THE CORRELATION BETWEEN STUDENTS’ WRITING SELF-EFFICACY AND ESSAY WRITING PERFORMANCE

Self-efficacy in writing is an important factor for a student-writer to accomplish a writing task. However, not all studies show that writing self-efficacy and writing achievement are positively related. The purpose of the study is to find out 1) the writing self-efficacy level of the second-year students of an English education study program in one of the private universities in Pasuruan Regency, East Java, Indonesia

Self-efficacy in writing is an important factor for a student-writer to accomplish a writing task.However, not all studies show that writing selfefficacy and writing achievement are positively related.The purpose of the study is to find out 1) the writing self-efficacy level of the second-year students of an English education study program in one of the private universities in Pasuruan Regency, East Java, Indonesia, and 2) the correlation between students' writing self-efficacy and their writing performance.The research uses a correlational design.The sample of the research was fifty students of English study program from one of private universities in Pasuruan regency, Easyt Java, Indonesia.The instrument used to collect the data was the writing self-efficacy questionnaire developed by Prickel in 1994, test, Primary Trait Scoring (PTSG) rubric, and documentation of the student's essay.The questionnaire consists of 25 items about self-efficacy in writing with a fivepoint scale of answers.The result shows that the students' self-efficacy level mainly falls in the moderate category (58%), while the rest fall in the high category (42%).The result of the correlation analysis shows that there is no correlation between the students' writing self-efficacy and their writing achievement (r= -.020, n = 50, p = .892).The result implies that writing teachers should continuously foster their students' writing self-efficacy by giving them sufficient practice opportunities, offering constructive feedback, as well as providing emotional and psychological support in time of difficulties.
It turns out that having strategies to solve the problems in writing is not a guarantee one is be able to write well because putting the plans into action (writing task) is not easy.Pajares and Valiante (2006) believe that the complexity of writing does not lay only in the thought process underlying the students' composition, but also in the way the students engaged with the text.They believe that one effort to understand the complexity is to focus on the students' writing self-belief which becomes the basis of motivation.This being the case, students who believe in themselves that they can write well are more likely to be motivated to write and have more chances to successfully accomplish the writing task.In other words, the accomplishment of these task demands in writing, thus requires beliefs in learners' capabilities (self-efficacy) since a lack of self-belief to carry out that task may inhibit the writing task accomplishment (Teng et al, 2018).Hadi and Boscolo (2006) believe that in difficult tasks such as writing which includes many recursively employed cognitive processes, self-efficacy plays a key role.Selfefficacy beliefs also can serve as a 'strong moderators' of academic performance, more specifically writing performance (Pajares, 2003;Pajares & Johnson, 1993).Thus, self-efficacy offer a promising area for informing writing instruction.
The term self-efficacy itself comes from Bandura's Social Learning Theory published in 1997.As Maddux (2012) points out, self-efficacy is the extent of one's belief in one's own ability to reach goals under certain conditions (Pajares, 2003).Bandura (1982) believes that the higher the level of one's self-efficacy, the higher the performance accomplishment will be.If self-efficacy is seen as a moderator, whether or not the plan is translated into action, it's execution will depend on one's degree of self-efficacy (Luszczynska et al., 2010).Shell et al. (1995) found that low self-efficacy learners tend to have unrealistic outcome expectancy to reading and writing while at the same time have lower self-efficacy, whereas high achievers have more realistic outcome expectancy, and are associated with higher self-efficacy.
Research findings have consistently shown that writing self-efficacy beliefs and writing performances are positively related.The studies conducted by the researchers mainly explore the relationship between self-efficacy beliefs and learners' writing performance.Interestingly, all of the results of these studies yield similar findings in which there is a significant positive relationship between writing self-efficacy and writing performance whether it is at the paragraph level and the sub-skill level in writing (Hetthong & Teo, 2013), or at a different writing apprehension level (Singh & Rajalingam, 2012).Further, writing self-efficacy makes an independent contribution to the prediction of writing outcomes and plays the mediational role that social cognitive theorists hypothesize (Pajares, 2003).
In an EFL context, some research is investigating the level of foreign language learners' (EFL) self-efficacy level.Kim et al. (2015) did research on profiling the EFL students' English self-efficacy belief in Korea.The research suggests that there are three profiles of EFL selfefficacy, namely high, medium, and low.The students with high and medium self-efficacy profiles are typically those who have spent more years studying English, and most are female students.They also report that the efficacious students use more Self-regulated learning strategies, unlike those with low self-efficacy beliefs.Sun and Wang (2020) investigate 319 sophomore Chinese students' writing self-efficacy and writing self-regulated learning strategies and how they are related to writing proficiency.The result of the research shows that Chinese EFL students' writing self-efficacy falls in the moderate category.
In the Indonesian context, research in English self-efficacy has gained much attention from teachers, researchers, and practitioners.The first study is conducted by Weda (2018).
Weda's research aims to find out whether there is an existing relationship between self-efficacy, motivation, and writing performance.The subjects under study are 50 students of the English Education Department in the Faculty of Language and Literature Universitas Negeri Medan.Weda (2018) reveals that there is a moderate and positive correlation between students' self-efficacy and writing performance, and there is a moderate and positive relationship between motivation and the students' writing performance.Secondly, Deviana et al. (2019) investigates a correlation between self-efficacy and creative thinking for writing skills.Their research reveals that there is a strong and positive correlation between self-efficacy and learning writing skills.Thirdly, Yulianawati (2019) conducts a case study on three senior High school students to investigate their self-efficacy beliefs when writing recount text.The result of the research suggests that the most important source for self-efficacy belief in the students' writing ability is their mastery experience.And the newest study is conducted by Anam and Stracke (2020) who investigate the role of self-efficacy when learning English in the Indonesian context for primary school students.The result shows that there is a relationship between self-efficacy and language performance.The result also suggests that motivation plays an important role in selfefficacy belief for learning English as a foreign language.Based on the theoretical and empirical stand, the researchers believe that writing self-efficacy can predict students' success in writing task accomplishment.It's crucial to remember that an association does not necessarily indicate a cause.Although self-efficacy and writing performance are related, it is unlikely that raising one's self-efficacy will inevitably result in better writing.
However, research in correlational study on writing self-efficacy and writing performance show a conflicting finding.Some research proposes that there is correlation between writing and self-efficacy (Anam & Stracke, 2020;Deviana et al., 2019;Hetthong & Teo, 2013;Rahimpour & Jahan, 2011;Singh & Rajalingam, 2012;Weda, 2018;Basaffar & Alzahrani, 2022), while other research shows different result (Khojasteh, Shokrpour, & Afrasiabi, 2016;Aldina, 2022).This present study is intended to investigate whether there is indeed a correlation between writing self-efficacy and writing performance.The result of this research is beneficial to enrich the body of knowledge on theoretical grounds in relation to writing performance and writing self-efficacy.Even though research on self-efficacy and language learning in the EFL context has flourished, not much attention is dedicated to self-efficacy for a specific skill, particularly for EFL writing in the Indonesian context.Furthermore, not much research is dedicated to investigating the students' writing efficacy in higher level education in Indonesian context as most research writing self-efficacy is directed in high school and primary school level.Therefore, this research is intended to analyse foreign language students' writing selfefficacy level when they learn to write an essay.The result of this research can add to the body of knowledge in the field of EFL writing in the Indonesian context and provide more information about what teachers or lecturers can do to help students become better writers.Research Problem: Is there any significant correlation between the students' writing self-efficacy and their writing performance?
Two hypotheses are presented as the Null (H0) and the alternative hypothesis (H1) or the researcher's hypothesis.The H0 is "there is no correlation between students' writing selfefficacy and the students' writing performance".While the H1 is "there is a significant correlation between students' writing self-efficacy and the students' writing performance".
them (Cresswell, 2012;Cresswell & Cresswell, 2018).The correlation coefficient can vary from -1 to 1, with 0 representing no association at all.The variables involved in this study are the students' writing self-efficacy (variable X) and the students' writing performance (variable Y).Thus, the main goal of this research is to find out the relationship between these variables, which are students' writing self-efficacy and their writing performance.

Population and Sample
The target population was all the second-year students majoring in English Language Teaching in a private university in Pasuruan, East Java, Indonesia.There were three parallel classes, with two classes consisted of 27 students, and one class consisted of 28 students.The researchers could not randomize the students into two independent group because of two reasons.The first reason was the department did not give any permission to do any random assignment because it would affect the students' flow of learning in other courses.And the second reason was there was a wide gap in the number of male and female students in all three classes.Therefore, the researchers decided to take two intact classes.The researchers applied random selection by using the lottery to decide which classes to use as the sample.The researchers used class A and C class for the sample of the research.As not all students join the class regularly, the researchers only took actively participated students (approximately 50 students from both classes) to be the research sample.At the time the research was conducted, the students took an essay writing course as a mandatory course.There were 6 males, and 44 females joined the research.Their age ranged from 19-20 years old.In sum, there were 67% of the total target population became the sample of the research.

Instruments
The researchers used three instruments to collect the data.The instruments used were tests, documentation, and questionnaire.Each is described as follows.

Self-Efficacy Questionaire
The researchers used the Writing Self-Efficacy Scale questionnaire developed by Prickel (1994).The self-efficacy scale is commonly understood as being very specific; that is, one can have firm self-beliefs in different domains or situations of functioning.The scale has 25 items with a 5-point scale ranging from A to E: (A) Strongly Disagree, (B) Disagree, (C) Unsure, (D) Agree, (E) Strongly Agree.Thus, the maximum score is 125, and the lowest one is 25.The Likert scale used in this study has a scale of 1-5.The number of items on the scale was 25.The researchers categorized the subjects into 3 groups, namely low, medium, and high.Thus, if the subject answered the lowest score of all, namely 1, then the possible score would be 1x25 = 25 (X min).Meanwhile, if the subject answers the highest score of all, namely 5, then the possible score is 5x25 = 125 (X max).Thus, the range of the data is 125-25 = 100.A normal curve consists of 6 standard deviations, each standard deviation of the value is 100/6 = 16.6.In a normal curve, the mean is always in the middle, so mean of the SE is (25 + 125) / 2 = 65.Before requesting the intended respondents to complete the surveys, the researchers did a try-out because there are many advantages to doing so.Apart from assuring the validity and the reliability of the questionnaire, it can also ensure the questions are worded, and that the respondents comprehend the questionnaire in the right way (Dahlberg & McCaig, 2015).In short, the try-out for the questionnaire was aimed at checking the clarity of the instructions, items, and its layout as well as eliminating ambiguities in wording which could lead to misunderstandings.
The try-out was administered to students with similar characteristics and backgrounds as the subjects of this study.The students of the English Study Program in who did not belong to the subjects of this research were taken as the subjects of the try-out.The think-aloud technique is used to collect feedback on the questionnaire.Think-aloud is "the verbalization of thoughts" (Nielsen et al., 2002, p. 5).Using the think-aloud technique enables the researcher to gain feedback directly from the students whether the instruction is clear, or whether any of the questionnaire items are confusing or difficult to understand.
The result of the try-out showed that the students had no problem filling out the questionnaire.the result of the computation shows that the Cronbach Alpha value is 0.785 which indicates a high internal consistency with this specific sample.Although the result of the try-out computation was a little bit lower than the original value of internal consistency when it was tested by Prickel (1994) which is 0.9249, the researcher believed that this self-efficacy (SE) scale developed by Prickel (1994) has served its function to be a reliable and valid scale for measuring writing self-efficacy level for adult students.Having the hard evidence of the reliability of the self-efficacy questionnaires, the researcher used this questionnaire for the present study to determine the students' self-efficacy level.

Writing Test
To get an appropriate writing test for the objective of the study, the researcher developed a test by doing some sequence of activities, such as (1) developing, (2) validating, (3) trying out, and (4) assembling the final form of the test.To make the writing test credible, the researchers used TOEFL test domain of writing framework.Thus, the purpose of the test in this study has a similar purpose to the TOEFL writing test which requires the learners to demonstrate their real writing competence within the academic context.The researchers adopted Test of Writing English (TWE) rhetorical functions and topic characteristics related to writing tasks.The topic of the writing test was 'Money and happiness'.This topic was chosen under the assumption that the topic was interesting and familiar to the students as they could support their claim by providing examples based on their life experience and background knowledge.

Scoring Guide
To score the students' compositions, this study employed a Primary Trait Scoring Guide (PTSG).Latief (2014:240) states that PTSG is simple and easy to use since it requires the rater to come up with only one single score to represent only the most essential component of a text.He further states that this scoring rubric enables the researcher to train the raters to rate the subjects' essays in a relatively short time.The scoring rubric used in this study was adopted from Latief's PTSG technique which was developed in 1990 for his doctoral dissertation on writing assessment.Latief's PTSG (1990) was considered appropriate for the following reasons, namely, the PTSG was developed to measure the students' argumentative essay on the idea development trait.Secondly, the rubric offers clear descriptors for the intended trait so it can guide the raters to score the composition accurately.And thirdly, it has been proven to be effective in asses the EFL learners' writing at the time it was developed.Latief's scoring rubric

Data Analysis
To analyse the students' self-efficacy level, the researchers did a reverse coding for some items.The scores allocated to each response option are reversed in reverse coding, so that the score with the greatest value corresponds to the most unfavourable reaction and the one with the lowest value corresponds to the highest score for the most favourable response.For example, item number 2.
I believe that errors in punctuation and grammar stop me from being a good writer.
The answer of item number 2 should be "strongly disagree" which is typically equal to 1, which seems to be strange since "strongly disagree" is indeed the intended response to the item.Therefore, to show positive direction, reverse coding should be made since the intention of the self-efficacy questionnaire is to show the positive direction of the respondents' self-belief in writing.Reverse coding would show otherwise.In the case of the respond item example above, the appropriate response should be "strongly disagree" which scores 5. Twelve items were reversed in coding, namely, items number 2, 5, 7, 9, 10, 12, 14, 16, 18, 20, 22, and 24.After the items were reversed in coding, the tabulation of each item began.To differentiate the students' self-efficacy level, the group category was not divided based on the median split.According to Schwarzer (2014), people should not be categorized into high or low selfefficacious for there is no 'cut-off score' between the two.Thus, the researchers classified the students' self-efficacy level in writing into three categories, namely low, medium, and high based on how the scores deviate from the mean in the normal curve.
To analyse the data from the students' writing test, the researchers invited two raters.The raters were the writing lecturers in the English Study program at one of the private universities in Pasuruan regency.The raters had two-year experience in teaching writing, have experience in developing writing tasks and assessing writing ability.The final score used is the average of the two raters.If there are any discrepancies more than five points, the third rater was invited to read the essay.To find out the reliability between the raters, the researchers used inter-rater reliability.The purpose of inter-rater reliability is to establish if both raters used the same set of criteria while evaluating the students' writing.In this study, inter-rater reliability was assessed using two methods: Pearson Product Moment Correlation and the Coefficient Alpha, which provided an indication of internal consistency.To analyse the correlation between writing self-efficacy score and the writing performance, the researchers used non-parametric correlation because of the inability to fulfil all the statistical assumptions for parametric test.The non-parametric correlation used was Spearman's rho correlation coefficient in SPSS 23.The correlation is said to exist if the significant level is equal or less than 0.005 (≤ 0.05).

The Students' Self Efficacy Level
Witing self-efficacy questionnaire is intended to measure the students' attitude attitude toward their self-efficacy in writing.However, not all items in Prickel's (1994) questionnaire show positive directions.Some response items in Prickel's (1994) self-efficacy questionnaire show negative directions as the words are negatively worded although the items say the opposite of what was intended.As stated earlier, the maximum score of each questionnaire item would be 125, and the lowest one is 25.The response of each item of every student was tabulated.Figure 2 shows the result shows the result of the students' self-efficacy level.The data shows that the students self-efficacy scores mostly fall in the medium category (58%), while the rest fall in the high category (42%).None of the students who participated in the study had low self-efficacy in writing.

Result of Writing Test
The mean score obtained from Rater 1 is 2,9600, and the mean score from Rater 2 is 3,000.It indicates that the mean score obtained from Rater 2 is higher 0,04 than the mean score from Rater 1.In terms of the standard deviation, the standard deviation from Rater 1 (0,75485) is considered smaller than the standard deviation of the scores from Rater 2 (0,75593).This indicates that the scores from Rater 1 are more tightly grouped around the mean than the scores from Rater 2.  As a result of involving two Raters to assess the work of the students' writing, it is important to know the inter-rater reliability between them.The goal of inter-rater reliability is to determine whether both Raters were using the same set of criteria when assessing the students' writing.In this present study, inter-rater reliability was measured by using two measurements, namely Pearson Product Moment Correlation between the first and the second-Rater (Table .3). and Coefficient Alpha, to provide an estimate of the internal consistency (Table 4).Table 4 shows that the reliability coefficient between the two Raters is 0.834.The average measure of the interclass correlation of .834indicates a very high agreement between the two Raters which shows that there is not a lot of variability between them during the scoring.The result of reliability coefficient (0.715) and coefficient alpha (0.834) indicates a fairly high level of reliability of the data.Therefore, it can be said that further confirmation of data analysis has been made based on those results.

Testing the Hypothesis
Based on the descriptive data, the students who have high self-efficacy scores do not necessarily have high writing scores.Yet, some students with low writing self-efficacy scores achieve high writing scores.Take for example, subject no 13, 17, 22, 25,44.These students score high in Writing Self-Efficacy Scores, but they get low writing scores.Some students who score low in writing (subject 3,11, 16, 23, 28,40) have moderate writing self-efficacy scores.The table presents an interesting fact about the students 'self-efficacy level and their writing performance.Based on the descriptive data presented, the students who were in the category of high self-efficacy were not always able to reach high scores in writing (4.High SE_ Moderate_ Writing/ 18%).The highest category fall in the Moderate SE_ Moderate Writing (24%).Surprisingly, no students have low self-efficacy.The data show that the students who joined essay writing courses have moderate to high writing self-efficacy even though their writing scores vary from low to high.

Testing the Hypothesis Fulfilling the statistical Assumption
The researchers need to fulfil five statistical assumptions before running the computation for correlational analysis.The first assumption is the variables should be at the interval and ratio.The first assumption is fulfilled because both the writing test and writing self-efficacy forms are in the interval scale.Thus, the first assumption is fulfilled.The second assumption is there should be a linear relationship between the two variables under study.The result of the assumption test for linearity shows that the dots do not form a straight line, or do not represent a straight line.This implies that the data is not linear, thus assumption number two is violated.The third assumption that should be checked is there should be no significant outliers.To check the third assumption, the researchers run an SPSS computation.The box plot shows that there are no outliers in the students' SE and writing scores.This implies that if the students' scores are visualized in the histogram, the score distribution is approximately normally distributed.Thus, the third assumption is fulfilled.The last assumption is that their data should be approximately normally distributed.There are two tests available for normality tests, namely the Kolmogorov-Smirnov and Shapiro-Wilk.Both normality tests for self-efficacy show that the significant value is higher than α > 0.5.This means that the data of Self-efficacy scores are approximately normally distributed.On the other hand, the computation of Kolmogorov-Smirnov and Shapiro-Wilk for writing shows that the sig value is lower than < 0.5.This implies that there is a difference in the scores, and they are not normally distributed.Thus, the normality assumption for the writing test is violated.As not all the assumptions for correlation computation are fulfilled, the researchers could not use the correlation for the parametric test.Therefore, to test the hypothesis, the researchers applied a non-parametric test to find out the relationship between the two variables.Table 9 shows the Spearman's rho correlation coefficient, r, is -.020, and that it is not statistically significant (p = 0.892).Thus, it can be concluded that there was no correlation between Writing and writing self-efficacy (r= -.020, n = 50, p = .892).This means that the researchers failed to reject the null hypothesis.

Discussion
In relation to the research question, the result of the research does not support previous research that found that writing and self-efficacy are correlated (Anam & Stracke, 2020;Deviana et al., 2019;Hetthong & Teo, 2013;Rahimpour & Jahan, 2011;Singh & Rajalingam, 2012;Weda, 2018).This study did not have enough evidence to reject the null hypothesis and to prove that the alternative hypothesis is true.However, the result of this study supports some research that found that there is no significant correlation between writing self-efficacy and writing performance.Khojasteh, Shokrpour, and Afrasiabi (2016) research shows that there is no significant correlation between students' self-efficacy and writing performance.The result of this study also supports Basaffar and Alzahrani (2022), who revealed that there is no correlation between students' self-efficacy and their writing performance seen in specific writing ability aspects (content, organization, vocabulary, cohesion, and writing mechanics).Similarly, Aldina (2022) also found that there is no significant correlation between writing selfefficacy and writing performance.Looking at the result of the correlational analysis, despite of being insignificant, a weak correlation (r= -.020, n = 50, p = .892)shows that some students with moderate or high self-efficacy did not necessarily have good writing achievement.This prediction, however, needs more evidence and clarification since it does not show a cause-andeffect relationship.
The findings of the present investigation reveal a positive correlation between students' self-efficacy and their writing performance, particularly in distinct facets of writing ability encompassing content, organization, vocabulary, cohesion, and writing mechanics (Astiantih & Akfan, 2023).This implies that heightened self-efficacy among students is associated with enhanced writing proficiency.This outcome aligns with the research conducted by Haerazi and Kazemian (2021), wherein it was established that self-efficacy serves as a determining factor Setyowati, Mabaroh, & Karmina The Correlation Between Students' ……….. Languages and Language Teaching, January 2024. Vol.12, No.1 | 83 in students' achievement of language skills.Therefore, the current study contributes to the existing body of knowledge by affirming and extending the understanding of the relationship between self-efficacy and writing performance, emphasizing its specificity in various dimensions of writing proficiency.

JOLLT Journal of
This research, however, yields two results.The first is the students' self-efficacy level, and the second is the result of the correlational analysis between self-efficacy and writing performance.The first result of this research supports Sun and Wang (2020), who state that EFL students tend to have moderate self-efficacy levels.More specifically, in the Indonesian context, most subjects in this study have a moderate level of self-efficacy belief.The result of this study also supports Li (2022), who found that college students' self-efficacy when writing in English was at a moderate level and lacked self-confidence when it came to writing in English.Li (2022) further suggests that writing teachers apply a process-writing approach, set achievable writing goals for the students, and help reduce the students' writing anxiety.In addition, the result of this research is similar to Aldina's (2022) research, investigating high school students writing self-efficacy.Adina's (2022) research reveals that the students' writing self-efficacy level is moderate level.Furthermore, the result of this research also confirms Rahman, Abduh, and (2022) study, which found that most college students writing self-efficacy falls in the moderate category.
Despite the failure to reject the null hypothesis, this study has pedagogical implications.The fact shows that no students in the present study have low self-efficacy level, and none of them also has low writing scores, English writing teachers or lecturers should pay attention to the students' self-efficacy in writing and how it might relate to their writing achievement.As stated by Bandura (1997) people learn from four sources, namely from the mastery experience, representative experience, social persuasion, and psychological state.Taken into this context, it is encouraged that the English writing teachers in the Indonesian context continually nurture the students' writing self-efficacy by giving more opportunities to experience writing and to get the feeling of success (mastery experience), to give more chances to learn from their peers through, modelling and discussion (vicarious experience), providing constructive and positive feedback through the student-teacher conference (verbal persuasion), giving sympathy and care for the student's psychological and affective state when they feel not confidence with their writing ability (emotional and psychological state).
In relation to feedback, feedback is regarded as one of the essential components for accomplishment in a variety of fields (Narciss, 2013).Research has shown that feedback from teachers can improve writing ability and self-efficacy in a similar manner (Cui, et al, 2021).Aside from that, the result of this current study also encourages the writing teacher to apply a more process-centered approach than the product one.The process-centered approach gives the students more opportunities to edit and revise before the composition comes to the final product (Setyowati et al., 2017;Setyowati et al., 2018).Furthermore, research indicates that the most compatible writing instruction that can improve the students' writing self-efficacy is the process-oriented approach (Zhang, 2018).Applying an appropriate writing approach can help the students become better writers.

CONCLUSION
Writing is a complex process.It requires not only ideas but the ability also to organize ideas and linguistic skills to form good writing.But writing also requires engagement to the text and the self-belief of one's ability to accomplish a specific task.The students in the present study happen to have a moderate level of self-efficacy belief.This means that they are a kind of a good writer and a kind of not when it comes to writing a composition in English as a foreign language.Unfortunately, the research fails to reject the null hypothesis as there is not enough Setyowati, Mabaroh, & Karmina The Correlation Between Students' ……….. Languages and Language Teaching, January 2024. Vol.12, No.1 | 84 evidence to prove that the researchers' hypotheses are correct.The correlational analysis shows that there is no correlation between the students' writing self-efficacy and their writing performance (r= -.020, n = 50, p = .892).

JOLLT Journal of
Some recommendations are addressed to future researchers and writing practitioners.Firstly, if future researchers wish to replicate the present study, it is suggested that future researchers use more sensitive rubric, such as the analytical one, to score the students' writing.The writing scoring rubric, which discusses each element of writing, will give a more fair judgment and comprehensive analysis of the student's writing ability.Secondly, future researchers need to elaborate on the role of self-regulation and how it relates to the students' writing performance and self-efficacy in the Indonesian context or the EFL context in general.Future researchers can also explore the type of writing instructions and feedback suitable for the students-writer in a foreign language context to help them improve the students' writing selfefficacy and writing performance.In sum, there are still more to be investigated to explore the students' writing in EFL context with the sole intention of helping the students become a better writer.
Setyowati, Mabaroh, & KarminaThe Correlation Between Students' ……….. JOLLT Journal of Languages and Language Teaching, January 2024.Vol.12, No.1 | 77 has a 4-point scale with the lowest score of 0 that describes the writer's inability to state a claim.
Figure 1.Result of the students' score of self-efficacy questionnaire

Figure 7 .
Figure 7.The box plot of SE and writing

Table 3shows
that the reliability coefficient is 0.715, which indicates a relatively high level of consistency between the first and second Rater.