FLIPPING THE CLASSROOM IN THE EFL READING FOR ACADEMIC PURPOSE COURSE DURING COVID-19: BARRIERS AND BOOSTERS

The Flipped Classroom model has been a debatable issue, especially due to its triumph in the new set during the current outbreak of Covid-19. This study attempts to assess the effectiveness and explore the barriers and boosters of an FC approach in the Reading for Academic Purpose course in a private university in NTT Province, Indonesia. Employing a mixed-method design, this study generated data from the pre-and post-test and students TOWS (Threat-Opportunity-Weakness-Strength) analysis. Two groups of fourth-semester students, comprising 35 in a Blended Learning setting and another 23 in a Flipped Classroom, were taught similar reading skills and tasks. The paired-samples t -test indicates both FC ( p =0.03<0.05, d =-0.70, medium effect size) and BL ( p =0.00<0.05, d =-1.06, large effect size) settings were statistically significant by comparison. The independent-samples t -test used to contrast their effectiveness shows no statistically significant difference between groups ( t [56]=1.11, p =0.27, d =0.29, small effect size). Although FC students benefited equally from the learning process in BL, they gained a lower mean score than their counterparts. TOWS analysis requires more improvement in FC due to students’ resilience to change, inability to self-learning management, poor collaboration and communication, technological setbacks, technical issues, and other socio-environmental constraints. ( p =0.00<0.05, d =-1.06, large effect size) have positive effect on students’ academic reading performance. The independent-samples t -test used to compare their effectiveness shows no statistically significant difference between the two groups ( t [56]=1.11, p =0.27, d =0.29, small effect size). Although FC students benefited equally from the BL learners, they gained a lower mean score than their counterparts. The TOWS analysis confirms that FC needs more development because of students’ resistance to change, failure to self-learning management, poor collaboration and communication, technological setbacks, technical issues, and other socio-environmental constraints.


Introduction
The abrupt outbreak of Covid-19 in late 2019 and early 2020 obliged all teachers and lecturers around the globe to reform their ways of teaching. Most of those in rural contexts are opposed to this rapid change due to a lack of preparation and insufficient resources. The Flipped Classroom (FC), having been long employed by world-leading institutions, has been one of few great applications to consider in Indonesia. Although FC as a type of Blended Learning (BL) has been recently utilised by universities in NTT, lecturers and students are triggered and still struggling to adapt to this learning model during new-normal. While learners in BL commonly have balanced, face-to-face instruction in class and online tasking complements outside of class, students in FC should do in an inverted way.
FC has become a buzzword in the last two decades, referring to flipping activities between face-to-face, traditional classrooms and independent study at home (homework). This classroom model requires students to attain background knowledge through materials outside of class before a direct class meeting and reserves class time for applying knowledge to solve real problems through discussion or problem-solving work facilitated by a teacher (Bishop & Verleger, 2013;Lewis, Chen, & Relan, 2018;Bohaty, Redford, & Gadbury-Amyot, 2020). Integrated with technology, FC helps towards student-centred learning and generates an engaging learning environment (Rotellar & Cain, 2016;Mehring, 2018;Brame, 2013). It provides students with ample opportunities for self-paced learning, knowledge retention, creativity and collaboration (Young & Moran, 2017).
On the contrary, the general application of the FL has drawn serious drawbacks and caused practical problems. Several pieces of literature claim poor conduct of FC due to negative attitudes (Abeysekera & Dawson, 2016), unfamiliarity (Pudin, 2017), technological issues (Jones, 2016), and almost having nothing to do with student assessment (Bishop & Verleger, 2013). In the English as a Foreign Language (EFL) context, although some scholars found the significant benefit of FC (Lockwood, 2014;Marsh, 2012;Ahmed, 2016;Lee & Wallace, 2018), some others reveal its backwards. For example, FC was sophisticated to apply at the first trial and made students rather uncomfortable and anxious (Pudin, 2017;Rintaningrum, 2018;Evseeva and Solozhenko, 2015). Furthermore, its material preparation and designing are often time-consuming (Lee & Wallace, 2018;Voss & Kostka, 2019) and need high self-discipline and efficacy from both teacher and students (Hamouda, 2013;Evseeva and Solozhenko, 2015;Shih & Huang, 2019). Moreover, integrating FC with technology mostly faces internet connection issues (Evseeva and Solozhenko, 2015;Brooks & Weaver, 2017) and that teachers or lecturers and students are not qualified enough to deal with technological devices (Goodwin & Miller, 2013;Shih & Huang, 2019).
Reflecting on the two sides of the FC applications above, it is crucial to reevaluate the effectiveness of the FC each time it applies to a new setting, especially during the pandemic era. This mixed-method study conducts to evaluate the effect and acceptability of the FC during the early breakout of pandemic compared to the BL in the undergraduate English Department at a private university in NTT Province, Indonesia. The research questions address the following concerns.
1. Is there any difference in student academic reading performance between the FC and the BL? 2. Do the students in the FC setting gain better scores than those in conventional BL? 3. What are the EFL students perceptions of the FC application in the Reading for Academic Purpose course during Covid-19?

Literature Reviews
The FC approach has been widely employed for years in some disciplines, including in the EFL context. Its broad applicability is compatible with several other well-known learning styles, such as inverted classrooms (Strayer, 2012;Moran & Young, 2013;Lage, Platt, & Treglia, 2000;Irianti, 2020) and peer instruction (Crouch & Mazur, 2001;Schell, 2012). Whatever term it might relate to, the focus of FC and the other two is on students gaining first exposure before class. This is in line with Bransford et al. (2000) assertion that "to develop competence in an area of inquiry, students must have a deep foundation of factual knowledge, understand facts and ideas in the context of a conceptual framework, and demonstrates knowledge in ways that facilitate retrieval and application" (p. 16).
Effective implementation of the FC in the EFL setting has triggered several scholars interest in field research and literature review. In the Indonesian context, few related pieces of literature can be referred to for having insights into each particular setting and context. Afrilyasanti, Cahyono, and Astuti (2017) tried to investigate the effect of flipped classroom models on the writing ability of EFL students across their differences in learning. The results confirmed a significant difference in the students post-test score (t-count=10.893, p-value=0.000) between the experimental and control groups. The flipped classroom method also apparently resulted in significant differences among the interaction groups on the post-test scores. Nonetheless, this study also presents several caveats for instructors when they plan to flip the class.
Furthermore, a study regarding the use of flipped instruction in teaching listening, especially language testing, has been examined by Khoiriyah (2021). She also explored student perspective toward their experience in having a flipped classroom. Under a mixed-method research design, she employed pre-test and posttest after treatment and the online open-ended questionnaire. The study revealed that student average scores improved in the post-test, indicating flipped instruction enhances the students listening comprehension skills. Also, most students provoked affirmative feedback on their flipped classroom experience.
Moreover, Fauzi (2020), by classroom action research, implemented the inclass flip model to enhance speaking skills and the learning process of the fifthgraders at an Islamic elementary school. The data collection process went through observation, interview, and rubric. The study result shows a significant level of improvement in students speaking skills, especially ones covering the area of vocabulary, pronunciation, and structure. Through group activities in the class, students were reported more active, motivated, responsible, and collaborative. In addition to that, Pudin (2017) has found that his Indonesian EFL students were more engaged in a flipped classroom rather than traditional classes. He also pointed out that flipped learning improved students interaction and communication skills and motivation toward learning English language skills. Nevertheless, he admitted confronting challenges at the beginning of the program in which some students were uncomfortable due to unfamiliarity with this model of teaching.
Despite those positive trends, another study highlighted the ineffective use of FC in improving students' reading skills. Maharsi, Wijayanti, and Astari (2021) tried to examine the implementation of the FC approach in an EFL private university in Indonesia. They explored how this approach impacted students' reading comprehension and how students perceived the FC in their learning process. By a mixed-method design, the results indicated that students' achievement score in the traditional classrooms outweighed their counterparts in the FC. The success story of controlled group students was due to the benefit of teacher-led instructions and scaffolding. Although some students perceived FC as promoting independence, the drawback was related to their inconvenience in using technology in learning, taskrelated time management, and technology-related workload.
Concerning the advantages and disadvantages of FC, several pieces of literature around the globe have reviewed it from various perspectives. On the one hand, flipped learning provides students more time to practice their language through exercises and activities in the classroom (Lockwood, 2014). On the other hand, students are deemed unqualified to deal with technological devices for downloading videos or submitting some homework (Goodwin & Miller, 2013). Also, several problems were encountered, such as preparing the materials and designing the flipped classroom are always time-consuming (Lee & Wallace, 2018). Several factors also count, among which are a poor internet connection, low self-confidence and selfdiscipline toward learning to do the required works appropriately (Evseeva & Solozhenko, 2015), less practice and unfamiliarity with the strategy (Rintaningrum, 2018), and students' low efficacy and technical problems to use technology (Shih & Huang, 2019).
Concerning the double-edged-sword pieces of literature above, it is substantial to explore more on the effectiveness of the FC each time it applies to a new setting. This current study aims to evaluate the effect and acceptability of the pilot FC experiment in comparison with the blended learning for undergraduate students enrolling Reading for Academic Purpose Course in the English Department at a private university in NTT Province, Indonesia.

Method
This present study was a mixed-method study design. It combined quantitative and qualitative approaches for the broader purposes of a single study (Creswell, 2012). This study employed Pretest-Posttest Control Group Design (Sugiyono, 2016, p. 112;Phakiti, 2014, p. 67), in which participants were randomly assigned into two groups. They were pretested on the dependent variable, and then the experimental groups received the experimental treatments while the control group received a typical condition such as what is normally practiced in the classroom. At the end of the treatment period, participants in both groups were tested on the dependent variable.
The focus of this study is on impact analysis of the FC approach, therefore, class division to the experimental and control group was considered necessary. Quantitative data were collected from pre-test and post-test of the FC and BL. Meanwhile, qualitative data were gathered from students' TOWS (Threat Opportunity Weakness and Strength) analyses after the learning process to provide shreds of evidence on how students perceived the FC in reading comprehension. The study obtained approval from the faculty and informed consent from all students. It was conducted within two years from January 2020 to June 2021.
A total of 58 students were involved in the study. A total of 35 fourth semester students enrolling on the Reading for Academic Purpose Course were set in the Blended Learning classroom (control group) in 2020 and 23 fourth semester students in Flipped Classroom (experimental group) in 2021. The pre-test and post-test were administered in the form of combined reading texts of TOEFL-like Reading Comprehension section from Longman Complete Course book and Reading Section for IELTS from Cambridge English IELTS book. The 50 questions in each test comprised 40 multiple choice questions and ten short answer questions. The pre-test was administered at the beginning of the semester; the post-test was eventually at the end semester as the final test.
The results of both tests were analyzed through a t-test using the SPSS instrument. The comparisons of learning outcome ratings across tests and groups were analyzed using paired sample t-test and independent sample t-test. Students' TOWS analyses were discussed with the instructor and classified based on barrier and booster themes. The thematic analyses were presented descriptively according to the central needed issues.

Findings and Discussion
The Procedure of the Study Students in both settings were provided with essentially identical objectives and handouts before class. They were also taught and facilitated by the same instructor. In the BL setting, students undertook various activities in class, either mainly having lectures, working on assignments, or having case-based small group discussions. On the other side, students in the FC setting were reinforced with homework problems to discuss rather than lectures. The following table is the summary of the educational setting chronology provided for students in the BL and FC settings. Students worked on and submitted their homework through Moodle and MsTeams application. They also were engaged in online discussion/ chats.
Prior to the class, the students joined online lectures through MsTeams and watched instructional videos, read online materials, finished quizzes related to the information provided from the video lectures, and did targeted exercises. All content-based information was delivered outside of class through Moodle. Campus Students received a further explanation for the teaching material and exercises and had problem-or case-based small group discussions. The instructor validated correct responses and clarified any misconceptions.
Entire limited class periods were devoted to a small amount of fast evaluation. The students discussed their understanding of the materials learned at home and had peer checking and instructor's assistance on their learning process. They tried to solve the problem and did knowledge internalization. Home (online) The students worked on and submitted their homework through Moodle and Microsoft Teams apps and studied from provided online learning sources.
Prior to the class, the students joined online lectures, watched guiding videos, read online materials, and finished quizzes related to information from the video lectures and reading materials.
Campus Students were given post-test Students were given post-test TOWS analysis of FC

Statistical analysis
To examine the statistically significant difference between pre-test and posttest derived either from the FC group or BL group, a paired-samples t-test was employed. The researcher compared the pre-test and the post-test after experimental treatment. Tables 2 to 5 below present the SSPS outputs of the paired-samples t-test. The paired-samples t-test analysis above indicates a statistically significant difference between the pre-test and post-test scores in the FC set. It shows that t[22]=-3.387, p=0.003<0.05, and d=-0.70, medium effect size (Cohen's d provides further evidence that will allow us to claim the effect of the experimental treatment). The finding indicates that the flipped classroom approach moderately helped increase the students' academic reading performance. Furthermore, based on the paired-samples t-test analysis above, it was found that there was also a statistically significant difference between the pre-test and posttest scores in the BL setting. It shows that t[34]=6.273, p=0.000<0.05, and d=-1.06, large effect size (Cohen's d provides further evidence that will allow us to claim the effect of the experimental treatment). The finding indicates that blended learning considerably increased the students' academic reading performance.
After that, the researcher used an independent-samples t-test to determine if one mean score is significantly different from another by comparing two means from the FC and BL group. The statistical analyses are as follows. The table above identifies students' background reading skills from the two separate groups at a similar level (p=0.400>0.05, d=0.21). The Cohen's d effect size for the pre-test shows a small effect. Similarly, the result of post-test scores shows that students from those two separate groups gained similar academic reading achievement levels (p=0.271>0.05, d=0.29). The Cohen's d effect size for the pretest also shows a small effect.
Therefore, it clarifies that the data set violates all the statistical assumptions for the alternative hypothesis. Levene's test for equality of variances ascertains that both groups had equal variance. The post-test was not statistically significant (p=0.56). It discovers that was no statistically significant difference between the two groups (t[56]=1.11, p=0.27, d=0.29, small effect size). Since the Cohen's d effect size was small, the experimental condition of flipped classroom approach was not effective in improving students' academic reading skills. Based on this independent-samples ttest, it can be concluded that there is no statistically significant difference between the two groups after the experiment.

The Students' Perceptions toward the Use of Flipped Classroom Approach
Students' perspectives on the FC approach during the Covid-19 times were synthesized in the TOWS analysis. The analysis was discussed together with the instructor to help avoid the data from unnecessarily repeated themes. This might provide very few topics based on students' experiences and understanding. The thematic analyses of threats and weaknesses are made clear in the following barrier topic and the opportunities and strengths in boosters. From the table of TOWS analysis above, it can be discussed some barriers affecting students reading performance and boosters for better improvement of FC in EFL students' Reading for Academic Purpose Course, especially during pandemic critical time.

Barriers
The ongoing threat of the Covid-19 pandemic has affected students learning situations and conditions. Due to limited time for the face-to-face teaching-learning process, the University has established more hours for the online lecture at home. That inspired the instructor of the RfAP course to implement flipped classroom approach. The application of FC indeed left some issues to consider, especially related to students' experience and perceptions during pre-class and in-class learning. Those issues redefine as the barriers of the FC application.
The first barrier is students' resilience to change. Students' overdependence on lecturing classroom makes them believe that their 'first exposure' only gains through traditional direct learning. They assume traditional lecturing is better than an FC or other remote learning in terms of creating lively peers interaction, supporting mutual teamwork, building social-emotional relations, and improving learning outcomes. Therefore, some students find FC useless for their learning process. They agreed to the idea of having additional video lecturing and e-material to prepare them before the class, but they saw no meaning behind that pre-class activity. Some introverted students also experienced isolation and found it inconvenient to adapt to collaborative in-class learning within FC. For some students with low motivation, their learning paces differ from their counterparts. They were not in favour of video lecturing and e-materials as the primary sources of knowledge transferring rather preferred listening directly to the instructor's explanation in an in-class setting.
The second obstacle to reflect on is self-learning inability during pre-class activities. Not all students were independent and convenient in the FC approach, especially during pre-class. Several students were astounded by the intense preparation in the pre-class activities, such as joining an online meeting with the instructor, watching video lectures, and learning or reading materials. Due to inadequate preparation, they often misunderstood the materials, were less motivated in getting the work done and panicked about the workloads and deadlines of reading tasks. Furthermore, pre-class activities via Moodle as a preliminary learning approach become an extra burden for several students. They tended to get bored eventually and confused about 'what to do next' even the application provided instruction. Too many weekly instructions, videos, and scheduled reading tasks posted on the Microsoft Teams and Moodle application caused students to miss some of it and be absent in completing them. They presumed that tasks were a waste of time and did not improve their understanding of the topics due to the very high Standard English language used in those academic texts. Moreover, individual and time limited-based tasks were frustrating for few students who were used to being collaborative in learning or overdependence on teamwork.
The third barrier to consider is poor peer collaboration and communication with the instructor during in-class activities. Classroom interaction should be more productive if students could communicate their problems face-to-face with the instructor and record their queries during pre-class to reflect them to the instructor and colleagues to seek a possible solution. However, it was indeed difficult for them to follow the active learning in the class due to limited time (40 minutes) and overload courses in only one-day meeting for all courses they took (8-12 courses a day per week during covid-19). Also, students were not ready for the in-class activities because they had not watched the videos and completed the homework yet. An FC could be successful if clearer, consistent and constant communication intertwined between instructor and students (Hsieh, Wu, & Marek, 2016). Students, indeed, found it hard to communicate with the instructor due to unknown reasons. Also, limited in-class hours caused the class to advance sluggishly. Teachers had to wait for students to finish their peer or group projects in addition to teaching and demonstrating. They were baffled during discussions and stressed due to the limited time provided by the University for this new learning model.
Another facet to contemplate is technological setbacks and technical issues. This drawback includes technological device possession and poor, unstable internet connection. Few students in this course did not owe compatible mobile phones or computers for learning applications. As a result, not all high-quality video materials were compatible and easy to access. The instructor found it challenging to create good compatible videos for all device types in pre-class activity. Technology may be a down pit for several instructors, yet they must be deemed to learn new and wellfounded things for their class. In addition, students who were not tech-savvy often had learning problems under pre-class. Moreover, the lack of internet data credit for personal use became a pressing matter for poor students. They faced difficulties in joining the pre-class activities that relied heavily upon technology and internet access.
Last but not least are the social-environmental constraints during the Covid-19 pandemic. Students learning in the FC require more time, resources, and active participation to achieve teaching-learning goals. Nonetheless, the current outbreak of the pandemic made them stressed during learning and giving more priority to health and personal/ family needs. Further, inconsistent electricity power in the Regency brought serious drawbacks to students learning connectedness in pre-class at home. Instability of learning time during pandemic also limited their physical interaction within in-class projects and tasks.

Boosters
Regardless of the barriers above, some aspects do account for increasing the positive or desirable quality of the FC. Despite its lower impact on EFL students' academic reading comprehension, the first boost is the perceived usefulness. FC offers flexibility, accessibility, and efficiency in learning for most students. This method combines technology and face-to-face learning that allow students to be the centre of the learning process. It was able to develop students' attention and interest in learning. Concerning the novel application of the national curriculum in higher education, this approach successfully ignited students' interest in learning, supporting the freedom of learning pace that is up to the learner. The accessible way of learning regardless of location is also an inspiring output of this learning model. Several students in this course claimed that they had more active learning opportunities and responsibility for learning.
The second boost is that FC provides students with self-efficacy to learn during pre-class. Pre-class activities through Moodle application motivated them to commit to self-regulation in completing assignments. Students can also voluntarily manage their spare time for studying, reading, thinking, and reviewing preparatory materials before the in-class discussion. Addressing multiple learning platforms through Moodle and MsTeams, learning materials, resources, and tasks were uploaded online and this reduced students' reliance on teacher explanations as sources of knowledge. The lecturing videos provided was an advantage for them in which they had total control of playing, pausing, and rewinding the learning videos to match their paces of learning. The pre-class activity eventually activates students' background knowledge and English reading skills.
The following boost is that FC increases students' readiness and facilitates thorough discussion during in-class reading activities. Some students acknowledged the meaningful activities because they found the in-class time more focused on feedback rather than the material explanation. Also, with the materials learned beforehand, students could feel more confident and readier when joining the in-class learning. It forced their mind to think and reflect more for optimal potential. In the end, peer and instructor feedback helped them on their reading competence and understanding.
Another fact to ascertain is that FC boosts students to gain technological skills integrated into reading comprehension learning. As the impact of Covid-19 is growing, hi-tech devices can take a central part for students to follow the online lecture and to join online quizzes (Arnold-Garza, 2014). Students, indeed, generate numerous online learning sources for reading practice and test simulation across the internet for their improvement in this course. With online quizzes, students can easily access their scores, progress, and feedback from the instructor in the online grade book. Moreover, FC enhanced online social interaction, communication, and collaboration. Some students preferred using video or chat to communicate their reading problems and share their approaches to those problems.
The last extra boost is that FC supports government policy and university commitment to help reduce the spread of Covid-19. The University authorities had empowered lecturers with the capacity building to recreate a digital learning environment and resources during the pandemic time to prevent crowd and physically direct contact. Besides online learning, the University facilitated students with limited face-to-face meetings. As this direct learning model took little time only, flipped teaching is one of the best applications to cope with that. Moreover, it is costeffective because of no need for hard copy reading materials or paper distribution during discussion and tests, which eventually help avoid physical contact among students and instructors. More time allocated for pre-class activities also admit parental and social involvement in monitoring students learning and moves during the pandemic critical times.

Conclusion
This current study compares the effect of a flipped classroom with blended learning and explores the barriers and boosters of flip learning in the Reading for Academic Purpose Course in a private university in Indonesia. To attain such goals, two groups of the fourth-semester students, composed of 35 students in a blended learning setting and 23 others in a flipped classroom, were taught similar reading skills and tasks. The result of paired-samples t-test indicates that both FC (p=0.03<0.05, d=-0.70, medium effect size) and BL (p=0.00<0.05, d=-1.06, large effect size) have positive effect on students' academic reading performance. The independent-samples t-test used to compare their effectiveness shows no statistically significant difference between the two groups (t[56] =1.11, p=0.27, d=0.29, small effect size). Although FC students benefited equally from the BL learners, they gained a lower mean score than their counterparts. The TOWS analysis confirms that FC needs more development because of students' resistance to change, failure to selflearning management, poor collaboration and communication, technological setbacks, technical issues, and other socio-environmental constraints.