Analysis Results ANALYSIS RESULTS AND DISCUSSIONS

63

CHAPTER IV ANALYSIS RESULTS AND DISCUSSIONS

This chapter is the part where the data are analyzed based on the justified methodology and comprehensively discussed. First, the analyzed data are presented under three subheadings: the ANCOVA test results, questionnaire results, and interview results. Second, all the data are overall discussed. In this chapter, the research questions are apparently answered whether the use of video- recordings is effective to facilitate the students’ speaking accuracy and fluency. In addition, what elements of video-recording use contribute to its effectiveness.

A. Analysis Results

The analysis results are classified into three subheadings: the ANCOVA test results, questionnaire results, and interview results. The two quantitative data present the analyzed data from ANCOVA test and questionnaires. The qualitative data present quotations from the students’ interviews. This every type of data shows how the research questions are finally answered. 1. The ANCOVA test Results The first quantitative result of the experiment conducted in this study was the gain scores of EG and CG See Appendix 08. The gain scores are actually the scores which are calculated by subtracting the post-test scores from the pre-test scores of each CG and EG See Appendices 5 and 6. The gain scores are commonly used to observe whether improvement takes place; therefore, they can be used to indicate whether the use of video-recordings is effective to facilitate the 64 students’ speaking accuracy and fluency. A further analysis should be made by computing the gain scores using ANCOVA test. ANCOVA test was used in this study because there are not only independent and dependent variable involved, but also covariate variable. Huck 2012: 344-345 states that “ANCOVAs involve a third variable called a covariate variable. Because the covariate is a variable on which the study’s participants are measured, it is more similar to the study’s dependent variable than to the independent variables” In other words, covariate is a variable which is related to the dependent variables. In this study, accuracy and fluency are two speaking elements which are related. In this study, both of them were treated at the same time. Initially, a normality test was conducted for the gain scores before they were analyzed using ANCOVA test for it is a prerequisite in parametric tests. A normality test is used to test whether the data is normally distributed. “A normal distribution means that most of the scores cluster around the midpoint of the distribution, and the number of scores gradually decrease on either side of the midpoint, and the resulting polygon is a bell-shaped curve” Hatch and Lazaraton, 1991. One of statistical tests for normal distribution is Kolmogorov-Smirnov Huck, 2012. Hence, the Komolgorov-Smirnov test was used in this study to test the normal distribution of the gain scores of fluency and accuracy. If the distribution is normal, the distribution of data approaches the bell-shaped curve. Therefore, the data can be analyzed using parametric ANCOVA test. Unless the data distribution is normal, non-parametric tests are required Bluman, 2012; Hatch and Lazaraton, 1991. 65 a Accuracy The study revealed that the students’ speaking accuracy in the video- recording-facilitated class is better than the students’ speaking accuracy in the regular class. The result is derived from the statistical technique, ANCOVA test. Initially, in order to be able to be analyzed using ANCOVA test, the gain score of overall accuracy should be statistically tested using Kolmogorov-Smirnov test. The result of the normality test for overall accuracy is presented in table 4.1 as follows: Table 4.1. Normality Test for Accuracy Gain Scores Kolmogorov-Smirnov Test ACCURACY N 42 Normal Parameters a,b Mean 2.540 Std. Deviation 1.6341 Most Extreme Differences Absolute .101 Positive .059 Negative -.101 Kolmogorov-Smirnov Z .658 Asymp. Sig. 2-tailed .780 To determine whether the data have normal distribution, the researcher sees the p- value. If p-value is greater than 0.05 p-value 0.05, the data distribution is normal, whereas if p-value is lower than 0.05 p-value 0.05, the data distribution is not normal. Table 4.1 indicates that the gain scores of overall accuracy have normal distribution since the p-value is greater than 0.05 0.780 0.05. Accordingly, the ANCOVA test can be conducted for accuracy. The results of the descriptive statistics and the statistical analysis of ANCOVA test for overall accuracy are presented in the following tables 4.2 and 4.3. 66 Table 4.2 Descriptive Statistics of ANCOVA Test for Accuracy Descriptive Statistics Dependent Variable: the gain score of accuracy group Mean Std. Deviation N control .048 .7229 21 experimental .990 .6188 21 Total .519 .8182 42 Table 4.3 The ANCOVA test Result for Accuracy Tests of Between-Subjects Effects Dependent Variable: the gain score of accuracy Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 19.318 a 2 9.659 46.355 .000 Intercept 1.218 1 1.218 5.847 .020 fluency 9.984 1 9.984 47.913 .000 group 1.532 1 1.532 7.352 .010 Error 8.127 39 .208 Total 38.760 42 Corrected Total 27.445 41 a. R Squared = .704 Adjusted R Squared = .689 The result of the test indicates that the p-value is 0.010. To conclude that there is a significant difference between EG and CG, p-value is lower than 0.05 p-value 0.05 Bluman, 2012. Here, the p-value is lower than 0.05 0.010 0.05; therefore, there is a significant difference between EG and CG. In other words, the alternative hypothesis is accepted meaning that the students’ speaking accuracy in the video-recordings-facilitated English class is better than students’ speaking accuracy in the regular English class. 67 b Fluency Similarly, the study showed that the students’ speaking fluency in the video- recording-facilitated class is better than the students’ speaking fluency in the regular class. The result was obtained from the statistical technique ANCOVA test, which was preceded by a normality test Kolmogorov-Smirnov test. The result of the normality test is presented in table 4.4 as follows: Table 4.4. Normality Test of Fluency Gain Scores Kolmogorov-Smirnov Test FLUENCY N 42 Normal Parameters a,b Mean .519 Std. Deviation .8182 Most Extreme Differences Absolute .128 Positive .128 Negative -.120 Kolmogorov-Smirnov Z .832 Asymp. Sig. 2-tailed .494 Table 4.4 points out that the gain scores of fluency have normal distribution since the p-value is greater than 0.05 0.494 0.05. Hence, the ANCOVA test can be conducted for the gain scores of fluency. The results of the group statistics and the statistical analysis of ANCOVA test for fluency are presented in the following tables 4.5 and 4.6. Table 4.5. Group Statistics of ANCOVA Test for Fluency Descriptive Statistics Dependent Variable: the gain score of fluency group Mean Std. Deviation N control .048 .7229 21 experimental .990 .6188 21 Total .519 .8182 42 68 Table 4.6. The ANCOVA test Result for Fluency Tests of Between-Subjects Effects Dependent Variable: the gain score of fluency Source Type III Sum of Squares df Mean Square F Sig. Corrected Model 19.318 a 2 9.659 46.355 .000 Intercept 1.218 1 1.218 5.847 .020 accuracy 9.984 1 9.984 47.913 .000 group 1.532 1 1.532 7.352 .010 Error 8.127 39 .208 Total 38.760 42 Corrected Total 27.445 41 a. R Squared = .704 Adjusted R Squared = .689 The result of the test indicates that the p-value is 0.010. To conclude that there is a significant difference between EG and CG, p-value is lower than 0.05 p-value 0.05 Bluman, 2012. Here, the p-value is lower than 0.05 0.010 0.05; therefore, there is a significant difference between EG and CG. In other words, the alternative hypothesis is accepted meaning that the students’ speaking fluency in the video-recordings-facilitated English class is better than students’ speaking fluency in the regular English class. 2. Questionnaire Results The elements of the questionnaire comprised seven indicators which might contribute to the effectiveness of the use of video recordings to enhance the students’ speaking accuracy and fluency. The indicators included self-correction, peer-feedback, one-to-one teacher evaluation, visual and auditory feedback immediacy and feedback preciseness or exactness, repetitionrefinement, flexibility, and video display. First, self-correction was the activity in which the students watched their own speaking videos and learned to find their speaking 69 weaknesses and strengths. Second, peer-feedback allowed every student to have a video exchange. One student let other friends to see his speaking video and to give some corrections and comments on his speaking performance. Third, one-to-one teacher evaluation was a segment where every student were given special time allocation to consult their speaking performance on the video with the teacher. The teacher and the students did some discussion on how to improve their speaking performance while watching the video. Fourth, visual and auditory feedback immediacy and feedback preciseness or exactness were believed to ease the students in doing the learning process. Video recordings provided the quality of seeming to happen right now able to see the action and hear the sound now although the action of speaking itself had been initially done, and accurate corrections for the students and teacher were able to directly point at the errors on the screen. Fifth, repetitionrefinement gave the students opportunities to do more than one video makings. Sixth, flexibility offered the students to study or learn conveniently as making and watching video-recordings were not limited by time and place. The last, video display was a time where the students showed their final fixed speaking performance on the screen in the class room to be watched by all of their friends. Table 4.7 presents general outcomes of the questionnaire results. It presents the elements enhancing the students’ speaking accuracy and fluency. The mean of every indicator is interpreted based on mean criteria in Table 3.5. See Appendix 11 for more detailed phenomena gathering items which constitute every indicator and their mean interpretation 70 Table 4.7 The Questionnaire Results: Indicators and Interpretations INDICATORS SPEAKING ELEMENTS Mean Interpretation self-correction ACCURACY 3.6 High FLUENCY 3.3 Fair peer-feedback ACCURACY 3.8 High FLUENCY 3.1 Fair one-to-one teacher evaluation ACCURACY 4.2 Very high FLUENCY 3.8 High visual and auditory feedback immediacy feedback preciseness or exactness ACCURACY 3.7 High FLUENCY 3.5 High repetition refinement ACCURACY 3.5 High FLUENCY 3.4 High flexibility FLUENCY AND ACCURACY 3.9 High video display FLUENCY AND ACCURACY 3.9 High Table 4.7 apparently reveals that the elements of video-recording-facilitated class which benefit the students’ speaking accuracy are self-correction 3.6, peer feedback 3.8, one-to-one teacher evaluation 4.2, visual and auditory feedback immediacy and feedback preciseness or exactness 3.7, repetitionrefinement 3.5, flexibility 3.9, and video display 3.9. Almost every element receives ‘high’ interpretation, except for one-to-one teacher evaluation which receives ‘very high’ interpretation. It means that students agree that those aspects of video- recordings help them improve their speaking accuracy. Differently, the elements of video-recording facilitated class which benefit the students’ speaking accuracy are one-to-one teacher evaluation 3.8, visual and auditory feedback immediacy and feedback preciseness or exactness 3.5, repetitionrefinement 3.4, flexibility 3.9, and video display 3.9. The five elements receive ‘high’ interpretation which indicates that the students agree that they are helpful to promote their 71 speaking fluency. However, the questionnaire analysis has found that self- correction 3.3 and peer-feedback 3.1 are less contributory to the improvement of students’ speaking fluency since both aspects receive ‘fair’ interpretation. The analysis results will be discussed in details in section B. 3. Interview Results In this study, two students were interviewed. They were a high achiever and a low achiever. The purpose of the interview is to obtain more information on the contribution of the elements of video-recording-facilitated class on their speaking accuracy and fluency. Thus, the information is used to additionally support the results of questionnaire. In the same way, the interview questions are based on the seven aspects of video-recordings: self-correction, peer-feedback, one-to-one teacher evaluation, visual and auditory feedback immediacy and feedback preciseness or exactness, repetitionrefinement, flexibility, and video display. The samples of interview results are presented in table 4.8. See Appendix 12 for more complete interview results and scripts Table 4.8 The Samples of Interview Results Indicators Interview Samples self-correction Watching my own video really helped me for I was able to learn how to evaluate my speaking performance. Finding my own mistakes appeared challenging. It made me curious. AL 07 Especially my pronunciation. When watching the video, I could see my face and its expressions. Also, I sometimes found that what I said was different from what I had thought. And, I spoke with er…er.. AL 08 I knew I spoke too much er..er.. and I knew that what I said was not understandable. But, I had no idea how to speak without er..er.., especially when I did not know the English words.AR 07 That could be very refreshing since it’s not like a class where the students always listen to the teacher most of the time. Watching videos was energizing and learning on my 72 own. Knowing my own mistakes made me aware of them. AR 09 peer-feedback I think it’s good since we can learn how to speak in good grammar. Sometimes other students are able to evaluate me more than I do. AL 10 For example, sometimes I did not how to pronounce a word correctly, but my friend did. My friend once said, “Alma, you did not pronounce the word correctly, and the structure of this sentence should be …”. So, it’s really helpful. AL 12 They only said that I should have spoken to much er..er..er. So I think they did not give me any tips to better my fluency. AL 18 It’s really useful because our friends were able to locate our errors and offer the correct ones. AR 13 The feedback is clear, for example, concerning grammar, they told me to use correct verbs. Regarding fluency, they told me to be more confident when speaking as there would be re-video making to fix the previous video. Well, just keep talking without worrying about making errors. I don’t think being confident is sufficient. I want to know how [to speak fluently] AR 14 one-to-one teacher evaluation That is even better since we know from the expert. AL 19 That is more favorable because we did not feel reluctant to ask the teacher. Sometimes, students do not like when there is a student who asks a lot of questions in the class as she seems to prolong the class. And, sometimes the other students may think she is a smart aleck. More importantly, one-to-one consultation would not make me feel shy. It’s different when we are evaluated in front of many students. AL 21 For example, the teacher always pressed the button ‘pause’ when she found my mistakes and said ‘The verb should be …. You pronounced the word incorrectly.’ Mostly the teacher evaluated my grammar and pronunciation. AL 26 Concerning fluency, the teacher said that I should have not memorized a text, but understood the points of what I would say. More to the point, when I forget a word, I need to search its synonym or define the word. So, I will not get stuck only to recall the word. AL 27 That was very helpful to improve my grammar and fluency since the feedback from teachers is true feedback. AR 15 Well, I think it’s enough. If I forget a word, the teacher told me not to stop talking. I should find another word which is easier. Then, I should speak with simpler language and avoid long sentences because it would make me speak unclearly. What else? That’s the only feedback I remember. It was difficult for me to do so though. I always forgot. AR 18 73 I felt free [to have one-to-one consultation] since there were only my teacher and me. I felt like she gave me private time [to study]. AR 19 flexibility It’s because the learning process did not always happen in the class and on the scheduled time. We were able to learn outside. I watched my videos anytime when I was not having my English class. AL 28 Yes, of course. Time flexibility is favorable as I could watch my video in my cell phone which I always bring wherever and whenever I go. When I wanted to learn, I watched it in my cell phone right away. AR 20 Both [to learn and to entertain]. When I watched it, I was thinking how I would improve it. AR 23 repetition refinement I repeatedly recorded my self around 4-6 times. AL 32 In order to get a good video because my friends were going to watch it. I would be shy if my video was not good. If I spoke in good grammar, correct vocabulary and pronunciation, it would be interesting to watch. And, because I repeatedly made the video, I knew what I was going to talk about. AL 33 Several times. AR 24 In order that I got better result than the first video. AR 25 It’s not bad because I had my friend video record my speaking for several times. When I made grammar, vocabulary, and pronunciation errors, I would repeat the video making until I made fewer mistakes. But, I think I made little improvement though. AR 26 video display Sometimes I felt shy since I still made some mistakes although I had repaired my speech. However, I had to make the video as best as I could so that my friends would not laugh at me. AL 36 When watching my friends’ videos, I learned from them whom I thought spoke very well. AL 37 I felt shy as I did not speak fluently and my grammar was still not very good. Yet, when I made my video, I needed to be careful enough so that I would not make a lot of mistakes. If I made a lot of mistakes, I would be very shy. AR 29 I was definitely able to see my friends’ speaking performance, especially those who spoke well. So, I knew which parts they were good at. Then, I compared them to mine. AR 30 visual and auditory feedback immediacy, feedback preciseness It is easier to evaluate [our speaking performance] using the video. As I said before that we could re-play the video and see where we made the mistakes. By watching the video, we knew what was wrong and learned from it. If we do not know that we have made errors, how can we learn then? If there was no video, we must have forgotten what and how we had said. So, we did not know which parts we had to 74 From the samples of interview results above, the researcher has found some general conclusions indicating that students have benefited the use of video- recording in learning speaking. First, every element of video-recording-facilitated class has given contribution to enhance the students’ accuracy. Only some elements have assisted the students to better their speaking fluency. Second, the researcher notices that the interviewed students agree that the use of video- recordings has given different learning atmosphere from what they usually have in a customary learning process. The detailed use of the interview samples will be found in the discussion to give supportive evidence on the questionnaire results.

B. Discussion