Using Metacognition in Lowering Writing Anxiety and Improving Writing Performance among Low-Intermediate ESL Students

Copyright: © 2020 The Author(s) Published by Human Resource Management Academic Research Society (www.hrmars.com) This article is published under the Creative Commons Attribution (CC BY 4.0) license. Anyone may reproduce, distribute, translate and create derivative works of this article (for both commercial and non-commercial purposes), subject to full attribution to the original publication and authors. The full terms of this license may be seen at: http://creativecommons.org/licences/by/4.0/legalcode


Introduction
English is an International Language. Thus, it is imperative for the students to master the language so that they can be competitive in the ever demanding and ever-expanding global market to become the most sought-after graduates. Many programs have been introduced to improve the quality of English among Malaysian ESL students namely MBMMBI (Memartabatkan Bahasa Melayu Memperkasakan Bahasa Inggeris), Oral Proficiency in English for Secondary School (OPS English) and Highly Immersive Program (HIP) Despite all these efforts, the quality of English in Malaysia remains poor especially in writing.
Researchers stated that writing is an arduous task (Mah & Khor 2015, Challob, Abu Bakar & Latiff, 2016 regardless when it is done in the mother tongue or in a foreign language. There are various reasons why students performed poorly in writing. Some of the reasons are low repertoire of vocabulary, difficulties in writing proper English sentences, feeling overwhelmed when writing introduction, develop ideas and conclusion as well as not having enough ideas about the given topics (Al Seyabi & Tuzlokova, 2014). This is clearly evidenced in the Kupasan Mutu Jawapan Bahasa Inggeris Kertas 1 SPM 2014 and 2016 where the team of markers suggested that the teachers should teach their students to write up to the required number of words especially in the Section B where the candidates are required to write a 350 words composition. Besides, teachers are also asked to stress on the grammar rules and spelling as students tend to commit simple grammatical error as well as spelling mistakes.
There are numerous researches have been conducted in this area. And this research is conducted based on the suggestion from a research done by Surat et al. (2014) where they encouraged educators to employ metacognitive strategy to enhance students' writing skill. Hence, this research aims to see the effects of metacognition in improving the low-intermediate students' writing performance. The objectives of this research are to: • Identify the level of writing apprehension among low intermediate ESL students • Observe the improvement in low-intermediate ESL students' performance in writing task after the intervention procedure (metacognition awareness)

Problem Statement
In performance dialogue conducted by English panel after each school examination it is evidenced that the students are heavily influenced by their mother tongues when writing where their sentences contain lexical, syntax and tenses from their mother tongues. This resulted in low performance in the composition. Normally, when students are facing difficulties in coming up with a proper English sentences, they tend to avoid writing the sentences which resulted in the scarcity of the number of words hence they are no able to write up the required number of words.

Significance of the Study
This research is conducted to give a valuable insight to the educators about the importance of metacognition awareness in lowering writing apprehension so that it helps to develop students' ability to write. Most of the time, the educators only pay attention to the low proficiency students and often neglected the needs of the low-intermediate students. Thus, it is hoped that through this research the educators will see that they need to change their current teaching strategies and employ metacognitive strategy into their teaching and learning process.
The ultimate goal of this research is to see changes in the low-intermediate students' writing performance. This is because standardized test has become a norm in Malaysia to test students' performance. Therefore, the educators need to equip their students with the appropriate skills to enable them to compete with the rest of the students. Otherwise it will forever be a bane to the less proficient students. While quite a number of researches have been conducted to see the effectiveness of metacognition awareness and metacognitive knowledge in writing, most of them are done by the researchers who are not teachers. Their only interest is to inform the educators of the importance of making students aware of their metacognition. Hence, this research will be conducted by the researcher on her own students hoping to see the same result will be achieved which will definitely help the students to perform better in writing.

Significance to the Theory
The findings of this study confirm previous studies that employing metacognition in writing task improves students' writing performance by lowering their writing anxiety level (Blasco 2016;Stewart, Seifert & Rolhieser, 2015). However, it is also noted that employing metacognition alone would not ensure improvement in writing performance as there is other factor that could affect students' writing performance which is their proficiency the language. Students' language proficiency has to be addressed first before other parts of writing strategies could be implemented in order to ensure success in the implementation of intervention.

Significance to the body of Language
In this study, the sample used is very specific where they are the low intermediate students who scored between 15-25 marks in their school-based examination. By selecting these respondents the researcher is able to see that making the respondents aware of the the metacognitive processes involved in writing an essay only improve their metacognitive strategies which does help in lowering their writing anxiety and slightly improve their writing performance. However, the researcher also found out that improving metacognitive strategies is not the sole factor to improve performance. There are other factors that need to be taken into consideration such as language proficiency. Flavell (1979) described metacognition as the awareness ones have about one's own thinking. Metacognitive knowledge consists of the knowledge that ones have about the factors affecting the outcomes of one's thinking. In other words, metacognitive knowledge can also be defined as what learners know about their own thinking processes (Stephanou & Mpiontini, 2017) such as declarative knowledge which is knowledge about oneself, procedural knowledge or knowledge about the given task, and conditional knowledge which is knowledge about the strategies of learning (Sahlan et al. 2014).

Literature Review Understanding Metacognition
Meanwhile, Stewart, Seifert and Rolhieser (2015) defined metacognition as the ways students try to comprehend their own writing process and how they adjust their processes to the requirement of the questions or tasks. Sadeghi, Hassani and Rahmatkhah (2014) stated that when students is aware of their own metacognition, they are monitoring their thinking process, checking whether they are making progress towards achieving an appropriate goal, making sure they are accurate as well as deciding if time and mental effort is used wisely. Shahlan et al. (2014) indicated that "students are more able to learn complex skills when they can think metacognitively, that is, when they think about their own thinking and performance so they can consciously monitor and change it." Apart from that, metacognition is also known as self-regulated learning where a learner is required to regulate his own process of thinking when given a task. Therefore, metacognition is a method which a learner uses to control his own learning (Briesmaster and Etchegaray, 2017). It encompasses the ability of a learner to plan, evaluate and monitor his own learning (Oguz & Ataseven, 2016;Wagaba et al. 2016;Zepeda et al. 2015).
Hence, it can be concluded that when students are aware of their own thinking it provides a mental picture of what are the steps they need to follow, strategies they need to use and what information they need to write when it comes to writing a well-constructed essay. By equipping the students with metacognition we are preparing them to be self-regulated learners where it is an important skill to acquire in the 21 st century learning.

Metacognition and Success in Language Learning
Metacognition has a significant role in impacting students' performance in their study. It is proven that the higher the students' metacognition the higher their self-regulatory learning style is (Stephanoe & Mpiontini, 2017). This is because when students are aware of their own cognitive processes they will become active learners who are able to manage and direct their own learning which will finally help them to evaluate the best methods to practice and strengthen their knowledge (Azizi, Nemati & Estahbanati, 2017). Some other studies have also yielded the same result. In a survey conducted by Shahlan et al. (2014) on 18 students from one secondary schools in Malaysian Education system (age between 14-16 years old) from one poor urban community and have low academic achievement. The finding of their study showed that Malaysian secondary school students did not apply metacognitive knowledge when writing. Hence, the writing that they produced are not growing in the sense that they failed to elaborate the content by relating it to current issues as well as to provide examples. It is also found that there is a positive relationship between students' writing performance and their metacognitive writing strategy used (Blasco, 2016;Briesmaster & Etchegaray, 2017;Wischgoll, 2016).
When students are made aware of their metacognition or when metacognition is explicitly taught to students, it showed positive result as they are aware of what, how and when they can learn (Oguz & Ataseven, 2016). Thus, it can be concluded that the use of metacognitive strategies such as planning, organising, evaluating and resourcing help to differentiate the effective and less effective learners (Zhang & Qin, 2018). Nonetheless, metacognition is a skill that needs to be explicitly taught to the students and this requires the teachers to train them to use metacognitive strategies effectively as this will enable the students to be better self-regulated learners (Pitenoee, Modaberi & Ardestani, 2016;Wagaba et al. 2016;Zepeda et al. 2015). Thus, teachers are required to model metacognition and explicitly teach their students the metacognitive strategies to further enhance their learning (Wagaba et al. 2016;Zepeda et al. 2015) as it will help to reduce students' writing apprehension and increasing their self-efficacy in completing their tasks.

Writing Apprehension/Anxiety
Emotion such as writing anxiety can have a deep influence on the strategy use by students and the outcomes of students writing task. One of the reasons why students produced low quality writing is due to their fear of writing or also known as writing apprehension (WA). WA is also known as writing anxiety which refers to the internal tendency to feel anxious when students are given a task which requires them to write (Blasco, 2016). It can be concluded that when students' writing anxiety is decreased it will positively affect their writing skills.
WA plays a potentially detrimental role in students' writing success (Blasco, 2016). A survey conducted by Stewart, Seifert and Rolhieser in 2015 aimed to examined the relationship between writing anxiety and writing self-efficacy and students' perceived use of metacognitive writing strategies. Their findings indicated that WA affects learning in two ways. One of it is that WA leads to fear of failure which in turn enables the students to perform better academically due to high motivation to achieved good result. Nevertheless, WA could also be the reason for students to underperform due to ineffective studying.
It is indicated in the literature that 'a particular amount of apprehension or anxiety is imperative for writers to succeed in writing However, some students display a higher than usual level of writing apprehension that may impact their writing in a negative way and display poor writer's performance' (Al-Shboul & Huwari, 2016). It is also evidenced that when students have WA they would often feel nervous, stress and these in turn cause poor writing performance. Studies showed that when WA is reduced the use of metacognitive writing strategies are increased (Balta, 2018;Stewart, Seifert & Rolhieser, 2015).
Based on the literatrure it is recommended that writing intervention should be conducted to reduce WA in order to increase students writing self-efficacy. To reduce WA, educators need to understand the importance of modeled or guided writing and the importance of giving constructive feedback. Apart from guiding the students, teacher should also create a learning environment that is non-threatening physically as well as mentally and emotionally in order to reduce writing apprehension and boost writing self-efficacy (Blasco, 2016).

Methodology Research Design
This research is using Action Research Design to study the effects of Metacognition in lowering students' writing apprehension and improving low-intermediate students' writing performance.

Population and Sample
The population of this research are 72 Form 5 students from a sub urban secondary school located in the Baram district. Out of the 72 students 12 low-intermediate students who scored between 14 -25 marks in Paper 1 Section B in their Mid-Term examination are chosen as samples of this research.

Data Collection Procedure
Action Plan Figure 5.1 shows the five phase action research cycle used for this research. The initial reflection phase The main objective of this phase is to identify issues, problem and to describe what changes have to be done to improve the situation.

The Planning Phase
The researcher address the problem in the first phase by coming up with a method to be implemented with the respondents in order to overcome their writing apprehension. The method chosen is metacognition, in which the respondents will be explicitly taught metacognitive strategies in order to lower their writing apprehension which in turns will improve their writing performance.
The action/implementation phase In this phase the researcher implement the action plans. The researcher is going to explain to the respondents what metacognition is and how to be aware of their thinking processes and how it would help the respondents to improve their writing. The respondent are taught about metacognitive knowledge which are declarative (knowledge about oneself), procedural (knowledge about the given task) and conditional (knowledge about the strategies of learning) knowledge explicitly to students as it it helps them to be aware of their metacognition (Balta 2018;Oguz & Ataseven 2016).

The observation and analysis phase
At this stage the researcher will make an observation of the respondents' progress and analyze the data gathered from the implementation phase.

The Refection and Assessment Phase
The final stage of the research is the researcher reflect and review the actions employed and the result from the observations conducted. The result of this reflection will be used for planning further actions for the following cycle if needed.

Instruments Questionnaire
A questionnaire adopted from Cheng's Second Language Writing Anxiety Inventory (2004) was administered to gather information to identify the students who are experiencing writing apprehension or writing anxiety.

Pre and Post Test
The writing test is based on the questions from Sijil Pelajaran Malaysia 2017 English 1/1119 section B, continuous writing. The respondents' writing will be marked using the scoring rubric provided by the Malaysia Examination Board.

Writing Journal
After each writing task, respondents are required to write in their diary what are the steps they have taken in the write up process as well as how they feel about the task. The diary will be inspected by the researcher every week and their progress will be observed closely in term of the steps the respondents take in writing their composition as well as the quality of their composition.

Semi-structured Interview
Semi-structured interviews will also be conducted to find out how the respondents feel about using metacognition in writing. The interview will be recorded and transcribed for data analysis purposes.

Data Analysis Procedure
For the purpose of this action research, descriptive analysis is used to analyse the data from the questionnaire. The answers to the questionnaire is analysed using SPSS based on the responses given by the respondents. Items 1,4, 7, 17, 18, 21 and 22 were negatively worded and entailed inverse scoring before being calculated to yield total scores. A higher level of ESL writing anxiety was determined by a higher score achieved. Scores above 65 showed a high degree of writing anxiety. Scores below 50 indicates low level of writing anxiety. A moderate level of writing anxiety referred to a total score in-between.
As for the respondents' level of writing apprehension it is analyzed based on their writing diaries and semi-structured interview. It is to see whether there is any changes in their level of writing apprehension after they are being made aware of their metacognition. Meanwhile for the writing performance, it is analyzed based on their score in their post-test. The scoring of the writing is based on the marking rubric provided by Malaysia Examination Board to mark English 1119 Paper 1 Section B for continuous writing.

Writing Anxiety Level: Pre and Post test result
The descriptive statistic of the respondents' anxiety level tested using Cheng's Second Language Writing Anxiety Inventory (2004) are as presented in Table 6.1 and Table 6.2.  Table 6.1 showed that the scores of the respondents in Cheng's Second Language Writing Anxiety Inventory (SLWAI) ranged from 42 to 87. The mean and standard deviation of the writing anxiety were 65.33 and 14.09 respectively. According to Cheng (2004), the score above 65 signifies a high level of writing anxiety. Thus, it can be concluded that the respondents have a high level writing anxiety. This answered the first research question which is "how does student feel when writing in English?". According to the respondents as stated in the semistructured interview there are two main reasons why they feel anxious when assigned a writing task. The first reason is they might not have any idea what to write and secondly, they lack the vocabulary needed to write a good essay.  Table 6.2 showed that the scores of the respondents in Cheng's Second Language Writing Anxiety Inventory (SLWAI) after the implementation of metacognition. It ranged from 42 to 85. The mean and standard deviation of the writing anxiety were 64.08 and 13.51 respectively. According to Cheng (2004), the score between 50 to 65 signifies an intermediate level of writing anxiety. Thus, it can be concluded that the respondents have an intermediate level of writing anxiety. There is a slight decrease in the respondents' anxiety level compared to before the implementation of metacognition. In the semi-structured interview conducted after the implementation of metacognition, the respondents are still very much concern that they lack the vocabulary to write an essay as well as they might not have any idea what to write on the given titles/topics.

Writing test score : Pre and Post Test result
Descriptive statistic of continuous writing pretest and post-test score are indicated in Table  6.3 and Table 6.4.  Table 6.3 the respondents' scores ranged from 13 -28 during pretest with the mean score of 19.00 and standard deviation (SD) of 5.93. In the post-test (Table 6.4) their scores ranged from 18 -30 with the mean score of 23.75 and standard deviation of 11.69. From the mean of the post-test scores it can be concluded that there is a slight increase in the respondents' performance in their continuous writing upon the use of metacognition. This answered the second research question which is "is there any improvement in low-intermediate students' writing performance when the teacher is making the students aware of their metacognition process?" Writing procedure: Prior and Post Metacognition implementation Writing procedures employed by the respondents prior to the implementation of metacognition. Figure 6.1 Writing procedures employed by the respondents prior to the implementation of metacognition.
From the data collected in the semi-structured interview and the writing diary, in the context of the writing procedures employed by the respondents prior to the introduction to Metacognition, it is stated that they did undertake certain pre-writing activities, such as trying to understand the question requirement, thinking about the topic and brainstorm ideas related to the topic.
During pretest for writing task, the respondents said that before producing a written text they have to activate their prior knowledge related to the question in order to come up with the ideas to write the essay. One of the respondents responded that: "Before I try to answer any essay, I have to think about the ideas that I have about some of the questions. If I don't have any idea, I will not be able to write anything" (Respondent 1). One of the strategies the respondents used in pre-writing activity is brainstorming for ideas. They normally brainstorm ideas by doing research on the Internet looking at sample essay. Respondent 11 stated in his writing diary that: "At first, I will think of the ideas myself. But if I still could not come up with any idea, I will go the the second step which is googling Mr Google. Sometimes, I also look at the sample answers from the answer sheet." As for the while-writing strategies, the respondents use online dictionary as well as Google translate to look up for unfamiliar words as they write they essay. This is stated by most of the respondents in their writing diaries: "When I started writing, I will get ready Google translate" (Respondent 6). Apart from unfamiliar words, respondents also use Google translate to check their sentences: "I will use Google translate if I do not know how to build [construct] some sentences" (Respondent 7).
Since the respondents are required to use at least three idioms when writing their essay, they usually use the Internet to look up related idioms based on the theme of their essay. This is stated by Respondent 1 in his writing diary, "I will find the idiom I can use on the Internet". His response is shared by Respondent 8 and 9 as well. Another respondent mentioned that during while-writing stage, when he encounters writer's block, he would refer to his friends' essay or ask them for ideas to continue his writing: "… if not, I will ask friend for idea and continue my writing" (Respondent 8).
Prior to the introduction to Metacognition, the respondents would count the number of words in the essay to make sure they reach the minimum 350 words and when they reached the word count, they would submit their essay. This is shared by respondent 5, 7, 9 and 11: "After finish writing, count the words then hand in to English teacher".
6.3.2 Writing procedures employed by the EFL students after the implementation of metacognition. Figure 6.3 Writing procedures employed by the EFL students after the implementation of metacognition.
In the context of the writing procedures employed by the respondents after the introduction of Metacognition the respondents of this study explicitly stated that they developed certain prewriting activities, which is drafting. The respondents are still doing the brainstorming of ideas as well as trying to understand the questions in their pre-writing routine, but they have added drafting into the stage. This includes making a list of the main points to be written in their essay, which helped the respondents to organize their ideas. One of the respondents stated, "Before I start writing, I write a draft to make sure I have the skeleton for my essay. This make it easier to write my essay" (Respondent 10).
There was a new strategy added into their while-writing strategies which is checking ideas from the draft done prior to writing. This showed that the respondents wrote their essay based on the initial points included in the first list they made. One of the respondents 5 in the interview stated, "While writing, I checked the draft I made earlier. This helps me to write my ideas in an organized manner." The respondents also added a few strategies into their post-writing strategies which are rereading their essay where they are trying to detect grammatical errors such as spelling errors and sentence structure. Apart from that, the respondents are also asking their peers to read through their essay and identify the error for them. The final step they take before submitting their essay is to rewrite the essay based on the error identified and suggestions by their peers.
Most of the respondents re-read their essay in order to ensure there are no spelling errors made. Respondent 6 stated in his writing diary that: "after I have finished writing my essay, I will reread the essay and if I found spelling mistakes will change it." While spelling errors are easier to detect, sentence structure could be challenging to identify just by re-reading. Respondent 10 revealed that she uses grammar checker to do so: "I use grammar checker to check some of the sentences I wrote. Sometimes I would check the whole paragraph. Then, I will correct the errors." The respondents of this study mentioned that they would ask their peers to read their essay and give suggestion on how to correct the errors found in their essays as a post-writing activity. According to them, this was conducted to double check that there are no errors that they have missed out after re-reading their own essays. As stated by respondent 9: "After I re-read and correct my essay, I will ask my friends to read to detect errors also to give suggestions to correct some of my sentences. Finally, after they have gone through re-reading and peer checking the respondents will correct their essay by rewriting it before submitting their essay to the teacher.

Writing Apprehension and Writing Test Score
Based on the findings, prior to the metacognition implementation it appeared that the respondents have high level of anxiety with the mean value of 65.33 on Cheng's SLWAI. The respondents' high level of anxiety has affected their performance in the writing pretest where their scores mean is 19.00. Based on the writing rubric for Paper 1 Section B, score of 19 is placed in band U(i) where the score is ranged between 14 to 19. It is considered weak. Band U(i) describes that the piece of writing is fairly clear which means the reader has to reread some of the sentences in order to understand the meaning. Apart from that, the piece of writing placed in this band has very frequent single word error (SWE) which causes blurring and impedes reading. There would also be serious vocabulary errors of many kind but mainly single word type and could be corrected without rewriting. Last but not least, the writing which falls under this band has very few accurate sentences as well as the sentences used are often simple and repetitive.
After the implementation of metacognition there is a slight decrease of the mean in SLWAI with the value of 64.08 which is considered as intermediate level of anxiety. The writing post-test mean score increased to 23.75 which is placed in Band E with score ranging from 20 to 25. In term of band placing, Band E is one band better than Band U(i). The score of 23.75 is considered a middle E band and of moderate performance. This piece of writing has frequent SWE which is serious to hamper reading but the meaning is never in doubt compared to the U(i) band where meaning could be blurry. Other than that, the sentences used in this piece of writing is accurate however the accuracy might not be sustained. In term of vocabulary it is rather limited or too simple.
The findings of this study reaffirms the study conducted by Blasco (2016) where he stated that students produced low quality writing when they have high anxiety when given writing task. Due to the respondents' high level of anxiety they were not able to apply effective writing strategies in their writing task (Blasco, 2016;Stewart, Seifert & Rolhieser, 2015). Despite having intermediate level of anxiety, the respondents showed improvement in their writing. This findings support the claim that by decreasing students' writing apprehension it could help to increase their writing skills and their performance. (Balta 2018). Based on the findings of current study, it is imperative that educators identify their students' level of anxiety in terms of writing and then use the information to devise a plan to help lower students' anxiety level so that they can improve their performance in writing.

Implementation of Metacognition in writing classroom
Looking at the strategies employed by the respondent prior to the implementation of metacognition is was clear that the respondents of this study are more familiar with the cognitive procedures of writing strategies compared to metacognitive procedures. It is evident in the findings where the respondents are only thinking about the topic through brainstorming, using online dictionary to check for spelling as well as looking up idioms to be used in their write up. These strategies mainly focused on vocabulary and content knowledge required to produce an essay. This findings showed that the respondents did not apply metacognitive knowledge when writing (Sahlan et al. 2014). This findings also prove that students would be less efficient in writing if they only use cognitive strategies in writing (Pitenoee, Modaberi & Ardestani, 2016) it is evidenced in this study where the respondents' mean score prior to the implementation of metacognition is lower than their mean score after the implementation which is 19 to 23.75 respectively.
After the implementation of metacognition, the respondents stated that they have employed more metacognitive strategies to complement their cognitive strategies in writing. This includes drafting in the pre-writing strategies which shows that they are trying to plan their ideas instead of merely thinking and brainstorming for ideas prior to the implementation of metacognition. The respondents see the importance of drafting to plan their paragraphs before they start writing. In while-writing stage, the respondents also added refer to draft which means that they are monitoring the organization of their ideas and making sure that they are not missing any points. The major improvement that the respondents made is in the post-writing strategies where they incorporate rereading of their essay, peer checking, checking grammatical error and eventually correct the errors identified before they submit their essays. This is where they are evaluating their finished essay in terms of vocabulary, sentence structure, tenses as well as ideas. That being said, it can be concluded that the respondents are still paying much attention on the accuracy of their lexical items.
The implementation of metacognition played a role in the improvement of the respondents' score in writing (Azizi, Nemati & Estahbanati, 2017;Balta, 2018;Blasco, 2016;Briesmaster & Etchegaray, 2017;Oguz & Ataseven, 2016;Wischgoll, 2016). In the writing post test the respondents' score improve from 19.00 to 23.75. This proved that metacognitive strategies help learners to obtain their writing goals which is illustrated in the findings of the present study. Employing metacognition encourages the respondents to use high order thinking, planning, monitoring, and evaluating which influenced the effectiveness of the metacognitive writing strategies. (Pitenoee, Modaberi & Ardestani, 2016).
Hence, the findings of this study echoed that it is important to teach metacognitive knowledge which are declarative (knowledge about oneself), procedural (knowledge about the given task) and conditional (knowledge about the strategies of learning) knowledge explicitly to students as it it helps them to be aware of their metacognition (Balta, 2018;Oguz & Ataseven, 2016). When students are aware of their own cognitive processes they will become active learners who are able to manage and direct their own learning which will finally help them to evaluate the best methods to practice and strengthen their knowledge (Azizi, Nemati & Estahbanati, 2017;Pitenoee, Modaberi & Ardestani, 2016).

Conclusion
The respondents in this study showed improvement in their writing performance when they employed metacognition in writing task. It also helps to lower their writing apprehension which helps them to write better. Being aware of their own thinking helps the students to practice the proper methods and strategies to enhance their writing skills. Although the respondents are still heavily focus on their cognitive strategy which focuses on the lexical items accuracy it helps to complement their metacognition and enable them to produce better writing.
It is noted that there is only a slight decrease in the anxiety level and minor increase in writing performance. It can be deduced that while implementing metacognition in writing do bring about improvement in the respondent's performance, students' proficiency in English and their mother tongue influenced the findings in this study. Therefore, the researcher believes that students' language proficiency needs to be addressed first in order to ensure they feel at ease with writing task. Combine with the knowledge on metacognition, the students would be able to achieve better result.
Ultimately, this study sheds light on how employing metacognition can help students to write better English composition. When metacognition is explicitly taught in writing lesson, it helps to lower students' anxiety level and as a result, it increases their writing performance.
As numerous studies have shown that metacognition is one of the key factors to help students improve their writing performance by lowering their anxiety level, educators are encouraged to explicitly teach metacognitive knowledge and awareness to students so that the student would employ metacognition in their writing task. When they are guided and supervised properly in practising metacognition, they would be able to be independent learners who are able to carry out high order thinking skill which is one of the criteria of 21st century students.

Recommendations
Time constraint is one of the limitations in this research, therefore, it is recommended that the future researchers take a full school year to conduct the study to ensure proper guidance are given to the respondents. Since this study is conducted with just 12 respondents, it is considered a small amount of sample. Future research should be conducted with bigger sample size so the findings will be more significant to the research. Next, instead of low-intermediate students, future researcher should consider using high proficiency students as sample because in this study it is found that low English proficiency affects the findings where there is not much difference in the respondents' pre and post writing score after the implementation of metacognition. One of the reason could be their English proficiency as it constitute a large role in giving the score.
In a nutshell, this current study complemented those of earlier studies which discovered that metacognition have to be taught explicitly and effectively to improve students' writing strategies. Since low writing anxiety could help to improve writing performance when combined with metacognitive awareness, the educators should teach and train students to use metacognitive strategies in language classroom especially in writing task so that they are able to improve their writing strategies and enhance their performance.