Exploring Pronunciation Teaching Practice in Emergency Remote Teaching: Lesson From Indonesian High School Context

: Regardless of its importance, pronunciation practice received a small portion in EFL teaching. This issue has attracted researchers to investigate how pronunciation practice is conducted in the EFL setting and what challenges are there. The results of the investigations have provided useful insights into what happens in the EFL classrooms. However, with the global pandemic outbreak that forced massive migration to emergency remote teaching (ERT), it remains unclear how pronunciation practice is carried out. This exploratory study attempts to shed light on the matter. It aims to portray the pronunciation practice during ERT and identify the challenges faced by the teachers. To limit the scope of the study, the researchers focused the investigation only on pronunciation accuracy. Data were collected through document analysis, non-participatory observation, questionnaire, and in-depth interview after gaining consent from two senior high schools English teachers in Indonesia. The collected data were then analyzed using the interactive model of analysis. This study reveals that during ERT, pronunciation practice received a small portion of attention, as earlier studies have found. The average time allocated is 10% of the total duration of learning. Most of the accuracy training was carried out with a similar pattern of activities, i.e., read aloud, listen-and-repeat, and teachers’ corrective feedback as the response of the student’s errors. Presented in ERT, pronunciation accuracy training experienced several problems, and the most prominent is related to the internet connection. A poor internet connection can be disastrous for accuracy training as implementing synchronous learning or uploading and downloading audio/video requires a good and strong internet connection. Another problem is the lack of students’ engagement to study due to the difference in sensation and excitement with face-to-face learning. These imply that adequate preparation is needed in applying pronunciation accuracy instruction, apart from the material itself.


INTRODUCTION
Pronunciation is one of the speaking skills aspects that received the least attention in English Language Teaching (ELT) compared to grammar and vocabulary. As Brown (1991) and Foote et al. (2016) corpus-based evidence of actual teacher behaviour with respect to the teaching of second language (L2 found, pronunciation teaching received a small concentration that is not more than 8% of the total duration in the English teaching process if not ignored. It is contrary to the fact that being able to pronounce English words well is crucial in speaking since it will make it easy for the interlocutors to understand and get the point of the message (Kelly 2002). The research conducted by Pennington and Richards (1986) and Celce-Murcia, Brinton, and Goodwin (1996) reported that pronunciation ability should be mastered by the students and taught by the teachers. Thomson and Derwing (2015)despite reports of significant improvement in many studies.
Possible explanations for divergent outcomes include learner individual differences, goals and foci of instruction, type and duration of instructional input, and assessment procedures. After identifying key concepts, we survey 75 L2 pronunciation studies, particularly their methods and results. Despite a move towards emphasizing speech intelligibility and comprehensibility, most research surveyed promoted native-like pronunciation as the target. Although most studies entailed classroom instruction, many featured Computer Assisted Pronunciation Teaching (CAPT argued that pronunciation instruction in the teaching and learning process would be improved by pronunciation instruction. By practicing pronunciation, it will train students' pronunciation to be more understandable. Since pronunciation ability is limited to communicating with the teachers in the learning activity and the others' language speakers (Yoshida 2016).
The significant topics of interest in pronunciation teaching in the past three decades were intelligibility, supra-segmental, and lingua franca. So far from that, in the nativeness principle in the 1960s, the target sound system of the majority English as Foreign Language (EFL) and English as Second Language (ESL) pronunciation teaching programs was nativelike mastery. However, in the 1990s, the aims of pronunciation teaching programs were changing from native-like to intelligible speech because nativelike pronunciation was not possible for all English learners, especially English as Foreign Language (EFL) learners (Abbas Pourhossein Gilakjani 2017).
In later years, the focus of teaching pronunciation has moved from segmental features to suprasegmental features (Derwing and Munro 2005;Pennington and Richards 1986). The researchers agree that factors such as stress, intonation, and sounds have more significance in accomplishing good pronunciation. Derwing, Munro, and Wiebe (1998) Hodges and Fowler (2020) showed the definition of ERT as a sudden educational procedure to keep conducting the educational process in the middle of lockdown. Nevertheless, ERT still has problems and is challenging for educators and learners (Mardiah 2020

Procedure
This exploratory case study was conducted for two months, and data were collected through lesson plan documents analysis, non-participatory classroom observation, questionnaires, and in- At the same time, the interview was implemented to gain more profound information and data about the teachers' motivation in teaching accuracy of pronunciation and the techniques used.

Participants
The participants were two English teachers, one female and one male, from a public senior high school in Indonesia. In this study, Joy will be used as the pseudonym of the female participant and Mark for the male participant to protect participants' identity and anonymity of information. The background profile of Joy and Mark is shown in Table 1.

Data Analysis
The accumulated data were analyzed using Miles, Huberman, and Saldaña's (2014) model in analyzing the qualitative data, which consisted of three steps; data condensation, data display, and conclusion and verification. The first is data condensation to select, simplify, and transform the data. In this step, the researcher converted and reduced raw data by selecting and sorting the collected data from the observation sheet, questionnaire data, and interview transcription. The second data analysis step is data display, aiming to make the data arranged into available and attainable form. The data which have been reduced and sorted were displayed in this step. The last step is conclusion and verification. In this step, the findings of the study were drawn from the accumulated and formulated data.  Table 2 and Table 3.  Table 2 shows that Joy and Mark planned to teach and evaluate pronunciation in their classes.
The strategies they chose were tailored to meet the basic competence and objectives of the materials that would be taught.  Table 3 shows the duration and the applied strategies by Joy and Mark during the teaching and learning process in ERT settings. It can be seen that both Joy and Mark did not always teach pronunciation in every meeting. In addition, Joy and Mark did not apply the strategy they planned as stated in the syllabi and lesson plans yet still taught pronunciation using their preferred approach.
The first findings show the duration of pronunciation instruction in ERT classes.
Observations reveal that pronunciation accuracy practice in Joy and Mark's classes is not taught for every meeting. In Joy's classes, pronunciation instruction was observed in the first, second, seventh, and eighth meetings, with a total portion was 16%.
While in Mark's classes, among ten observation meetings, pronunciation teaching duration was 5% which was found in the third, fourth, and seventh meetings. So, the average duration of pronunciation accuracy practice was 11% of the total learning time. pronunciation instruction relies on error correction through corrective feedback. Lyster, Saito, and Sato (2013) and Saito and Lyster (2012) have found that corrective feedback is effective for language improvement in learners.
The findings in this study partially confirm and reject earlier results reported by Shah, Othman, and Senom (2017) in terms of focus on pronunciation teaching. Shah, Othman, and Senom (2017) found that through observation, the English teachers observed did not focus and emphasize any specific pronunciation features during lessons in the classroom. Furthermore, the teaching did not focus on only suprasegmental or segmental aspects of pronunciation. It is because the focus of pronunciation teaching is determined by what the learners need. In contrast, this present study showed that the pronunciation teaching clearly focused on the suprasegmental aspect but with an uncertain focus for each lesson. Also, the teaching was based on teachers' initiative instead of students' needs.

Challenges in Pronunciation Teaching During ERT
T h e n o n -p a r t i c i p a t o r y o b s e r va t i o n , questionnaire, and in-depth interview applied to the research participants showed the teachers' challenges in teaching pronunciation during ERT.
The challenges could be from the students, media, and the teacher themselves.

Internet connection
The gathered data from observation during the teaching and learning process showed that Joy's problems are primarily due to poor internet been used to involve parents and help their children study due to poor internet connection. In Portugal, a partnership between schools and the postal service helped deliver printer learning resources immediately delivered to the students' homes. From the students' perspective, the technical requirements are also the most significant disadvantage of conducting ERT.
For several families, providing device equipment with cameras, microphones, and good internet connection for their children to study and parents to teach or work is impossible (Nenakhova 2021).

Student's motivation and engagement
The teachers have to be as creative as possible to design learning materials and media to increase students' motivation and engagement. Joy said that she has tried teaching approaches to know which one is suitable to get student's attention and increase their This fact almost proves that pronunciation will be a hidden factor and a source of unsystematic variation in assessments and examinations (Levis 2006).
• A large number of students In one class, the number of students ranged between 30-32 students. With those numbers and the learning duration only 30 minutes per lesson hours, the teachers faced difficulties managing the time when they wanted to evaluate students' pronunciation. Joy assessed her students' pronunciation asynchronously by giving assignments. Her students practiced and recorded their pronunciation by themselves, and then Joy scored the recordings one by one. She claimed that this way of assessment is the best way to save time. However, it is an ineffective way to provide feedback. In addition, she also carried out the pronunciation and speaking evaluation simultaneously via WhatsApp Video Call. This kind of evaluation took much time, and she could not finish evaluating all of the students even though they were divided into groups.
This finding agrees with Asikin and Ibrahim (2020) and Darcy, Ewert, and Lidster (2012) findings that showed the class size could also be challenging, with a large number of students making it impossible for the teacher to evaluate every student's pronunciation ability. Therefore, teaching and evaluating pronunciation in a short period can be very challenging for teachers. Moreover, Zhang (2020) confirmed that the participants in his research struggled to do pronunciation assessment because they did not find effective ways to conduct a wholeclass assessment. The teachers also did not find suitable ways to create direct interaction and communication with students due to the lack of time in ERT and a large number of students.

CONCLUSION
The present study was designed to investigate pronunciation accuracy practice in an ERT setting.
From the findings, the conclusions are as follows: (1) the duration and portion of pronunciation accuracy instruction during ERT is 11% of the total duration of lesson hours; (2) the most used strategy is corrective feedback as the response of students' mispronounce and stress is the least attention among three pronunciation accuracy features, also the teachers faced challenges in conducting teaching and evaluating students' pronunciation, which are a terrible internet connection, students' motivation and engagement, time management, the unavailability of pronunciation rubric assessment, and a large number of students in one class.