Interactional Approach in Improving Students’ Writing Skills as A Strengthening of The Ability of Religious Moderation at State Islamic Institute of Kediri

This article aims at investigating the effect of interactional approach in improving students’ writing skill as a strengthening of the ability of religious moderation at State Islamic Institute of Kediri. By employing a quasi experimental research design, two groups of students participated in this study. The experimental group was taught by using interactional approach, then the control group was taught by using non interactional approach (self-correction). Direct writing test was used to collect the data, and ANCOVA was used to analyze the data. The finding shows that both approaches can improve students’ writing skill. It can be seen from the improvement of students’ writing mean score from pretest to posttest. The experimental group got 19.18 improvement, and the control group got 16.45 improvement. It seems that the experimental group achieved higher improvement. However, based on the result on ANCOVA calculation, the sig.value obtained was 0.418; it was higher than the significant level (0.05). It means that there was not enough evidence to reject the null hypothesis. There was no significant difference on writing skill between the both groups. Those approaches affect on improvement of students’ writing skill. The implication is that both approaches can be used to improve students’ writing skill. By having good writing skill, they have more chances to do religious proselytizing (dakwah) through producing or writing religious books or written products in English that can be read by many people in the world, and as the form of religious moderation in Islamic higher education.


Introduction
Acquiring foreign language including English language is really important for Islamic students (santri). It is one of the forms of religious moderation in Islamic higher education. As Karim stated that the main constancy and flexibility in Islamic moderation become the principles in adjusting with any situation in all times (Karim, 2019, p. 1). By having good English language acquisition, students can be able to adjust with the current situation in which globalization is influencing entire nations in the world. Moreover, English is being international language now; therefore, acquiring English as international language becomes necessary to come up with the current condition. Islamic students will be able to receive, understand and get more knowledge and information if they have good proficiency in English, since many sources are written in English. In addition, Islamic students will have more changes in doing religious proselytizing (dakwah) in worldwide by using English orally or written if they have good speaking and writing skills. Therefore, teaching and learning English especially on speaking and writing skill is really important for Islamic students.
Writing is one of compulsory course that must be followed by all English students at English Study Program of State Islamic Institute of Kediri. Moreover, the students should have good writing skill in order to be able to graduate from this institution. They must demonstrate their writing skill especially in writing thesis as one of the requirements to graduate. Considering the importance of writing for students, there are some writing courses offered in English Study Program. Those are Sentence and Paragraph Writing, Essay writing, Academic writing, Thesis Proposal Writing and Seminar, and the last is Thesis.
Writing as one of productive skills is still considered as a difficult skill to be mastered by second and foreign language students. Studies on writing in Indonesian context have found that it is a hard skill to be learned (Muth'im, 2010;Widiati & Cahyono, 2006). The complexity of writing and the differences of the language convention between the first language and written English make students confront barrier with writing. Because of the difficulties in writing, students very often make errors in their writing.
Error on students' writing becomes something common in English as a foreign language. It is most often represented as an inadequate knowledge of rules for the L2 or FL students. Based on a behaviorist perspective, error is a bad habit which should be changed through learning (Ravand & Rasekh, 2011, p. 1136. Through teaching and learning process, the students can be more aware on their errors and overcome their errors. In instructional process, teacher may give response to students' writing to help students identify their errors. The teachers' response on students' writing is called as feedback. The provision of feedback in the writing process is widely believed that it is important. It can improve students' writing skill (Bitchener & Knoch, 2010;Wahyuni, 2018). Feedback in the writing process makes students concentrate more on what is being learned (Muth'im & Latief, 2014). Based on Lewis, there are five roles of giving feedback. The first is providing information for teachers and students. Next, it is providing learning advice for students. Third, language input is provided through feedback for students. Fourth, motivation can be formed also through feedback provision. Last, it can lead students to be more autonomous (Lewis, 2002, pp. 2-3). In brief, feedback is a kind of information telling students about their learning task performance with the purpose to improve their performance.
However, there is a debate on the roles of feedback. Some studies found that comments on feedback are unhelpful, badly timed, and do not address what they want them to address (Sadler, 2010;Urquhart et al., 2014). Feedback is just considered as information transmission or "telling" or language input (a cognitivist point of view) without any dialogue or interaction between provider and receiver. It is in contrast with a socio-constructivist point of view that feedback should be dialogic and help to develop ability of students to monitor, evaluate and regulate their learning (Ajjawi & Boud, 2017). There should be interaction (dialogue/interactional approach) between the feedback provider and receiver.
Interactional approach is a teaching approach emphasizing interaction/dialogue in the instructional process. Interactional approach in the teaching of writing, the focus is placed on meeting the need of students both in obtaining control over the English written conventions and chances for self expression (Lestari, 2008, p. 44). It allows students to get feedback from both teacher and peers during the writing process. Feedback can be provided to any aspects of composition such as content, organization or language use. In the interactional approach, the teacher's feedback is commonly in paragraph-level written comments. It is not in phraselevel comments or oral comments or in the form of graphic devices, such as underlining, circles, arrows, etc. Then the peer may give feedback in all aspects of writing.
Furthermore, according to Ajjawi and Boud, feedback should be considered not a language input but more on dialogical process (Ajjawi & Boud, 2017). It is in line with a socio-constructivist point of view that feedback should be dialogic. It should be able to help students develop their ability to monitor, evaluate and regulate their learning. Ravand and Rasekh said that it is a need to continue to take a multi-dimensional view of feedback (Ravand & Rasekh, 2011). The nature, content, mode and quality of feedback provision should be separately and interactively considered along with the differences of situational and individual in the instructional context. Therefore, this study tries to investigate the effect of interactional approach in improving students' writing skill as a strengthening of the ability of religious moderation at State Islamic Institute of Kediri.

Interactional Approach in Writing Classroom
Many researches on feedback have been conducted; however, research on interactional approach in providing feedback on students' writing (feedback as a dialogic process) is still limited. Studies on feedback have been conducted by Bitchener and knoch and Wahyuni; they found that the feedback provision can improve students' writing skill (Bitchener & Knoch, 2010;Wahyuni, 2018). However, the other studies found that feedback comments are unhelpful, badly timed and do not address what they want them to address (Sadler, 2010;Urquhart et al., 2014). Therefore, the feedback provision should be more dialogic (interactional) to make it more appropriate with the students' need.
Studies on interactional approach have been conducted by some researchers ( e.g. Jones, Garralda, and Lock, 2006;Lestari, 2008;Ravand and Rasekh, 2011;Ajjawi and Boud, 2017). Jones, Garralda and Lock found that interactional dynamics makes writers have more improvement on their writing. However, there is difference between interactional dynamics in on-line and face-to-face peer tutoring sessions. Face-to-face interactions made more hierarchal encounters in which instructors took more control of the discourse; however, in on-line interaction, it was more egalitarian, in which clients took more control on the discourse (Jones et al., 2006).
In line with Jones, Garralda and Lock (2006), Lestari also found that the interactional approach in the teaching of writing allows students to get beneficial in getting feedback from both teachers and peers. The feedback provision makes the four hypotheses in second language acquisition (the input, monitor, noticing and output hypotheses) work simultaneously during the writing process for better output or the students' final draft of writing can be achieved (Lestari, 2008). Furthermore, Ravand and Rasekh found in their study that it is a need to continue to take a multi-dimensional view of feedback. The nature, content, mode and quality of feedback provision should be separately and interactively considered along with the differences of situational and individual in the instructional context. The interactions among task characteristics, instructional context, and students' characteristics are potential to be investigated more (Ravand & Rasekh, 2011, p. 1142).
Ajjawi and Boud's study has similar finding with the previous studies above. They found that interactional approach can enable insight into feedback undocumented aspects previously for instance the interactional features that promote and sustain dialogue of feedback. Feedback should be considered not a language input but more on dialogical process. It is in line with a socio-constructivist point of view that feedback should be dialogic and help to develop the ability of students to monitor, evaluate and regulate their learning (Ajjawi & Boud, 2017). : post-test A : treatment for experimental group by using interactional approach B : treatment for control group by using non-interactional approach (self-correction) Two classes of the second semester English students of IAIN Kediri are taken as the sample of this study. They are assigned as the experimental group and the control group. Based on the result of pre-test, both groups have quite similar in their writing quality. However, not all students in both classes participated completely. Only 27 students in class B and 11 students in class C joined this study. There are some reasons of this condition. First, the classroom interaction was done through online (Google Classroom) since there is Covid-19 Pandemic, and offline classroom meeting is forbidden. Second, only the students who can join the treatment and submit their writing on time can be the participants of this study. The students join the online classroom from their home, and really often they get problem on their connection. Therefore, only the students who have no problem in their online connection, can join the treatment, and submit the writing on time can be the participants of this study.

Methods
To achieve the purpose of this study that is to investigate the effect of interactional approach in improving students' writing skill as a strengthening of the ability of religious moderation at State Islamic Institute of Kediri., the present study used a quasi-experimental research design since it was conducted during the regular class, and randomized the participants was not allowed. Therefore, the existing classes of students were used in the present study. The experimental design of this research can be seen in Table 1.
If they are late, they are eliminated as the participants.
The experimental group is taught by using interactional approach, and the control group is taught by using non-interactional approach. For the experimental group, there is interaction (dialogue) between provider and receiver of the feedback. However, for the control group, there is no interaction (dialogue) between provider and receiver; the students do self-correction on their own writing. To collect the data on students' writing skill, direct writing test is used as the instrument of the research. The writing test is asking the students to write expository writing. The test is divided into two: pre-test and post-test. In scoring the students' writing, an analytical scoring rubric adopted from Jacob for writing is used (Jacobs, 1981). It consists of five components: content, organization, vocabulary, language use and mechanics. Then to analyze the data, Analisis of Covariate (ANCOVA) is used.

Results
The purpose of this study is investigating the effect of interactional approach in improving students' writing skill as a strengthening of the ability of religious moderation at State Islamic Institute of Kediri. Based on the result of analysis, the findings of this research are divided into some parts. Those are the result of pre-test and post-test for the both groups, experimental and control groups, fulfillment of ANCOVA assumptions, and the result of ANCONA analysis.

The Result of Pre-Test
Pre-test was given for both groups, experimental and control groups before the treatment was conducted. As stated previously, in the pre-test, the students were asked to write an expository paragraph on a given topic in 60 minutes. Then their writing products were score by using an analytical scoring rubric. The summary result can be seen in Table 2 Table 2 shows that both groups, exprimental and control groups have quite similar characteristics. In the mean score of the pre-test, the experimental group gets 56.52; meanwhile, the control group gets 56.73. Eventhough the both groups have different minimum scores (37 for the experimental group and 42 for the control group), the both groups have the quite similar maximum score (78 for the experimental group and 76 for the control group). The far difference is in the standar deviation. The experimental group has smaller standard deviation than the control group. It implies that the experimental group has more homogeneous score than the control group.

The Result of Post-Test
Similar to pre-test, post-test was given for both groups, experimental and control groups; it is conducted after the treatment was given. As mentioned previously, in the post-test, the students were asked to write an expository paragraph on a given topic in 60 minutes. Then their writing products were score by using an analytical scoring rubric. The summary result can be seen in Table 3.  Table 3 shows that both groups, exprimental and control groups do not have same characteristics. In the mean score of the post-test, the experimental group gets 75.70; meanwhile, the control group gets 73.18. Eventhough the both groups have quite similar minimum scores (62 for the experimental group and 63 for the control group), the both groups have different maximum score (98 for the experimental group and 86 for the control group). Similar to the result of pre-test, the far difference is in the standar deviation. However, the experimental group has bigger standard deviation than the control group. It implies that the control group has more homogeneous score than the experimental group.

The Result of ANCOVA Assumptions Fulfillment
To go further with ANCOVA, there are some statistical assumptions needed to be fulfilled. Those are data normality distribution, homogeneity variances, homogeneity regression, and linier relationship between covariates and dependent variable.

Assumption of Data Normality Distribution
The normality distribution testing of the data was the first assumption that should be fulfilled. The criteria of the data normality testing in this research used the general level of significance in Kolmogorov-Smirnov scale (Dörnyei, 2011). Distribution of the data was normal if Sig. > .05. The result of the computation is presented in Table 4. From Table 4, it shows that the highest obtained value was .217, and the lowest obtained value was .091. The whole obtained values were higher than the level of significance in Kolmogorov-Smirnov (.05); it means that the data did not deviate from the normal distribution. In other words, all the data produced by the samples looked like a bell-shaped curve. As the data distributions were normal then the data fulfilled the criteria to be used for testing the hypothesis by using parametric test.

Assumption of Homogeneity Variances
The second statistical assumption is homogeneity variance testing. It aimed to determine whether the variant of sample was homogeneous or not. Levene's test for equality of variances was employed to fulfill the assumption of homogeneity of variance in the present study, The variances of the data were equally homogeneous if Sig. > .05. The result of homogeneity variances testing is presented in Table 5. From Table 5, it shows that the result of homogeneity variances testing was .094; it is higher than .05 (the significance level of Levene's Test). It means that there was enough evidence to state that the variance was homogenous. In other words, the underlying assumption of homogeneity variance had been met; the variant of sample was homogeneous. From the results of data normality testing and homogeneity testing of variance, the two assumptions were fulfilled. Therefore, it can be proceed to analyze the data using parametric test.

Assumption of Homogeneity Regression
Homogeneity regression testing aimed to estimate the interaction of covariate (pre-test) and independent variable (teaching approach/ interactional and non-interactional approach) in predicting the dependent variable. In analyzing data using ANCOVA, the covariate must be no interaction with independent variable. The significant value must be higher than the significant level (Sig. > .05). The test result of homogeneity regression is presented in Table 6.  Table 6, it can be seen that the significant value was 0.956. It was higher than 0.05. It means that there is no interaction between covariate (pre-test) and independent variable (teaching approach/interactional and non-interactional approach). Therefore, this assumption can be fulfilled.

Assumption of Linier Relationship Between Covariates and Dependent Variable
The last assumption was assumption of a linier relationship between covariate (pre-test) and dependent variable (post-test). The significant value obtained must be smaller than the significant level (õ < .05). The covariate is included to control the differences on the independent variable in the analysis. Evaluating the relationship between the covariate and the dependent variable is the primary purpose of covariate testing. The result of a linier relationship between covariate and the dependent variable testing is presented in Table 7.  Table 7 shows that the significant value obtained was .007. It was smaller than the significant level (.007 < .05). It means that there is a linier relationship between the covariate and the dependent variable. Therefore, the assumption was fulfilled, and it can be continued to proceed with ANCOVA.

The Result of ANCOVA
As presented previously, the testing results of the assumptions for ANCOVA had been met the requirements. Therefore, the present study can proceed with ANCOVA analysis. The use of ANCOVA analysis in this study is to examine the hypotheses of the study. The hypothesis of this study is presented as follow.

Ho
: There is no significant difference on writing skill between students' taught by using interactional approach and those taught by using non-interactional approach.
Ha : There is significant difference on writing skill between students taught by using interactional approach and those taught by using non-interactional approach.
If the significant value obtained is higher than the significant level (p > 0.05), the null hypothesis (Ho) cannot be rejected. However, if the significant value obtained is smaller than the significant level (p < 0.05), the null hypothesis can be rejected. The result of ANCOVA computation is presented in Table 8.  Table 8 shows that the significant value obtained was 0.418; it was higher than the significant level (0.05). It means that there is not enough evidence to reject the null hypothesis. In other words, the null hypothesis cannot be rejected. There is no significant difference on writing skill between students taught by using interactional approach and those taught by using non-interactional approach. Those approaches may affect on improvement of students' writing; however, there is no significant diference on both approaches in affecting the students' writing.

Discussion
After analyzing the data taken from the research instrument, the result shows that there is no significant difference on writing skill between the students taught by using interactional approach and those taught by using non-interactional approach (self correction) at English Study Program of State Islamic Institute of Kediri. The significant value obtained was higher than the significant level (0.418 > 0.05). Therefore, there is not evidence to reject the null hypothesis; in other words, the null hypothesis cannot be rejected.
Both groups, experimental and control groups have improvement on their writing skill; however, there is no significant difference on writing skill between the two groups. The average score of the experimental group in the pre-test was 56.52; meanwhile, the average score in the post-test was 75.70. The gain score of the experimental group from pre-test to post-test was 19.18. Then the average score of the control group in the pre-test was 56.73; meanwhile, the average score in the post-test was 73.18. The gain score of the control group from pre-test to post-test was 16.45. From these results, there was different gain score between the experimental and control group. It seems that the experimental group achieved higher average score than the control group. However, statistically, there was no significant difference between those two groups on their writing skill.
There are some possible reasons of insignificant result of this study, insignificant difference between the experimental and control groups on their writing skill. Those are the readiness in implementating online teaching and learning process and the problem in internet connection. Those two possible reasons also turned up the following problems.
The first problem is the readiness in implementating online teaching and learning process. Both lecturer and students were not common in implementing online classroom. They were accustomed to use face-to-face teaching and learning process. Moreover, in the treatment, both lecturer and students were needed to give feedback through online application, Google Classroom. They might be not common in using the application. The treatment needed longer time and steps, since they needed to open their friends' file, then give some corrective feedback on their friends' writing and then resend it again to their friends. These steps would be longer and failed when they got problem in internet connection. These activities made them frustrated in joining online teaching and learning process, especially in joing the treatment, then it influeced the quality of feedback given. This condition should be solved since online classroom is still implemented during Covid-19 Pandemic. Additional technology training may be needed for the lecturer and students. It is also in line with Kim and Bonk's study. They revealed that one of the factors that can improve online learners' success was the provision of additional technology training (Kim & Bonk, 2006, p. 26). By giving technology training for both lecturer and students, it will make them easier in assimilating technology in the classroom. It is similar to Siregar, Fauziati, and Marmanto's study. They found that the 21 st -century pedagogical competence focused on how teachers assimilate technology in the classroom and how to facilitate the students' creativity due to the widely opened sources of information which positively affect students' proficiency. Therefore, it is suggested to provide more courses and training about confronting the 21 st century education (Siregar et al., 2020).
The second possible reason is internet connection problem. During the treatment, many students excused to leave or could not join the online classroom because they got problem in internet connection. This problem was caused two reasons namely the area problem and financial problem. During Covid-19 pandemic, the students went home and joined the online classroom from their home, and some of them come from rural area which had not good internet connection. Furthermore, the students needed to provide extra quota for their internet since all courses were conducted through online. They needed to provide extra money to buy internet quota which was not cheap. Some students excused not to join the online classroom since they had spent theit internet quota. This condition need some supports from institution especially in term of more quota support for students to be able to join online classroom. This is also line with Kim and Bonk's statement that some sort of training and support from the institution were very needed to make the students be ready for online teaching and learning process (Kim & Bonk, 2006, p. 29).
The finding of this study was in contrast to Jones, Garralda and Lock' study. They found that interactional dynamics makes writers have more improvement on their writing. Face-to-face interactions made more hierarchal encounters in which instructors took more control of the discourse; however, in on-line interaction, it was more egalitarian, in which clients took more control on the discourse (Jones et al., 2006). These findings was opposite with the finding of the present study in which during the online treatment the lecturer lost control on the participants. The lecturer could not completely control the quality of feedback given and whether the participants really revised the writing draft based on the feedback given. The dialogic process of giving feedback could not completely be controlled.
The finding of this study was also not in line with Lestari's study. She found that the interactional approach in the teaching of writing allows students to get beneficial in getting feedback from both teachers and peers. The feedback provision makes the hypotheses in second language acquisition (the input, monitor, noticing and output hypotheses) work simultaneously during the writing process for better output or the students' final draft of writing can be achieved (Lestari, 2008). The different findings between the two studies might come from the different mode of the treatment. Lestari's study was conducted through face to face treatment in which the reseacher had completely control to the participants. Meanwhile, the present study was conducted trough online treatment in which the researcher could not completely control to the participants.
Furthermore, the finding of this study was not in line with Ajjawi and Boud's study. They found that interactional approach can enable insight into feedback undocumented aspects previously for instance the interactional features that promote and sustain dialogue of feedback. Feedback should be considered not a language input but more on dialogical process. It is in line with a socio-constructivist point of view that feedback should be dialogic and help to develop the ability of students to monitor, evaluate and regulate their learning (Ajjawi & Boud, 2017). These point of view were completely correct when the reseacher could completely control the situation; however, in online classroom, it was quite difficult. Moreover, Google Classroom was text based application. It means that there was no oral dialogue or communication between provider and reciever of the feedback given. All activities of the treatment was conducted through text based in which it might rose other problems especially in understanding the written feedback given.
However, the finding of this study supports the study of Ravand and Rasekh. They found in their study that it is a need to continue to take a multi-dimensional view of feedback. The nature, content, mode and quality of feedback provision should be separately and interactively considered along with the differences of situational and individual in the instructional context. The interactions among task characteristics, instructional context, and students' characteristics are potential to be investigated more (Ravand & Rasekh, 2011, p. 1142. It is similar to the finding of the present study which revealed that the quality of feedback given and the online instructional context might influences the result of the study. The finding of this study was also in line with Wahyuni's study. She found that both types of corrective feedback, peer correction and self correction, affected the students' writing quality. Both students who conducted peer correction and the students who conducted self correction improved their writing in the post-test (Wahyuni, 2018). It is in line with Alroe's study also found that error correction can produce significant benefits (Alroe, 2011). However, Wahyuni also found that there was no significant difference on the writing quality between the students who conduct peer correction and the students who conduct self correction. If they had same correction guide, they can do the same best action in correcting and revising their writing (Wahyuni, 2018). Eventhough, the both studies have different mode of the treatment, face to face and online treatment, they have quite similar results. In addition, Wahyuni's study was conducted on argumentative essay, then the present study was conducted on expository paragraph.
The positive effect of feedback types on the improvement of students' writing skill found in the present study was also supported by Bitchener and Knoch. They found that the provision of feedback on students' writing can improve their writing skill (Bitchener & Knoch, 2010). Making a good writing product needs a process, then the provision of feedback during the process will really help Islamic students to have better writing quality and improve their writing skill. In line with Bitchener and Knoch, Syafi'i and Ramdhan stated that by giving appropriate feedback, students will learn well regarding the progress of learning (Syafe'i, & Ramdhan, 2019, p. 292). Furthermore, Muth'im and Latief also stated that students will focus more on learning when they are given feedback (Muth'im & Latief, 2014). However, it is in contrast with some studies. They found that corrective feedback is not helpful and badly timed. It does not address what they want it to address. Feedback is just considered as information transmission or language input (a cognitivist point of view) without any dialogue or interaction between provider and receiver (Sadler, 2010;Urquhart et al., 2014).
The present study has proved that the provision of corrective feedback on students' writing can improve students' writing quality. It is an appropriate method to improve students' writing. As Wahyuni also stated that an appropriate teaching method can give significant contribution in solving students' problems in learning (Wahyuni, 2014, p. 4). Different types of corrective feedback can be used by teacher as various methods in improving students' writing skill. As Budhi Ningrum and Widyawati stated that teachers should be able teach with various methods to make them happy and motivated to learn (Budhi Ningrum & Widyawati, 2015, p. 398). Furthermore, Akbar also found that there is significant positive effect of the use of teaching method on students' achievement (Akbar, 2014, p. 239). Therefore, teachers need to use appropriate method in teaching their students.
The implication of this study is that different types of corrective feedback, both interactional and non-interactional approach can be used to increase the writing skill of Islamic students. The teachers and students may consider these teaching and learning methods as an effort to acquire English language especially in writing skill. By having good skill on English writing, students will have more opportunities to do religious proselytizing (dakwah) in global area. They can produce or write many religious books or written products in English which can be read by many people in the world. Students can spread Islamic values widely, and attract people to love Islam. Akhmadi stated that one way to implement religious moderation is by having open minded in receiving the different culture and language (Akhmadi, 2019, p. 52). Open minded as the value of religious moderation can be a basis in learning foreign language including English language in Islamic higher education.

Conclusion
In accordance with the research findings and the discussions of the study, it can be concluded that there is no significant difference on writing skill between the students taught by using interactional approach and the students taught by using non-interactional approach (self-correction). The significant value obtained was higher than the significant level (0.418 > 0.05). Thus, there is not any evidence to reject the null hypothesis; in other words, the null hypothesis cannot be rejected. Statistically, there was no significant difference between the two approaches in affecting students' writing skill. The implication is that both approaches can be used to increase the writing skill of Islamic students. By having good writing skill, Islamic students will have more chances to do religious proselytizing (dakwah) through producing or writing religious books or written products in English. Therefore, their product of writing can be read by many people in the world. The implementation of religious moderation in Islamic higher education by acquiring foreign language especially English language will give much benefits not only for the students, but also for institution, society and the nation.