EFL Secondary School Teachers’ Conceptions of Online Assessment in Emergency Remote Teaching: A Phenomenographic Study

The investigation of teachers’ conceptions of online assessment during a global pandemic has received relatively scarce attention in the current literature. Situated in an emergency remote teaching, this phenomenographic study aims at identifying and describing EFL secondary school teachers’ qualitative different ways of understanding online assessment. A cohort of fifteen EFL teachers from different Indonesian secondary schools were recruited purposively using a set of criteria. They were invited to involve in online semi-structured interviews to explore their online assessment conceptions. Then, the interview data were analyzed qualitatively in an iterative process to discern categories of description and an outcome space. The findings point out five categories of teachers’ conceptions. The participants view online assessment in emergency remote teaching (ERT) as a means of (1) measuring knowledge, (2) checking learning targets, (3) enhancing online interaction, (4) facilitating authentic tasks, and (5) reflecting the teaching and learning process. Furthermore, an analysis of relationships among those categories is reported hierarchically, ranging from accountability to enhancement assessment purposes. Some implications promote our understanding to consider several mediating factors affecting teachers’ conceptions. Although emergency remote teaching situation is provisional, the findings bring out the possibility of implementing assessment for learning, as an alternative to assessment of learning, in the context of online assessment after the pandemic.


INTRODUCTION
As an impact of the Covid-19 outbreak at the beginning of 2020, people worldwide had to adjust many aspects of their lives, from restricting physical interactions, working from home, and even cancelling face-to-face teaching and learning activities. Responding to that global health crisis, the Indonesian Ministry of Education and Culture urgently instructed all education institutions to close the schools and universities as a mitigation attempt to minimize the virus contagion in the education environment (Indonesian Ministry of Education, Culture, Research, and Technology, 2020). Due to the closures of all schools during the pandemic time, instruction delivery was shifted from face-to-face mode to distance online mode abruptly. This sudden transformation compelled teachers to integrate online learning pedagogy into their instructions without adequate preparation (Cahyadi, Hendryadi, Widyastuti, & Suryani, 2021;Erdem-Aydin, 2021).
The term Emergency Remote Teaching (henceforth, ERT) is gradually used in the current literature to address a temporary solution for possible instructions implemented in unexpected conditions, such as in pandemic crises time or warfare (Hodges et al., 2020). Unlike online learning or distance learning formed in careful, systematic, and well-planned instructional design and ecosystems, ERT is a sudden and interim online instruction that learners can feasibly access during the pandemic time (Hodges et al., 2020;Rahiem, 2020). Cahyadi, Hendryadi, Widyastuti, Mufidah, et al. (2021) propose five principles of ERT, i.e. it should be feasible, accessible, showing empathy, flexible, and affordable. Therefore, the ultimate objective of ERT is not to re-design a fixed instruction. Instead, it is a dynamic instructional design, as a response to crisis circumstances, that is possibly adjusted and contextualized based on the socio-cultural conditions, online readiness, and many other factors (Cahyadi, Hendryadi, Widyastuti, & Suryani, 2021;Kusuma, 2022;Luthfiyyah et al., 2021).
Shifting from a physical learning environment to ERT in a rapid time is demanding for many teachers, especially the ones who have no experience in conducting a distance learning (Hodges et al., 2020). Numerous studies have highlighted how teachers from different countries faced many challenges during ERT (Cahyadi, Hendryadi, Widyastuti, Mufidah, et al., 2021;Cahyadi, Hendryadi, Widyastuti, & Suryani, 2021;Erdem-Aydin, 2021;Rahiem, 2020). However, there is a paucity of information on discussion of online assessment practice during crisis time, especially how teachers understand and experience online assessment in ERT. We argue that exploring further discussions in online assessment is worth doing since it contributes significant inputs to the practices of assessment in the online environment after the pandemic.
Online assessment shares similar purposes with face-to-face assessment. It has dual functions, to measure students' achievement and to improve students' learning progress (Rovai, 2000). However, the implementation of assessment in an online environment tends to be more challenging since it is limited to physical distance interactions to directly monitor and provide timely feedback on students' works (Cheng et al., 2013;Hung & Chou, 2015). A number of studies have discovered some typical issues in online assessment during pandemic Covid-19. For instance, Meccawy et al. (2021) reported the result of an online survey distributed to 547 undergraduate students. It reveals that the biggest problem of online assessment during pandemic situations is students' low academic integrity. Cheating and plagiarism practices have become more prevalent problems due to a lack of security in assessment. Besides, Luthfiyyah et al. (2021) invited five experienced EFL teachers in Indonesia to reflect on their online assessment experiences in ERT. They found that students' workload and teachers' online readiness become other concerns in online assessment during ERT. Those unfavorable circumstances can decrease the primary function of online assessment as the medium to assist students in achieving their learning goals (Guangul et al., 2020;Meccawy et al., 2021).
The identified constraints of online assessment in the distance and remote situations led to an agreement to urge teachers to adjust and modify their assessment design and practices that meet the learners' needs in an online environment, especially in ERT (Cheng et al., 2013;Rovai, 2000;Zhang et al., 2021). Some scholars have explicitly suggested that formative assessment or assessment for learning is a more robust strategy for online assessment than a test or one-shot summative examination (Byrne et al., 2021;Cheng et al., 2013;Şenel & Şenel, 2021). Online formative assessment is not a fixed entity. However, it should be regarded as an integral part of learning activities that can assist learners in achieving the learning targets with the assistance of constructive feedback (Şenel & Şenel, 2021). Moreover, it is believed that online formative assessment or assessment for learning activities, such as online discussion forums and online collaborative writing, can promote social, teachers, and cognitive presences in online instruction and increase students' engagement in remote interactions (Byrne et al., 2021).
Although much evidence has distinctively pinpointed the advantages of implementing formative assessment in an online environment, the teacher, as an active doer in the classroom, plays a vital role in deciding the strategies for designing and practicing online assessment in their contexts (Mimirinis, 2019). Furthermore, there is a remarkable consensus among scholars pointing out a relationship between the teachers' conceptions and their instructional decisions, including in assessment practice (Brown & Gao, 2015;Gebril, 2017;Mimirinis, 2019). For instance, the way teachers assess their students in the online environments is influenced by their view and understanding of the concepts of online assessment and the concept of emergency remote situations. As such, we conclude that teachers' conceptions of online assessment in ERT are regarded as the prominent factor influencing their assessment practices in the ERT context.
The discussion on assessment conceptions has received enormous attention in the current literature (Brown & Gao, 2015;Dayal & Lingam, 2015;Gebril, 2017;Mimirinis, 2019;Postareff et al., 2012). Dayal and Lingam (2015) and Postareff et al. (2012) contend that teachers possibly have multiple and complex conceptions of assessment influenced by the socio-cultural factors they work with and individual R. Luthfiyyah, B. Y. Cahyono, F. M. Ivone & N. Suryati, EFL secondary school teachers' conceptions of online assessment in emergency remote teaching: A phenomenographic study | 1044 factors they have. However, most of the previous studies prefer to report conceptions of assessment in face-to-face environments in higher-education levels rather than in online environments in secondary level settings. A study with a closer setting was conducted by Mimirinis (2019). He examined e-assessment conceptions of university academics from different disciplinary backgrounds using a phenomenographic study. The result of the study suggests elaborating further on a similar topic in fully online instructions. Responding to the previous studies' discrepancies, this study intends to explore EFL secondary school teachers' conceptions of online assessment in emergency remote settings. The research questions are formulated as follows: 1. What are EFL secondary school teachers' different ways of understanding and experiencing online assessment in ERT? 2. How is the hierarchical relationship among those categories of description?

Online Assessment in Pandemic Covid-19 Situation
It has been widely known that assessment is a pivotal part of the teaching and learning system that aims to collect and interpret information about students' performance and help them improve their learning. The principles of traditional assessment do not change in online assessment (Benson, 2003;Rovai, 2000). However, assessment challenges in online distance environments become more critical due to the limited interactions to monitor the students directly and provide effective feedback (Rovai, 2000). Moreover, when the instructional design is not prepared well due to the rapid and unprecedented transitions, assessing students in online remote situations has endangered significant challenges in the practices, such as the issues of cheating and plagiarism (Guangul et al., 2020;Meccawy et al., 2021). Ninković et al. (2021) contend that teachers must possess sufficient assessment literacy and technological knowledge to design and conduct an effective online assessment. Assessment literacy refers to knowledge and skills to construct, assess, and interpret the assessment data (Coombe et al., 2020;Inbar-Lourie, 2017;Stiggins, 1991). Meanwhile, technological knowledge in this study refers to the competency of selecting and utilizing technology for assessment purposes. Responding to those points, we argue that teachers should update and adjust their pedagogical and technological literacies based on the various contexts.
The nature of the educational assessment system in Indonesia generally focuses on high-stake examination culture aiming to improve the national education standard (Law & Miura, 2015). However, as a response to the crises, the Indonesian Ministry of Education and Culture mandated teachers to specifically implement on-going assessment activities to diagnose students' learning problems and assist them in improving their learning (Ministry of Education and Culture Act No. 719/P/2020). The regulation of assessment in emergency remote situations aligns with the literature discussing assessment strategies in the online environment. Many studies show a prevalence of formative assessment practices in an online environment (Gikandi et al., 2011;Rahiem, 2020;Şenel & Şenel, 2021;Zou et al., 2021). For instance, Şenel and Şenel (2021) reported that practicing formative assessment which provides helpful feedback can be the potential assessment strategy during ERT. In addition, Gikandi et al. (2011) argue that providing authentic formative assessment in an online or blended context can reduce the issues of academic integrity (e.g., cheating, dishonesty, and plagiarism). In this case, precise regulation and guidance from the institution are prominent to support the assessment strategies in a crisis time. In the same vein, it is strongly suggested that teachers be creative in designing and improvising the assessment strategies that fit online ERT settings (Rovai, 2000;Zhang et al., 2021).
Furthermore, online assessment in ERT requires a stable and sufficient infrastructure and technology resources, such as software, hardware, digital devices, and fast and affordable internet connection. Although the advancements of technology are massively used, the fact reveals that many places in Indonesia still have a lowspeed internet network (Cahyadi, Hendryadi, Widyastuti, Mufidah, et al., 2021). The insufficient infrastructure can hinder the process of online learning during ERT, especially in conducting an online assessment. Therefore, teachers should consider the conditions of each student by ensuring that they have good access to join the online assessment. Teachers need to provide a flexible and alternative assessment for students with special unsupported conditions (Benson & Brack, 2010;Burgos et al., 2021). Providing various types of online assessment, such as online quizzes, e-portfolio, discussion forums, video presentation, simulations/gaming, and setting the assessment into the asynchronous mode, can be the alternative ways to facilitate students from different conditions (Pu & Xu, 2021;Rahiem, 2020) In short, those endeavors can lead to a fair online assessment for all students and not disserve students with the assessment procedure in emergency remote situations.
Given those challenges of practicing online assessment in ERT, teachers from different schools might have various experiences in assessing their students. We argue that some influential factors, such as the institution assessment policies, infrastructures, digital sources, teachers' knowledge on assessment and technology, can contribute to the way teachers understand and view the concept of assessment in the remote online environment. Consequently, this study intends to examine EFL teachers' conceptions of online assessment in ERT in secondary school contexts.

Teachers' Conceptions of Online Assessment in ERT
Teacher conceptions generally refer to the way teachers view, believe, interpret, and interact with the teaching environment (Brown et al., 2009;Marton, 1981). In the assessment context, teachers' conceptions of assessment are defined as how teachers understand and experience assessment. Numerous studies have been done to examine teachers' conceptions of assessment in different countries (Azis, 2015;Brown & Gao, 2015;Dayal & Lingam, 2015;Gebril, 2017;Mimirinis, 2019;Postareff et al., 2012). The studies suggest that teachers' conceptions of assessment are gradually formed through their experience toward a phenomenon (Gebril, 2017;Postareff et al., 2012). Further, it becomes a mechanism of how they react or respond to the phenomenon (Brown & Gao, 2015). In the same vein, teachers' experiences in conducting online assessment in ERT shape how they understand and view that phenomenon, and teachers' conceptions influence their assessment practice in the online remote environment.
While some previous studies investigate teachers' conceptions of assessment in traditional face-to-face context using longitudinal study or general case-study, a recent study from Mimirinis (2019) reports university academics' conceptions of assessment R. Luthfiyyah, B. Y. Cahyono, F. M. Ivone & N. Suryati, EFL secondary school teachers' conceptions of online assessment in emergency remote teaching: A phenomenographic study | 1046 in an online environment using a phenomenographic study. Instead of examining eassessment from the first-order perspective, he intends to discover a given phenomenon from second-order perspectives. His study seeks the twenty-one academics' conceptions of integrating technology to support summative and formative assessment purposes in online settings. The findings identify four categories of description of the ways participants view e-assessment and the relationships among those categories. The academics discern e-assessment as a means of the efficient way to manage and streamline the assessment process, facilitate dialogue and engagement, enhance student's learning, and develop digital identity and community. However, it should be taken into account that the study was conducted in a conducive ecosystem where the research settings had implemented online instructions in the advanced stage.
Referring to the previous academic work, this study aims to extend the investigation on teachers' conceptions of the online assessment. Nevertheless, some discrepancies are highlighted in this paper. First, the contexts where both studies were conducted are distinct. While the earlier study was conducted in well-prepared online instructions, the present study portrays the teachers' experience and understanding of online assessment in a crisis time. The transition is abrupt and not intentionally planned. Thus, the situation causes many significant challenges for teachers to conduct the online assessment. Second, unlike the previous studies, which are predominantly situated in higher education contexts, we prefer to explore the online assessment conceptions from EFL secondary school teachers due to the sparsity of discussion.

Research Design
The study employs a phenomenographic approach. It is a qualitative approach which aims at discerning people's different ways of understanding a certain phenomenon (Åkerlind, 2012;Hajar, 2021;Marton, 1981). Adopting a non-dualistic ontology stance, the focus of phenomenography is neither on the phenomenon itself nor the individual's experiences on a particular phenomenon. Instead, it identifies and interprets the relational meaning between individuals (within a group and as a group of individuals) and a particular phenomenon presented in a limited number of qualitatively different categories. In this study, we generally intend to seek and describe EFL secondary school teachers' qualitatively different ways of understanding online assessment during Emergency Remote Teaching (ERT).
Another basic tenet of phenomenography also acknowledges second-order perspective rather than first-order perspective (Hajar, 2021;Marton, 1981;Marton & Pong, 2005). Marton (1981) explains that instead of investigating a phenomenon from a researcher's point of view (first-order perspective), a phenomenographer explores individuals' conceptions about a particular phenomenon that they have experienced (second-order perspective). The aspects of a phenomenon are described as it is understood and experienced by the individuals. In addition, highlighting the epistemological assumption of this approach, it is believed that each individual has various conceptions and experiences about a particular phenomenon, and they can change over time. However, those multiple understandings can be described, communicated, and understood by others in a limited number of qualitatively different ways (Sjostrom & Dahlgren, 2002). Hence, the outputs of the phenomenographic study are the variations of conceptions (categories of description) and the hierarchical relationship among those categories of description (outcomes space) (Åkerlind, 2012). In this study's context, we explored empirical evidence of how teachers understand, experience, and reflect online assessment during ERT in different ways. The phenomenographic data can portray teachers' understanding, represent complexity, and reflect different ways of experiencing online assessment in a disruptive setting. The data can further provide information for stakeholders and policymakers to amend the quality of online assessment and devise more meaningful and feasible regulations.

Context and Participants
The study was conducted in the context of an emergency remote teaching in Indonesia. Due to the outbreak of Covid-19, teachers were urged to shift their instruction swiftly into a fully online learning mode without any sufficient preparations. They strived to adjust their instruction, particularly in online assessment, to meet students' needs and conditions. On the other hand, the national assessment system in Indonesia adheres to test dominated culture, which identically assesses the students through a test. Consequently, the online assessment practices in ERT met several constraints, such as limited pedagogical and technological knowledge, exam culture, academic integrity, and lack of sufficient infrastructure. Furthermore, diverse backgrounds of participants, such as context, experience, age, and gender, are taken into consideration to maximize the variations of individuals' experiences of online assessment in ERT.
However, to obtain potential sources, the participants should be purposefully selected based on a set of specific criteria. They are experienced teachers who have more than five years of teaching experience. They actively join the assessment and technology-enhanced language training. The participants also must have three months of experience conducting fully online learning amidst pandemic time. At the first step, we distributed an online form to a group of EFL teachers who joined a national online webinar held by an association of technology enthusiasts. Then, fifteen EFL secondary school teachers who fulfilled the criteria were recruited as the participants in this study (the demographic data is available in Table 1). The invitations to participate in semistructured interview were sent to their email. Then, all participants signed an electronic informed consent form confirming their agreement to participate.

Data Collection
One of the most preferred techniques of gathering data in a phenomenographic study is interviews, for the technique can provide rich and deep data from the participants' perspective (Åkerlind, 2012). A semi-structured interview was conducted online using Zoom Meeting from May 2020 to November 2020 to investigate EFL secondary school teachers' understanding and experience of online assessment during ERT. The schedules of interview sessions were set based on the participants' convenience. Furthermore, a predetermined question was prepared to probe the teachers' experience of online assessment during ERT. All the participants received the same questions to convey the purpose and how they conducted online assessment during ERT, yet the interviews ran flexibly. Some followed-up questions were addressed to the participants to clarify, elaborate, or confirm their insufficient responses. During the interview sessions, it should be taken into account that the researchers should bracket their insights, knowledge, and experience to gain the participants' experience of online assessment during ERT (second-order perspective). Each online interview session lasted 40-60 minutes. The interview data were video recorded and transcribed verbatim.

Data Analysis
The main objectives of data analysis in phenomenography are to identify and depict a set of categories description of online assessment during ERT and the hierarchical relationship of those categories of description (outcome space). The analysis process started when all the interviews were conducted and transcribed. Following the data analysis procedures from Sjostrom and Dahlgren (2002), we analyzed the data through seven stages. The first stage is familiarization. The researchers read the interview transcripts iteratively to get a good sense and familiarity with the details of the data. After getting familiar with the data, the next stage is a compilation, in which the most significant and essential data are identified. The third stage is the reduction of the condensation stage. The researchers selected the most relevant data associated with online assessment during ERT and removed the most redundant data at this stage. It was followed up by categorizing the similar answers into the same group. Then, each group of categories was compared to establish borders among them. In the sixth stage, the categories of description were labelled or named to highlight the essence of each category. Finally, the last stage is arranging the outcome space, in which the categories of description were arranged systematically to see the relationship among them.

RESULTS
To answer the research questions, we first present five categories of descriptions discovered from EFL secondary school teachers' experiences in conducting online assessments in ERT. Second, to obtain a more profound and meaningful insight about the categories, we then reveal the hierarchical relationship among those categories into three conceptions.

EFL Secondary School Teachers Different Ways of Understanding Online Assessment during Emergency Remote Teaching
From the data analysis process, we discovered five different ways of understanding and experiencing from the transcripts. The EFL secondary school teachers view online assessment during ERT as a means of (a) measuring students' competencies; (b) checking the learning targets; (c) enhancing online interaction; (d) facilitating authenticity in assessment; (e) fostering teaching reflection. Those categories of description are elaborated further in the following subsections.

Measuring students' competencies (Category A)
The first category represents EFL secondary school teachers who perceived online assessment during ERT as an activity for measuring students' competencies. The teachers admitted that the objective of an online assessment is similar to that of a traditional assessment, that is, to measure students' competencies (P02-SHS, P03-JHS, P10-JHS, P13-VHS, and P14-JHS). They have a responsibility to regularly record and report the students' achievement for administrative tasks, as reported by one of the participants: (1) "…. I think I have the same purpose in online assessment. I need to assess and score my students' competencies. I know that the assessment regulations in emergency time instructed us (teachers) to focus on the learning process. However, my institution (still) asked for quantitative reports to fulfil the administrative documents at the end of the semester". P02-SHS On the other hand, P04-JHS and P15-SHS also contended that giving an online test was the easiest way to measure students' competencies. Some online quiz apps assist the teachers to measure students' understanding and to obtain instant results. However, they found that the results of online quizzes are less valid and reliable due to academic dishonesty practices. Unfortunately, some of the participants could not find the strategies to prevent dishonesty issues (P03-JHS, P10-JHS, P04-JHS and P15-SHS).
(2) "I prefer to give online quizzes or tests to measure my students' ability using assessment tools, such as Google Form or Kahoot. The tools help me get the results of my students' work quickly. …. Well, I know it is possible for my students to search for the answers (of the tests) from the Google search engine or ask for the answers from their friends or parents. However, I don't know how to solve the problems". P03-JHS

Checking the learning targets (Category B)
We also found that the EFL secondary school teachers view online assessment during ERT as a practical activity to check the learning targets. The teachers frequently provided regular assessments, such as online quizzes, tests, and project tasks to check students' learning achievement (P05-SHS, P06-VHS, P10-JHS). They emphasized that they have a responsibility to ensure that the students have achieved the learning outcome based on their levels. Furthermore, the assessment in a remote online environment focuses on cognitive skills rather than affective and psychomotor skills because of the limited assessment literacy. The following are excerpts from the participants.
(3) "I regularly give my students online quizzes, question and answer, or a simple task simply to check their understanding of the materials and ensure the learning objectives have already been achieved". P05-SHS (4) "My responsibility is to ensure that the learning targets have been achieved. However, it is hard for me to assess my students' affective and psychomotor skills in an online setting; I don't know how to assess those skills. Thus, during the pandemic time, my assessment practices focus only on students' cognitive skills". P10-JHS

Enhancing online interactions (Category C)
The next category reflects the participants' understanding of online assessment during ERT to enhance online interaction among students and teachers (P07-JHS, P09-JHS, and P11-VHS). Besides, online assessment activities could engage students in online learning (P01-JHS and P05-SHS). P07-JHS contended that students could share their ideas through ongoing assessment activities, such as discussion forums or quick surveys. She encouraged students to respond to a given topic posted in WhatsApp or Padlet. It is one type of interactive online board where every student can post their ideas or answers and give comment or vote to their friends' ideas.
(5) "I posted a reading text in WhatsApp group or Padlet and gave my students time to comprehend the text. Then, I invited them to comment on what they understood from the text and posted it to Padlet. Actually, I prefer using a Padlet to WA group because it eases me to track my students' answers and to give feedback. Besides, both teachers and students could comment or vote on their friends' work directly on the interactive board". P07-JHS Furthermore, by providing audio or video feedback to the students, P02-SHS found that she could guide the students to improve their work and engage more in the interaction. It is noted that the teachers integrated the assessment activities with the teaching and learning process, rather than giving homework or a test, to reduce students' workload and maintain students' well-being (P05-SHS, P09-JHS, and P12-JHS).
(6) "I frequently gave audio feedback by recording my voice or take a short video in giving difficult explanations. Interestingly, I found that it can substitute the face-to-face interactions between teacher and students to virtual interactions during online learning. Besides, audio and video feedback can save my time in giving individual feedback and help my students understand my feedback better". P02-SHS (7) "During ERT, the students were overwhelmed with the homework from almost every subject. Thus, I preferred to give online quizzes during the teaching and learning process. I tried to minimize giving homework to my students". P12-JHS

Facilitating authenticity in assessment (Category D)
Some participants found that online assessment during ERT could facilitate authentic tasks (P02-SHS, P04-JHS and P11-VHS). They admitted that the tasks related to students' real-world context could reduce the students' nerves in doing online tasks and stimulate their creativity and higher-order thinking skills. Students reflected on their daily activities and solved the problems associated with their current situation, particularly the issues of health and the environment. For instance, P04-JHS reported that students demonstrated how to cook some viral foods and drinks during the pandemic (Garlic Cheese Bread, Nori Roll Rice, Dalgona Coffee, etc.) or described daily activities when they had to stay at home. Given the authentic and familiar topics, students worked on their projects excitedly. On the other hand, giving a real-world task to students could minimize the possibility of cheating and plagiarism practices (P02-SHS and P11-VHS). The students worked on their tasks based on the context they had. So, they produced more meaningful and specific reports. Here are some excerpts from the participants.
(8) "I realized that we faced a hard time teaching online during ERT. Therefore, I attempted to simplify the task for my students and reduce their anxiety, yet it was meaningful and fun for them. I gave some projects related to their environment. For instance, they had to record their voice, reporting their daily activities on weekends". P02-SHS (9) "I gave a simple project to my students related to their favorite food. They should write a recipe for their favorite food and make a short video using VivaVideo for a cooking demo. My students did their project enthusiastically. They demonstrated the procedures to make Dalgona Coffee, Garlic Cheese Bread, or Nori Roll Rice, etc." P04-JHS

Fostering teaching reflection (Category E)
The last category represents EFL secondary teachers' understanding of online assessment during ERT as a teaching reflection. All participants admitted that their teaching experience in the first semester of ERT was very challenging. They had uncertain situations and insufficient knowledge and support to move their instructions to an online environment. However, P01-JHS, P05-SHS, P11-VHS, and P15-SHS revealed that their sensitivity and awareness to reflect and evaluate their teaching quality are more gradually increased during ERT than their experience in face-to-face (F2F) learning. They could reflect their teaching from the student's assessment results and students' learning reflections. For instance, when P01-JHS found that the results of students' works were far from their expectations, they would find the solutions to adjust their assessment strategies.
(10) "Teaching in a fully online mode is new for us. I learned a lot from my experience. For instance, I used to utilize the Quizizz application to conduct online quizzes. However, I stopped using that app since I knew that my students had a trick to finding the answer keys online easily. So, I tried to find another app and adjust the types of my questions". P01-JHS R. Luthfiyyah, B. Y. Cahyono, F. M. Ivone & N. Suryati, EFL secondary school teachers' conceptions of online assessment in emergency remote teaching: A phenomenographic study | 1052 Moreover, P02-SHS, P07-JHS, and P11-VHS directly invited students to reflect on their learning by distributing online exit tickets. The data obtained from students' reflections could be meaningful and valuable information for teachers to rearrange and improve their online instructions.
(11) "Once I gave my students an online task, I needed to ensure that all the instructions were clear and easy to understand. Otherwise, my students would misunderstand and get lost in doing the task. It happened in my class twice". P02-SHS (12) "Teaching during ERT was challenging for all of us. I often questioned myself. Did I teach well? Did my students learn something? Why did my students tend to be passive in the discussion? etc. Therefore, I frequently asked my students to fill out an exit ticket, in which they can write what they like and dislike from the learning process". P11-VHS

Hierarchical Relationship among the Categories of Description
The second outcome of the phenomenographic study represents the categories of descriptions in an outcome space. The outcome space is presented in a diagrammatic representation to show a finite set of hierarchically structured categories of description of a given phenomenon and show the logical and inclusive relationships among categories (Åkerlind, 2012). The outcome space of this study can be seen in Table 2 and is explained rigorously in this subsection. We identified five conceptions of online assessment in ERT during the pandemic time in Indonesia from 15 EFL secondary school teachers involved in this study. Category A and Category B are considered less subtle than other categories since they simply focus on adopting accountability to measure and check the students' cognitive skills. Meanwhile, the following two categories (C and D) represent higher complexity than the previous ones. The focus is not merely on measuring students' cognitive skills but emphasizes the ongoing assessment activities that facilitate instruction, social, and content presence in online learning interaction. Through online assessment, students are expected to be more engaged and motivated in online learning. Finally, Category E, online assessment to facilitate teaching reflection, has the most advanced conceptions compared to others. This category generally includes all elements from the previous categories. In this case, teachers use the information elicited from assessment activities to reflect the whole online assessment process and devise a better online instruction.
Furthermore, the logical nexus among the categories elaborates further using a framework of referential aspect and structural aspect of online assessment in ERT. The former aspect refers to the meaning of experience ("what" of online assessment in ERT), and the latter describes the structure of that experience ("how" of online assessment in ERT). Those two aspects are linked to contextualize the conceptions (Marton & Pong, 2005). This study discerns three broader structural aspects that are associated with referential aspects, i.e. online assessments focusing on accountability, the learning process, and enhancement. Accountability deals with measuring and checking students' competencies. Categories A and B represent teachers' online assessment that simply measures and checks students' knowledge, whereas learning improvement is given little attention. On the other hand, Categories C and D emphasize that online assessments facilitate online learning. It portrays participants' conceptions that online assessments are a part of the learning process that can foster students' engagement in online learning through meaningful interactions and authentic tasks. At last, Category E highlights enhancement in the online learning process. It does not merely focus on students' scoring and online learning process, but it requires high awareness and deep thinking from the teachers to trace back their online instructions and devise them better.

DISCUSSION
The study generally attempts to extend a previous report by Mimirinis (2019) to identify and describe teachers' conceptions of the online assessment. However, this study provides a more specific context called emergency remote teaching (ERT). Instruction delivery was imperatively shifted from face-to-face to a fully online learning context without sufficient preparation and poor facilities due to a global crisis (Hodges et al., 2020). Having that particular setting, we found that all participants agreed that assessment plays a pivotal role in traditional and online environments. However, their conceptions of online assessment are multiple and range in different levels of conceptions, from simple to more sophisticated ways of understanding. We identified five qualitatively different ways of understanding online assessment in ERT perceived by EFL secondary school teachers in Indonesia. Those five categories are then presented in an outcome space which is mapped into three dimensions: focus on accountability, learning process, and enhancement. The findings broadly tune with the previous studies related to assessment conceptions (Mimirinis, 2019;Postareff et al., 2012).
It is believed that teachers' conceptions are generally affected by their sociocultural factors (Brown et al., 2011;Gebril, 2017). For instance, teachers who teach in a context with an examination-oriented culture tend to have conceptions of accountability (Brown & Gao, 2015). In the same vein, the finding also reveals that most of the participants in this study perceived assessment as a means of measurement and checking learning targets through online tests or quizzes (Categories A and B). They contend that there is no significant difference between strategies in face-to-face and online assessment. The typical findings are also found in the previous studies with a similar context (Azis, 2015;Gebril, 2017). Luthfiyyah, B. Y. Cahyono, F. M. Ivone & N. Suryati, EFL  On the other hand, teachers find that online assessment is more challenging. They cannot control academic integrity and do not know how to assess the students' affective and psychomotor skills due to their limited online assessment literacy. The findings corroborate the previous studies by Guangul et al. (2020) and Meccawy et al. (2021), revealing that cheating and plagiarism become prevalent in online assessment in ERT. Meccawy et al. (2021) proposed a multilevel approach as the strategy to overcome academic dishonesty. The strategies are developing students' academic integrity, providing creative and meaningful tasks that assess students' holistic skills, rather than information recall, and giving extensive training for teachers related to various assessment tasks, academic integrity, and assessment security. Those endeavors can improve teachers' online assessment literacy and enlarge their understanding to prevent identical issues.
The second finding shows a more developed understanding of online assessment. Teachers perceive online assessment in ERT as a medium to promote the online learning process. They enlarge and adjust the concepts of online assessment, from testing factual knowledge to integrating assessment as a part of the learning process. Through online assessment activities, students can communicate with their friends and the teacher (Category C). They feel the presence of their teachers and friends during the learning process. The finding is in line with the framework by Anderson and Garrison (1998), highlighting that providing student-student, student-teacher, and student-content is prominent in effective online instruction. In addition, the findings also echo one of the categories in Mimirinis' (2019) work. In a conducive online environment, he finds that technology can facilitate dialogue and students' engagement.
EFL secondary school teachers view another benefit of online assessment during ERT. They admit that online assessment can facilitate authenticity in assessment (Category D). Several works have pointed out the advantages of authentic assessment (Ashford-Rowe et al., 2014;Huang & Jiang, 2021;Villarroel et al., 2018). In the context of ERT, teachers contend that they provide assessment activities that allow students to solve problems related to their real-life problems. Providing authentic tasks can minimize students' tension in assessment, stimulate their creativity and problemsolving skills, and engage students in online learning. Besides, Gikandi et al. (2011) report that authentic formative assessment can reduce cheating practices.
The last category (Category E) represents the most advanced conception of online assessment in ERT among the other four strategies. The conceptions require complex and deep thinking from teachers to evaluate their instructions and make better instruction. The participants contend that moving on assessment practice from the physical classroom to the online environment is such a tough experience during ERT. They have to struggle to find valid and reliable yet engaging online assessment strategies that meet the intended learning purposes. However, they claim that the emergency remote situation increases their awareness and willingness to reflect on their online assessment practices and to change or improvise the assessment strategies. The findings echo a current report from Zhang et al. (2021). They argue that teachers adjust and redesign classroom assessment practice during pandemic time based on their reflection and teaching experiences.
Based on the findings gleaned from the present study, some implications can be yielded to support assessment practice in an online environment. Although emergency remote teaching is provisional, we believe that online assessment practices have a high possibility to be more prevalent in the classroom-based assessment area. The first implication is related to teachers' assessment and technology literacies. Since the transition of instruction is abrupt, we cannot expect that teachers have sufficient knowledge and capability to design various assessment strategies and to integrate various assessment tools that fit the online environment in ERT. However, due to the continuity of online instruction, teachers must update and enrich their knowledge and skills associated with online assessment and technology-enhanced assessments. Secondly, the study results can assist the government, institutions, and stakeholders in depicting teachers' clear understanding and experience in conducting online assessments during ERT. It is expected that they can devise bottom-up regulations that fit teachers' and students' socio-cultural contexts. Lastly, this study shows a silver lining to optimize the implementation of assessment for learning as a classroom-based assessment trend in Indonesia, especially in secondary schools.

CONCLUSIONS
The study highlights the unprecedented transition of online instructions due to Covid-19 spread. We identified and described EFL secondary school teachers' qualitatively different ways of understanding and experiencing online assessment during emergency remote teaching. The study identifies five different conceptions of online assessment and a hierarchical relationship among those conceptions. The findings reflect that EFL secondary school teachers in Indonesia have multiplied and complex understanding and experience in online assessment practices. Their various conceptions are generally affected by influential factors, such as regulations, policy, readiness, knowledge, and infrastructure.
At last, this study has limitations on using the small number of participants and unbalanced gender ratio. Future researchers are suggested to further explore online assessment practices in the post-pandemic time, especially at the secondary education level. The investigation should cover more than perceived online assessment, but it can go deeper to seek the actual practices of the online assessment.