Improving Speaking Skill of the Tenth Grade Students through Mingle Model

The purposes of this study are to unveil the application of Mingle Model to enhance students’ speaking skill at tenth grade of Vocational High School Number 1 Simpangkatis, Bangka and to clarify whether there was or not a significance difference in student mastery of speaking skills between two classes; experimental class where Mingle Model was implemented and non-experimental class where there was no new model given by teacher. Observation and test were the techniques used in collecting the data. This study used quantitative approach with quasi experimental design. There were 72 students from two classes taken as the sample purposively; TKI 1 and TKI 2. This research applied Mingle Model as the treatment given to the experimental class. This study considers some aspects in speaking; comprehension, pronunciation, vocabulary, grammar, and fluency. From the finding, it showed that the students who were given Mingle Model treatment reached better or higher score in the post test than the students who were taught by Non-Mingle Model. In this case, the alternative hypothesis (Ha) was accepted as t-obtained, 3.56, was higher than the critical value of t-table, 2.00. Based on the evidence above, the research concluded that between the experimental and control group there was a considerable difference of students’ speaking performances. Therefore, the findings of this study confirmed that Mingle Model was effective to increase the students’ speaking ability.

EISSN: 2655-3716 https://jurnal.lp2msasbabel.ac.id/index.php/sci Model as a treatment and the students who were taught by using conventional method?

Method of Research
The method of this study was quantitative approach. As the purpose of this study was to investigate whether Mingle Model could significantly develop the students' skill of speaking, the experimental research was proposed (Rasinger, 2013). Therefore, this study used the quasi-experimental design as one of type experimental research in quantitative approach Fraenkel & Wallen, 2009).The basic characteristic of this quasi experimental research is the dependent variable are put in groups; experimental group which will be given treatment and control group which will be given no treatment, no new model or technique during the research ((0. Creswell, 2012).

Data type and source
The population of this study was all of the students of the first graders of Vocational School No. 1 Simpangkatis, Bangka Tengah in academic year of 2017/2018.

Table 1.Population of Study
Resources: SMKN 1 Simpangkatis (2017 The researcher considered two classes of the tenth grade as the samples of this research. In this study, the English teacher of Vocational High School Number 1 Simpangkatis who taught classes of the ten grade suggested to take X TKI 1 and X TKI 2 as the samples. She recommended those classes because she thought that they had low speaking ability but had bigger potential than others to get a good result in speaking development. It would be very useful for students and school. Then, according to the school teacher, students' speaking ability in X TKI 2was lower than X TKI 1. Because of that statement, the researcher decided X TKI 1 as control and X TKI 2 as experimental group.It means that, the lower group should get treatment to improve their speaking achievement. Experimental group would be given a treatment by using mingling model to improve their speaking skill. The sample is figured out in the table below.

Technique for Collecting Data
To answers the research problems (problems of study), the instruments are needed. For research problem number one related to the implementation of Mingle Model, observation is needed and to answer research problem number two, tests are acquired to be given tho samples of the study.

a. Observation
The observation is needed to see students' reaction toward the treatment given during the study.
According to Fraenkel, observation should be conducted for two reasons; the first is in order to see the object of the research, in this case is the students, respond to the technique and the measurements of all students activities during the observation (Wallen & Fraenkel, 2001). During the observation, teacher can use observation sheet to measure the students' activeness, participation, presence, attention, dealing with questions and answers and many others. The observation sheet should be filled by the researchers every meeting during the research either for experimental or control group. b. Test The second instrument used in this study was pre-test and post test of speaking ability. This test was used to measure the students mastery of speaking ability; comprehension, vocabulary, pronunciation, fluency, and accuracy (Marczyk, DeMatteo, & Festinger, 2005). The pre-test was given before treatment, and post test was given after treatment. The main score of these kinds of test would be analyzed using statistical analysis, T-test. The purpose of using T-test is to see whether there was a significant different for students' speaking performances before and after treatment, and the differences of the mean scores of speaking ability between those two groups, experimental and control group.

The Result of Observation
In this part, the researcher would like to discuss the result of the observation both experimental and control group.

a. The Result of Observation in Experimental Group
These were the result of observation in experimental group. It showed the students' attendance, attention, activeness, and doing all activities. Based on the result of the table above could be read as a scale of value in this following table: Table 4. The Scale of Value The table above showed the observation result of experimental group during twelve meetings. There were four aspects. Those were attendance, attention, activeness in asking and answering question and doing all the activities. The result of students' attendance, the pre test and post test was done by all 36 students. Also, the eighth, ninth, tenth and eleventh meetings were attended by all students, 100% was 36 students. But in the other meeting, there were some students not attending the class. It was around one up to four. Over all, it was still in very good category.
The second aspect was students' attention. The students' attention was very unpredictable. The second, fourth up to eighth were in very good category. It was around 88%-97%. It showed that most of students showed their attention to the lesson. In the tenth and eleventh meeting was in good category. It was 80% and 83%. But in the third meeting, it was the lowest attention of the students. It was average category, 66%.

<53% Low
Next is students' activeness aspect. It showed 77% in the second meeting, 75% in the second meeting, 83% in the third meeting, 77% in the forth meeting, 83% in the fifth meeting, 66% in the sixth meeting, 80% in the seventh meeting, 83% in the eighth meeting, 63% in the ninth meeting and 75% in the tenth meeting. The sixth and the ninth were average category. The others were in good category.
The last was doing all activities aspect. Based on the observation result there was no meeting have low category. There were eight meetings had good category and others had good category. It showed that almost of students wanted to do all activities was required. The class reached good and very good category which 75%-91% of students were active enough for doing all activities.

b. The results of pre-test and post-test in experimental group
In order to see the students' speaking skill after being taught by Mingle Model in experimental group which was X TKI 2, the researcher conducted pre-test and post test. Every student retold their stories about My Holiday, My Bad Experience and My Good Experience. The scores were given by raters after listening to the students' recording about those topics above. The experimental group scores of pretest and post-test and the score category were presented in appendix.

c. Statistic analysis of pre-test and post-test in experimental group
This was the result of paired sample statistics in experimental group:  Table 6 provides the mean scores of pre-test an post-tests of students from experimental group. The mean of pre-test score was 44.0556, while post-test score was 69.222, much higher than the scores before the treatment was given. For paired sample test result, the paired differences was showed above that the mean between pre-test and pots-test in the experimental group was 25.16, standard deviation was 6.93, standard error mean was 1.15 and t-obtained was 21.77 , at the significant .000 for two tailed and degree of freedom 35. Based on those result above the sig(2 tailed) .000 was lower than the value of probability .05. It meant that there was a significant difference for students' speaking skill after being taught by using Mingle Model in experimental group.

d. The statistics analysis of students' post-test between control and experimental group
This part would discuss about the statistic of the students' post test between control and experimental. Tabel 9 showed us the statistical analyses of independent sample test where it compares between the post-test from two classes. It was obviously seen that t-obtained was higher than t-table, meaning that the Mingle Model was successfully implemented. And as a result, the students' speaking ability developed significantly.

Discussion
In this part, the researcher would like to discuss the interpretation of the implementation and observation, and the interpretation of statistical analyses.

The Interpretation of Implementation and Observation
In this part, the researcher would like to describe in detail some of the interpretations of the research result. In the experimental group showed an improvement during twelve meetings which had been finished. This class consisted of 36 students who had been cooperative did everything needed together with the researcher. The researcher had explained that the researcher taught the class by using Mingle Model in order to improve student's speaking skill. Almost in every meeting, the student practiced their speaking through going around the class and completed the difference list of question in meeting. The main topics were about holiday, bad experience and good experience.
For mingle model, the steps were very easy to do by students. This model gave every student a chance for practice their speaking in every meeting. They had chance for asking and answering each question with their partner. In the first meeting, they did not seemed interesting in it. But, after the researcher explained clearly and showed them the example how to do it, they really wanted to try. In that meeting, they asked the teacher a lot of question about how to say some words in English, when they wanted to answer their friend question. In every meeting, they improved little by little. They brought dictionary in every meeting to find the word they did not know, in order to answer their friend's question faster. The class was noisy because of their laugh and their speaking but they could enjoy the class. They used three topics above. They had less confident when speaking in English.
By this model, they could practice their speaking from the nearest people in the class. Then some students came forward to present their story. That time, the researcher did little bit of error as knowledge for the students about the aspect of speaking. It could be the pronunciation, grammar, fluency, comprehension or vocabulary. The result showed that they could speak better than before and the grammar was improved after having Mingle Model during learning process.
Based on the observation sheet, the percentage of students in every meeting was unpredictable. The students were practiced to participate in every meeting. They were trained to be brave to talk during research. They had to speak in the chance given. The presence of students was in very good category during meetings, both of classes, control and experimental group. One of meeting had average category in attendance. It was in the ninth meeting of control class. For the attention part, there was very good, good and average category both control and experimental. For students' activeness were good in the eight meetings and average category in two meetings for experimental group. Then, in the control group, there were four in low category, there were one in very good category and five meetings in good category.
The students of experimental group were active enough for asking their partner. But there were some meeting got average category. It was the seventh and tenth meeting. Others had good category in their activeness. They felt afraid to make mistake. They were not confident saying any word in English. So that, the teacher taught them some expressions before practicing mingle model. Then they practiced mingle model by trying to use English it in the conversation. Everyone was trying in every meeting. For doing all activity, all students were able enough to do the activities during meetings. The ninth meeting in control group was the lowest. It was average category.

The Interpretation of Statistical Analysis
The statistical analysis provides data through mathematical process and the output might be in the forms of numbers, tables, graphs, and figures (Kothari, 2004). The experiments of both classes were conducted 12 (twelve) times. Pre-test was given in the first meeting and post-test was done at last meeting. To see the differences of the students' pre-test and post-test scores, statistical analyses, T-test was used. To see the differences between pre-test and post-test of the two classes, Independent T-test was implemented, while to EISSN: 2655-3716 https://jurnal.lp2msasbabel.ac.id/index.php/sci measure the post-test between the two groups, Paired Sample T-test was used. The mean score of the pre-test in the experimental group was 44.0566 and the post test was 69.222, much higher than the pre-test score before the treatment of Mingle Model was given. And when the post-test of control group and experimental group was analyzed using T-Test of statistical analysis, it was obviously seen that the t-obtained (3.56) was higher than the critical value of t-table a(2.00). It can be inferred that there was a significant difference between the two classes, one which Mingle Model was implemented and other which was taught by using conventional method.
There were some studies who confirmed that Mingle Model was effective in developing students' mastery of speaking skill, among of them were study conducted by Elena Borzova (Borzova, 2014) and research implemented by (Darmayenti & Nofiadri, 2015), Borzova argues that Mingle Models covers three important tasks; language (form-focused function), communicative function, and meaningful mingle. Language means the teacher tries to enhance and increase the student's skills, specifically for grammar and vocabulary related to topic learnt by the students. The second stage is communication function where the students practice what they have got in the form-focused function.
Given a simple situations, the students tried to achieve the communicative function in using expressions, grammar and vocabularies they know. Those three stages proposed by Borzova in implementing Mingle Model are in line with some statements, opinions and theories related to communicative competence and performance proposed by Gunn, Pakula and Jack Richards (Gunn, 2003;Pakula, 2019;Richards, 2001). Furthermore, Borzova claimed that Mingle Mode is a very challenging strategy in developing the students' speaking skills because it can create a convenient learning atmosphere where it can increase the quality of communicative competence and performance of the students' speaking ability.
Students must learn English as one of international languages is the most widely used language in the world (Abrar et al., 2018;Marzulina, Pitaloka, & Yolanda, 2019;Mukminin et al., 2019). In Indonesia, English as foreign language is learned by students of Elementary school, Junior High School, Senior High School and Vocational School , until University level . While Erlina argues that English as one of the compulsory subject that is need to be learnt either for its receptive or productive skill . Students have to master English orally and written form to pass their English examination .
The second research was done by Darmayenti and Nofel Nofiadri to the first-year students of State Institute of Islamic Studies Imam Bonjol Padang, academic year 2012/2013. According to Darmayenti and Nofiadri, Mingle Model is a unique strategy in increasing students' speaking ability because it can make the students' accuracy and fluency becomes much better (Darmayenti & Nofiadri, 2015). Mingle Model provides real world communication, a meaningful conversation which promotes their motivation in speaking. One kind of activities which can be used while Mingle Model was implemented is role play. According to Harmer and Nation, role play is one of activities that can be proposed in evaluating students' speaking skill (Jeremy Harmer, 2007;Nation, 2008).
One of the characteristics of Mingle Model is students have to talk with their pair, or in small groups. Students who are usually reluctant to talk in big group will be more relaxed to discussed with their pair or in small group (Brown, 2003;Frazier & Brown, 2001). In addition, Harmer and Hedge also stated that Mingle Model could create a cozy learning atmosphere, lowering the affective filter, and providing effective motivation strategy from students (J. Harmer, 2004;Jeremy Harmer, 2007;Hedge, 2003). The teaching and learning English in Indonesia has been designed to help students to use English both in written and spoken forms (Gan, 2013;Indrawati, 2017;Marriam Bashir;Muhammad Azeem;Ashiq Hussain Dogar, 2016;Richards, 2010; "THE IMPORTANCE OF TEACHING LISTENING AND SPEAKING SKILLS," 2018). Students have to master English competence and know how to perform it in daily life in order to be well communicated in this foreign language (Indrawati, 2018;Pakula, 2019).

Conclusion
The author should clearly explain the important conclusions of the research highlighting its significance and relevance.

Conclusions of the Study
This shows you everything simpler about the result of research finding elaborated in the previous chapter. In experimental group, every student had chance to talk at class and got the information while completing list of questions in the cards from some students. After that some of them did presentation in front of class. The implementation of Mingle Model in the first grade of Vocational High School Number 1 Simpangkatis, Bamgka was accepted and done well by almost all students' of TKI 2 class. It was proven by the observation sheet analysis. The attendance was very good category. They did pay attention for the materials and instructions. Then, they were active enough in the class. At last, they did the almost all of activities well. In experimental group, their attendance all was very good in every meetings, their attention was one meeting average, two meetings good and seven meetings very good, the activeness during meetings was nine meetings good and one meeting average, and doing activities were two meetings good and other very good category. All points were mostly 'good' and 'very good' categories during the meetings based on above explanations..
Based on the statistical analysis of the evidence in the previous chapter, the researcher concludes that there was a significant difference in students' speaking skill, between students who were taught by using Mingle Model and those students who were taught by Non-Mingle Model. The students who were taught by using Mingle Model in experimental group got higher score than the students who were taught by Non-Mingle Model in control group. Based on the result of statistical analysis, it showed t-obtained score 3.56. Then, the critical value of t-table was 2.00. That score was clearly higher than critical value of t-table. Therefore, the alternative hypothesis (Ha) was accepted and null hypothesis was rejected. It means that there was a significant difference in speaking skill between students who were taught by using Mingle Model and students who were taught by using Non-Mingle Model. Based on the evidence above, the writers states the Mingle Model was effective to improve students' speaking skill of the tenth grade of Vocational High School Number 1 Simpangkatis.

Suggestions of the Study
The suggestion related to the conclusions above is that Mingle Model is recommended to be applied by the teacher in the class for improving students' speaking skill. This model gives every student a possible chance to practice their speaking every meeting and creates comfortable atmosphere for the students during the learning process.