by using instrument with objective test multiple choice tests and the total items provided were 30 questions. For each question, the correct answer will be graded 1 and the incorrect answer will be graded 0. Hence, the maximum score will be 30, while minimum score will be 0. Respondents treated by Conventional method with high reading interest were 40 students. The empiric score stated that the highest score was 83, the lowest score was 70. Furthermore, mean was 75.19, median was 74.50, mode was 73, standard of deviation was 4.199 and variance was 17.629. The complete result gained from calculation can be seen below. The mean 75.19 indicated that the average score for the students was relatively fair. The standard of deviation 4.199 indicated that the answers given by students using Conventional method with low reading interest were relatively the same. To make it clear, it can be seen the display of histogram and polygon presented below:

2. The Prerequisite Test for Data Analysis

Before analyzing the regression or testing the hypotheses, it was needed to do the prerequisite analysis test of X1, X2, and Y variables. Moreover, the requirement analysis test is one of the requirements that must be accomplished in order to make the regression analysis or hypotheses test done well. Thus, the normality test and the homogeneity test must be done first before the ANOVA test conducted. Normality test was applied to the representative research sample. This test was done as hypotheses test which was required that the sample must be normal. The next one was homogeneity test, which was assumed that the score of dependent variable Y was categorized based on the equation of independent variable scores X1 and X2. The result of the test is presented in the following:

a. Normality Test

Normality test is used to find out whether the spreading data is distributed normally or not. In this study, the normality test used Kolmogorov- Smirnov method in which the significance level α = 0,05 as the rule to accept or reject the normal test. The normality test is done to both experimental and control groups by using statistical hypotheses formula stated as follows: H = sample data is distributed normally H 1 = sample is not distributed normally The computation is performed with the assistance of SPSS version 20 for windows. Based on the criteria of this program, the data is normal if p value Sig 0.05 which means H is accepted and on the contrary H 1 is rejected data is distributed normally. The score of p value Sig is the number on the column of Sig from the table of normality test outcome by using SPSS program. In this case, the method used is Kolmogorov-Smirnov . The computation of normality test can be seen on the table below: Table 4.10.Recapitulation of Normality Test One-Sample Kolmogorov-Smirnov Test A1B1 A1B2 A2B1 A2B2 N 16 16 16 16 Normal Parameters a,b Mean 77.63 73.75 73.38 75.19 Std. Deviation 6.032 2.049 5.584 4.199 Most Extreme Differences Absolute .231 .268 .277 .261 Positive .231 .268 .277 .261 Negative -.159 -.239 -.148 -.187 Kolmogorov-Smirnov Z .925 1.071 1.107 1.045 Asymp. Sig. 2-tailed .359 .201 .172 .225 a. Test distribution is Normal. b. Calculated from data. According to Table 4.2 above, it is seen that the scores on Sig column by using Kolmogorov-Smirnov method for each group are mentioned consecutively: 0.925, 1.071, 1.107, and 1.045 which means all the p value score for each group are bigger than 0.05. Hence, H is accepted and H 1 is automatically rejected. In other words, it may be concluded that all data from the sample of this research have been distributed normally.

b. Homogeneity Test

Beside normality test, one prerequisite test mostly needed to analyze the data using ANOVA is homogeneity test. The purpose of this test is to find out whether the designed groups are homogenous or not. In other words, we have to find out the homogeneity of the groups we designed. The homogeneity test for the data of reading comprehension is performed by using Levene‟s test in the significant level of 5. The result of homogeneity computation can be seen on the Table 4.3 below: Table 4.11.The Computation of Homogeneity Test for Reading Comprehension Levenes Test of Equality of Error Variances a Dependent Variable: Y F df1 df2 Sig. 3.306 3 60 .026 Tests the null hypothesis that the error variance of the dependent variable is equal across groups. a. Design: Intercept + A + B + A B The hypotheses for homogeneity test were set as follows: H : Data comes from homogenous population H 1 : Data comes from non-homogenous population The criteria were set as follows: If the Sig value Levene’s test 0.05 means that H is accepted and H 1 is automatically rejected. On the contrary, the Sig value Leven e‟s test 0.05 means that H 1 is accepted and H is automatically rejected. Refer to table 4.3 above, it can be seen that the Sig p value for reading comprehension was 0.026. It means that p value is smaller than 0.05. It means that H 1 is accepted and H is automatically rejected, which implies that data comes from non-homogenous population. Even though came from non-homogeny variance, the data can be processed by using contrast test in Anova Multi-factor t- test that can be treated for both homogeny and non- homogeny variance Agung, 2004, p.19. According to both normality test and homogeneity test revealed above, it can be concluded that the prerequisite test which are needed before processing the data by using ANOVA test are already fulfilled. 3. The Testing of Hypotheses Hypothesis testing was intended to determine the proposed null hypotheses H tested at a certain significance level. Two way ANOVA analysis was performed and, because in this study to be obtained was how much influence that occurs between the two independent variables and the dependent variable. Hypothesis testing was done consecutively, starting from the first hypothesis, The Directed Reading Thinking Activity was more effective than Conventional method toward reading comprehension, the second hypothesis was the students who have high reading interest have better reading comprehension than those who have low reading interest, the third hypothesis was there was interactional effect between teaching method and reading interest toward reading comprehension. The analysis of reading comprehension variable is performed by using two tailed ANOVA test, with the assistance of SPSS version 20 for windows. The result of ANOVA test then continued to extended test to find out the level of significance among groups significantly simple effect. In other words, the extended test was performed to find out which group contributes more to be students‟ reading comprehension according to the teaching method and the level of reading interest.