ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 323 THE EFFECT OF COOPERATIVE INTEGRATED READING AND COMPOSITION LEARNING METHOD TOWARDS FABLE WRITING SKILLS Desri Arwen Department of Arabic Education Universitas Muhammadiyah Tangerang, Indonesia E-mail: desriarwen72@gmail.com Saiful Haq Department of Management Education Universitas Muhammadiyah Tangerang, Indonesia E-mai: saifulhaq@yahoo.com APA Citation: Arweb, D. & Haq, S. (2021). The effect of cooperative integrated reading and composition learning method towards fable writing skills. English Review: Journal of English Education, 10(1), 323-336. https://doi.org/10.25134/erjee.v10i1.5721 Received: 12-07-2021 Accepted: 07-01-2021 Published: 30-12-2021 INTRODUCTION Learners can improve their thinking skills and thoughts through writing and it is crucially needed so that they can express their feeling, knowledge, and ideas in a piece of writing (Asiah, Ardian, and Amri, 2020; Jayanti, 2019). Moreover, writing can be a complex activity because the students convey the ideas into written form and notice several aspects of writing, such as good grammar, spelling, clear point, etc (Purnamasari, Hidayat, & Kurniawati, 2021). "Writing is used to reports or informs, influences, and such aims and objectives can only be achieved properly by people who can organize their thoughts and express them clearly (Rusyana 2012; Morsey 2008; Tarigan, 2008; Hamalik, 2013) A number of language experts believe that language learners can use Cooperative-Integrated Reading and Composition (CIRC) as an integrated method which includes reading and writing activities (Prajogo, 2020; Nolia, 2021). CIRC was first developed by Slavin, Farnish, Stevans and Madden. The main reason for developing this method is because they are concerned about the traditional teaching of reading, writing and language arts by teachers. Pangesty, Nursirwan, Marliah, Yasa, & Hartono (2021) defined CIRC as one of the cooperative learning models that actively engages students in their learning activities and combines writing, reading, discussion and presentation activities. This method emphasizes on the group work either in pairs or small groups to practice reading Abstract: The purpose of this study was to determine the effect of using the cooperative integrated reading and composition method on learning outcomes to write fables in class VIII students at MTs Daarul Muqimin Buaran Jati, Tangerang Regency. students think learning to write fables is boring, students don't think creatively in learning to write fables, students find it difficult to find creative writing ideas, use less effective learning media, and inappropriate writing learning methods. The method used in this study is a quasi-experimental method by conducting a pretest and posttest in two groups. This type of experimental research method uses a nonequivalent control group design, where the experimental class is treated and the control class is not treated. This nonequivalent control group design is almost the same as the pretest and posttest control group designs, only in this design the experimental and control groups are not chosen randomly. Data collection techniques using tests. The data analysis technique was started from normality test, homogeneity test, and continued with hypothesis testing. From the research results obtained tcount 2.061> ttable 1.9987. So it can be concluded that there are differences in learning outcomes between the control class and the experimental class. Thus, the results of the study using the CIRC (cooperative integrated reading and composition) method showed an increase in the ability to write fables for the students. Keywords: CIRC; learning to write fables; composition; writing; reading. mailto:desriarwen72@gmail.com mailto:saifulhaq@yahoo.com Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 324 and comprehending texts. Cooperative learning is known as group learning. But cooperative learning is more than just group learning or group work because in cooperative learning there is a cooperative structure or task that allows open interaction and effective interdependent relationships among group members. The stages of CIRC are orientation, organization, concept recognition, publication, strengthening and reflection (Shoimin, 2014). Meanwhile, according another researcher, Kessler in Ristanto et al. (2018) the learning steps are as follows: (1) Students are grouped into several heterogeneous groups, (2) Each group reads an article or a reading, (3) Students rewrite the result of discussion on a worksheet, (4) Each group displays or presents their results, (5) The best group is rewarded. It is an integrated composition of reading and writing cooperatively in groups so that students learn and work in pairs (Ngalimun, 2013). There are basic elements of cooperative learning that distinguish it from the division of groups that are carried out at random. It is believed to be helpful in learning, especially in learning to read and write, because this model helps students be more interactive in learning and is able to help students who are weak in improving skills and easy to apply at the primary and secondary education levels, and are able to train students to study in groups (Telaumbanua, 2020). The implementation of cooperative learning model procedures will actually enable educators to manage the class more effectively. Hutagalung and Tanjung (2021) mentioned that CIRC has fun activities where students are given chances to process the information enabling students to choose different pairs of groups. On the other side, as Abdalrahman (2021) stated, reading and writing are inseparable acts. It is the instructor’s duty to explain this to language learners that writing effectively depends on effective reading. CIRC type cooperative learning in terms of language can be interpreted as a cooperative learning model that integrates a reading as a whole and then composes it into important parts. The CIRC program consists of three main elements, basic activities, hands-on teaching in reading comprehension, and integral language/writing arts. The main goal of CIRC is to use cooperative teams to help students learn broadly applicable reading comprehension skills. Several elements of the CIRC are indeed geared towards this goal. During the follow-up period, students worked in pairs to identify five important features of each narrative story: characters, background events, problems, efforts made, final solutions. The main objective of the CIRC program developers towards writing and language arts lessons is to design, implement, and evaluate a writing process approach to language arts and language lessons that will make the most of the presence of classmates. Specifically, in writing and language arts, a major objective of the developers of the CIRC writing and language arts program was to design, implement, and evaluate a writing-process approach to writing and language arts that would make extensive use of peers (Zainuddin, 2015). Responses from peer groups are a typical element of writing process models, but peer involvement is rarely the central activity. However, in the CIRC program, students plan, revise, and edit their essays in close collaboration with their teammates. The teaching of language mechanics is truly integrated as well as being part of the writing lesson, and the writing lesson itself is integrated with the teaching of reading comprehension lessons, both with the integration of writing process activities in the reading program and with the use of newly learned reading comprehension skills in teaching writing lessons. CIRC consists of three important elements, related basic activities, direct teaching of lessons in reading comprehension, language arts, and integrated writing. In all these activities, students work in heterogeneous teams. All activities follow a regular cycle involving presentations from the teacher, team exercises, independent exercises, peer pre-assessment, additional exercises, and tests. With cooperative learning, it is expected that students can improve creative thinking, and foster a high social sense in language skill, especially in writing (Jayadi, 2021). According to Saifulloh (in Budiyanto 2016) said that there are several advantages of the Cooperative Integrated Reading and Composition (CIRC) learning model: (1) Students' learning experiences and activities will always be relevant to the child's level of development. (2) Activities that are selected according to and depart from the interests of students and children's needs. (3) All learning activities are more meaningful for students so that the learning outcomes of students will last longer. ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 325 (4) Integrated learning can develop children's thinking skills. (5) Integrated learning presents activities that are pragmatic (useful) in accordance with problems that are often encountered in the child's environment. (6) Integrated learning can foster student learning motivation towards learning that is dynamic, optimal and effective. (7) Foster children's social interactions such as cooperation, tolerance, communication, and respect for other people's ideas. (8) Generating learning motivation, broadening the insight and aspirations of teachers in teaching. METHOD The research is included in experimental research, where the researcher gives a treatment or experimental conditions and then observes the effect caused by the treatment. The research method used is a quasi-experimental research method. This method is used to find the effect of certain treatments on others under controlled conditions. This study uses the nonequivalent control group design research design, meaning that in this design there are two groups, namely the experimental group and the control group. The experimental group was given treatment in the form of learning using the Cooperative Integrated Reading Competition (CIRC) learning method, while the control group used conventional learning methods. Then the two groups were given a pretest to determine whether there was a difference between the experimental group and the control group in the initial state. Population or universe is an area or place where objects/subjects are studied, whether people, objects, events, values or other things that have certain quantities and characteristics to obtain information (Riyadi, 2014). The population in this study were all eighth grade students of MTs Daarul Muqimin from Jati, Tangerang Regency (334 students) registered in the odd semester of the 2020/2021 academic year which is divided into nine classes. In other words, researchers only take a part that can represent the population. Therefore, as explained above, of the nine classes the researcher only took two classes that would be used as research samples. Sampling was not done randomly because the researcher used a quasi-experimental method with a nonequivalent control group design which was a weak experimental group, therefore the samples were chosen freely for both the experimental group and the control group. Of the several classes there are only two classes that will be sampled, namely the experimental class which is given treatment and the control class which only uses conventional learning methods. The data obtained from the research instrument used to measure the ability to write fables. The ability to write fables is seen from understanding and creative ideas, the ability to solve the questions given in the form of tests. To determine the effect of students' fable writing ability on the use of the CIRC learning method, data is needed. The tools used to collect data are in the form of pretest and posttest devices. The pretest was carried out before the learning activity to find out the initial description of the two classes, while the posttest was carried out after the learning activity was carried out to determine whether or not there was an increase in students' fable writing ability after participating in the learning process. The things that need to be considered in collecting the data are the variables studied, namely the independent variable consisting of the CIRC learning model and the dependent variable consisting of the ability to write fables. The implementation of the fable writing ability test is assessed with a fable writing ability assessment format consisting of several main criteria, namely, theme, orientation, complication, resolution, code or mandate. The learning stages carried out for data collection refer to the following steps: (1) The first step is preparation, which aims to design learning to write fables. (2) Conditioning students in accepting previously planned learning. (3) Before the researcher gave the material about writing the fable text, the students were given a pretest. (4) After students are given a test at the beginning of learning, then students are given treatment using (CIRC) learning method. (5) Doing the final test (posttest). RESULTS AND DISCUSSION The data described are data obtained from the results of filling out tests using developed instruments. The presentation of the data is intended to provide an overview of the size of the data concentration and the size of the data spread. The data is presented in the form of a frequency distribution table and in the form of a graph to make Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 326 it more communicative and easy for readers to understand. Description of pretest data for control class and experiment class Control class pretest data Based on the results of the pretest of the control class at MTs Daarul Muqimin, Jati Buaran, Tangerang Regency, the data obtained from the respondents are as follows: The lowest pretest score obtained in the control class was 30 and the highest score was 70. Students who scored 30-35 were 3 students. The value of 36-41 is 5 students. The score is 42-47 as many as 6 students. The score is 48-53 as many as 8 students. The score is 54-59 as many as 6 students. The value of 60-65 as many as 4 students. The value of 66-71 is 3 students. For more details, see the frequency distribution table below. Table 1. Control class pretest frequency distribution Class Interval Class Class Edge Absolute Frequency Frelative (FR)% 1 30–35 29,5–35,5 3 8,57% 2 36–41 35,5–41,5 5 14,29% 3 42–47 41,5–47,5 6 17,14% 4 48–53 47,5–53,5 8 22,86% 5 54–59 53,5–59,5 6 17,14% 6 60–65 59,5–65,5 4 11,43% 7 66–71 65,5–71,5 3 8,57% Total 35 100% From the frequency distribution table, the data is then presented in the form of histogram graphs and polygon graphs. The form of data presentation that describes the ebb and flow of statistical data. The presentation of this data is so easy for readers to understand with the data obtained in the control class as follows: The data range is 29.5-35.5 as many as 3 students. The data range is 35.5–41.5 as many as 5 students. The data range is 41.5–47.5 as many as 6 students. The range of data is 47.5–53.5 as many as 8 students. The data range is 53.5–59.5 as many as 6 students. The data range is 59.5–65.5 as many as 4 students. The range of data is 65.5– 71.5 as many as 3 students. For more details, see the table below. Figure 1. Control class pretest histogram Figure 2. Control class pretest polygon Furthermore, the data is presented in the form of an ogive graph. This data is made to find out the data is less or more than a certain value. Based on the results of the pretest obtained in the control class, the following data were obtained: data less than 29.5 with 3 frequencies, less than 35.5 with 5 frequencies, less than 41.5 with 6 frequencies, less than 47.5 with 8 frequencies, less than 53.5 with 6 frequencies, less than 59.5 with 4 frequencies, less than 65.5 with 3 frequencies, and data less than 70.5 with 35 frequencies. For data more than a certain value as follows: data more than 29.5 as many as 35 frequencies, more than 35.5 as many as 27 frequencies, more than 41.5 as many as 21 frequencies, more than 47.5 as many as 13 frequencies, more than 53, 5 as many as 7 frequencies, more than 59.5 as many as 3 frequencies, more and data more than 65.5 as many as 0 frequencies. For more details, see the graph below: ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 327 Figure 3. Ogive pretest control class Experimental class pretest data Based on the results of the experimental class pretest at MTs Daarul Muqimin, Jati, Tangerang Regency. The data obtained from the respondents are as follows: The lowest pretest score obtained in the experimental class is 40 and the highest score is 80. Students who get a score of 40-45 are 4 students. Value 46-51 as many as 5 students. Values 52-57 as many as 6 students. The score is 58-63 as many as 10 students. The score is 64-69 as many as 5 students. The value of 70-75 is 3 students. The value of 76-81 is 2 students. For more details, see the frequency distribution table below Table 2. Experimental class pretest frequency distribution Class Interval Class Class Edge Absolute Frequency Frelative (Fr)% 1 40–45 39,5–45,5 4 11,43% 2 46–51 45,5–51,5 5 14,29% 3 52–57 51,5–57,5 6 17,14% 4 58–63 57,5–63,5 10 28,57% 5 64–69 63,5–69,5 5 14,29% 6 70–75 69,5–75,5 3 8,57% 7 76–81 75,5–81,5 2 5,71% Total 35 100% From the frequency distribution table, the data is then presented in the form of histogram graphs and polygon graphs. The form of data presentation that describes the ebb and flow of statistical data. The presentation of this data is so easy for readers to understand with the data obtained in the experimental class as follows: The data range is 39.5-45.5 as many as 4 students. The range of data is 45.5–51.5 as many as 5 students. The data range is 51.5–57.5 as many as 6 students. The data range is 57.5–63.5 as many as 10 students. The data range is 63.5–69.5 as many as 5 students. The data range is 69.5–75.5 as many as 3 students. The data range is 75.5–81.5 as many as 2 students. For more details, see the table below: 0 2 4 6 8 10 12 Absolute frequency Absolute frequency Figure 4. Experimental class pretest histogram Figure 5. Experiment class pretest polygon Furthermore, the data is presented in the form of an ogive graph. This data is made to find out the data is less or more than a certain value. Based on the results of the pretest obtained in the experimental class, the following data were obtained: data less than 39.5 with 4 frequencies, less than 45.5 with 9 frequencies, less than 51.5 with 15 frequencies, less than 57.5 with 25 frequencies, less than 63.5 with 30 frequencies, less than 69.5 with 33 frequencies, and data less than 75.5 with 35 frequencies. For data more than a certain value as follows: data more than 39.5 as many as 35 frequencies, more than 45.5 as many as 26 frequencies, more than 51.5 as many as 20 Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 328 frequencies, more than 57.5 as many as 10 frequencies, more than 63, 5 as many as 5 frequencies, more than 69.5 as many as 2 frequencies, and data more than 75.5 as many as 0 frequencies. For more details, see the graph below 0 5 10 15 20 25 30 35 Less than More than Figure 6. Ogive pretest experimental class Description of posttest data for control class and experiment class Control class posttest data Based on the results of the control class posttest at MTs Daarul Muqimin, Jati, Tangerang Regency. The data obtained from the respondents are as follows: The lowest posttest score obtained in the control class was 35 and the highest score was 80. Students who scored 35-41 were 6 students. Value 42-48 as many as 5 students. The score is 49-55 as many as 11 students. Value 56-62 as many as 5 students. The value of 63-69 as many as 4 students. Value 70-76 as many as 2 students. The value of 77-83 as many as 2 students. For more details, see the frequency distribution table below. Table 3. Control class posttest frequency distribution Class Interval Class Class Edge Absolute Frequency Frelative (Fr)% 1 35–41 34,5–41,5 6 17,14% 2 42–48 41,5–48,5 5 14,29% 3 49–55 48,5–55,5 11 31,43% 4 56–62 55,5–62,5 5 14,29% 5 63–69 62,5–69,5 4 11,43% 6 70–76 69,5–76,5 2 5,71% 7 77–83 76,5–83,5 2 5,71 % 35 100% From the frequency distribution table, the data is then presented in the form of histogram graphs and polygon graphs. The form of data presentation that describes the ebb and flow of statistical data. The presentation of this data is so easy for readers to understand with the data obtained in the control class as follows: The data range is 34.5-41.5 as many as 6 students. The data range is 41.5–48.5 as many as 5 students. The range of data is 48.5–55.5 as many as 11 students. The data range is 55.5–62.5 as many as 5 students. The range of data is 62.5– 69.5 as many as 4 students. The data range is 69.5– 76.5 as many as 2 students. The data range is 76.5– 82.5 as many as 2 students. For more details, see the table below. Figure 7. Control class posttest histogram Figure 8. Control class postest polygon Furthermore, the data is presented in the form of an ogive graph. This data is made to find out the data is less or more than a certain value. Based on the posttest results obtained in the control class, the following data were obtained: data less than 34.5 with 6 frequencies, less than 41.5 with 11 ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 329 frequencies, less than 48.5 with 22 frequencies, less than 55.5 with 26 frequencies, less than 62.5 with 31 frequencies, less than 69.5 with 33 frequencies, and data less than 76.5 with 35 frequencies. For data more than a certain value as follows: data more than 34.5 as many as 35 frequencies, more than 41.5 as many as 24 frequencies, more than 48.5 as many as 13 frequencies, more than 55.5 as many as 8 frequencies, more than 62, 5 as many as 4 frequencies, more than 69.5 as many as 2 frequencies, and data more than 76.5 as many as 2 frequencies. For more details, see the graph below: Figure 9. Ogive postes control class Experiment class posttest data Based on the results of the experimental class posttest at MTs Daarul Muqimin, Jati, Tangerang Regency. The data obtained from the respondents were as follows: The lowest posttest score obtained in the experimental class was 60 and the highest score was 90. Students who scored 60-64 were 2 students. The value of 65-69 as many as 3 students. The value of 70-74 is 4 students. The score is 75-79 as many as 6 students. The value of 80-84 as many as 4 students. The score is 85-89 as many as 10 students. The score is 90-94 as many as 6 students. For more details, see the frequency distribution table below. Table 4. Experimental class posttest frequency distribution Class Interval Class Class Edge Absolute Frequency Frelative (FR)% Interval 1 60–64 59,5–64,5 2 5,71% 2 65–69 64,5–69,5 3 8,57% 3 70–74 69,5–74,5 4 11,43% 4 75–79 74,5–79,5 6 17,14% 5 80–84 79,5–84,5 4 11,43% 6 85–89 84,5–89,5 10 28,57% 7 90–94 89,5–94,5 6 17,14% 35 100% From the frequency distribution table, the data is then presented in the form of histogram graphs and polygon graphs. The form of data presentation that describes the ebb and flow of statistical data. The presentation of this data is so easy for readers to understand with the data obtained in the experimental class as follows: The data range is 59.5–64.5 as many as 2 students. The data range is 64.5–69.5 as many as 3 students. The range of data is 69.5–74.5 as many as 4 students. The data range is 74.5–79.5 as many as 6 students. The data range is 79.5–84.5 as many as 4 students. The data range is 84.5–89.5 as many as 10 students. The data range is 89.5–94.5 as many as 6 students. For more details, see the table below. Figure 10. Experimental class posttest histogram Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 330 0 2 4 6 8 10 12 Absolute Frequency Absolute frequency Figure 11. Experiment class postes polygon Furthermore, the data is presented in the form of an ogive graph. This data is made to find out the data is less or more than a certain value. Based on the posttest results obtained in the experimental class, the following data were obtained: data less than 59.5 with 2 frequencies, less than 64.5 with 5 frequencies, less than 69.5 with 9 frequencies, less than 74.5 with 15 frequencies, less than 79.5 as many as 19 frequencies, less than 84.5 as many as 29 and data less than 89.5 as many as 35 frequencies. For data more than a certain value as follows: data more than 59.5 as many as 35 frequencies, more than 64.5 as many as 30 frequencies, more than 69.5 as many as 26 frequencies, more than 74.5 as many as 20 frequencies, more than 79, 5 as many as 16 frequencies, more than 84.5 as many as 6 frequencies, and data more than 89.5 as many as 0 frequencies. For more details, see the graph below: 0 5 10 15 20 25 30 35 40 Less than More than Figure 12. Ogive posttest experiment class Measures of concentration and distribution of pretest control and experimental classes The measure of data concentration is a description that provides an explanation that the data tends to converge or gather. Concentration measures that are often used are the average to determine the student's average score. Median to find out the middle value of a data. Standard deviation and variance to determine the size of the spread of statistics that measure how the data is spread out. Based on the results of the control class pretest research conducted at MTs Daarul Muqimin Buaran Jati Tangerang Regency in the control class obtained the following data: mean value (mean 80.71), (median 49.4), (mode 50.5), (standard deviation (sd ) 9.38) and (variant (sd²) 88.04). While the data obtained in the experimental class are the values: mean (average 58.61), (median 58.7), (mode 599.9), (standard deviation (sd) 10.26), and (variance (sd²) 105.24 ). For more details, see the table below: Table 5. Measures of concentration and spread of pretest control and experimental classes Class Mean Median Mode Sd Sd² Control 59,35 49,4 50,5 9,38 88,04 Experiment 58,61 58,7 59,9 10,26 105,24 The measure of concentration and distribution of posttest data in the control class and the experimental class Based on the results of the posttest control class research conducted at MTs Daarul Muqimin Buaran Jati, Tangerang Regency in the control class, the following data were obtained: mean value (mean 59.0), (median 51.5), (mode 51.5), (standard deviation (sd ) 9.8) and (variant (sd²) 96.6). While the data obtained in the experimental class are the values: mean (mean 80.71), (median 83.0), (mode 88.1), (standard deviation (sd) 9.38), and (variance (sd²) 88.04 ). For more details, see the table below: Table 6. Measures of centralization and distribution of control and experimental class posttest data Class Mean Median Mode Sd Sd² Control 54,00 51.5 51.5 12,20 148,75 Experiment 80,71 83.0 88.1 9,38 88,04 Testing data analysis requirements Normality test of pretest data for control class and experiment class Normality test of control class pretest data ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 331 One of the conditions that must be taken to perform the t-test is the normality test of the data. In this data normality test, the researcher used the chi square normality test. With the provision that if x² count < x² table, it can be concluded that the data is normally distributed. From the calculation of the chi square normality test for the control class pretest data, it is obtained that x² count is 1.379 if the value is consulted with the x² table (0.05: k – 1) then the x² table is 12,592. Because x² count < x² table = 1.379 < 12,592. It can be concluded that the control class pretest data is normally distributed. (data in appendix 12). Normality test of experimental class pretest data From the calculation of the chi square normality test of the experimental class pretest data, it is obtained that x² count is 2.413 if this value is consulted with the x² table (0.05: k – 1) then the x² table is 12,592. Because x² count < x² table = 2,413 < 12,592. It can be concluded that the experimental class pretest data is normally distributed. (data in appendix 13). For more details, see the table below: Table 7. Normality test of control and experimental class pretest data Class X² count X² table Classification Control 1,379 12,592 Normal Distribution Experiment 2,413 12,592 Normal Distribution Normality test of posttest data for control class and experiment class Normality test of control class posttest data One of the conditions that must be taken to perform the t-test is the normality test of the data. In this data normality test, the researcher used the chi square normality test. With the provision that if x² count < x² table, it can be concluded that the data is normally distributed. From the calculation of the chi squared normality test for the control class posttest data, it is obtained that x² count is 3.320 if this value is consulted with the x² table (0.05: k – 1) then the x² table is 12,592. Because x² count < x² table = 3.320 < 12,592. It can be concluded that the control class posttest data is normally distributed. Normality test of experimental class posttest data From the calculation of the chi squared normality test of the experimental class posttest data, it is obtained that x² count is 6.944 if this value is consulted with the x² table (0.05: k - 1), then the x² table is 12,592. Because x² count < x² table = 6,944 < 12,592. It can be concluded that the posttest data of the experimental class is normally distributed. (data in appendix 15). For more details can be seen in the table below. Table 8. Normality test of posttest data for control class and experiment class Class X² count X² table Classification Control 3,320 12,592 Normal Distribution Experiment 6,944 12,592 Normal Distribution Homogeneity test After the two samples in this study were declared from a normally distributed population, then the homogeneity of variance test of the two populations was then carried out using Fisher's exact test. The homogeneity of variance test is used to determine whether the two samples come from the same population (homogeneous) or different (heterogeneous). The test criteria used are that the two groups are said to be homogeneous if Fcount < Ftable is measured at a certain level of significance and confidence. Pretest data homogeneity test Based on the comparison of statistical data between the experimental group and the control group in the initial test or pretest, the largest variance was the control group, which was 114.62, while the smallest variance was the experimental group, which was 105.24. Then from the calculation results obtained Fcount = 1.09. With 34 degrees of freedom in the numerator and 34 in the denominator, we get Ftable = 1.77 with = 0.05. If Fcount is compared to Ftable, then Fcount < Ftable is obtained, namely 1.09 < 1.77, meaning that the pretest data for both classes is homogeneous. The results of the homogeneity test can be seen as follows: Table 9. Pretest data homogeneity test results Class S2 Fcount Ftable Conclusion Control 114.62 1,09 1,77 Both data are homogeneous Experiment 105.24 Posttest homogeneity test Based on the comparison of statistical data between the experimental group and the control group in the final test or post-test, the largest variance was the control group, which was 148.75, while the smallest Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 332 variance was the experimental group, which was 88.04. Then from the calculation results obtained Fcount = 1.69. With 34 degrees of freedom in the numerator and 34 in the denominator, we get Ftable = 1.77 with = 0.05. If Fcount is compared to Ftable, then Fcount < Ftable is obtained, namely 1.69 < 1.77, meaning that the posttest data of the two classes is homogeneous. The results of the homogeneity test can be seen as follows: Table 10. Results of homogeneity test of posttest data Class S2 Fcount Ftable Conclusion Control 148,75 1,69 1,77 Both data are homogeneous Experiment 88,04 Pretest control class and experiment class Based on the research findings above. As stated in the pretest frequency distribution table for the control class, the highest frequency lies at intervals of 48–53 at 22.86%, with an average value obtained in the control class as much as 50.16, the frequency is above the average of 14.29%. While below the average value of 11.43%. In the pretest frequency distribution table for the experimental class, the highest frequency lies in the interval 58–63 by 28.57%, with the average value obtained in the experimental class as much as 58.61, the frequency is above the average of 14.29%. While below the average value of 11.43%. Table 11. The pretest frequency distribution Class Location of the highest frequency Most frequrncy Mean Value above mean Value below mean Control 48–53 22,86% 50,16 14,29% 11,43% Experiment 58–63 28,57% 58,61 14,29% 11,43% Based on the table above, the differences between the two classes can be seen. The highest presentation was in the experimental class of 28.57%, and the control value was 22.86%. The value above the average for the control class is 50.16% while in the experimental class it is higher at 58.61%. The value below the average for the control class is 14.29%, and the value below the average for the experimental class is 14.29%. So it can be seen that the values in the control class are slightly lower. The difference is in the value above the average. While the histogram graph in the description above, it can be seen that the histogram graph of the highest control class pretest graph is located at a value of 47.5 as many as 9 respondents. The highest score was 71.5 with 2 respondents, while the lowest score was 29.5 with 3 respondents. In the histogram graph of the experimental class the highest graph is located at 68.5 with a total of 14 respondents. The highest score was 84.5 with 2 respondents, while the lowest score was 49.5 with 4 respondents. Posttest control class and experiment class For the posttest frequency distribution, the highest frequency is in the 50-55 interval of 36.36%, with an average value obtained in the control class as much as 52.32, the frequency is above the average of 15.15%. While below the average value of 12.12%. In the posttest frequency distribution table for the experimental class, the highest frequency lies in the 80-83 interval of 27.27%, with the average value obtained in the experimental class as much as 76.65, the frequency is above the average of 18.18%. While below the average value of 18.18%. Table 12. Posttest frequency distribution Class Location of the highest frequency Most frequrncy Mean Value above mean Value below mean Control 50–55 36,36% 52,32 15,15% 12,12% Experiment 80–83 27,27% 76,65 18,18% 18,18% Based on the table above, the differences between the two classes can be seen. The highest presentation was in the experimental class of 27.27%, and the control value was 36.36%. The value above the average for the control class is 52.32% while in the experimental class it is higher at 76.65%. The value below the average for the control class is 15.15%, and the value below the average for the experimental class is 18.18%. So it can be seen that the values in the control class are slightly lower. The difference is in the values above the average and below the average. While the histogram graph in the description above, it can be seen that the histogram graph of the posttest control class, the highest graph is located at a value of 49.5 as many as 12 respondents. The highest score is 79.5 with 1 respondent, while the lowest score is 37.5 with 7 respondents. In the posttest histogram graph of the experimental class the highest graph is located at a value of 79.5 with 9 respondents. The highest score is 87.5 with 6 respondents, while the lowest score is 59.5 with 2 respondents. ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 333 Based on the results of the pretest of the control class and the experimental class, the mean value (average) is a measure that gives an idea of the concentration of a data to determine the average value of a statistical data. The mean (average) value for the control class is (50.14), while in the experimental class the mean (average) value is (64.88). Meanwhile, during the post-test of the control class and the experimental class, the mean value in the control class was (52.32), and in the experimental class the mean value (76.65). For more details, see the table below: Figure 13. The mean value of the pretest and posttest for the control class and the experimental class 50,14 52,32 64,88 76,65 0 50 100 150 Pretest Postest Experiment Control Based on the graph above, the pretest data for the control class and the experimental class showed a significant difference. It can be seen from the graph height which is slightly far from the average. This is influenced because at the time of the pretest the researcher did not use treatment so that the students did not understand how to write descriptive text. From the post-test data the control and experimental classes showed a significant difference. It can be seen from the graph height of the two classes. The experimental class graph is higher than the experimental class graph giving treatment using the cooperative integrated reading composition method, while the research control class does not provide treatment so that there are differences in learning outcomes in the control class and in the experimental class. Furthermore, to find out whether there is an effect on the pretest data of the control class and the experimental class. The researcher conducted a t- test to find out whether there were differences in the two classes. After doing the calculations, it was obtained that tcount 1.1615 < ttable 1.9987, so it can be concluded that there is no difference in learning outcomes between the control class and the experimental class. This is because at the time of the pretest the researcher did not use the media (treatment). So that students are less able to write descriptive text. Likewise, the results of posttest research conducted a t-test to find out whether there were differences in the two classes. After calculating the results obtained tcount 2.061> ttable 1.9987. So it can be concluded that there are differences in learning outcomes between the control class and the experimental class. CONCLUSION The ability to write fables for the control class students who were not given any treatment was lower than the fable writing abilities of the experimental class students. By using the cooperative integrated reading composition method, students are more creative and more interested in writing fables. So it can be said that there is an effect of the cooperative integrated reading composition method on the ability to write fables of class VIII students. Mts Daarul Muqimin Buaran Jati, Tangerang Regency After performing the t-test with the results of the pretest class tcount (1.615) < ttable (1.9987), tcount is smaller than ttable, this indicates that there is no significant difference between the experimental class and the control class on the ability to write fables in the pretest. After the pretest was carried out, then a posttest was carried out, showing the results of tcount (2.061) > ttable (1.9987), tcount is greater than ttable, this indicates that there is a significant difference between the experimental class and the control class on the skills of writing fables in the posttest. REFERENCES Abdalrahman, K. K. (2021). Teaching and learning writing skills through literature. Canadian Journal of Language and Literature Studies, 1(2), 1-10. Akhadiah, Sabarti, G. A., Maidar, H. R. (2012). Pembinaan kemampuan menulis bahasa Indonesia. Jakarta: Erlangga. Aljannah, I. (2021). Pembelajaran kooperatif CIRC. Diakses dari https://izzaaljannah55.wordpress. com /2013/05/19/model-pembelajaran- cooperative-integrated-reading-and-composition https://izzaaljannah55.wordpress/ Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 334 Aqib, Z. (2010). Profesionalisme guru dalam pembelajaran. Surabaya: Insan Cendikia. Arikunto, S. (2010). Prosedur penelitian suatu pendekatan praktik. Jakarta: PT Rineka Cipta Arikunto, S. (2012). Dasar-dasar evaluasi pendidikan. Jakarta: Bumi Aksara Ariyana, A., Enawar, E., Ramdhani, I. S., & Sulaeman, A. (2020). The application of discovery learning models in learning to write descriptive texts. Journal of English Education and Teaching, 4(3), 401-412. Asiah, N., Ardian, E., & Amri, S. (2020). A study on the student’s factor difficulty in writing narrative text ar VII grade of MTS Sabilal Muthadin Tembilahan. J-Shelves of Indragiri (JSI), 1(2), 84–97. Aunurrahman (2012). Belajar dan pembelajaran. Bandung: Alfabeta. Budiyanto, A. K., (2016). Syntax 45 learning models in student centered learning (SCL). Malang: Universitas Muhammadiyah Malang. Fithriyani, I. (2019). Peningkatan perhatian, aktivitas, dan keterampilan menulis cerpen melalui model pembelajaran berbasis masalah dan media audio visual. Lingua Rima: Jurnal Pendidikan Bahasa dan Sastra Indonesia, 8(2), 11-23. Goziyah, G., Sulaeman, A., & Suherman, A. (2021). Pelatihan inovasi pembelajaran di era kenormalan baru pada guru-guru SMK Islam Baidhaul Ahkam Sepatan. Jurnal Abdimas Prakasa Dakara, 1(1), 12-17. Hudhana, W. D. (2019). Peningkatan keterampilan menulis cerita pendek berbasis karakter menggunakan media pembelajaran VideoScribe. Seminar Nasional Pendidikan Bahasa dan Sastra, 265-270. Hudhana, W. D., & Sulaeman, A. (2019). The development of video scribe in improving short story writing skill at tenth grade students of sman 1 tangerang regency. In the Proceeding of Annual International Seminar on English Language Teaching-AISELT, 3(3). Hutagalung, I. W. & Tanjung, H. R. (2021). Pengaruh penerapan model pembelajaran CIRC (Cooperative Integrated Reading and Composition) terhadap keterampilan menulis puisi modern siswa kelas X SMA Negeri 1 Sibabangun. Jurnal Bahasa dan Sastra Indonesia, 1(2), 1-10. Jayadi, U. (2021). Penerapan metode pembelajaran cooperative integrated reading and composition (CIRC) untuk meningkatkan hasil belajar dalam menemukan kalimat utama pada siswa kelas IV SDN 22 Mataram tahun pelajaran 2020/2021. Berajah Journal, 1(1), 21-42. Jayanti, A. D. (2019). Students’ writing ability on English descriptive text at grade VIII. Academic Journal of English Language and Education, 3(1), 71–94. Jihad, A. & Haris, A. (2013). Evaluasi pembelajaran. Yogyakarta: Multi Pressindo. Kette, E. S. S., Pratiwi, Y., & Sunoto, S. (2016). Pengembangan bahan pelatihan menulis cerita pendek bermuatan nilai karakter untuk guru SMP negeri mata pelajaran bahasa Indonesia se-kota Kupang. Jurnal Pendidikan: Teori, Penelitian, dan Pengembangan, 1(4), 698-704. Khulasoh, K. (2019). Meningkatkan keterampilan menulis puisi menggunakan media audio visual pokok bahasan melengkapi puisi anak berdasarkan gambar model pembelajaran tematik. Dinamika Jurnal Ilmiah Pendidikan Dasar, 9(1), 13-22. Ngalimun. (2013). Strategi dan model pembelajaran. Yogjakarta: Presssindo. Nolia. (2021). Model cooperative integrated reading and composition (CIRC) untuk meningkatkan hasil belajar Bahasa Indonesia materi surat resmi kelas VI SD. Sirok Bastra, 9(1), 23-24. Pangesty, D. A. R., Nursirwan, H., Marliah, A., Yasa, L. N., Hartono, R. (2021). The influence of Cooperative Integrated Reading and Composition (CIRC) model on students' written mathematical communication skills in primary school. Technium Social Science Journal, 22, 249-254. Prajogo, S. (2021). The use of cooperative integrated reading and composition (CIRC) method to improve students’ reading comprehension. Konstruktivisme: Jurnal Pendidikan dan Pembelajaran, 13(1), 78-85. Purnamasari, D., Hidayat, D. N., & Kurniawati, L. (2021). An analysis of students’ writing skill on English descriptive text. English Education: Jurnal Tadris Bahasa Inggris, 14(1), 101-114. Riadi, E. (2014). Metode statistika parametrik & non parametrik untuk penelitian ilmu-ilmu sosial dan pendidikan. Tangerang: Pustaka Mandiri. Ristanto, et.al. (2018). The potential of cooperative integrated reading and composition in biology learning at higher education. International Journal of Educational Research Review, 3(2):50- 56 Sharan, S. (2014). The handbook of cooperative learning. Yogyakarta: Istana Media. Shoimin, A. (2014). 68 innovative learning models in curriculum 2013. Yogyakarta: Ar-Ruzz Media. Sugiono (2012). Metode penelitian pendidikan kuantitatif, kualitatif dan R&D. Bandung: Alfabeta. Sugiono (2012). Statistika untuk penelitian. Bandung: Alfabeta. ENGLISH REVIEW: Journal of English Education p-ISSN 2301-7554, e-ISSN 2541-3643 Volume 10, Issue 1, December 2021 https://journal.uniku.ac.id/index.php/ERJEE 335 Sulaeman, A. & Goziyah.(2019). Metodologi penelitian bahasa dan sastra. Jakarta: Edu Pustaka. Sulaeman, A. (2019). Motivasi pembelajaran bahasa Indonesia bagi penutur asing (BIPA) mahasiswa peminatan BIPA prodi PBSI FKIP Universitas Muhammadiyah Tangerang. Simposium Nasional Mulitidisiplin (SinaMu), 1. Sulaeman, A., & Dwihudhana, W. (2019). Hubungan motivasi belajar dengan hasil pembelajaran Bahasa Indonesia bagi Penutur Asing (BIPA) pada mahasiswa Semester 7 program studi Pendidikan Bahasa dan Sastra Indonesia FKIP Universitas Muhammadiyah Tangerang. Silampari Bisa: Jurnal Penelitian Pendidikan Bahasa Indonesia, Daerah, dan Asing, 2(1), 59-70. Sulaeman, A., Enawar, E., Sori, S., & Suherman, A. (2021). Skimming reading techniques on the ability to identify intrinsic drama text elements. Journal of English Language and Literature (JELL), 6(1), 25-38. Supardi (2013). Aplikasi statistika dalam penelitian konsep statistika yang lebih komprehensif. Jakarta: Change Publication. Susanto, A. (2012). Teori belajar dan pembelajaran di sekolah dasar. Jakarta: Fajar Interpratama Mandiri. Tarigan, H. G. (2008). Menulis sebagai suatu keterampilan berbahasa. Bandung: Angkasa Bandung. Teulaumbanua, M. (2020). The application of the cooperative integrated reading and composition (CIRC) learning model in improving the ability to find elements of the nonfiction book for class VII students of SMP Negeri 1 Ma'u Nias Regency 2020/2021 learning year. EJI Education Journal of Indonesia, 1(1). Zainuddin. (2015). The effect of cooperative integrated reading and composition technique on students’ reading descriptive text achievement. English Language Teaching, 8(5), 11-21. Desri Arwen & Saiful Haq The effect of cooperative integrated reading and composition learning method towards fable writing skills 336