Ethiopian Secondary School EFL Teachers’ Classroom Assessment Conceptions and Practices from an Activity Theory Perspectives

Teachers’ language assessment conceptions play a vital role in the effectiveness of classroom assessment practices. Drawing on activity theory perspectives, the researchers intended to present a picture of English language teachers’ classroom assessment/testing conceptions and how their conceptions reflect their practices. To this end, 15 English language teachers whose teaching experience is more than three years were purposely selected for interview. The obtained data were coded, categorized into themes, and analysed qualitatively. The results indicated teachers’ assessment conceptions reflect that students and schools are highly accountable for English language classroom assessment. Teachers’ classroom assessment practices do not reflect their assessment conceptions as they use summative oriented assessment tools, and the activity system is dominated by teachers themselves.


INTRODUCTION
Assessment helps both teachers and students to achieve the intention of language teaching and learning. Assessment serves multiple purposes such as providing information about student learning and progress, teaching quality, and program and institutional accountability; thus, the study of teachers' conceptions about assessment is crucial (Opre, 2015). As far as teachers implement and work closely with assessment stakeholders, especially students, teachers' assessment conceptions matter the manner and purposes of classroom-based assessments. In line with this, Stiggins (2016) argues every teacher in every classroom needs to know why they are assessing, what they are assessing, how to create a quality assessment for that particular context, and how to communicate results effectively. Stinggins argued that these standards are not negotiable; hence, teachers' assessment conceptions affect any teaching practice including assessment (Fernández-Ruiz & Panadero, 2020). Accordingly, teachers need to be equipped with a strong background in language assessment (Gebril, 2017).
On the other hand, the way teachers conceive of assessment purposes and practices is an important issue (Remesal & Brown, 2015;Popham, 2017). That is teachers' evaluations of learners' deeds and performances (i.e., assessment) are affected by the conceptions teachers hold regarding educational facts such as teaching, learning, assessment, curriculum, and teachers' efficacy (Brown, 2004). In the same vein, purposes of assessment, teachers' assessment orientation influences their classroom practices. Regarding this, Fernández Ruiz and Panadero (2020) explicate that those teachers with a traditional vision of learning …will consider assessment a summative tool; contrary to this. "Those teachers with a constructivist conception of learning will be inclined toward a formative concept of assessment" (Eren, 2010, p. 40).
In particular, how teachers conceive of assessment shapes how they use assessment results and frame their selection and development of assessment tools in their classroom (Brown, 2008). As teachers' assessment conception is mandatory to conduct the right classroom assessment in order to achieve the purposes of assessment, they should equip themselves with assessment awareness/knowledge. They need to develop a better understanding of what they already do before they can start about which aspects of classroom-based assessment practices could be developed (Hill, 2017).
In this study, Activity Theory is used as a theoretical framework guiding data collection and analysis. Gedera (2016) argues activity theory considers sociocultural settings like a classroom, school context or society. The sociocultural perspectives of activity theory consider the roles of both the individual and the social activity (Mwanza, 2001;Feryok, 2012). The activity theory framework uses activity as the basic unit as the basic human practices so that activity or 'what people do' is reflected through actions as people interact with their environment (Mwanza, 2001). It is a philosophical and cross-disciplinary framework for studying different forms of human praxis as developmental processes, both individual and social levels interlinked at the same time (Jonassen & Land, 2000).
The concept of activity theory has three generations. The first generation was proposed by Vygotsky (David & Land, 2000;Mwalongo, 2016), and it considers subject, object, and tools in the human activity system. In this generation, mediated activity is a process of making meaning whereby humans use tools to interact with an object in order to achieve a particular goal (Engestrom, 1999). Leont'ev 1975, according to Mwanza (2001, extended Vygotsky's concept of mediated action as the second generation. He added another element that is the division of labour in a collective activity and the individual completing particular actions (Mwalongo, 2016). Later on, the third generation, which the current study is based on, and which is relatively comprehensive, was proposed by Engeström in 1987 according to Razmjoo and Barabadi (2015). Accordingly, Engestrom's human activity system model shows how individual activities are interwoven with other individual activities and where individual thinking emerges in context. Engeström (1999) proposed an activity system with six interacting elements: subjects, tools, objects, community, rules, and division of labour.
The subjects are individuals who engage in the activity to achieve an object/goal. Here, classroom assessment includes subjects (teachers and students) and the action of assessment is undertaken by these subjects. On the other hand, the action of teachers and students are led by motives (objective). For example, for this study, the classroom assessment is expected to improve students learning and then their performance (Carr and Harris, 2001;Brown, 2008); and classroom assessment needs to be aligned to curriculum. According to Kelly (2004), the assessment (and evaluation) tail will always wag the curriculum dog. In addition, the actions undertaken by the teachers and the students during the activity are mediated by tools. For the context of this study, tools are assumed to be the techniques that teachers use to collect assessment information from the students. For further explanation, tools that address the expected competencies (language skills and language areas) are required. Further, within a community of actors, there are bound to be rules and regulations that affect the means by which the activity is carried out (Mwanza, 2001). Classroom assessment should be aligned with the curriculum (curriculum framework, syllabus, teachers guide, and other expected rules). Teachers have big roles (division of power) in classroom assessment that they need to discharge to lead the classroom community (students) to learn and improve their performance.
Then, the purpose of this paper is to explore English language teachers' conceptions of classroom assessment/testing/examinations as their conceptions affect what and how they practice as well their tools selection. Regarding this assumption, it is argued that there is compelling evidence that beliefs and conceptions can affect teaching even greater than experience and socioeconomic context (Pishghadam &Shayesteh, 2012). Teachers' assessment conceptions are an essential part of the teaching-learning process as they influence teachers' classroom assessment actions.
Since 1994, the issuance of Education and Training policy, in Ethiopian education system attempts were made to design educational curriculum in order to shift the practice from traditional ways of assessment (teacher-centred) to modern (studentcentred) (MoE, 1994;MoE, 2008;MoE, 2009;MoE, 2013). In line with this, it is expected of teachers to adjust themselves to the newly introduced education system and implement accordingly. As part of the new curriculum, classroom assessment (English language classroom assessment/testing for the current study) need to be students centred. Assessment should not merely be an activity that takes place at the end of the year/term, but it should be a spontaneous and natural part of the teaching-learning process (MoE, 2012). The implication is that currently in the Ethiopian context, the assessment method should be student-centred oriented, so there is a need to explore whether teachers' classroom assessment conceptions align with their real classroom assessment practices. The Ethiopian Ministry of Education (MoE, 2013) sets English language teachers' assessment conceptions/knowledge that the teacher understands and uses multiple uses of assessment to monitor learners' progress. English language teachers make a continuous assessment using different assessment techniques.
There are a number of studies conducted in the area of classroom English language assessment in the Ethiopian context (Yigzaw, 2013;Mekonnen, 2014;Tulu et al., 2018). Accordingly, Yigzaw (2013) reached on conclusion that English teachers did not properly practice continuous assessment in their schools; in the same manner, the qualitative results of Mekonnen's (2014) study indicated that most of the English language teachers used assessment for the purpose of assessment of learning, and they hardly ever used the feedback form classroom assessment in adjusting their teaching techniques. In this context, Tulu et al. (2018) reported in their study that assessment of learning and assessment for learning are the most common and dominant forms of assessment across language, sciences, and social sciences of Ethiopian secondary school teachers. On the other hand, teacher conceptions of assessment are influential mediators of how assessment policy initiatives are implemented in schools (Brown & Michealidis, 2011). Besides, Fernández-Ruiz and Panadero (2020) concluded that the teachers reported having a formative conception, but this was incoherent with their assessment methods characterized by a scarce use of self-assessment and peer assessment and hardly formative feedback. However, little is known about teachers' assessment/testing conceptions and their alignment with classroom practices in the Ethiopian context. The target is that teachers' conceptions or beliefs become key factors, being regarded as essential determinants of the instructional activity and for the students' learning process (Opre, 2015). More clearly argues that teachers own a variety of beliefs/conceptions and this influence the quality of their performance. Accordingly, the study intended to answer the following research questions: How do Secondary school English language teachers conceive classroom assessment? How do teachers' assessment conceptions influence their classroom practice and selection of assessment methods?

METHODS
It was a qualitative study that obtained information from secondary school English language teachers who teach grades 9 and 10 during the time of data collection. Altogether, 15 English language teachers from 5 secondary schools of East Haraghe zone were interviewed. The participants were purposively selected considering years of experience as experience plays determinant roles in the context of teaching, so participants with diverse years of experience were considered. In order to get relevant data, teachers with more than three years of experience were selected because newly employed teachers might not have attended pedagogy (methodology) courses that directly/indirectly affect teachers' classroom practices. Regarding this, Zolfaghari, and Ashraf (2015) concluded that English language teachers with more years of teaching experience demonstrated to be more knowledgeable in assessment-related matters, and they further argue that the more teaching experience increases the more teachers' assessment literacy increases too. On top of the participants' years of teaching experience, the researchers made use of the contact they had before. To make it clear, the interviewees and the researchers had a studentteacher relationship because almost all of these participants were Haramaya University graduates. It was this condition that helped the researchers to easily strengthen rapport and friendly discussion with the participants, which in turn facilitated the data collection process.
A semi-structured interview, which is open for probing and fully open-ended questions to evoke the participants' thoughts, was the prominent data collection tool for the current study. Accordingly, to conduct an interview, the researchers should identify key participants in the situation whose knowledge and opinions may provide important insights regarding the research questions (Algozzine & Hancock, 2017). Prior to the interview, the participants' interest was assessed, and the researchers explained the purpose of the study. The interview, which lasted on average 35 to 40 minutes for each participant, was made at each research site where was convenient for the interviewees.
The researchers themselves conducted the interview, and it was audio recorded. Since the participants were English language teachers, interviews were conducted in English, and the researchers made enough explanation to clear in case there was confusion regarding the interview guides.
It is a common trend that qualitative data analysis is concurrent with data collection and management (Saldana, 2011). Regarding Denscombe (2007) emphasizes, in the explanation of data analysis steps, familiarity with the data by reading and rereading the text data helps the researcher to become immersed in the details of what was said, what was done, and what was observed. Listening to the recorded interview frequently, it was transcribed and its summary was prepared for coding and thematically organized. So, the researchers read the interview data several times. Then, after the process of coding was secured, it was given to colleagues for comment.

RESULTS
The study intended to explore English language teachers' assessment conceptions and their intended classroom practices. It meant to infer the teachers' classroom assessment practices and what they do/perform and how their conceptions influence the selection of assessment tools. While the researchers interviewed the participants, detailed interviews in the form of extended probing tried to explore teachers' assessment conceptions and practices. After the interview data were read repetitively, the researchers scrutinized them carefully to reach on conclusion about teachers' English language classroom assessment conceptions and practices. The results are discussed as follow:

Assessment Makes Students Accountable for Learning
The participants indicated that classroom assessment makes students accountable for their own achievement. Accordingly, one of the participants responded, "I believe that after I provide necessary explanation and guideline for my students when I feel confident that I have discharged my responsibility, I expect more from my students so that they can make their own marks" (T3).
The other participant expressed his thought in the same way, "assessment helps students prepare themselves to achieve their goals and this is by scoring grades/marks which indicates success for the next level" T5. However, one of the participants differently forwarded his thought, "students must be free that they should not bother about making grades/marks, but their understanding matters so that grades/marks are number games" T4. Contradictory to this issue, one of the informants said, "Both understanding and grades/score matter because grades may indicate students' understanding".
It can be said that almost all participants have similar concepts regarding students' accountability for their own learning. Another respondent confirmed as follows: Actually, classroom assessment forces students to be responsible in the teaching-learning process to determine their tomorrow's life, and in the Ethiopian context, most of the time, students are judged by their marks/grades. So, it is important to encourage students so that they become responsible for their own learning (T7).
Still, there are other respondents who consider that assessment leads students to take responsibility for their own learning, but at the same time, the respondents blame the students for not discharging their responsibility. For example, "I understand that some of our students come to class without being prepared when there is assessment and sit for tests as well as for other assessments simply because they are told to do so" T9. "At the end of the day, not only teachers but students are also accountable for what they do whether they are successful or not" T8. Therefore, the participants' substantial explanation underlines that assessment is a mechanism by which students become accountable for their learning process in addition to teachers' effort.
From the Activity Theory perspective, in the English language classroom assessment, teachers are the subjects as they are the main implementers. The primary goal of assessment which is improving students learning, can be achieved by attracting students to be engaged in assessment planning, motivating students to set their own goals, guiding students to actively take part in classroom discussions, inviting students for self and peer assessment, assessing students on continuous bases, and other constructs increase students' accountability. Respondents claim that students are expected to be responsible for classroom assessment/testing/ for every aspect. On the contrary, the respondents themselves confess that the aforementioned constructs are not practical in the real classroom activities, and this, in turn, shows the misalignment of teachers' classroom assessment conceptions and practices. On the other hand, the participants did not give due attention to teachers' instructional improvement which indicated the tendency of their conceptions inclined to one direction.
The notion of working together that is division of roles is not applicable may be because of teachers' traditional experience to show direction for the students. Teachers desire to be entirely authoritative was deduced from the data because they consider their role in the activity system as teaching (source of knowledge) and then assessing so that whatever happens in the process of assessment, they made students accountable. This affects the labour division (power relations) of the activity system. Practically, in the activity system, teachers (the subjects) would have been played the greatest role through making the students understand the essences of classroom assessment rather than externalizing themselves from being accepting responsibility. May be that is why the students relate classroom assessment only to scoring. On the other hand, this can affect the outcome (goal) of classroom assessment.

Assessment Evaluates the Quality of Schools and Teachers
Participants were asked to explicate 'what is assessment for school?' which has helped the respondents to generate their concepts regarding classroom assessment particularly on the English language.
Accordingly, most of the respondents clearly showed that school accountability is the antecedent to school quality. One of the participants indicated: "Schools need to be responsible for every procedure that happens in school including classroom assessment. She said that there must be strict follow-up from the side of the school about the processes of assessment. School management, particularly the assessment/exam committee, should know how and when teachers assess their students. In addition, test/exam profile (techniques) need to be examined based on the subject nature by the exam supervision committee. These enhance school quality and advances both teachers' and students' performances" T3.
Others noted that schools use assessment results for different purposes. "They communicate to higher officials about the overall students' performance which indicates schools' semester/annual report, and assessment is one of the many criteria that school efficiency is judged in our country" T2. Similar to this concept, "Schools need to endeavour to improve assessment quality by providing necessary infrastructures which function as inputs for assessment quality; for example, stationary and duplicating machine in general" T5.
Finally, one of the respondents uncovered, "Schools are expected to work with parents because parents are our close stakeholders, and their participation increases students' achievements although the trend is absent in our school" T11.
The aforementioned are then teachers' assessment conceptions which they expressed from their assessment understandings, and it contributes to school accountability and quality. Therefore, respondents argue that school accountability comes before school quality.
"Whenever the school is accountable for its procedures including assessment, quality follows. Therefore, as assessment is one of the school improvement packages, working on classroom assessments contributes to the overall school quality" T11.
It is understandable from the participants' discourse that they wanted to impose huge accountability on students and schools. "We give homework/assignments, but there is no established system to manage those students who do not do home works and those who do not submit an assignment on time", said one participant. Many other participants have such arguments. For instance, another participant argued that they had been reporting to their schools about misbehaving students and homework problems, but he did not see while corrective measures were taken (T13). Confirming this, still, there was complaints from the side of teachers. "If students accept assessment as their business, things will be alright, but if they are careless, things will be confusing" (T10). He added that schools must work on the students' understanding of classroom assessment/testing.
The above extracts imply that the participant teachers argued schools and students are not discharging their roles of classroom assessment/testing.

Assessment for Managing Students' Discipline
The interview data reflect that the teachers use classroom assessment to monitor students' misbehaviour. Accordingly, one of the participants said, "I suddenly ask misbehaving students, and if she/he could not respond, I devaluate his/her scores" T3. She explicated that the objective is to monitor the students so that they give due attention to the lesson. On the other hand, another participant said, "To monitor absenteeism, I use classroom assessment at some intervals. On market days, most students miss classes to make money. This is a serious problem for our school. So, on such days to attract our students to school, most teachers give tests" T14.
Other participants found the use of classroom assessment to monitor discipline issues as positive. One of them said, "This is the strategy we created to keep students in schools thinking that students may give little value for assessment… then it is a mechanism to control them" T1. The last one said, "Discipline is not an easy thing and in case there is a misbehaving student in a class, the lesson may get dangerous so that a teacher's role becomes managing discipline" T6. She further explains that to have a healthy classroom process, sometimes we use classroom assessment as a controlling mechanism.
The implication of the above discourse is that the participants considered the use of classroom assessment for discipline issues positively. However, this is not the intended goal of classroom assessment. The source of this conception might be emanated from teachers' background/past assessment culture. For example, they rose about how they were assessed when they were students.
One of the participants said that he had a model teacher; you could not believe how he used to teach grammar and his assessment such as multiple choice and grammar content (T14). The other one reported that when he was a student, there was a test every Friday afternoon so that the students did not miss classes (T7). This was the culture of the school. So, teachers' (subjects) conception affected the primary goal/motive (improving students' learning) of classroom assessment.
Here, contrary to Activity Theory perspectives, the extracts showed that teachers' conceptions led them to the traditional approach (teacher-centred). Activity Theory by its nature, encourages collaboration (multi-voicedness), i.e., individuals and community. Furthermore, teachers' conceptions influence the system's 'mediators' 'to this context' it meant assessment tools. Evidently, series of tests, mid, and final examinations are the frequently reported tools to achieve the goal teachers intended to achieve and these are highly dominated by objective items. On the other hand, the integrations of teachers' conceptions intended to violate English language classroom rules and regulations because one of the regulations is that English language teachers need to work toward developing the four language skills competencies as well as grammar and vocabulary. However, these competencies cannot be addressed only by the summative type of assessment tools and objective items.

Teachers as the only Source of Feedback
Classroom assessment by its nature is a collaborative activity. Teachers are expected to engage students in the process of classroom assessment. This means there is a division of labour (role). Students can take part in the process of classroom assessment from different angles: peer assessment and peer feedback can be some examples. However, emanating from a teachercentred approach, teachers consider themselves as the only source of feedback. This led teachers to conceptualize as if they were the only source of knowledge. They reported that: "I told you that teachers stay only for forty minutes per day with one class. How can I manage to observe the feedback that the students give to each other? I fear that they mislead one another. Actually, when I was a student, I did not have such an experience. Our teachers did it. After I became a teacher, I made a little attempt, but I observed that it did not work. It does not have meaning for me except wasting time" T9.
There are also other participants who share the same motive and whose discourse showed similar conceptions which leads to the absence of the activity.
"We work for the success of our students on the national examinations, so we do not want our students to get confused. Therefore, we manage the provision of feedback ourselves. On the one hand, we save our time. On the other hand, when I rarely assign them to give feedback for one another, I observe they are engaged with other issues" T7.
One of the participant's reports was very specific and to the point of her action of feedback in the activity system. She said that as to her, feedback focuses on grammar, and her reason was about the content of the national examination. She confessed that other teachers' feedback was similar to her practice (T6). The other participants said that they only provide written feedback on tests/examinations. The essence of the discussion from Activity Theory perspective is that only teachers (the subjects) control feedback provision. On the other hand, a teacher's feedback is important; student's peer feedback is also important too. The motive behind teachers' feedback provision and students (peer feedback) is different. So, a mismatch between the objective and the practice is revealed. On the other hand, teachers believe that peer feedback is one way of improving students learning, but the practice is missed in the context of the current research. Further, here the interview data indicated that the reason for the mismatch is a lack of trust in studentto-student feedback. For instance, one of the participants argued that students do not know how to provide feedback, they must be trained (T12). However, showing the way is one role of classroom teachers. Finally, because of the hegemonic action of teachers, students are not practising their roles. This hampered students' involvement in the classroom assessment process.

English Language Teachers' Classroom Assessment Conceptions and Tool Selection
The interview data revealed that English language teachers' classroom assessment conceptions influenced their tool (methods) and items selection. Hence, under this category, the participants have got the opportunity to discuss the assessment techniques and the types of items they often use in EFL classroom assessment. So, from the entire data discourse, English language teachers' conceptions led them to practice teacher-centred classroom assessment. Regarding this, the participants reported in the following ways: "Most of the time, to achieve the purpose of the assessment, we use tests. I have said before that there is no assessment on the four language skills. The content of or assessment is actually grammar and passage on the final examination. The reason is that students like grammar" T10.
Another participant also reported that they rarely use peer and group assessment and the hindrances are like sitting table. He identified that most of the time they use individual assessments: tests, mid, and final examinations are the established assessment culture in their school.
As can be understood from these extracts, the culture of practising peer assessment, group tasks, portfolios whose modalities go to the formative nature of classroom assessment and function for the improvement of students' learning and improvement of teachers' instruction is less practical in the context of this study. The established assessment tools are purely summative types. Even most of the participants assert that grammar in the form of summative type is deep-rooted assessment practice in their contexts. The respondents indicated, "We should have focused on all language skills and integrate with grammar, but because of different reasons, English language classroom assessment in our school is limited to grammar" (T2). Another participant also confirmed, "Tests, mid/final examination are used as the tools of classroom assessment which mainly focus on grammar" T5. This is also repeated, "Alternative items (choice, matching, true-false) and grammar content more practical in English language classroom assessment in our context" T8.
In the activity theory perspective, tools mediate the subjects' (teachers) interaction with the object/ive (developing learns learning and teachers' instructional improvement). In the above extracts, the use of a limited number of classroom assessment tools that are not insufficient to address assessment objectives/goals rooted in the summative type of assessment and emanate from the traditional teacher-centred approach.
Classroom assessment attributes to teachers' quality in teaching-learning processes. The participants relate this concept with their own regular activities. "Classroom assessment is part of our everyday activity, and we obtain experiences, each day we come across new events because of dynamic students' behaviour" T2. Other respondents replied that preparing tests, asking questions for crosschecking for formative assessment and dealing with different types of assessments enhances their teaching skills" T5. Still, the respondents' reservation is that they were narrating, but they were doubtful about the effective implementation of classroom English language assessment/testing. The reason is that most of them were not confident enough when they were asked about their own classroom assessment action. For example, one of the participants indicated as follows: I have attended in-service training at Haramaya University and I have got the opportunity to take assessment related courses. What I have explained above is also because of this experience whether I implement or not in the real classroom" T6.

DISCUSSIONS
The study intended to examine the Ethiopian secondary school English language teachers' assessment conceptions and practices and to investigate if the assessment tools teachers use to align with their conception or not, drawing on Activity Theory perspectives. After investigating the participants' assessment conceptions and asking whether they implement (practice) their conceptions, the researchers come up with findings. The study proposes that English language teachers' classroom assessment/testing conceptions reflect that students and schools are accountable for English language classroom assessment (Brown, 2008). The context in which the participants expressed students' accountability was to indicate the less attention students give for assessment/testing. So, it is to indicate that students' responsibility, but the participants did not overlook their own effort in the overall activity system.
On the other hand, the informants' intention to discuss school accountability was a burden to school management/leaders with many of classroom assessment/testing responsibilities expecting school managements to arrange any necessary situations regarding classroom assessment/testing. In addition to this, the participants were bold enough to argue that the school administrators' contribution (administrators in general) to classroom assessment is insignificant. This is to postulate that the role of administrators for effective classroom practice is mandatory so that they need to act to their best level. As far as the stakeholders in the activity system (classroom assessment) fail to discharge their roles, it affects the process of the entire activity system. Fernandez and Panadero (2020), in their study entitled "Conceptions and assessment practices among secondary education teachers", concluded the incoherence between teachers' classroom assessment conceptions and practices. So, they recommended the need for working jointly with teachers with an emphasis on consistency between conceptions and practices.
Despite English language teachers believes that assessment/testing improves their instruction and at the same time, improves students' learning which is the goal of classroom assessment/testing, this does not align with their classroom practices. Here, the subjects have developed the culture of a teachercentred approach disregarding students' engagement. The teachers, in this study, had understood as assessment improves students learning and teachers' instruction, but the problem was that they did not work towards their conceptions. This contradicts the study by Brown and Gao (2015), which claims that there is an alignment between Chinese teachers' conceptions and their practices.
As a consequence of their assessment conceptions, teachers act as the only source of knowledge and source of feedback. On the other hand, this shows that one of the elements of activity theory in the assessment system, i.e., division of labour is missing from the process. Moreover, this indicates the absence of students' involvement or considering students as recipients.
On the one hand, as the results of the study indicate, teachers may have theoretical knowledge of assessment. They can talk about various aspects of classroom assessment/testing. However, the problem is on its implementation because the participants were not fully confident when they were asked about their classroom assessment actions. Thus, most of the participants mentioned classroom assessment tools associated with summative; for example, paper-pencil tests are dominant tools. This shows that there are differences between what the participants explained and what they practically implement in the real classroom context because the majority of them assess their students either at the end of a week, a semester, or a chapter. These included a narrow measurement view of assessment (Dayal & Lingman, 2015). In the same manner, as a rule, special attention is given to ongoing assessment (MoE, 1994;GEQIP, 2008;Curriculum Framework, 2009;Arega et al., 2014). However, teachers' interviews confirmed the absence of the practice. As the components of the activity system of this study, in the students' textbooks, teachers' guide, curriculum framework and syllabus, all language skills and language areas should be equally treated. Nevertheless, the study revealed that grammar takes the lion's share in the English language classroom assessment, and other skills are marginalized. This also showed that classroom assessment is not occurring as per the rule.
Assessment can be seen as a mechanism by which teachers can improve student learning and improve their own teaching (Brown, 2008). The results of the study uncovered the importance of classroom assessments for both teachers and students, and there was no basic difference among the interviewed teachers that classroom assessment is fundamental for students' learning and teachers' instructional improvement. Some of the participants claimed that they consider assessment in the English language classroom as a mirror for that they get feedback for their own instruction and to identify the understanding of the students. Furthermore, Fernández-Ruiz & Panadero (2020) compare two positions: those inclined toward an assessment based on improving learning and those inclined toward an assessment based on certifying the students' knowledge. In the context of the current study, teachers' classroom assessment knowledge as well as their classroom practice inclined to the later position. However, the participant teachers' interview reflects that teachers might have a formative assessment concept (which improves students learning) though it is not adequate, but no attempt of its practice. Widiastuti (2016), in the context of her study, indicated that teachers conduct a formative assessment based on their own understanding of formative assessment.
Regarding students' accountability issues, in Ethiopian secondary schools, it is undeniable that teachers strive to prepare students for the national examination. In line with this, the respondents clearly uncover that the nature of their assessment resembles the one that is prepared by national examinations agency. To make it clear, the participants explained that teachers adapt classroom tests, mid-exams, and final examinations from the national exam paper. These assessment tools promote the assessment of learning than assessment for learning. The discussion made by the participants of the current study inclined to assessment results/ the final mark/grade/ that helps the students for a requirement to be employed when they finish secondary school. To be more precise, in the same way as Brown's (2008) explanation, students' accountability is related to certifying students' performance through encouraging them to score better results.
The result of the present study goes with previous findings in that it came up with a conclusion 'teachers hold multiple conceptions of assessment' (Brown, 2004;Dayal & Lingam, 2015;Barnes, Fives & Dacey, 2017). Regarding this, Popham (2017) argued that although teachers like to teach, they rarely like to test (assess). However, he confirmed that well-conceived classroom assessment will almost always lead to bettereducated students. In the present study, dozens of reasons were mentioned that restricted teachers to the assessment of learning (summative) and fewer implementations of continuous assessment. This is similar to the study conducted by Gebril (2017), which concluded pre-service teachers preferred traditional techniques to formative assessment methods.

CONCLUSION
Drawing on activity theory perspectives, the current study attempted to investigate English language teachers' classroom assessment conceptions and how their conceptions align to the implementation and to assessment tool selection. Case in point, English language teachers consider that assessment contributes a lot for the improvement of students' learning and helps teachers get feedback from it so that they sustainably rebuild their classroom instructions. Despite the fact that they conceptualize classroom assessment in such a way, the result showed that teachers did not confidently express that they implemented it accordingly. The result clearly indicates a mismatch between teachers' classroom assessment conceptions and implementation, i.e., the teacher-centred approach is a deep-rooted culture. On the other hand, the study revealed that students and schools are highly accountable for English language classroom assessment. This result is inconsistent because teachers have mentioned that they benefit from classroom assessment in one or many ways, which makes them accountable, too. More or less, in the process of the activity system (English language classroom-based assessment), teachers control every aspect. In general, the result of the current study does not reflect the components of the activity theory model. Finally, investigating the impact of teachers' assessment conceptions on classroom teaching-learning practices can be areas that need further work.