Effects of Corrective Feedback on Students’ Linguistic Errors

Linguistic Error; Writing Accuracy; Corrective Feedback. The research aimed at finding out (i) Whether corrective feedback could minimize students’ linguistic error; (ii) Which type of corrective feedback had more permanent effect on students’ writing accuracy. The research applied single-subject experiment design. Data analysis in single-subject research typically was based on visual inspection and analysis of graphic presentation. The step are writing test, ratio, scoring and compare. In this study, six treatment groups and no control group were used. All treatment groups received in different types of CF on their writing tests. Treatment 1 (T1) received Direct CF, treatment 2 (T2) received indirect CF, treatment 3 (T3) received metalinguistic CF, treatment 4 (T4) received reformulation CF, treatment 5 (T5) received focused CF, Treatment 6 (T6) received unfocused CF. The findings indicated that (i) CF could minimize students’ linguistic errors except focused CF on vocabulary; (ii) Based the result of the data analysis using analytic rating scale and composite rating scale showed that direct CF had the most effective in minimizing students’ linguistic error in vocabulary, language use and mechanics than the other types of CF, so direct CF could be categorized having more permanent effect on students’ writing accuracy.


A. INTRODUCTION
Writing is the most complex and difficult skills than another three language skills (listening, speaking and reading) because writing process has several steps. In expressing and delivering information from mind into piece of paper, writing needs process of thinking systematically. This mean that when you first write something down, you have already been thinking about what you are going to say and how you are going to say it. Then after you have finished writing, you read over what you have written and make changes and corrections. Linguistic is incredible complexity of language and very important to improve students' writing accuracy. Politzer (Politzer & Ramirez, 1981), Evans (Evans et al., 2011), and Boggs (Boggs, 2018) as cited in Bitchener & Ferris (Bitchener & Ferris, 2012) errors should be corrected because it was expected to help learners identify their own errors and discover the functions and limitations of the syntactical and lexical forms of the target language. Error is the important thing to be corrected and giving feedback can reduce the students' errors.
Correction can come from teacher, peer, and self-correction. Corrective feedback was important to help learner to achieve target language. Teacher feedback on student writing was a critical part of writing instruction and could have a great influence on student writing (Bitchener & Ferris, 2012) as cited in Li Zhan (Zhan, 2016). In other words, teacher feedback was effective to influence students' writing. Written teacher feedback played an important role in improving students' writing skills, especially for second or foreign language writing. Not only did it provide a valuable opportunity for individualized, text-based, contextualized instruction from teachers but also it was likely to be taken seriously by students than feedback given to groups of students (Bitchener & Ferris, 2012). as cited in Li Zhan (Zhan, 2016). Teacher feedback was more effective than students feedback. Corrective feedback has important role to raise students, awareness of their errors and it has been debated for more than 15 years. Bitchener (Bitchener, 2008) posit that direct CF with oral and written explanation is effective in minimizing students' linguistic error. Bitchener & Knoch (Bitchener & Knoch, 2009) contended that direct CF, with and without written and oral metalinguistic explanation is effective to increase students writing accuracy. Bitchener & Knoch (Bitchener & Knoch, 2010) affirm that direct CF is efficient method for students' attention to errors. Writing corrective feedback (CF) is one of effective methods in minimizing students' linguistic error.
A number researchers have investigated different aspects of corrective feedback such as the effects of writing corrective feedback (WCF) on students' linguistic accuracy. The drawbacks of all research I had read that most studies had investigated and compared between direct, indirect and meta-linguistic CF and focused and unfocused CF. There was only one researcher who has investigated the effect of six types of CF but he has done research in practitioner for IELTS writing test 2 program. They found different results so there were no certain types of WCF could suggest for the most beneficial to use long-term to ESL learners especially in University level. So it was a novelty in my research. The purpose of this research is to find out whether corrective feedback can minimize linguistic errors and to find out the types of corrective feedback that have more permanent effect on students' writing accuracy. This study was quantitative method. Quantitative was based on a single-subject experimental design using alternating treatments design, which involved six experimental groups.

B. METHODS
This research used alternating treatments design. It is single-subject experimental design in quantitative method. This method was chosen to investigate the comparative effectiveness of six types of CF (direct CF, indirect CF, metalinguistic CF, reformulation CF, focused CF and unfocused CF) in minimizing students' linguistic error and to know which type of corrective feedback had more permanent effect on students' writing accuracy in vocabulary, language use and mechanics (Arikunto, 2010).
The data collection was analyzed through the following techniques: 1) Writing test, the data was obtained through writing test which consisted of four writing tests. 2) Ratio, students' linguistic errors was made by ratio using symbol as follows: Total Errors (TE) X 100%, Total Words (TW), and Total errors (TE) was divided to total words (TW) to get percentage of the students linguistic errors. 3) Scoring, In scoring students' writing test, researcher used analytic rating scale and composite rating scale. Analytic rating scale was used to analyze students' writing score on vocabulary, language use and mechanics. The result of four students' writing tests was analysed by manual. The composite rating scale was used to analyze students' writing score in composite. The data of analyzing students' linguistic errors by using analytic rating scale and composite rating scale (please see Table 1). For scoring, the researcher scored the students' writing test based categories of scoring system by (Henning, 1990). 4) Compare writing test, Students' linguistic errors and score were measured by using six types of corrective feedback. To investigate the effects of corrective feedback on students' writing, the researcher compared the students' linguistic errors and score using analytic rating scale and composite rating scale (please see Table 1).

Error dealing with language use
The present study showed that direct CF had the highest percentage in reducing students' linguistic errors on language use from the first writing test to the fourth writing test than the other types of CF. The students' errors percentage in direct CF from the first writing (31.66%) to the fourth writing (8.79%) decreased sharply 22.87% and the score increased sharply 29 points. It mean that direct CF was the most effective in minimizing students' linguistic errors (please see Table 1).

Errors percentage dealing with vocabulary
Students' vocabulary errors for all groups were fluctuation and their score were stable. The students' vocabulary errors only decreased slightly. Indirect CF had the highest percentage (1.86%) in minimizing students' linguistic errors on vocabulary than the other types of CF from the first writing test (1.90%) to the fourth writing test (0.72%), and the score rose fluctuated 25 points. Based the data analysis above, indirect CF was the most effective in minimizing students linguistic errors on vocabulary than the other types of CF (please see Table 1).

Error dealing with mechanics
The students' errors percentage in direct CF from the first writing (38.61%) to the fourth writing (11.62%) decreased sharply 26.99% and the score increased 40 ponts. Direct CF showed the highest percentage in reducing students' errors on mechanics than the other groups. It mean that direct CF feedback was the most effective in minimizing students' errors on mechanics than the other types of CF (please see Table 1).

Scoring dealing with composite
The students' errors percentage in Direct CF from the first writing to the fourth writing decreased sharply 26.99% from 17.20% to 5.41% and the score increased sharply 29 points. Direct CF showed the highest percentage in reducing students' linguistic errors using composite rating scale than the other groups. It mean that direct CF feedback was the most effective in minimizing students' errors on mechanics, language use and mechanics using composite rating scale than the other types of CF (please see Figure 1 and Table 1). The discussion was based by research questions and the researcher discussed about the answer for research questions as follows: a. Could corrective feedback minimize students' linguistic errors? Based my research, corrective feedback could minimize students' linguistic errors. This finding could be proved by the research finding was taken from the teaching process which consisted of the result of the data analysis (please see Figure 1, Figure2, Figure 3, Figure 4 and Table 1). Although the result of the students' linguistic errors (dealing with language use, vocabulary, and mechanics using analytic scale and composite) showed that all groups' linguistic errors were fluctuation, but almost the percentage of students' linguistic errors declined except focused CF on vocabulary. 58 | IJECA (International Journal of Education and Curriculum Application) Vol. 1, No. 2, August 2018, pp. 53-59 Although the students' linguistic errors in unfocused CF increased sharply in the third writing to the fourth writing on language use, but the students' linguistic errors declined from the first writing to the fourth writing. It could be concluded that unfocused corrective feedback could minimize student's linguistic errors on language use, and focused CF' errors increased slightly 0.04% from the first writing 0% to the fourth writing 0.04% on vocabulary. It could be concluded that focused CF could not minimize students' vocabulary errors. b. Which type of corrective feedback had more permanent effect on students' writing accuracy? Based the result of the data analysis using analytic rating scale and composite rating scale (please see Figure 1, Figure 2, Figure3, Figure 4, and Table 1) showed that direct corrective feedback could be categorized having more permanent effect on students' writing accuracy.

D. CONCLUSION AND SUGGESTIONS
There were six types of corrective feedback in assessing students' writing linguistic errors in language use, vocabulary and mechanics namely direct CF, indirect CF, metalinguistic CF, reformulation CF, focused and unfocused CF. Students' linguistic errors was measured by using six types of corrective feedback to find out whether corrective feedback could minimize linguistic errors and find out the types of corrective feedback that had more permanent effect on students' writing accuracy.
Based analyzing students' linguistic errors on vocabulary, language use and mechanics by using alternating treatment design found that Direct CF could minimize students' linguistic error and Direct CF was effective in using long-term to the teacher and students. It mean that Direct CF is the most effective in minimizing students' linguistic error and have more permanent effect on students' writing accuracy than the other types of CF. This finding could be proved by the research finding was taken from the teaching process which consists of the result of the data analysis (please see Figure 1, Figure 2, Figure 3, Figure 4, and Table 1)The researcher advice to add explanations to use corrective feedback for more effective inreducing students' linguistic errors and rose students' writing accuracy.
Thanks to all my students lift of 2015 in F class. Thanks also to head of Makassar Muhammadiyah University has given his permission to do a research to the researcher. Thanks to Ummi Khaerati Syam, S.Pd., M.Pd. as a head of English Language Study Program, Nunung Anugrawati, S.Pd., M.Pd., and Muhammad Zia Ulhaq S.Pd., M.Pd. have given me a support and help my research in Muhammadiyah Makassar University. My deep gratitude goes to all lecturers of Graduate Program of State University of Makassar, especially to all lecturers of Language Education Program for their invaluable knowledge, advice, and encouragement support, spirit and motivation given to me. My special thanks go to the late my mother "Nahra" had prayed my life during the time of his life and the late my father "Arfin" had given me inspiration in crossing my life, as well as my both my sisters Marwah, S.Pd and Murniati, S.Pd always give me motivation in completing this article. The last, the researcher hope that all help who have given by all of them can be valuable religious service and get merit from Allah Swt. May Almighty Allah always be with us, Amin.