The Effectiveness of Portofolio Assessment in English Writing Skill of Islamic Junior High School

Article Info ________________ History Articles Received: 16 January 2021 Accepted: 21 February 2021 Published: 30 April 2021 ________________


INTRODUCTION
One of the goals in mastering language is to improve students' communication achievement, either in oral or written form. Whereas both skills show what is encoded from. In term of the skills which is called as one of the micro abilities that is salient for effective communication (Aprianto and Zaini, 2019). The first function of language is a tool of communication. Communication is a process of forwarding speakers' intention to others by using certain channel either in oral or written. Therefore, having ability of productive skill might be beneficial to work with in terms of communication among the people in the world.
In the Indonesian curriculum-13 Curricullum is stated that English is considered as a tool for expressing meanings. Based on the concept and the function of English, the teaching of English at secondary school aims at developing the four language skills: listening, speaking, reading, and writing skill. These four skills must be applied in balance and thematic ways.
One of English skills that must be improved is writing skill. Writing skill is the capability in using language to communicate in written form. By writing, someone can elaborate his/her idea to get their purposes. Hafiz, HS. (2020) states having writing skill is complex and unique one. It needs some knowledge as well as skill. It means that we cannot argue against the grammar only but also, we must express the idea obviously, united, significant, logic, and coherent sentences or paragraph. The process of learning writing advantages in developing students' skill to get success. Considering the needs of mastering writing skill as one of the ways to elaborate of idea in written form, teachers should have many kinds of method applied to teach their students' writing skill. Teachers' strategy in teaching and learning process will be engaged in as tutor and writing assistant. In this regard, Tarigan (2013) investigates that student should be given the chance to write several drafts and develop their ideas because writing skill is not easy and a programmed activity.
The time, which is given for the process to work, along with the appropriate feedback from the readers, such as the teacher or other students will discover new ideas, new sentences and new words as planned, writes the first draft, and revises what he/ she has written for a second draft. Consequently, the teacher's assessment procedures should be adapted in such a way that they faithfully reflect teaching practices.
Referring to Elia Santi, et.al. (2019) in the 13 curricula, teachers must not only evaluate aspects of knowledge, but also evaluate aspects of attitudes and skills. Such situations, Aprianto (2017) emphasizes on the use of authentic assessment which is commonly as a comprehensive measurement used to the students resulted in the aspects of attitude, skill, and knowledge outcomes. Assessment is usually an ongoing strategy through which students' learning is not only monitored but by which students is involved in making decisions about the degree to which their performance matches to their ability. It involves collecting evidence of learning over an extended period, using a variety of assessment methods both conventional tests and alternative methods of assessment. Porsedo (2019) states that conventional language tests no longer meet the needs of language monitory students acquiring English as a foreign language since they have often been limited to assessment of the students' outcomes at a specific point of time and have provided little information about teaching/ learning process. The recent wave of instructional reform reflects revolutionary ideas concerning the nature of assessment and its purpose. The main purpose of language assessment is to help us gain the information we need about our students' abilities and to do so in a manner that is appropriate, consistent, and conductive to learning. Thus, assessment has become an integral component of the instructional system (Efendi, et.al.,2017). In addition, a recognizable change about assessment process is the moving of assessment from a judgmental role to a developmental role. Regarding to the assessment process as one major element of school program, portfolio assessment should be regarded as basic part of teaching, and then it can be used for instructional objectives.
The observation shows at the current setting of research, the evidence from teachers' reports and final tests show that writing is the weakest aspect of students' work in English. This is also supported by the results of the semi-structured interview given with a group of students. Referring to informants' responses, the students underwent having problem in English writing. Furthermore, memorization is given the focus. Once students type the memorized composition, the teacher can easily grade them because there is an identifiable structure. Students are not shown the way to develop or explore ideas on their own. Thus, they play no role about their own learning. Considering to the way of assessing students' writing products, the traditional paper and pencil tests, which are usually shared twice during the term, still dominate the writing classes. According to the conventional grading system, writings are usually checked, given grades, and returned while students are passive participants in the assessment process. Therefore, test scores cannot be trusted as a basis of making decisions concerning the student' actual writing performance. Note that the students' low writing skill may be due to the instructional practices used in teaching and assessing writing.
The method used is called product based. Writing process is somewhat ignored. Teachers attend to the product: its clarity, originality, and correctness but they do not attend to the writing process. Nor do they attend to the writers themselves. In such a method, the teacher is dominant and errorhunter while he/she should be facilitator and assistant. Boumediene (2016) explains with such type of process means more interaction and cooperation both the student and the teacher. Approximately, it is a single of the activities used over time in the development of written products. So, the researcher would like to argue for improving the process orientation in teaching and assessing writing skill. While this is the current situation of teaching and assessing writing at MTs NW Aik Ampat, elsewhere there are new trends that consider the students' needs and give emphasis to their strengths rather than to their weaknesses. These trends are the alternative forms of assessing what the students know and can do with the language, the alternative forms of assessing the students' achievement are portfolio assessment.
Portfolio assessment is recommended as alternative assessment approaches. The assessment role is expected to be good ways in building students' writing skill. Referring to the effect of portfolio assessment on students' ability in writing, Sharifi and Hassaskhah (2011) proved that portfolio assessment had a favorable role in improving students' writing ability in general and the writing aspects after examining the effect of portfolio assessment on writing in regular test. Portfolio is a collection of the past work. However, in the context of assessment, portfolio does not represent only a mere collection of the past work and purposeful collection of student work that tells the story of the student's effort, progress, or achievement in given areas (Kemdikbud,2019). Portfolio can be viewed as a systematic and organized collection of evidence used by the teacher and student to monitor the growth of student's knowledge, skills, and attitudes in a specific content area. Portfolios offer the benefit of involving students in the assessment process. Portfolio assessment is not teacher driven as is common in conventional assessments. In keeping with the trend toward student-centered classrooms, portfolio assessment is a shared responsibility. It requires the involvement of students, parents, as well as teachers, in establishing the assessment standards, criteria, and selecting contents of the portfolio.

METHODS
The research is quantitative study which is analyzed with numerical data approach then being statistically counted. (Azwar, 2012). The study is in terms of preexperimental method. The design of the research was a pretest and posttest design, it can be seen in Figure 1. In this design, observation was done two times; they were before and after experiment. Observation was done before experiment (O 1 ) was called pretest and observation had been done after experiment (O 2 ) was called posttest.
The population of the study is the eighth graders of MTs. NW Aik Ampat which consists of three classes. The following table 1 is the description of population of the study. The sample is randomly selected as experimental class in the way the researcher randomizes the three classes, they are the eighth grade of A, B, and C. As a result, the eighth grade of A is chosen as the sample of study.

English Writing Test
The English writing test is developed to assess the students' writing skill in English, the instrument is subjective test which asked the students to write a simple descriptive text. In English writing test, the students must pay attention sub-writing domains are measured: purpose, content, organization, vocabulary, sentence structure, and mechanics.
The English writing test is one paragraph writing on a randomly selected prescribed topic that is familiar to students. Test materials are represented in descriptive style of writing materials for the eighth-grade students. In pretest and posttest, students must write a descriptive paragraph.
The test refers to some important aspects as criteria point out competency standards and basic competencies. The writing assessment criteria consist of 1) Main Idea/purpose; 2) content organization; 3) Vocabulary; 4) Grammar; and 5) Mechanical (Elia Santi, et.al.2019). The English writing test is scored analytically by using an analytic scoring rubric which is developed by researcher. Performance on each criterion is judged along four levels of performance. The teacher gives each student a score on each of the criteria (sub-writing domains) out of 60 (sixty).
An instrument is urgent of finding out the result of study, so an instrument must be well-prepared. This topic presents about tryout of instrument and validity of instrument.

Instrument of Try Out
To know whether the instrument is suitable or not, the researcher tries out the instrument. The try out is held on MTs.NW Aik Ampat. The result of try out is useful for the following: To decide the time allocation. To find out whether the instruction is understandable or not. To gather evidence to support the instruments validity and reliability.
Based on the result of the try out, the students can answer the test based on the instruction. As the result, the students do not ask too much about how to answer the test. It means the instruction of the test is not needed to be revised because it is understandable for the students. From the result of the try out,

Experimental
Group O1 X O2 the researcher gets the data needed to count the validity of the instrument.

Validity of the English Writing Test
An instrument will be valid if it is able to measure what should be measured (Arikunto, 2020). It is also stated that to assess the right assessment must use the valid instrument (Hakim and Zulkifli, 2021). In this study, the researcher used two kinds of validity, namely, construct validity and content validity.
Construct validity According to Tentawa (2018), construct validity refers to the extent to which the results of the data collection process can be interpreted in terms of underlying psychological construct. The instrument has construct validity if there is a relationship between theory and concept of the competence needed to accomplish the tests in this study. The instrument was consulted to the teacher, and a course book is being used for Junior High School level. So, it is categorized into a valid test.
Content validity According to Hendryadi (2017), content validity refers to the extent to which data collection process measures a representative sample of the subject matter or behavior that should be encompassed by the operational definition. Related to this study, the English writing test content is descriptive text.

Portfolio Assessment Model
The portfolio assessment model is developed to improve as well as to assess subjects' writing product beside the English writing test as a single measured is incapable of estimating the diversity of skills, knowledge, processes, and strategies that combine to determine students' progress. It was the treatment given to the experimental class.
To measure students' level skill and ability, it commonly uses rubrics, (Brown in Syafrizal, et.al.2020). The scoring rubric of writing skill prepared as a guide which consists of evaluative criteria, definition of quality, and assessment strategies which is for assessing students' language ability to evaluate the progress of language teaching. The evaluation given is in terms of portfolio assessment of writing descriptive text.
During the implementation of portfolio assessment model, the teacher regularly reviews the students' portfolios contents evaluating every writing assignment separately and providing feedback. No grades are reported on the first, second, and third drafts; grading is delayed until the final version of each chosen assignment is submitted. The teacher responds to those drafts not to provide a grade but to provide suggestions for revision as well as some general commentary about the students' development as writers. The scores are recorded secretly to keep track of students' writing improvement.
At the end of the experiment, portfolios are submitted for final assessment. Entries are not graded separately. Instead, a grade is assigned for the portfolio using scoring rubric.

Data Analysis
The data are statistically counted through the computer package SPSS 17 for Windows. Descriptive statistics such as mean score (M) and standard deviation (SD) of students' score are computed, t-test (2-tailed test) is used to determine if there are any significant differences in the mean scores between the students' achievement on the pretest and posttest of the English writing test (before and after the treatment). For statistical analysis, the alpha level of significance 0.05 of confidence is set. Pearson product-moment correlation is also used to assess the correlation between the subjects' (experimental class) means of scores.
To get mean score of the students' score on pretest and posttest manually, it can be used the following formula:  (Riduwan and Sunarto, 2013) Furthermore, to analyze the hypothesis testing manually which is aimed at knowing whether the alternative hypothesis is accepted or not, the researcher finds r-test first before analyzing t-test. To find out r-test, it can be used the following formula: (Riduwan and Sunarto, 2013) Meanwhile, to analyze the hypothesis testing, it can be used the following formula: Where: r = Correlation score of X 1 and X 2 n = Number of samples = Mean score of pretests = Mean score of posttests s 1 = Standard deviation of pretest s 2 = Standard deviation of posttest S 1 = Variance of pretest S 2 = Variance of posttest Riduwan and Sunarto (2013) The hypothesis criteria, if: t table ≤ t test, it means that Ha is accepted, and Ho is rejected, t table ≥ t test, it means that Ha is rejected, and Ho is accepted.

Research Findings
The research finding in this study is description about students' writing improvement for the eighth graders of MTs.NW Aik Ampat after following pretest, treatment, and posttest. As mentioned above, there are 34 students of Grade A as the sample whom they are given treatment by using portfolio assessment. The collected data in this research are about English writing test and portfolio assessment. Before giving treatment, the researcher gives pretest to the students in order to know the students' writing achievement before portfolio assessment implemented. After pretest is done, the treatment is implemented in teaching and learning process. At the end of treatment, the researcher gives posttest to the students and collects students' portfolio to know whether the portfolio assessment is effective in writing skill for the eighth graders.
From the study, the researcher gets the mean score, standard deviation, and variance of pretest and posttest score of the experimental class. The mean, standard deviation, and variance of the score are computed by SPSS 17.0 for Windows, it can be seen on Table 2. = Variance Based on the data, it can be seen there is difference between students' achievement in writing before and after treatment implemented. Mean of the students' score after portfolio assessment is higher than mean of the students' score before implementing treatment.
To analyze whether there is a significant difference between students' achievement in writing before and after treatment implemented, the researcher uses ttest method to get answer the hypothesis. The hypothesis criteria, if: t table ≤ t test, it means that Ho is rejected and t table ≥ t test, it means that Ho is accepted. The result of ttest is to know whether the hypothesis is accepted or rejected. The hypothesis in this study is portfolio assessment significantly effective in writing skill. The result of hypothesis testing after analyzed by using SPSS 17 for Windows, it can be seen on Table 3: After analyzing the data by using ttest at the level of significance () 5%, the researcher gets rtest is 0.671 and ttest is 14.997, it shows that there is a significant difference between the students' achievement in writing before and after portfolio assessment implemented.

Discussion
Based on the result of the study, there are several data found namely students' achievement in writing skill before conducting treatment, students' portfolio, and students' achievement in writing skill after conducting treatment. In the data of students' achievement in writing skill before treatment conducted, it is found mean score of pretests is 31.18 and mean score of posttests is 44.97.
Students' score after treatment implemented better than before implemented in writing skills. The subjects show a satisfactory improvement with reference to each writing product skill (purpose, content, organization, vocabulary, sentence structure, and mechanics). The high gains obtained on the post administration of the English writing test can be due to the nature of the portfolio assessment model. It is noticed that the portfolios are effective in students' writing skill.
The results also shows from hypothesis testing, the researcher gets r-test = 0.671 and ttest = 14.997 at the level significance () 5%, it shows that there is a significant difference between the students' achievement in writing before and after portfolio assessment implemented. It means that portfolio assessment is significantly effective in writing skill and such Sugiyono (2014) states that if interval coefficient 0.60-0.799, it means that there is positive correlation between portfolio assessment and students' writing skill. The results also indicate that providing the subjects with chance and time to practice writing improvement of students' writing skill.
Furthermore, the implemented portfolio assessment model provides the students with ample opportunity to deal with a variety of pieces of writing, thereby develop each writing skill. During the experiment of the current study, the subjects (experimental class) are involved in the processes of creating the portfolio. Conferencing is an effective strategy to encourage the subjects to share their writing, provide feedback, and develop their writing skills. Generally, developing or exploring such processes through portfolio improved the students' writing skill and created an active atmosphere in class.
Portfolio is usable for teacher and students. Teacher can use portfolio to analyze which material that should be explained again and which instruments that should be repaired. Teacher can send portfolio to parents to show the students' learning progress, so parents know how to help the students to learn at home.
Besides, portfolio is very important for students. Students can see their development time to time. Students can find feedback of their effort in learning. According to Farr and Tone (Vangah, et.al., 2016), over the alternative assessments, portfolio mixes instruction with assessment of self-reflection and self-evaluation. It includes in collecting samples to denote the increase through record keeping and conferencing with teachers and peers. Thus, it can make students to be independent learners to have great progress. Dimyati and Mudjiono (2013) states that students will be more motivated in learning if the students know the result of work as a feedback. In the implemented treatment, a teacher asks the students to make revision of their work, so the students are accustomed to make a best work. Because the students' work is documented, the students can see their work time to time, so they know what their weakness is, and it motivates them to study hard and better. Then, it will be a tool for teacher and parents to help the students in learning English when the students meet such difficult materials.
Based on the explanation above, the researcher concludes the students are taught with portfolio assessment will be prompted to have intrinsic motivation to learn English hard and better because they get enough feedback from the works that are documented in a file, so their learning achievement are better than the students' learning achievement teach with.

CONCLUSION
Based on the findings of present study, the researcher can conclude that there is significantly different in the mean score between students' writing before treatment conducted and after it conducted. It also shown from the result of hypothesis testing, it means that Ho is rejected because 0.000 < 0.025 (2-tailed) and based on t-test criteria, if ttest is bigger than t-table, it means that Ha is accepted, and Ho is rejected. Besides students' writing skill achievement with portfolio assessment is better than the student writing skill achievement without portfolio assessment at the level of believe 95%. Portfolio assessment has positive correlation with students' writing skill, it shows from the result of r-test in this study is 0.671, meaning that the level of correlation between portfolio assessment and students' writing skill is strong. Based on the findings it can be recommended that the assessment of teachers should have well-prepared materials, assignments, and other activities being appropriate time and it is better implemented in small class because it can be easily managed. The headmaster should take a close look at teacher's salary because this is out of proportion to teacher's duty in complementing portfolio assessment.