UNDERGRADUATE STUDENTS’ PERCEPTIONS TOWARD SELF-ASSESSMENTS

In the current education context that encourages students to be independent learners, self-assessments have gained more attention as one of the effective ways to support self-learning. However, teachers in Indonesia seem hesitant to use self-assessments because they assume that students cannot understand and perform self-assessments. Therefore, the research is aimed to know about undergraduate students’ perceptions toward self-assessments, which can result in knowing students’ comprehension of self-assessments. The research was conducted at an English Language Education Program (ELEP) from one of the private universities in Central Java, which in volved 42 students from the 2018 academic year who had passed Teaching Skills Courses (Teaching English for Young Learners, Teaching English for Adult Learn -ers, and Teaching English for Specific Purposes). A questionnaire consisting of close-ended and open-ended questions was used to collect the data, and the data were analyzed using statistical analysis and content analysis methods. From the re -sults, it was found that the participants reacted positively toward self-assessments. The participants felt content and helped when doing self-assessments, and they also knew the benefits and the importance of it. Furthermore, it was found that they were able to carry out self-assessments appropriately and showed their preference for self-assessments, such as comparing with criteria or making reflections. Therefore, it can be concluded that the teachers’ previous assumptions can be considered inac-curate. The study is hoped to help teachers consider implementing self-assessments more in their courses and help them design suitable self-assessments.


INTRODUCTION
Self-assessments have grown importance in the education context over the years for its various benefits toward students. Self-assessments can be used by students to self-evaluate the quality of their work and choose strategies for work improvement (McMillan & Hearn, 2008). It can needs to be improved for their skills (Desjarlais & Smith, 2011). It means that self-assessments are not merely about the current task, but it also determines performance of the future tasks as students try to improve their skills. Brown and Harris (2014) categorized self-assessments into three main categories based on the levels of performance and students' thought process complexity. The three levels of self-assessments were a basic level, an intermediate level, and an advanced level.

Levels of self-assessments
In the basic level, Brown and Harris (2014) stated that students at this level do self-assessments by simply determining how well they do their tasks. It can be from reviewing their performance or task, such as counting how many correct answers students can get in a test or quiz (Jones et al., 1977;Wan-a-rom, 2010; as cited in Brown & Harris, 2014). This activity requires a simple thought process, hence why it is categorized in the basic level. However, this activity does not really develop students' critical thinking because it relies on the basis of guessing, and students create their own criteria to determine their improvement. Therefore, improvement of ability may be very limited in this level, which does not optimally serve the purpose of self-assessment.
In the intermediate level, Brown and Harris (2014) stated that students do self-assessments with the assistance of external sources (such as teachers' evaluations and models). The models can be in the form of videos, others' works, or even teachers' explanations of the work process. Self-assessments can also be done by comparing the current task with the previous task. As long as students can correct their work based on decided standards through the determined examples, it can be considered as self-assessments. However, Panadero and Alonso-Tapia (2014) argues that students may not be able to self-assess if teachers do not explicitly teach them. When there is no explicit instruction to do self-assessments, students may not know which aspects of their work they need to fix. Even, they may not self-assess themselves at all because they think it is not required. Thus, the purpose of self-assessments in the intermediate level has not also been fully optimized.
In the advanced level, Brown and Harris (2014) suggest that students need to be given rubrics or criteria that contain rating scales or checklists to influence them in making a deep analysis for self-assessments. At this level, students are required to think deeply about whether their works have matched with the description of their desired ratings or checklists. Students could also make reflective journals in this level as they undergo deep and critical thought of their performances. Looking at the subject, which is undergraduate students, rubrics are the most necessary one in contrast with the action of comparing students' works by their own made-up rating or with given models. This is because university students need to think more critically and deeply about their performances, which can be done through rubrics that give explicit criteria (such as the characteristics that their work needs to possess in order to get satisfactory scores in terms of writing organization).

The effects of self-assessments
The effects of self-assessments vary for everyone, there may be positive or negative effects on the students. This section is going to discuss previous studies from various researchers that found the negative and positive effects of self-assessments. The effects are in terms of the students' performance and self-regulation in learning.
According to the study by Rolheiser and Ross (2013), self-assessments can be used for future improvement of students' work through matching their work with provided descriptions or criteria in the rubric. By comparing the criteria with their work, they can determine their quality of work and try to revise it based on their expected criteria. Moreover, self-assessments include the process of self-evaluation where students can identify their skills' forte and limitations (Shatri & Zabeli, 2018). When knowing their limitations, they can focus more on improving their concerning skills and reach breakthroughs that match their expectations. Therefore, students can significantly increase the quality of their work as their strengths can be highlighted, and their weaknesses can be reduced.
However, there are also arguments against the effective use of self-assessments regarding their effects on students' performance. Epley and Gilovich (2005) claims that students can overestimate their skills as they deny their lack of ability, thus resulting in poor performance (as cited in Brown et al., 2015). It may be the case when talking about short-term effects, but overestimation can still result in students' improvements in long-term effects. Students may overestimate their work because they are too confident, so they are still likely to persist when meeting failure. Butler (2011) claims that students with a positive bias toward their work significantly improve their scores after a long term, while negative bias is the one that should be concerned more as it can lead to undeveloped abilities for low achievers. However, negative bias can also have a different case for high achievers because it can prompt them to improve their abilities and performances.
Second, self-assessments can promote self-regulation by giving students a sense of control over their learning. According to Punhagui and Souza (2013), self-assessments include the process of students progressively monitor their own learning. They can revise their works through evaluation with criteria at any time and place they want to (Fahimi & Rahimi, 2015). This means that not only they can correct their work at their own pace, but they can also control their learning processes. Therefore, self-assessments may most likely play a fundamental role in students' self-regulatory learning.
However, giving self-assessments to the students may mean more things that they need to consider in producing their work. Students may see self-assessments with various criteria inside of it as burdensome and overwhelming. A study by Panadero and Romero (2014) finds that students can be overwhelmed with self-assessments, leading to avoidance behavior of their tasks. The avoidance behavior can escalate to the extent where students abandon their tasks until there are externalities that drive them to do the task, such as deadlines or their teacher's pressures. Therefore, self-assessments can make students feel overwhelmed, which in turn can make them avoid self-regulating themselves.

Students' perceptions toward self-assessments
There are clashes of views regarding self-assessments, some students can react positively toward the effects of self-assessments, and some can disagree. Such as a study done by Andrade and Du (2007) found that students felt improvements in their skills because they knew what to expect from their work. They are confident to make better academic work because they know what they will receive for their desired criteria. Another study by Wang (2016) showed that students perceived self-assessments through rubric as beneficial as it can make them realize what they should do to reach the standard criteria in their task. However, Wang also states that some of the students disagree as they feel that the description is not detailed enough to make them have significant improvements. Nevertheless, most of the students perceive self-assessments as one of the most effective ways to increase their academic ability. In regards to students' self-regulation, a study by Ratminingsih et al. (2018) showed that students feel a growing responsibility for their work through the built habit of revising their work. In other words, students can feel that they need to take their learning into their own hands from the responsibility. Furthermore, a study by Andrade and Du (2007) shows that students feel they plan more on what they should do for their work with self-assessments through rubrics as the trigger. Therefore, students feel they need to plan and execute the plan by their initiatives. However, Andrade and Du also point out that some of the students disagree as they feel they are only following the standards of their teacher. They further explain that because the teacher determines the grade in the end, students fall into thinking about what is suitable for their teacher's taste rather than focusing on improving themselves. Therefore, even though students do the work independently, they still think they rely too much on their teacher to make good quality work. What is certain, students are not of a single identity, and each student has different perceptions of self-assessments.

METHODS
The study was conducted at the English Language Education Program (ELEP) from one of the private universities in Central Java. A quota sampling method was used to determine the participants. In other words, the participants were chosen because they could pass the determined criteria (Perry, 2011). The participants were 42 undergraduate students from ELEP 2018 academic year who had passed the Teaching Skills Courses (which were Teaching English for Young Learners, Teaching English for Adult Learners, and Teaching English for Specific Purposes). The reason for using these criteria was because the participants have gone through several courses which apply self-assessments as a part of learning and have ample experience in designing various assessments during the chosen courses. Therefore, the participants were able to give comprehensive opinions about self-assessments.
A questionnaire was used to collect the data from the participants. It was developed based on the literature review, and it contained a combination of close-ended questions (fourteen questions) and open-ended questions (five questions). After piloting the instrument, the questionnaire was revised according to the answers of the pilot participants. The data were analyzed using a statistical analysis method for the close-ended answers and a content analysis method for the open-ended answers. The gathered data were categorized based on the following categories: Students' feelings in doing self-assessments, students' opinions toward self-assessments, and students' behaviors when doing self-assessments. Answers from the closed-ended questions were presented in a form of bar graphs and used as a basis for students' views toward self-assessments. Lastly, answers to the open-ended questions were presented in two ways, which were in a form of tables (for the total of occurrences found in the answers), and descriptions to provide elaborations behind their views based on the categories.

FINDINGS AND DISCUSSION
The study is aimed to know undergraduate students' perceptions toward self-assessments. The analysis is divided into three main themes: Students' feelings in doing self-assessments, students' opinions toward self-assessments, and students' behaviors when doing self-assessments.

Students' feelings in doing self-assessments
This section explores students' feelings when they are doing self-assessments. To explore the students' feelings, there were two questions provided. The first question was a close-ended question, "Do you enjoy doing self-assessments?" and the second question was an open-ended question, "What do you feel when doing self-assessments?".

Figure 1. Students' Feelings in Doing Self-assessments
As shown in Figure 1, 92.86% of the participants reacted positively toward self-assessments. It was found that almost every participant enjoyed doing self-assessments. Responses of agreement were the highest, with 32 participants agreeing that they enjoy self-assessments (76.19%). Furthermore, responses of strong disagreement were the lowest (0%), with no participants extremely disliked doing self-assessments. Another evidence that students liked self-assessments was shown in the open-ended responses (Table 1). From the number of occurrences the participants gave, 70.69% of the occurrences were found to be positive feelings. For the rest responses, 29.31% of the total occurrences were participants' negative feelings toward self-assessments.
Most of the participants gave positive words such as feeling content, helped, and enthusiastic. Frequently given answers were related to how self-assessments gave feelings of contentment (32.76%) and support (24.14%) to the participants in knowing their abilities and accomplishment. On the other hand, the participants gave negative words related to their feelings of apprehension, disappointment, and confusion when doing self-assessments. Participants' negative responses were frequently about their feelings of fear for not being able to understand their works (15.52%) and disappointment for not reaching the expected criteria (8.62%).
Below were some responses from several participants. S15 shared that self-assessments were helpful to evaluate and know what the participants had achieved. S24 and S27 also shared their feelings of contentment when doing self-assessments.
I feel helped with self-assessments, it is essential and useful in order to know my strengths or weaknesses, and I can evaluate my accomplishment. (S15/translated by the researcher) The participants also expressed some concerns in the open-ended answers. S14 expressed her feelings of fear when doing self-assessments and S18 felt confused in understanding their abilities when doing self-assessments.
(I feel) afraid if I cannot pass the standards' criteria, and (I feel) unsure about the results of my works when I self-assess. (S14/translated by the researcher) Terkadang bingung memahami sifat, karakter, maupun kemampuan sendiri.
Sometimes (I feel) confused in understanding my traits, character, and abilities. (S18/translated by the researcher) While sometimes, participants had mixed feelings toward self-assessments. They could have negative feelings on one side and positive feelings on the other side. They could feel apprehensive or disappointed, but they also realized that self-assessments were useful. The evidence can be seen in the responses of S30, S33, and S41.
Saya merasa sedikit gugup karena harus menilai diri saya sendiri dan saya juga belum tau apakah penilaian saya terhadap diri sendiri sudah tepat atau belum. Tapi saya juga senang karna saya bisa mengevaluasi diri saya sendiri jika hasil dari self assessments saya sendiri kurang tepat dengan apa yg saya kira. From the above results, it could be proven that most of the participants felt positive in doing self-assessments. While they might have mixed feelings about self-assessments, it was undeniable that self-assessments were helpful for them to evaluate their capability. These results support the existing evidence from Brantmeier et al. (2012), who found that most students liked self-assessments because it gave them the ability to appraise and increase their academic performance. Similarly, the participants also felt content, helped, and enthusiastic because they realized the use of self-assessments in evaluating and improving their work.

Students' opinions toward self-assessments
This section attempts to reflect the nature of self-assessments based on the students' opinions. There were thirteen questionnaire items used to attain students' opinions toward self-assessments, which contained ten close-ended Likert scale items and three open-ended questions. There are two main points discussed in this section: Students' definition of self-assessments, and the benefits and challenges in self-assessments.
An open-ended question was used in order to investigate how students define self-assessments. All of the participants' answers were similar, in the way they defined self-assessments as an activity of self-evaluation. However, 42.86% of the participants had not explicitly stated their purpose for doing self-assessments, with the rest clearly stated their purposes for doing it (57.14%).
The responses from S17, S22, and S39 could illustrate the point on the lack of purposes in the participants' definition of self-assessments.
Self-assessment can be in a form of assignments given by the teachers or a prediction of the grade that will be obtained in an assignment before the teachers give it. (S22/transalted by the researcher) Penilaian atau memperkirakan target tugas mandiri kita, dan melakukan evaluasi mandiri.
Assessing or predicting our own tasks' target and doing self-evaluations. (S39/ translated by the researcher) It could be seen that the participants did not explicitly state on which aspects they should evaluate, such as evaluating performances, abilities, and other aspects. It could mean that several participants were not aware of the self-assessments' purposes. To evaluate our performances/tasks 7 25.93% 3.
To increase our performances/abilities in the future 6 22.22% 4.
To evaluate our understanding of certain materials 4 14.81% 5.
To know our own progress in learning 1 3.70% Table 2 presented the purposes found in the participants' open-ended answers. The highest occurrences were related to the purposes of self-assessments to evaluate participants' abilities (29.63%), with the other top answers were about self-assessments to evaluate participants' performances/tasks (25.93%) and to increase participants' performances/abilities in the future (22.22%). Participants' responses below could illustrate on self-assessments were used to evaluate participants' abilities (S9), to evaluate participants' performances/tasks (S23), and to increase participants' future performances/abilities (S6 and S34).
In my opinion, self-assessment is an action that we take for assessing or determining how appropriate or correct the task we are doing. (S23/translated by the researcher) Self-asssessment sendiri menurut saya adalah tempat dimana saya dapat melihat kembali seberapa jauh kemampuan saya didalam bidang studi yang sedang saya jalani ini. Dimana saya juga bisa membenahi diri saya untuk menjadi lebih baik.
In my opinion, self-assessments are where I can look back on how far my abilities are in the field of study that I am currently having. I can also improve myself to become better. (S6/translated by the researcher) Self assessment adalah bentuk refleksi diri. Dimana saya merefleksikan apa yang telah terjadi dan membuat peningkatan di kemudian hari.
Self-assessments are self-reflections, where I reflect on what has happened and make improvements in the future. (S34/translated by the researcher) From the above results, it could be inferred that the majority of the participants were already aware of self-assessments' purposes, and they could explicitly state the purposes. In contrast, some participants might still be unaware of it. Without the definition of self-assessments, its importance was not able to be highlighted. The participants might not know why they should carry out the activity of self-assessments, which means they might not know which aspects of themselves they need to improve on. Similar to what had been discussed by Andrade (2019), which concerned on the lack of self-assessments' purposes from the definition of other researchers, especially the lack of self-assessments purpose for increasing performances. She further stated that without the act of improving performance, self-assessments would possibly be a futile activity.
However, looking in a positive light, it could be determined that participants had already known that doing self-assessments meant they needed to carry out self-evaluation. This supported the existing claim that self-assessments needed to include an evaluation of performances (Andrade & Du, 2007;Brown et al., 2015). It could signify that the participants were on the right track, as they know that self-assessments were not merely a task given by their teachers, but it is for their own use in determining their capabilities.
Next, we will discuss the benefits and challenges that the students faced when doing self-assessments. The benefits and challenges were investigated mainly on two aspects, which were in terms of students' performance and self-regulation. improved from my assignment. Table 3 indicated that most of the participants believed in the benefits of self-assessments for their performances. They believed that self-assessments were useful to identify and evaluate their abilities (90.48%). This led to the belief in improved work quality (90.47%) and long-term improvement of their abilities (88.09%). Furthermore, participants believed that they became more confident when doing self-assessments (80.95%). In addition, it was found that 69.04% of the participants understood the criteria given by the teacher as they considered it clear enough. Further evidence was shown in Table 4, which contained the participants' responses to the open-ended question. From 32.81% of the total occurrences, it became apparent that many students viewed self-assessments as useful in determining and evaluating their strengths and weaknesses. The second highest occurrences showed that self-assessments could be used to hone their abilities (20.31%). Below, S1 and S5 mentioned the benefits of self-assessments in determining their potential and honing their performance.
Improving the ability in analyzing and looking for opportunities, improving self-standards in doing things, and improving learning outcomes. (S5/translated by the researcher) From the excerpts, it was known that self-assessments were helpful in identifying the participants' strong and weak points. With the identification, the participants could design strate-gies for improvement of work and ability. This is in line with Shatri and Zabeli (2018), which claimed that self-assessments were used for students to consider their strengths and weaknesses. In return, it could be beneficial for students' future improvements (Rolheiser & Ross, 2013). Conversely, participants found challenges in self-assessments, as illustrated in Table 3 and Table 5. Based on Table 5, the top occurrences were students' confusion in identifying their work appropriately (33.33%) and their subjectivity in self-appraising themselves (18.75%). Another evidence was shown in Table 3, which around 30.96% of the participants revealed that they did not know which aspect of their work that they need to improve on and they needed clear criteria. Challenges that the students had in self-assessments were reflected in the answers from S26, S16, and S1.
The difficulty is, if there are no criteria, it will take a long time because I need to reflect back on each of my processes. If there is a rubric, it might be much easier to remember what I have worked on in doing certain things. (S1/translated by the researcher) From the S26 and S16, it was implied that students' confusion and biased markings were because of participants' inability to self-evaluate themselves accurately. It is also noted in S1's answer that students could evaluate their work more efficiently with a rubric. It was already known that a rubric could provide detailed and explicit criteria (Brown & Harris, 2014). It means that the challenges of self-assessments were mainly on students' accuracies in self-evaluating their work. To know the gravity of the problems, there was a close-ended question regarding the participants' opinions of their accuracies in self-evaluation.

Figure 2. Students' Opinions toward Self-assessments
Figure 2 checked on students' accuracy in self-evaluation. The data showed that 40.48% of the participants stated that the scores they gave in self-assessments for their works were the same as the scores given by their teachers, meaning that they could appropriately self-evaluate themselves. This could signify that the problem of subjectivity did not frequently occur to the majority of the participants. The second highest data showed that 30.95% of the participants gave lower scores to their work than the teachers, which meant that they sometimes looked down on their own work. Interestingly, the challenges in self-assessments sometimes could also have a positive side when they underestimated their work. It can be seen in S14's response where the participant became more motivated to learn after finding a challenge in self-assessments.
The difficulty is within me, I am not confident in my work so I have to check it several times and even I need to ask my friend. Also, if there is something not understandable during self-assessments, I need to look up the materials first to understand more and can appropriately improve my works. (S14/translated by the researcher) Referring to S14's response, it could be found that challenges in self-assessments might trigger her to seek more understanding of the materials. In other words, she underwent the process of deep analysis of her work, which was still beneficial for the participant. Therefore, the feelings of underestimating one's work might bring a positive side to the students in the learning process. This result was in line with Butler (2011), which stated that when the students underestimated their work, sometimes it might encourage them to enhance their performance. While in terms of self-regulation, it was previously found that participants viewed self-assessments as beneficial in increasing their self-regulatory process (as seen in Table 6). When they did self-assessments, they took their work into their own hands without being instructed to (88.09%). It also stimulated them to become more independent (90.47%) and more initiative (90.48%). Furthermore, it was found that participants viewed self-assessments as not burdensome (78.57%). However, it was found that participants considered teachers' expectations more than their own expectations when doing self-assessments (54.76%), which might be a problem as they thought more of what others expected for their work.
As also illustrated in Table 4, participants mentioned that self-assessments were beneficial for their self-regulation. Around 9.4% of the occurrences were mentioned that self-assessments helped them to become motivated in handling their work, and 6.3% of the occurrences were related to how students could finally set their strategies of improvement independently using self-assessments. S20's response could illustrate the point on self-assessments motivated them to learn, and S31 could illustrate the point on self-assessments were used to set strategies for improvements.
I can identify my own strengths and weaknesses, so that I can determine what strategies I can use to make me learn materials effectively. (S31/translated by the researcher) From the above discussion, it could be determined that self-assessments increased participants' self-regulation. Self-assessments fostered feelings of independence within the participants. Their motivation emerged because of self-assessments, and they were finally able to set their own strategies to improve. The results fit with the theory that self-assessments supported students to take their learning independently, including the process of reflecting and strategic planning (Punhagui & Souza, 2013).
On the other hand, referring to Table 6, participants agreed that they think more of what the teachers expected rather than their own expectations for their work (54.76%). This was the challenge that participants found in doing self-assessments, particularly related to their self-regulation. Further evidence was found in the responses of S20, S29, and S6.
The difficulty of self-assessments is that, if there are no comments from outsiders at the beginning, it is very difficult to carry out self-assessments. (S20/translated by the researcher) Kesulitannya jika saya tidak mendapatkan feedback kesalahan saya diri sendiri dari dosen maupun orang lain.
Self-assessments are difficult things to do if I do not get feedback about my own mistakes from teachers or other people. (S29/translated by the researcher) Kekurangan self-assessment sendiri yaitu saya harus mengikut standar yang ada dikarenakan oleh itu kebebasan serta ke kreatifitasan juga akan terbatas.
The challenge of self-assessments itself is that I have to follow the existing standards. Because of that, my freedom and creativity to create my works will also be restricted. (S6/translated by the researcher) From the excerpts, it was revealed that the participants strongly relied on extrinsic help in order to create their works. The participants needed others' comments or feedback in order to do self-assessments, such as feedback from their teachers and peers. Furthermore, S6 expressed that she needed to restrain her freedom and creativity in her work. These statements implied that students followed more the expectations given by others than what they actually wanted to have in their work. This evidence supported the existing study of Andrade and Du (2007), which the students stated that they relied on thinking of their teachers' expectation instead of what they expected for their own work. It could be a problem for students' self-regulation as they needed extrinsic triggers in order to revise their work.

Students' behaviors when doing self-assessments
This section is intended to find out on which levels undergraduate students currently are at based on the levels of self-assessments by Brown and Harris (2014). Previously, it has been discussed that there are three levels of self-assessments, which are basic, intermediate, and advanced levels. To explore students' behaviors, there were three questions provided, which were two close-ended questions, and an open-ended question. The questions would find out how frequently students would do self-assessments, and what kind of self-assessments were needed and done according to the students. Figure 3 showed the frequency of the participants doing self-assessments when they have an assignment. The top answers were that students did self-assessments once to twice when they got the assignment (47.62%), and the second top answers were thrice to five times after getting the assignment (45.24%). Therefore, it could be concluded that students repeatedly did self-evaluation when they were given an assignment from their teacher.

Figure 4. Students' Behaviors When Doing Self-assessments
Figure 4 was used to investigate the types of self-assessments that participants' needed. The most chosen answers were students' needs to match their assignment with the criteria (69.05%) and the teachers' models (61.90%). Criteria and teachers' models had one thing in common, which was a clearer and explicit standard that is expected by the teachers. This implied that students' preference for clearer standards in self-evaluation on an assignment. Predicting the work result 3 5.66% Table 7 included students' answers in an open-ended question. The table was presented in order to give more clearance of students' preferred methods in self-assessments. Based on the total occurrences, most of the time, the participants did self-assessments by matching their work with criteria given by their teachers (30.19%). The self-assessments that they used were checklists and rubrics, as could be seen in the responses of S33, S12, and S11.
After getting an assignment from the teachers, I usually do self-assessments. I usually predict my grade by reviewing the material and comparing my work with the assessment criteria given by my teacher. (S33/translated by the researcher) Dengan mencocokkan jawab pada rubric yang disediakan oleh dosen I do self-assessments by matching the answers using a rubric provided by the teacher. (S12/translated by the researcher) Kalau di kelas, di beri pertanyaan lalu kita menceklis pertanyaan yang ada untuk penilaian self assessment.
If (I am) in the class, we are given questions and then we checklist the questions for self-assessments. (S11/translated by the researcher) Occasionally, students might not do self-assessments by matching the given criteria. However, they could still do self-assessments through self-reflections on their work results (30.19%). Several participants mentioned they do self-reflections by writing journals and contemplating their work results. S2, S18, and S5 showed the activities they had done for self-assessments. Another essential point to note, participants could carry out a series of activities in self-assessments. They might match their works with the examples and criteria, and they might also combine other activities of self-assessments. Below, S1 and S29 shared their usual methods for self-assessments.
Usually I do self-assessments by making comparisons of my works with my friends' work results, my previous works' results, and with a rubric from the teacher. (S1/ translated by the researcher) Biasanya saya melihat kesalahan yang saya lakukan pada saat mengerjakan sesuatu kemudian mencoba mencatatnya. Dan juga saya membandingkan antara pekerjaan sebelum dan sesudah saya sesuai kriteria yang ada.
Usually, I analyze the mistakes that I have made while working on things and then take notes of them. Moreover, I compare my works based on the existing criteria before revising and compare them again after revising. (S29/translated by the researcher) Saya menggunakan perkiraan diri untuk menilai pekerjaan saya sebelum dikumpulkan dan mengevaluasi diri saat pekerjaan saya sudah diberi nilai.
I self-estimate my works' grades before it is submitted, and I also do self-evaluation after my work has been graded. (S26/translated by the researcher) From the above discussion, it was concluded that the participants were no longer in the basic stage of self-assessments. In other words, the participants were already in the intermediate level and the advanced level of self-assessments. It could be seen in their answers of what should be done in self-assessments (Figure 4), which were matching their works with the criteria and teachers' examples. However, results found from the open-ended question (Table 5) showed that students' frequent activities in self-assessments were matching their work with the criteria and doing self-reflection of their performance. Therefore, it is highly probable that the participants were in the advanced level. The mentioned activities of matching with the criteria and doing self-reflection fell under the category of advanced level, in which the students deeply analyzed their performance. This is in line with Brown and Harris (2014) that the end goal of self-assessments should be about students' ability to critically evaluate their work based on the explicit criteria.

CONCLUSION
The problem with the rare use of self-assessments persists in the education context. One of the main reasons can be because of teachers' assumptions of students' limited comprehension of self-assessments. Without a doubt, self-assessments need to be understood from undergraduate students' perceptions in order to know how they view self-assessments. The result of the study showed students' positive responses toward self-assessments. First, it was found that most of the students enjoyed self-assessments, and there were no students who hated self-assessments. Even several negative comments were mixed with positive feelings, such as students' disappointment in getting bad grades but still believing that self-assessments were vital for them. Second, it was found that students were aware of self-assessments' use. They understood that the activity of self-assessments was for evaluating and increasing their academic performance. They can also analyze the benefits of self-assessments in other aspects, such as increasing their confidence, motivation, and independence. Although there was a problem with the subjectivity in self-assessments, the problem did not commonly occur in the students' learning process. It could even prompt the students to learn more from their materials. Lastly, it was found that students were already either on the intermediate or the advanced level. Most of the students did self-assessments by matching their works with criteria given by their teachers, and they also did a reflection on their strengths and weaknesses. Furthermore, students sometimes did a series of actions in self-assessments, such as matching work with their previous works and with the criteria.
The study concludes that the implementation of self-assessments' is viewed favorably by undergraduate students. The results show students' capability of understanding and carrying out self-assessments, which challenge teachers' previous assumptions. It is recommended that teachers start to treat their students as independent learners and give necessary support (such as feedback and clear instructions) for the students in their self-evaluation process. However, the results were obtained from students who previously had tried several kinds of self-assessments activities in offline and online classroom situations. There might be different responses from students who do self-assessments only in online classroom situations as there are several problems faced, such as harder comprehension of works criteria because of limited interactions with their teachers, internet connection problems, et cetera. Furthermore, the researcher believes that this study has not yet explored on what students view as precise and objective criteria in self-assessments. This is an important point to be researched to improve the reliability of self-assessments in increasing academic performance, which in hope results in criteria that the students can use as a measurement of their work. This can potentially solve the problem of subjectivity in which teachers and students can have the same perceptions of what should be assessed and how to assess the performance with measurable standards.