INCORPORATING RECIPROCAL QUESTIONING STRATEGY AND NUMBERED HEADS TOGETHER IN READING

Abastract This study aimed at investigating the effect of reciprocal questioning strategy combined with numbered heads together on reading comprehension of the eighth grade students in SMP Negeri 2 Singaraja. This study was an experimental study with post – test only control group design. The population of this study was the eighth grade students in SMP Negeri 2 Singaraja in academic year 2017/2018 with the sample of 73 students who were selected by using cluster random sampling. The data were collected using reading comprehension test. The data were analyzed by using independent samples t-test assisted with SPSS 24. The result showed that there was an effect of reciprocal questioning strategy combined with numbered heads together on reading comprehension of the eighth grade students in SMP Negeri 2 Singaraja. The result of independent samples t-test shows that the difference of reading comprehension between reciprocal questioning strategy combined with numbered heads together and reading guide strategy was significant (p<0.05). In effect size analysis, the result was 0.5. It means that the effect of reciprocal questioning strategy combined with numbered heads together is categorized as medium.


INTRODUCTION
Reading is one of four basic skills in learning language that should be mastered by the students. Reading skill is needed by the students to improve their knowledge. Rivers (1981) states that reading is the most important skill in which the students get new information and strengthen other areas of language in any language class. By reading, the students can get new knowledge and information. Reading also can develop one's mind, point of view, and perception (Zuchdi & Budiasih, 2001). Thus, reading is important skill for the students because by reading the students can get more new knowledge and information and it will develop the students' mind, point of view, and perception.
Reading is an activity for obtaining the information. This activity involves the reader and the text (Grabe, 2009). It means that the students as the readers read the text from the text the students get new knowledge and information. To get new knowledge and information, the students have to comprehend the text. As Supriyadi (1995) states that the real reading does not only read and pronounce the words in the text, but it is also about comprehending the text precisely. It means that besides of decoding written words in the text, the students have to understand the text. If the students only read the text without understanding the text, the students will not get any information from it. Therefore, it is important to note that reading without comprehension is just an activity to decode the text without understanding the text.
Comprehension is the goal of reading. As Nunan (2003) expresses that the goal of reading is comprehension. The students are expected to be able to comprehend the text. The students require to comprehend the text besides of decoding the written words. On the contrary, Zafarani & Kabgani (2014) declares that many students still do not know what they have read although they have read the words in the text. Reading is not simple activity (Moreillon, 2007). They do not only need to read the text, but they also need to achieve the comprehension about the text by activating their prior knowledge and match them with the information in the text.
Based on Peraturan Menteri Pendidikan dan Kebudayaan Nomor 21 Tahun 2016 about Standar Isi in primary and secondary education, reading activity in Junior High School in Indonesia is an activity that asks the students to read the text such as descriptive, recount, narrative, procedure, and report text. Then, the standard of reading competence in syllabus in the first semester of the eighth grade is to learn the simple meaning of interaction and interpersonal skill from the text, in formal and informal situation. The students have to understand the meaning of the text that they read related to the types of text in junior high school.
Although the standard of reading competence is stated in the curriculum, there is still a problem. The result of PISA (Programme for International Student Assessment) which was conducted by Organization for Economic Co -Operation and Development (OECD) in 2015 showed that Indonesia got 397 in reading performance. The OECD average in 2015 was 493. It can be seen the score 397 was categorized as low score. It was below the OECD average. Moreover, based on the preliminary interview with English teacher who teaches the eighth grade class in SMP Negeri 2 Singaraja, it was found that the students were not enthusiastic in reading. It was proven by most of students were lazy to read the text when the teacher asked them to read. In one class, there were only five students who wanted to read the text. In consequence, the students did not understand the text and difficult to get the information from the text. Alexander (1983) explains that the factors that influence the students in comprehending a text are the way of teaching reading, students' personality, motivation, habit, and social economy. It can be seen that motivation is one of factors in comprehending the text. Besides of motivation, teacher's way of teaching reading is another factor. There are some strategies that can be used in teaching reading such as KWL (Know, Want, and Learn), Reciprocal Questioning (ReQuest), and Question Answer Relationship (QAR). In this research, the researcher is interested in using reciprocal questioning strategy on reading comprehension.
Reciprocal questioning strategy is a strategy that helps the students understand the text through questions. Hamilton (2009) explains that reciprocal questioning strategy is to make the students actively engage in the text through questions. Moreillon (2007) claims that reciprocal questioning strategy make the students participate actively in the learning process because the students will be involved in such a think aloud process. This strategy makes the students participate actively with the teacher by creating and responding the questions. Both the teacher and the students have to create and respond questions. According to Hales (2009), reciprocal questioning strategy helps the students learn to create effective questions. The students learn to develop their own questions about the text so the students can understand the text independently. Since questioning as known can make the readers find their own way to comprehend the text. In short, reciprocal questioning strategy is a strategy that makes the students participate actively in the text through questions by involving the teacher and the students in creating and responding questions. This strategy also can help the students create effective questions about the text that they are reading.
In Indonesia, the class is categorized as big classroom size. Mostly, one class consist of 30 or more students. This condition makes the reciprocal questioning strategy is difficult to be implemented. It will take much time if the teacher should respond students' question one by one. Considering to this condition, the researcher wants to combine this strategy with numbered heads together where the students working in group to ask and answer questions. Numbered heads together will make the students work in group but all the members in the group will know their group work. The researcher chooses numbered head together is to avoid only one student or two students know their group work.
Numbered head together is a cooperative learning technique. This strategy asks the students to work in group to reach the answer and each member of the group must understand the answer because their number may be called to share their answer (Kagan & Kagan, 2009). The students are asked to form group and each member in the group has their own number. As Lie (2002) explains the students are form into group of four to five so the number of each member in the group from one to five. It depends on the group member. Rayanto (2017) explains that numbered heads together is one of the way to ensure equitable response opportunities by stop calling on the students who raise their hand. In summary, numbered heads together is a strategy that divides the students into group of four to five and each member has their own number from one to five depend on the group member. All the member in the group must understand their answer because their number may be called to share their answer, not the students who raise their hand have chance to share the answer.
Empirically, some researchers have conducted the similar study about reciprocal questioning strategy and numbered heads together. Such as Afrilia (2012) found Numbered heads together gave significant effect on reading comprehension achievement. Efriza, Mukhaiyar, & Radjab (2013) found that numbered heads together gave effect on reading comprehension in descriptive and narrative text of students with low and high reading motivation. Ardemelia (2013) found that the combination of DRTA strategy and REQUEST strategy gives positive effect and can improve the students' reading comprehension. Yuniarti, Rufinus, & Wijaya (2014) found that reciprocal questioning strategy was effective in teaching reading comprehension. Novita (2014) the combination of give one and get one strategies can develop students' critical thinking and reading comprehension. Both reciprocal questioning strategy and numbered heads together have good effect on reading comprehension. In this present study, reciprocal questioning strategy was combined with numbered heads together. This strategy was implemented in the eighth grade students of SMP Negeri 2 Singaraja.
Regarding to the previous explanation, theoretically reciprocal questioning strategy and numbered heads together can help the students to comprehend the text. Reciprocal questioning strategy which takes more time to be applied in big classroom size was combined with numbered heads together to save time. Moreover, reciprocal questioning strategy and numbered heads together have been proven effective to be applied in learning process by other researchers. Thus, the researcher investigated whether or not the reciprocal questioning strategy combined with numbered heads together has significant effect on reading comprehension of eighth grade students in SMP Negeri 2 Singaraja.

METHODS
This study was an experimental research. The design of this study was post-test only control group design. There was two group in this research, experimental group and control group. Experimental group was treated by using reciprocal questioning strategy combined with numbered heads together, while control group was treated by using reading guide strategy.
The population in this study was the eighth grade students in SMP Negeri 2 Singaraja academic year 2017/2018.
The sample of this study was taken by using cluster random sampling. The lottery was used in choosing two classes as the sample of this study. Then, another lottery was drawn to decide which class to be the experimental group and the control group. The sample of this research was VIII 14 and VIII 15 in SMP Negeri 2 Singaraja. Class VIII 14 with 38 students became the control group and class VIII 15 with 35 students became the experimental group.
The data was collected before giving treatment and after giving treatment. The data was collected by using reading comprehension test in the form of objectives test. Both groups were given test before giving treatment, it was to know the condition of both groups and to make sure whether both of groups are equivalent or not. While, the test after given treatment was to identify whether there is difference or not in both group after treated by using different strategy.
The data was analyzed by using two forms of statistical analysis in SPSS 24. Those two forms were descriptive analysis and inferential analysis. Descriptive statistic was used to describe the basic features of the data. Those basic features are mean, median, range, variance, and standard deviation. Descriptive statistic helps to show what the data shows trough tables or graphs. Inferential statistic was used to determine the significant different between two groups, experimental group and control group. In inferential statistic, the data of this research was analysed by using independent sample t-test. Independent sample t-test was used for comparing the score of two group which are not connected to each other. However, before analysing the data by using independent sample t-test, the normality and the homogeneity of the data were analysed by using normality test and homogeneity test. After independent sample t-test, effect size was administered to know to quantify the size of the difference between two groups.

RESULTS AND DISCUSSION
The class VIII 15 which was the experimental group was treated by using reciprocal questioning strategy combined with numbered heads together. The picture below is the picture of the experimental group during the treatment. The picture was taken during the treatment in the experimental group. In the picture, the students were working in the group in creating questions related to the text.
The result statistics analysis of pretest in both group experimental group and control group can be seen in Table 1. The normality test of pretest showed in Kolmogorov -Smirnov, the significant value of control group was 0.152. While, the significant value of experimental group was 0.200. The significances of both group were higher than 0.05. It meant that the data of both classes was from normal distribution.
The homogeneity of pretest showed that the significant value of based on mean was 0.101. The value was higher than 0.05. It meant that the data of both classes were homogenous. The data was assumed homogenous. In independent samples test of pre -test, the significant value is 0.768. It was higher than 0.05. It indicated that there was no significant difference on reading comprehension between the mean of both groups, control group and experimental group.
The result of statistics analysis of posttest can be seen in Table 2. The mean scores in statistics analysis of showed that both groups were not significant different while in posttest, the experimental group performed better. The data distribution of pretest and post test is shown in Figure 2. The normality test of posttest showed that in Kolmogorov-Smirnov was 0.122, while the significant value of experimental group was 0.200. The significant value of both group exceeded 0.05. It meant that the data of both group was distributed normal.
The homogeneity test of posttest showed that the significant value in based on mean was 0.712. The value exceeded 0.05. It means that the data of both group was homogenous.
In independent samples t-test, the significant value in equal variances assumed need to be read. The significant value is 0.027. The significant value did not exceed 0.05. It means that there was difference on reading comprehension between students who were taught using reciprocal questioning strategy combined with numbered heads together and the students who were taught using reading guide strategy. Thus, null hypothesis was rejected. It indicates that there was a significant effect of reciprocal questioning strategy on reading comprehension of the eighth grade students.
The result of effect size was 0.5. In Cohen's effect size guideline, the value 0.5 was categorized as medium. Thus, the effect of reciprocal questioning strategy combined with numbered heads together on students' reading comprehension was categorized as medium.
This study is to investigate the effect of reciprocal questioning strategy combined with numbered heads together on reading comprehension of the eighth grade students. Based on descriptive statistical analysis of the post test, it was showed that the students who were taught by using reciprocal questioning strategy combined with numbered heads together performed better than those who were taught by using reading guide strategy. It was proven by the result of reading comprehension test (post-test). The mean score of students who were taught by using reciprocal questioning strategy was 74.29 while the mean score of students who were taught by using reading guide strategy was 68.34.
Moreover, both of the groups, experimental group which was class VIII 15 and control group which was class VIII 14 were given pretest before the treatment was conducted. The normality and homogeneity test was administered for the pretest. In normality test, the significant value in Kolmogorov -Smirnov was 0.152 for the class VIII 14 while for class VIII 15, the significant value was 0.200. It was higher than 0.05. Both of groups came from normal distribution. In homogeneity of variance based on mean in Levene's statistics, the significant value was 0.101. It exceeded 0.05. It meant that both of groups were homogenous. The independent samples t-test was also administered. The significant value of pretest was 0.768. The significant value is higher than 0.05. It means that both groups were not significantly different before giving treatment. It was proven by the result of descriptive statistical analysis. It showed that there was no class which performed better. The mean of class VIII 14 was 67.58 while the mean of class VIII 15 was 66.77. Thus, before the treatment was conducted in each group, both of groups were equal.
Reviewing the previous studies, this study has similarities and differences. The similarity between this study and the previous studies is to investigate reciprocal questioning strategy and numbered heads together on reading comprehension. The difference is this study combines the reciprocal questioning strategy and numbered heads together. Some researchers had combined reciprocal questioning strategy but with different strategy not with numbered heads together. Ardemelia (2013) conducted a study combined reciprocal questioning strategy with direct reading thinking activity. In different year, Novita (2014) combined the reciprocal questioning strategy with give one and get one strategies. From all those studies, the result show that reciprocal questioning strategy gives good effect on reading comprehension In this study, the experimental group was treated by using reciprocal questioning strategy combined with numbered heads together. The class which was taught by using reciprocal questioning strategy combined with numbered heads together consisted of 35 students. Because of a large class, the students worked in group to create and answer questions related to the text that they have read in treatment given. Word (2011) explains that reciprocal questioning strategy is a strategy that involves the teacher and the students in creating and responding questions. During the treatment, the students created question that related to the text. The teacher answered the students' questions. Afterward, the teacher asked the questions to the students and the students answer the teacher's questions. Thus, not only the teacher ask question but the students asked question. And not only the students answered the questions, the teacher also answered the questions.
During the treatment, the students also were motivated because the students did many activities. The students did not only ask to read the text but they were challenged to create appropriate questions to be asked to the teacher. They also wanted to read the text in order that they could answer the questions that were given by the teacher. This is in line with the statement by Moreillon (2007) that reciprocal questioning strategy is to make the students participate actively in the learning process because the students will be involved in such a think aloud process. The student did not only think for answering questions but the students also had to think how to create questions that related to the text.
This strategy asks the students to create questions that related to the text. By asking the students to create appropriate questions, one advantage of implementing reciprocal questioning strategy is it helps the students how to create effective questions (Hales, 2009). During the treatment, the students were given a text and instructed to ask questions that related to the text. The students were also given questions' guideline in order they could create questions effectively. They had example of the questions. They knew what questions must be created and had limitation in creating question in order that the students' question was not out of the goal.
Reciprocal questioning strategy was combined with numbered heads together. During the treatment, the students were asked to work in group and each member in group had different number. All the member in the group had to know their group question which was asked and all the member in the group also had to know their group answer. The student's number was called randomly to ask question or answer the question so not only the group's leader knew the question or the answer of their group. In line with the statement by Kagan & Kagan (2009) that each member in the group must understand their group work because their number may be called to share their group work.
Students in experimental worked in group in creating and answering question. Each member tried to share their question to the other members. The other members tried to hear carefully and choose the best question for their group. It was the same as answering questions, each member tried to share the opinion to the other members until choose the right answer. They tried to speak sharing their opinion and hear the other members' opinion can make the students more productive in learning. The students did not only become good speaker but they had to become good listener. As stated by Kagan & Kagan (2009), numbered heads together is designed for students to share information, being good listener, and try to speak in order that the students can be more productive.
One advantage of number heads together is the member in group helps themselves and their whole group, because the given response belongs to whole group not just to the group member who shares it (Jacobs & Hall, 2006)To help themselves, the students seemed often read their dictionary to find words that they did not understand. They also tried to find information by themselves besides of discussing something that they did not understand with their friends. The member in group tried to solve their problem with their friends instead of asking the teacher. Until the students finally solved the problem in their group, the number of student was being called randomly. The students whose number was being called must be ready because they had responsibility to share their group work to all the students in the class. The given answer belongs to whole group not just to the group member who shares it.
Numbered heads together also gives opportunity for the students to build togetherness and appreciate the diversity. In line with the statement by Krismanto (2003) the advantage of implementing numbered heads together are to teach the students to cooperate with other students and appreciate other students opinion, to train how to be good tutor for their friends, to build togetherness, and to make the students used in diversity. It could be seen in creating and answering questions, the students worked with their friends who were in same group. There were more than one opinion in a group. To get the final question and answer used for their group, they accepted all the opinions from every member but they discussed together to choose the correct one. This situation put the students in diversity. Thus, they appreciated it to make decision.
In this study, the treatment was conducted six times for each group. Before conducting treatment, both group experimental group and control group were given pre-test. The inferential statistics of pretest showed that both group came from normal and homogeny distribution. The independent samples test was also administered. It showed that both group were not significantly different before giving treatment. The treatment in both group was conducted in whilst activity. At the beginning of the treatment, both group were introduced about recount text such as generic structure, social function, and language features because it was the first time for the students learning about recount text. In the next meeting, at the beginning of the treatment the students were only reminded about the recount text.
In whilst activity, the experimental group was taught by using reciprocal questioning strategy combined with numbered heads together. Began with asking the students to form a group consists of four to five members. Then, they divided number for each member from one to five or one to four depend on the number of the group. Every member in the group got different number. In line with the statement by Lie (2002) that the students are form into group of four to five so the number of each member in the group from one to five. It depends on the group member. Following activity, the researcher told the procedure of reciprocal questioning strategy combined with numbered heads together to the students. Meanwhile, the control group was treated by using reading guide strategy which a strategy was used by English teacher in SMP negeri 2 singaraja. This strategy asked the students to read the text by themselves and find the meaning of the difficult words. During the treatment in both group, six teaching scenarios for each group were used in order to cover the learning objective for both groups.
In the experimental group, all the students got their group and their own number. The students were given a recount text and questions guideline. The students were instructed what they should do. Then, the students read the text silently continuing with creating question appropriate with the question guideline. The students in each group showed enthusiasm to share their question. They discussed together in a group which question that they could not answer would be asked to the teacher. Activity continued by calling the student's number randomly. The students whose number was being called asked the question for their group. The questions were written on the whiteboard. As what Hamilton (2009); Kagan & Kagan (2009) say that the students actively engage with the text through questions in which the students work in group in creating and answering questions and all the member must understand their group work because their number may be called to share their group work.
After all the group asked the questions, the researcher answered the questions but before that the researcher gave chance to the students to answer first. Some questions were able to be answered by the students but the rest were answered by the researcher. When all the questions from the students had already been answered. The students were instructed to put aside their text. The researcher asked questions. As what Manzo & Manzo (1995) explain that after the students ask questions to the teacher. The teacher has the turn to ask questions to the students which cover all the information of the text.
All the students in the group seemed really motivated to discuss for the answers because all the students in the group must know their group answer in order that they could answer when their number was being called. Then, the student's number again was being called. The students with that number answered the given question as the representative for their group. The students whose number was not being called were also really excited for answering the question. It was proven by all the students raised their hand. However, only the students whose the number was being called share their group work. In line with the statement by Rayanto (2017) that this technique is one of the way to ensure equitable response opportunities by stop calling on students who raise their hand. In the end of the treatment, the students were given exercise in the form of multiple choice related to the text that they have read.
On the other hand, the control group was treated by using reading guide strategy. The treatment began by giving recount text and list of difficult words to the students. The students were instructed to find the meaning of difficult words. After all the students answered the difficult words, the class discussed together the meaning of difficult words. Then, the students read the text by themselves. After all the students finished reading the text, the students were asked questions about the generic structure, language feature such as the synonym of some words, and specific information. Some students were active to answer the questions but mostly the questions were answered by same students. In the end of treatment, the students in control group were also given exercise same as the experimental group related to the text that they have read.
The obstacles during the treatment was on the first meeting in experimental group the students got confused what activity that they must be done. By giving more instruction carefully and gently, the students could more understand what they should do. It was proven in the next meeting they knew what activity that they must be done moreover they have done the activity in the first meeting. In creating group, the students were difficult to be controlled. They were too much talking. It made them took much time in creating group. The researcher had to be more firm in instructing them. The students also got difficulty in creating questions that related to the text. They was confused creating appropriate questions. They asked a lot of questions to the researcher and took long time in creating questions. The students had given the questions guideline but on the first meeting the students were difficult to create questions. Thus, the researcher gave more explanation and other examples. Later on the second to the sixth meeting, the students showed improvement. They took shorter time in creating appropriate questions. On the first meeting, when the student's number was called randomly. They got confused with their number and they did not know what they must do. After one of the student was done the activity, they started to understand. The next meeting, more students understood what they had to do and they did not get confused with their number again.
After the treatment was conducted in both group, both of group were given post-test. The inferential analysis was administered in posttest. In inferential analysis, it analysed the normality and homogeneity. In normality test, the significant value of control group in Kolmogorov-Smirnov was 0.122, while the significant value of experimental group was 0.200. The significant value of both group exceeded 0.05. It meant that the data of both group was distributed normal. In homogeneity of variance based on mean in Levene's statistics, the significant value was 0.712. It was higher than 0.05. It meant that both of group was homogenous. After the homogeneity and normality test was conducted, the independent samples t-test was administered to test the hypothesis. The test showed that the significant value was 0.027. The significant value did not exceed 0.05. It indicated that there was difference between the students who were taught by using reciprocal questioning strategy combined with numbered heads together and the students who were taught by using reading guide strategy. It meant that the null hypothesis was rejected. The effect size analysis was conducted after independent samples test. The result of effect size was 0.5. In Cohen's effect size guideline, the value 0.5 was categorized as medium. Thus, the difference between students who were taught by using reciprocal questioning strategy combined with numbered heads together and the students who were taught by using reading guide strategy was categorized as medium.
Considering to the result of the study and the previous explanation, it can be concluded that reciprocal questioning strategy combined with numbered heads together was effective for students' reading comprehension. Ardemelia (2013) conducted a research about combining reciprocal questioning strategy with direct reading thinking activity. She found that reciprocal questioning strategy combined with direct reading thinking gives positive effect on reading comprehension. Other researchers also conducted researches without combining with other strategy. In 2014, Yuniarti, Rufinus, and Wijaya conducted a research about the effect of reciprocal questioning strategy in teaching reading comprehension. The result is this strategy gives good effect in teaching reading comprehension. Risnaldi, Usman, & Diana, (2016) conducted a research about the impact of implementation of numbered heads together on reading comprehension. It was found that numbered heads together gives positive impact on reading comprehension. The result of both descriptive and inferential statistics analysis also strengthen the previous statement that reciprocal questioning strategy combined with numbered heads together was effective for students' reading comprehension in which the null hypothesisthere is no significant effect of reciprocal questioning strategy combined numbered heads together on reading comprehension of the eight grade students was rejected. It meant that there was significant effect of reciprocal questioning strategy combined with numbered heads together on reading comprehension of the eighth grade students in SMP Negeri 2 Singaraja.

CONCLUSION
Based on the result of data analysis through descriptive and inferential analysis, it can be concluded that reciprocal questioning strategy combined with numbered heads together has significant effect on reading comprehension of the eighth grade students in SMP Negeri 2 Singaraja. The result of effect size was 0.5. In Cohen's effect size guideline, the value 0.5 was categorized as medium. Thus, Reciprocal questioning strategy combined with numbered heads together gives significant effect on reading comprehension of the eighth grade students in SMP Negeri 2 Singaraja. The effect is categorized as medium.
In line with conclusion above, some suggestion can be proposed to teachers, students, and other researchers.
a. For the teachers The teachers are recommended to implement reciprocal questioning strategy combined with numbered heads together in teaching English especially in teaching reading. This strategy can make the students more active in the class and make the students understand the text easier. This strategy makes the students engage actively in the text through questions. They can understand the text easier and learn to create proper and related questions about text. b. For the students The students are recommended to learn in understanding the text through question and creating effective question. Through questions, the students will be easier to understand the text. c. For other researchers The other researcher who are interested in the area of teaching English. Reciprocal questioning strategy combined with numbered heads together can help in understanding the text. It is also recommended for other researchers to conduct experiment about the use of reciprocal questioning strategy in Senior High School. The other researchers can combine reciprocal questioning strategy with the other strategy or combine numbered heads together with the other strategy.