“Should I change myself or not?”: Examining (Re)constructed language teacher identity during the COVID-19 pandemic through text-mining

Abstract

On the basis of interviews with 20 Chinese EFL teachers and text-mining analysis, this study investigates the reconstruction of language teacher identity during the COVID-19 pandemic in terms of context, agency, and practice. The word co-occurrence analysis and collocation statistics with modal verbs coded by KH Coder 3.0 demonstrate that English teachers face two types of identity reconstruction processes: (a) situational context related to various changes and (b) interactional context that causes communication difficulties with students. How teachers enact agency is influenced by their teaching experience. Through practice, teachers reconstruct their identity as solution seekers, positive learners, and confident professionals.

1. Introduction

Since the turn of the 21st century, researchers have proposed a shift in the field of second-language teacher education from a product-oriented perspective to a process-oriented stance (Crandall, 2000). The latter perceives teachers as knowledge sources (Freeman, 2002), resulting in the strong attention toward Language Teacher Identity (LTI) research. LTI is directly connected to how teachers view themselves, particularly how they comprehend what they should learn and how they should behave as educators, which undermines language teaching methodology and philosophy (Kincheloe, 2004; Varghese et al., 2016).

A language teacher is not a neutral participant in the classroom, and the relationships between teachers’ and their students, and the teaching context are vital in understanding teaching practices (Varghese et al., 2005). After COVID-19 outbreak worldwide, establishments, including schools, were closed. This revolutionary change has made normalized distance and online learning teaching modes in all educational sectors, including foreign language education. In this new teaching circumstance, although teachers face greater challenges implementing inclusive education, they have the opportunity to reflect on how contextual changes influence their teaching beliefs and practices as well as teacher identity.

The current study enriches previous research (Zhang & Hwang, 2021), which revealed a significant statistical difference in LTI in the English as a Foreign Language (EFL) context before and after the COVID-19 outbreak. The study demonstrates that teachers strived to develop new teaching practices, shift their understanding of teaching, and reconstruct their identity as language teachers during the COVID-19 pandemic. On the basis of previous findings, this study aims to further investigate how LTI interacts with online teaching through an in-depth analysis of qualitative interviews using a text-mining approach. The following research question guides this study: How is LTI disrupted, challenged, and reconstructed during the COVID-19 pandemic in online teaching environments in terms of context, agency, and practice?

2. Literature review

2.1. Deconstruction of LTI

As the foundation for LTI, identity is considered the unique characteristic of an individual regarding the perceptions and characteristics of others (Pennington, 2015). These perceptions are determined by differences in individuals' physical characteristics, abilities and skills, societal position, and affiliations with different groups (Gee, 2000). Thus, identity encompasses more than an individual's physical attributes; it is, the sense that people have of themselves, including the self-image and self-awareness of the properties they take to construct who they are. This sense of self-image or self-awareness is based on an individual's values and beliefs regarding how people should conduct and behave in a given context (Richards, 2015). This implies that a person establishes their identity on the basis of a set of appropriate behaviors, and this self-perception may vary over time.

Although identity is deeply personal, it is closely related to the social context. According to Pennington and Richards (2016), identity reflects the context or activity in which an individual is situated. It can be understood as a type of self-performance within a specific context (Davies & Harré, 1990; Hall, 1997). Therefore, different teaching contexts or approaches may assume various roles for teachers, which may lead to conflict with their sense of identity (Pennington & Richards, 2016). For example, a teacher who is required to teach in a new environment may experience identity stress when adapting to the constraints of the new environment. Nonetheless, these constraints influence teachers to develop new teaching methods to address teaching problems. During this process, the teacher's identity is constructed in response to constraints.

2.2. Theoretical framework

Given the multiple, dynamic, and contradictory nature of LTI, this study adopts a post-structural perspective, which regards LTI as constructed and reconstructed through practice and discourse within a particular context (Varghese et al., 2005). On the basis of Trent's (2015) integrated framework this study analyzed how LTI was reconstructed through synchronous online teaching during the pandemic. Fig. 1 shows how Trent (2015) analyzes LTI from the perspectives of practice, discourse, and agency in intrapersonal, interpersonal and institutional contexts.

Fig. 1.

Integrated framework for investigating teacher identity (Trent, 2015).

Identity-in-practice refers to an action-oriented approach to understanding identity. It emphasizes the need to study identity as socially and historically constructed through concrete practices and tasks through three modes of belonging—engagement, imagination, and alignment (Trent, 2015). Engagement facilitates a community's recognition of an individual's competence, through which identity is partly formed. This engagement is crucial for language teachers to learning the teacher community (Danielewicz, 2001). Imagination is another powerful force in identity construction, which refers to the image creation of a world beyond the present time and space. Alignment means expanding an individual's activities to broader contexts (Trent, 2015). Teachers who experience professional identity conflicts during reforms must shift their identities (Bao & Feng, 2022; Huang et al., 2018; Liu & Xu, 2011).

For a multifaceted understanding of LTI, Varghese et al. (2005) called for attention to “identity-in-discourse.” The crucial role of discourse in identity construction lies in providing individuals with particular subject positions from which they “actively interpret the world and by which they are themselves governed” (Weedon, 1997, p. 93). According to Bucholtz and Hall (2005), naming identity categories by individuals is the most explicit way to discursively position oneself (e.g., new teachers, traditional teachers). Fairclough (2003) claimed that an individual's commitments to such positionings are significant identifiers that can be assessed on the basis of modality and evaluation. The use of modal verbs (e.g., should, must) and modal adverbs (e.g., probably, possibly) exemplifies modality, which describes what individuals commit to in terms of obligation and necessity. Trent (2015) explained that evaluation refers to the judgement of what is considered good or bad.

LTI is an ongoing and dynamic process from a post-structural view, which leads to research into teacher agencies. Teacher agency refers to people's ability to do things that affect their social relationships (Layder, 2006). A teacher's agency is reciprocally related to teacher identity. When teachers construct their identity in teaching contexts, they exercise their agency and take actions that they believe are associated with this identity construction. These actions, and how they are perceived by others, influence ongoing teacher identity construction. Therefore, Trent's integrated framework provides a theoretical lens to study the reconstruction of LTI during the pandemic.

2.3. Research of LTI before and after COVID-19

Studies on LTI before COVID-19 can be roughly divided into three phases chronologically. First, the study of second/foreign LTI did not begin until the late 1990s. Early studies mainly focused on how non-native speakers formed and negotiated their LTI in teaching contexts (Amin, 1997; Brutt-Griffler & Samimy, 1999; Pavlenko, 2003; Tang, 1997). For example, Pavlenko (2003) investigated the linguistic biographies of 29 teachers and argued for the complexity of the linguistic identities of non-native Teaching English to Speakers of Other Languages (TESOL) teachers. This research area primarily focuses the distinction between native and non-native speakers and the relevant influence on their professional identity construction (Braine, 1999; Liu, 1999). However, these early studies neglected the social aspects of language teachers. With in-depth research on non-native English teachers, scholars have found that teachers' social characteristics, such as their gender, race, and religion, are closely related to their professional identity (Kubota, 2001; Norton, 1997). Second, in the first decade of the 2000s, studies in this field expanded the focus by considering the sociocultural identities of language teachers, studying the interaction between language teachers' sociocultural identities and their professional identities (Lasky, 2005; Motha, 2006). To explore the development of teacher identity, sociocultural theory and a communities of practice framework are used (Miller, 2009; Tsui, 2007; Varghese et al., 2005). Additionally, under the influence of post-structural approaches, recent research mainly focused on the plurality and composite nature of LTI in various contexts (Ashton, 2022; Barkhuizen, 2017; Huang & Varghese, 2015; Metz, 2021; Vallente, 2020). As opposed to exploring a static and universal form underlying teachers' behavior, a post-structural approach to LTI searches for the fluid and complex framings of teacher identity (Block, 2007). Studies from a post-structural perspective associate teachers’ racial and cultural identities with their teacher identity, presenting LTI as multiple, dynamic, complex, and shifting.

The COVID-19 outbreak has necessitated online teaching. Consequently, teachers' expectations of being language teachers are challenged. This sudden shift has resulted in diverse LTI research approaches. Ashton (2022) studied the agency of four language teachers when teaching online amid the COVID-19 pandemic. Ashton (2022) also revealed that teachers exercise their agency in line with their teacher identity and called for teacher programs to prepare teachers for diverse teaching situations. On the basis of interviews with four university faculty members, El-Soussi (2022) uncovered that teachers experienced changes in teaching beliefs and practices due to the shift to online teaching. Yan and Wang (2022) also examined the transition of three Chinese, junior secondary school English teachers’ online teaching, presenting three stages of online shift and providing strategies to facilitate emergency online teaching. They contributed to understanding what and where teachers must learn in this revolutionary paradigm shift; however, studies of LTI in the COVID-19 pandemic era ignore the dynamic process of LTI transformation. To fill this research gap, this study aims to investigate how this process influences online teaching practice through teacher agency. To further the previous study, it also seeks to determine the influence of teaching experience on teacher identity reconstruction and online teaching practices.

The COVID-19 pandemic has increased the demand for educational technology in teaching. In online teaching environments, teachers who had previously taught in face-to-face settings had to adapt to new teaching methods and approaches while creating a new teacher identity (Fu & Zhou, 2020; Gao and Zhang, 2020, Zhang & Hwang, 2021). Therefore, how language teachers reconstruct their teacher identity along with the use of new technology in online teaching environments warrants further exploration.

3. Method

In this qualitative study, semi-structured interviews were conducted to gain an in-depth understanding of how English language teachers (re)construct their teacher identity, and how it influences their online teaching practices. On the basis of multi-dimensional framework of teacher identity, this study further divided contexts into three levels: micro, meso, and macro. At the micro-level of context, participants were asked to describe and reflect on how COVID-19 influenced their perspectives on English teaching. At the meso-level, participants discussed their relations with students, colleagues, and institutional policies. At the micro-level, participants recounted their beliefs about how English should be taught and reflected on the pedagogies they adopted in the pandemic era. The interview protocol is presented in Table 1 .

Table 1.

Interview protocol.

Context		Questions
Macro-level	Q1	How would you describe yourself as an English teacher before and after the pandemic? Has that description changed?
	Q2	Describe the most significant influence of COVID-19 on your teaching. Why do you think so?
	Q3	What were the most difficult aspects of teaching during the pandemic? What were the easiest? What do you think of these aspects now (after the pandemic)?
Meso-level	Q4	Describe a really good experience you had in online classes during the pandemic. How and why was it a good teaching experience?
	Q5	Describe a really bad experience you had in online classes during the pandemic. How and why was it a bad teaching experience?
	Q6	When teaching during the pandemic, did you receive any feedback on your teaching from students, colleagues, or others? What do you think of the feedback?
Micro-level	Q7	When teaching online, how do you think you present your unique style (your personality, management style)? What are the biggest differences between online and face-to-face teaching?
	Q8	What qualities do you believe contributed most to your success as an English teacher after the pandemic?
	Q9	Do you enjoy teaching online now? Why or why not?
	Q10	What two pieces of advice you would give to a new English teacher?

3.1. Participants

A purposeful sampling strategy was adopted. As part of a netnographic sampling strategy, an invitation was sent to three WeChat groups whose members had completed a previous online questionnaire for our study (Zhang & Hwang, 2021). Thirty teachers gave feedback in the affirmative. In total, 20 were interviewed for saturation, which means that the data collection stopped when the themes or categories were saturated (Charmaz, 2006). All 20 participants had no synchronous online teaching experience before the COVID-19 outbreak. By the time of the interview, 20 participants had one or two semesters of synchronous online teaching experience. Participants’ identities were protected by using pseudonyms selected from a list of common English names. Table 2 exhibits the detailed information of each participant, including years of teaching experience, current position, and educational background.

Table 2.

Interview participants’ information.

#	Pseudonyms	Years of Teaching	Title	Teaching Institution	Degree
1	Ada	2 years	Lecturer	University	Master
2	Fiona	3 years	Lecturer	University	Doctor
3	Tina	3 years	Lecturer	University	Doctor
4	Heidi	5 years	Lecturer	College	Master
5	Isabel	7 years	Lecturer	University	Master
6	Judy	8 years	Lecturer	University	Master
7	Rebecca	8 years	Lecturer	University	Master
8	May	9 years	Lecturer	University	Master
9	Nina	9 years	Lecturer	University	Master
10	Linda	10 years	Lecturer	High School	Master
11	Gloria	14 years	Lecturer	University	Master
12	Olivia	14 years	Lecturer	University	Master
13	David	15 years	Associate Professor	University	Doctor
14	Caroline	15 years	Associate Professor	University	Master
15	Emily	17 years	Associate Professor	University	Master
16	Kate	17 years	Lecturer	High School	Master
17	Peggy	17 years	Associate Professor	University	Master
18	Vicki	21 years	Professor	University	Master
19	Daisy	24 years	Professor	University	Master
20	Shirley	28 years	Associate Professor	University	Master

Given teaching experience, 20 participants fell into four groups: less than 3 years (1 participant), 3–5 years (3 participants), 5–10 years (6 participants) and more than 10 years (10 participants). Two participants were professors, 5 were associate professors, and 13 were lecturers. Among the 20 participants, 2 had doctoral degrees and 18 had master's degrees. As for their workplace, 18 worked in university or college and two taught in high school. Given that teaching experience is an important factor, it has been added at the end of each pseudonym in the results section.

3.2. Data collection and analysis

The interview was conducted during the fall semester of 2021. The interviews lasted from 35 to 40 min. To minimize language barriers, interviews were conducted in the participants' mother tongue, Mandarin Chinese. With the participants’ consent, the interview process was audio-taped, transcribed, and translated into English.

Three phases were included to guarantee translation validity. First, two translation applications were used: Google Translate and Baidu Translate. Second, the machine-translated version was revised, which was further checked by a professional English translator. Third, the translated drafts were sent to the participants to reduce any possible misinterpretation. The drafts were edited until an agreement was reached.

The data analysis included two steps. First, to obtain a holistic view of the participants, KH Coder 3.0, which supports English input and combines natural language processing techniques with R statistical software, was used. Specifically, word occurrence analysis was first used to identify the participants' attitudes toward the shift of the teaching mode. To avoid misunderstanding caused by negative words, words, such as hasn't, haven't, didn't, doesn't, and not are picked up. To further analyze differences among teachers with various teaching experiences, the correspondence of words was analyzed to identify different themes, which helped to determine whether participants were responding to one code idea similarly or differently. Next, content analysis was used to identify the relevant emerging themes which were coded on the basis of the revised multi-dimensional framework of teacher identity. Before the analysis, the recordings were reviewed to uncover missing details. Moving back and forth, data analysis was conducted in a recursive manner, from codes to the revised multi-dimensional theory of teacher identity construction.

4. Results

4.1. Reconstructed LTI in contextual changes

To obtain a holistic view of the interview data, the co-occurrence network of words was first analyzed using KH Coder 3.0. By seeing groups of frequently occurring words that are often used together, a word co-occurrence network enables researchers to read the main themes of the data. Fig. 2 displays different emerging themes of the interview.

Fig. 2.

Co-occurrence network of frequently occurring words in the interview.

In Fig. 2, two major emerging themes are identified. Network 1 presents teachers' attitudes toward teaching during the pandemic. The co-occurrence of excerpted words, such as job, work, understanding, and change, implies that due to the influence of online teaching, participants’ perceptions of English teaching shifted. This result is also illustrated by the responses to Q1 (How would you describe yourself as an English teacher before and after the pandemic? Has that description changed?).

“I think the pandemic has made many big changes to my career. One is the changes in teaching methods.” (David, 15 years)

“The epidemic has inevitably reformed and changed the teaching mode…I think online teaching during the epidemic has changed my way of thinking, including modern teaching methods.” (Daisy, 24 years)

“The epidemic has caused some changes in my understanding of work.” (Shirley, 28 years)

“After the epidemic, my understanding of my work has changed.” (May, 9 years)

“The epidemic has caused some influential changes in my understanding of work.” (Vicki, 21 years)

In addition to the five participants above, eight teachers directly expressed that the pandemic had brought great changes to their work and/or understanding of English teaching. Seven teachers, including Fiona (3 years), Gloria (14 years), Heidi (5 years), Judy (8 years), Linda (10 years), Olivia (14 years), and Peggy (17 years) directly claimed that the pandemic had little or no influence on their understanding of teaching. However, during the interview, the seven teachers emphasized the importance of lifelong learning for language teachers and the need to keep up with the times. Conclusively, three types of shifted perceptions were identified in this situational context: (a) teachers’ self-perception and perception about teaching, (b) decision-making, and (c) attitude toward learning.

Faced with dramatic changes, participants expressed their willingness to adopt to fit the change in terms of modality. In this study, modality was heard and displayed using three modal verbs: need, must, and should. The statistics of collocation were used to identify words that were strong related to the three model words. Fig. 3 exhibits that the words teacher, teachers, and teaching are the most frequently used words in association with need, must, and should. For example, the word teacher appears 17 times in a position one word (L1) before the model word should. By using these three modal verbs, the participants highlighted their commitment to fit to the new teaching environment.

Fig. 3.

Collocation Statistics with Modal Verbs (should, need, must).

Participants' commitment to fit the change was verified by the following excerpts: Ada (2 years) expressed her belief in teachers' ability to continue learning after the pandemic. She emphasized that the role of a teacher was to effect changes during the pandemic. The changing necessity of her teacher identity is discursively constructed by her use of the model verbs need and should. For Olivia (14 years) and Nina (9 years), teachers’ ability to use teaching technology is part of their professional knowledge which is a requirement and must constantly to be improved. David (15 years) and May (9 years) emphasized the role of teaching technology in increasing student support and engagement. This linguistic strategy found in the interview data suggests that the effect of positioning the change in teacher identity is a necessity.

“We need to change … Teachers should make timely adjustments to teaching in accordance with the requirements of the overall teaching context.” (Ada, 2 years)

“… the teacher’s knowledge should incorporate a wide range…teacher’s technology ability needs constantly to be improved.” (Olivia, 14 years)

“… we must keep pace with the times.” (David, 15 years)

“Teachers need to know the function of accessing information and sharing information through mobile phones, including some online software. Teachers need to learn them.” (May, 9 years)

Not all 20 participants expressed willingness to fit to the situational change. Six (Fiona, 3 years, Gloria, 14 years, Heidi, 5 years, Judy, 8 years, Olivia, 14 years, and Peggy, 17 years) presented their resistance and/or dislike toward online teaching. These six teachers also claimed that teaching during the pandemic had little or no influence on their understanding of being a teacher. Only Linda (10 years) expressed a willingness to accept online teaching, but she insisted that the pandemic had little effect on her understanding of teaching.

Another type of identity reconstruction process was identified in the interactional context. Specifically, Fig. 4 shows how participants reconstructed their teacher identity when interacting with students online.

Fig. 4.

Co-occurrence network of frequently occurring words in the interview.

In Fig. 2, Network 2 of the occurrence words, such as feel, think, online, classroom, biggest, difficulty, and interaction, indicated that interaction in the online classroom posed the greatest challenge for teachers during the pandemic. This result implies that LTI has been reconstructed and, disrupted in terms of the interactional context. This difficulty was frequently observed in their expressions.

“The biggest difficulty that online teaching has brought to me is interacting with students. Half of my students do not have technological support for online teaching …” (David, 15 years)

“Teaching is like a one-man show. I cannot receive feedback from students in time, and I am not sure what they are doing over there … It means that there is less interaction with students …” (Emily, 17 years)

“The biggest difficulty in online teaching is that there is no interaction between teachers and students. There was no enthusiasm for either side. It is difficult for students to reach the teacher …” (Kate, 17 years)

David has15 years of teaching experience. He expressed that communicating with students was his prime concern when teaching during the pandemic. He quickly found a lack Internet access was the main reason for the poor connection between him and his students. In addition, Emily (17 years) and Kate (17 years) mentioned that when teaching online the role of being an English teacher was different from teaching in the traditional classroom. Emily (17 years) felt that teaching online was a one-man show due to the lack of instant response from students, whereas Kate (17 years) found the teacher's and students' low enthusiasm hindered online classroom interaction.

In summary, networks of the word co-occurrence coded by KH Coder 3.0 and collocation statistics with modal verbs show that English teachers face two types of identity reconstruction processes: (a) situational context related to various changes and, (b) interactional context that causes difficulties in communicating with students. Contextual changes, especially regarding how to interact with students in online environments, cause tension in the ongoing process of teacher identity construction. Meanwhile, although some expressed their willingness, others showed resistance to contextual changes, which draws new attention to the role of agency in the possibility of meaning negotiation and identity reconstruction depending on different variables.

4.2. Reconstructed LTI in agency

A frequently used definition of agency refers to “the socioculturally mediated capacity to act purposefully and reflectively on [one's] world” (Rogers & Wetzel, 2013, p. 63). This definition emphasizes that agency is guided by contextual mediation and considers the interaction between personal and social factors. It plays an essential role in developing one's identity, and navigates agentic actions (Vitanova, 2004).

To further track differences in teachers’ reactions and attitudes among participants with various teaching experience and agentic behavior in a certain context, the correspondence of words was calculated via the Jaccard coefficient. The result categorizes the data into four groups as presented in Fig. 5 . This analysis used 129 frequently occurring words and filtered words through chi-square values to focus on 40 words whose frequency of occurrence varied significantly between groups. Uncharacteristic words uniformly found in the four groups were plotted near the origin (0,0) (e.g., online, teaching, teacher, class), whereas words with strong characteristics were located away from the origin (e.g., better, progress, voice, method, work, technology). This demonstrates that various themes emerged within the participants depending on their distinct teaching experiences.

Fig. 5.

Correspondence analysis of words in different teaching groups.

As presented in Fig. 5 and excerpts from the participants, teachers with less than 3 years of teaching experience showed a change in their belief in language teaching.

“Teaching in the pandemic changed my understanding of language teaching. I thought it could not be possible to teach a language online, but now I believe teachers can do it.” (Ada, 2years)

Teachers with 3–5 years teaching experience reported that voice connection posed the greatest challenge when interacting with students in online classes.

“I felt that this kind of voice communication decreases language classroom efficiency” (Tina, 3 years)

“Teaching English online is difficult especially for the speaking part. I need to listen to my students to assess their expression, but the voice connection costs more time” (Heidi, 5 years)

For teachers with 5–10 years of teaching experience, online teaching was a time for them to learn and advance. Adopting a teaching platform makes online teaching more convenient for them.

“My teaching entered a period of job burnout before the epidemic. I hope to improve myself through learning. The epidemic just provided me with such an opportunity to learn as a language teacher” (Isabel, 7 years)

“I had no online teaching experience before the pandemic. I would say that teaching online, especially teaching a language online, is a big challenge for me. I had lots to learn but it was the most thrilling thing for me.” (Linda, 10years).

Teachers with more than 10 years of teaching experience commented that an online teaching mode enriched their language teaching content. In addition, they expressed that combining online and offline teaching was better than teaching online or offline alone.

“Given the interaction with students, teaching English online is more difficult than a face-to-face classroom. Nonetheless, it did provide more convenience.

From the original traditional teaching model to online teaching, now I think the combination of them is better.” (Vicki, 21years)

“This teaching experience allowed me to reconsider the way of teaching English contents. I began to accept new possibilities, and I think the combination of online and offline would give insights into language teaching in the future because I can introduce language contents more easily via online and apply them in the classroom.” (Emily, 17 years)

Moreover, teachers with different teaching experiences enacted agentic behavior and found solutions in distinct ways. All four teachers with 2–5 years of teaching experience prioritized technological issues and classroom interaction.

“As a new teacher, teaching is not that easy, let alone teaching online. The most important thing for me is to make sure the synchronous teaching platform works well” (Ada, 2 years)

“I cannot get immediate response from my students. Group work did not go smoothly either. I spent lots of time figuring a better way to help students engage (Tina, 3 years).

Heidi argued that numerous issues, such as poor Internet signals and poor voice connection with students were beyond her ability. She mentioned repeatedly that an online teacher should attractive and supported by the institution's excellent infrastructure. Give that she could not change her physical appearance, or obtain supported infrastructure, she did nothing to deal with the technology issues during the pandemic. Heidi's agentic behavior is closely related to her LTI. When asked about her opinions toward teaching English online, she presented her resistance and expressed that, as a language teacher, it was not her responsibility to handle technological issues.

“I don’t like teaching English online. I think universities should assign professionals to handle the technological issues. This is not my responsibility. (Heidi, 5 years).

The other three teachers (Ada, 2 years, Fiona, 3 years, Tina, 3 years) made efforts to increase effective communication with students rather than relying on voice connection.

“We had a WeChat group as our backup channel if the online teaching platform broke down” (Ada, 2 years)

“I sent the materials to our teaching platform for students to download before class. When the Internet did not work, students could read and learn by themselves, and then we found some other time for make-up classes.” (Tina, 3 years).

Regarding the work plans, these four teachers expressed that they were provided specific teaching contents, but they had the freedom to rearrange or include additional teaching content. However, they did not modify or add teaching contents because they focused on how to complete the required teaching curriculum. In addition, organizing class activities and drawing students' attention in online classes were their most significant problems. For example, Ada (2 years) expressed that she spent most of her time designing classroom activities to capture students’ engagement. She rearranged the lecture time so that she had ample time to engage students in the online class, which was the most significant change she made compared to traditional classroom instruction.

“I did not consider showing my personal characteristics online. As I said, teaching English online meant that I could not get students' response especially when we practice listening and speaking. So, I reorganized the teaching time, leaving 15–20 minutes every class for students to interact with me and with each other” (Ada, 2 years).

Fiona (3 years) expressed her desire to finish the required teaching content because teaching English online decelerated the progress. Meanwhile, Tina (3 years), was inventive in her interaction with students, but she did not add extra teaching content or modify the work plan.

“I just taught step by step because the curriculum guided us what to teach. Compared to face-to-face classroom, I spent more time to ask students speaking English online. So, the teaching progress was slow.” (Fiona, 3 years)

“Online communication is really limited … I let the students type in the chat box…Students say that such interaction is like playing games…I did not add extra materials because it was not easy to finish the required teaching content. I focused on students' interactions through which I could know how much they had grasped some language points.” (Tina, 3 years).

Contrastingly, teachers with more than 10 years of teaching experience behaved differently in addition to the agentic behaviors mentioned above. Their actions included engagement with other teachers and utilization of technology (i.e., teaching platforms). Peggy (17 years), Emily (17 years), and Vicki (21 years) all emphasized the necessity to communicate with their colleagues. Peggy (17 years) claimed that communicating with colleagues helped to discuss and solve similar problems they encountered. Emily (17 years) collaborated with her colleague to make short videos presentations, which helped them rearrange their teaching plans.

“When I met teaching problems, I preferred to communicate with my colleagues. Sometimes we had similar problems, and discussing with them always enlightened me.” (Peggy, 17 years).

“We all found online communication was limited. To improve teaching efficiency and offered students more opportunities to speak English, my colleagues and I rearranged our teaching plan by recording grammar and vocabulary videos. (Emily, 17 years).

Engagement with colleagues significantly contributed to solving matters effectively. Through this engagement, teachers can “share problems, viewpoints, and ideas, working together towards solutions” instead of dealing with issues alone (Hunzicker, 2011, p. 178). In addition, rather than focusing on the disadvantages of teaching platforms, the teachers in this group effectively used online teaching platforms. For example, to better support the blended learning environment, Peggy (17 years), David (15 years), and Emily (17 years) used the surveillance system of the teaching platform to track students’ learning and encouraged them to keep pace.

“I tried to make better use of our teaching platform to overcome some teaching difficulties. For example, I used the surveillance system to track their learning pace and gave feedback on their assignments” (David, 15 years).

Overall, the majority of the participants with more than 5 years of teaching experience displayed a strong sense of enacting agency, especially on pedagogical changes. Comparatively, teachers with less than 5 years of teaching experience taught within more tightly organized work plans, lacking creative behaviors for alternative teaching activities. They enacted their teaching agency in two ways. First, teachers over 5 years of teaching experience used educational technology to implement the best pedagogy for their courses instead of “slapping classroom content online” (O'Neil et al., 2008, p. 18). Next, they made addition and/or modifications on the basis of their interpretation of students' requirements, providing a learning environment that accommodates students' learning styles. Additionally, they redesigned their teaching plans and tasks according to students' interest and abilities. All teachers' agentic behaviors are exercised within the spaces of the work plans. However, this space is far more limited for teachers with less than 5 years of teaching experience than for those with more than 5 years of teaching experience. In this study, how teachers enacted their agentic behavior was closely related to their LTI. How they perceived themselves as English teachers determined their teaching practice. Three transformation processes were identified as presented in the following section.

4.3. Reconstructed LTI in practice

Practice is crucial to understand how LTI is reconstructed during the pandemic. This section examines the transformation process of LTI from the three components of practice proposed by Wenger (1998): engagement, alignment, and imagination.

4.3.1. Engagement: from a “problem finder” to a “solution seeker”

Engagement entails teachers to establish and maintain community membership and negotiate meanings by connecting with other community members. Teacher identity is considered partly constructed through a teacher's relations with community members (e.g., colleagues, school authorities, learners, and mentors) which is manifested through a teacher's professional commitment in this study.

“The teaching effect is not as good as that in a traditional classroom which made frustrated. Later, I took some measures, including requiring students to send notes immediately after class.” (Daisy, 24 years)

“I mainly present classes through synchronous communication tools. There are also recorded videos … This saves plenty of time for our discussion.” (May, 9 years)

“Online communication is really limited … I let the students type in the chat box … The interaction with students is very good…Although students sometimes give wrong answers, I still feel very relieved and very happy.” (Tina, 3 years)

As presented in the above excerpts, Daisy (24 years) found that their teaching difficulty was caused by the low effect of online platforms. She uncovered that students were easily distracted in an online environment. Shen then provided a solution by requiring students to take notes or make video calls. Daisy (24 years) was fully aware of the pedagogical purpose of taking these measures, which can draw students' attention in online classes. May (9 years) used a different method to improve the low teaching efficacy. She used recorded videos with the synchronous teaching platforms to guide students’ independent learning. Tina (3 years) utilized the teaching platforms by using the chat box as a bullet screen. In modern-day China, the bullet screen is a popular online format for receiving real-time comments from viewers. This suggests that teachers can quickly identify their teaching difficulties and then make informed decisions to solve teaching problems. In this dynamic process, participants experienced a process from being a problem finder to a solution seeker.

In summary, engagement offers a perspective for investing what English language teachers do in online during a pandemic. Scholars argued that internal and external factors shape teachers' perceptions of being a teacher. Thus, teacher identity is constructed through the interaction between person and context, which manifests through a teacher's professional commitment (Kelchtermans, 2009; Olsen, 2008). In this study, teacher's professional commitment is explained as a process of participation in teaching, providing a lived sense of who they are in the online teaching environment. Problems of low teaching effect were first identified by teachers. To improve the online teaching effect, the participants used different pedagogical practices. This revealed that when faced with tension between teachers' professional competence and social construction, teachers reconstruct their identity as solution seekers.

4.3.2. Imagination: from a “passive adapter” to a “positive learner,”

Imagination is a creative process that transcends time and space, which enables individuals to reconstruct the dimensions of a community and self-images and envision new interpretations (Wenger, 1998). In this study, teachers' reflections encourage them to reimagine their pasts and reconstruct their teacher identities on the basis of the inspiration of teaching in the pandemic era. The following excerpt demonstrates how Caroline reconstructs her identity using technology.

“I had no choice [to teach online]. I did worry about teaching online before I started. But I also had a longing for online teaching … I hoped to improve myself by learning … The pandemic provided me with such an opportunity … I think in the future, we should combine online and offline teaching. We should never stop learning.” (Caroline, 15 years)

Caroline (15 years) reflected on her attitude toward online teaching before COVID-19. She accepted online teaching because she was forced to do so, and expressed her concerns. Nonetheless, her teaching experience and feelings later changed her attitude by emphasizing the significance of learning to her professional development. By envisioning the future of teaching, Caroline (15 years) recognized the role of online teaching and technology in transforming her teacher identity. Teaching online during the pandemic has helped her learn, improve, and become a positive learner.

Another participant, Emily (17 years), portrayed an image of belonging as illustrated in the following excerpt.

“I began to worry when teaching during the pandemic that if online teaching would affect the teacher’s dominant position … But now through online teaching, I realize … A teacher plays a guiding role, and the dominant position will not be affected … I feel that after the epidemic, requirements for teachers become higher … Both online and offline teaching skills are required.” (Emily, 17 years)

Imagination connects Emily's teaching experience with a new understanding of teaching in the future. She also expressed a promising direction for her professional commitment. Before teaching online, Emily (17 years) envisioned the problem of playing a teacher's role in online classes and then explained her understanding of a teacher's role and responsibility by reflecting on her teaching experience during the pandemic. After identifying the main problem of teaching online, Emily (17 years) explored new possibilities to tackle it. She envisioned and reflected on her future professional commitment. From this perspective, the process of imagination entails teachers' professional commitment and determination, which offers them new teaching possibilities.

Teaching during the pandemic has enabled English language teachers to see the direction of their future professional commitment through imagination lens. Through imagination, learner identity is strongly emphasized, especially the requirement for learning new educational technology. The combination of online and offline teaching modes is considered a new teaching trajectory for English language teachers in the future. Teachers' reflections on learning and their efforts to seek new teaching possibilities reveal that teaching during the pandemic is an ongoing learning process. This learning process concerns language teachers’ perceptions of what they want to be. Thus, the learning process involves teacher identity construction and reveals identity development from a passive adapter to a positive learner.

4.3.3. Alignment: from an “uneasy layperson” to a “confident professional”

Wenger (1998) defined alignment as the coordination of an individual's activities within broader contexts and practices. According to Wenger (1998), how people align themselves determines a sense of belonging, engendering identity formation. In this study, the research participants established personal and professional connections beyond the immediate teaching context and transcended them in macro contexts of participation. A sense of increased awareness of using teaching technology (such as online platforms) emerges in the interviews that account for the reconstruction of LTI. This is represented in teachers' reflections on the influence of online teaching. During the pandemic, teachers argued that mastering the new online teaching mode allows them to improve their professional profiles.

In the following excerpt, Gloria (14 years) reflected on teaching during the pandemic and expressed the necessity to learn new things, such as how to utilize education technology. She then commented on the effectiveness of combining online teaching and classroom teaching, highlighting its supervising and supplementary functions. The data suggest that Gloria (14 years) has developed a sense of learning new technologies and new teaching modes, which facilitated her own sense of professional alignment.

“After the epidemic, I found that teachers also need to learn new things. Teachers need to learn more online teaching methods … The combination of online and classroom teaching can more effectively improve the teaching effect …” (Gloria, 14 years)

The emphasis on learning new teaching technology is also referenced in the interview given by Ada and Isabel as shown in the following excerpt.

“I use this opportunity to consciously read some teaching-related theories and lectures.” (Ada, 2 years)

“I think it [computer] is of great help to my teaching because computers and cellphones are very close to students' life. For example, students use mobile phones every day, and now they can use them to learn more.” (Isabel, 7 years)

For Ada (2 years), the teaching experience allows her to learn, facilitating her own development. Ada used this teaching opportunity to practice the relevant teaching theories by communicating with colleagues. Isabel (7 years) also expressed her understanding of technology as reliable for her teaching. She explained that she used cellphones to teach because they were helpful and convenient for students’ learning.

The interview data above are a good source of explanation of professional alignment, which has provided insights into understanding the reconstruction of LTI due to the pandemic's influence. The evolution of language teachers' understanding of teaching beyond local teaching contexts enables the act of being English teachers to extend beyond a particular teacher community to encompass broader enterprises. The reason is because teaching during the pandemic creates new modes of communication for teaching.

5. Discussion

On the basis of interviews of 20 Chinese EFL teachers, the current study further explores how LTI is reconstructed through context, agency, and practice in the COVID-19 pandemic. It also demonstrates how teachers' teaching experience plays a vital role in teachers’ agentic behavior. This section discusses how our findings contribute new insights into the LTI field.

5.1. LTI and context

In the co-occurrence network coder by KH Coder 3.0, words, such as job, understanding, work, and change, are connected by lines to express participants' changing feelings about teaching in the pandemic era. This finding reveals that teaching during the pandemic altered teachers’ beliefs toward language teaching. It is in line with the poststructuralist perspective of teacher identity that teacher identity shifts across time and space (Beijaard et, al., 2004; Miller, 2009; Varghese et al., 2005). Difficulties such as online communication with students and technological issues, were the main teaching constraints. Faced with dramatic changes and unexpected difficulties, participants expressed their willingness to adapt to the change, which was heightened in terms of modality in the use of three modal verbs: need, must, and should. Two types of identity reconstruction processes were identified: situational context related to various changes, and interactional context that causes difficulties in communicating with students. However, not all the participants experienced changes in their teacher identity, and some participants, such as Heidi and Kate presented unwillingness to adapt to the changes. This demonstrates that changes in the teaching context do not always cause identity reformation (Vähäsantanen, 2015).

Another characteristic of teacher identity—being individually constructed is also revealed. This study uncovered that teaching experience is a personal factor that caused significant differences in teachers' communication with students, perceptions of online teaching and classroom activities. Teachers with 5–10 years of teaching experience used various applications most frequently. Based on the interview data, the correspondence of words for teachers with 2–5 years of teaching experience indicated that, the difficulty of teaching in the pandemic era and establishing a voice connection with students were their greatest challenges. Meanwhile, teachers with more than 5 years of teaching experience considered online teaching as an opportunity to learn and develop. Another important difference caused by teaching experience was found in teachers' classroom behaviors. These findings indicate that language teacher reconstruction during the pandemic era is related to teachers' teaching experience, even though all participants had no online teaching experience before the pandemic. The preceding findings regarding the first research question indicate that teacher identity is an integration of the teacher's inner and outer selves.

5.2. LTI and agency

What attitudes participants hold and how they behave are determined by teacher agency. Thus, teachers' agentic behaviors influence their online teaching practices. Although all the participants encountered technology issues, teachers with less than 5 years of teaching experience prioritized technological challenges. Teachers with more experience sought to develop creative pedagogies and were more concerned with the personal development of their students by focusing on making English lessons more enjoyable and meaningful. The pedagogical innovations of teachers with more than 5 years of teaching experience were not only confined to textbooks but also included teaching activities and class management. These findings are consistent with that of Beijaard et al. (2000), who found that over time, teacher identity shifts from being subject focused to pedagogical practices focused. Teachers' teaching experience is an essential factor for various agentic behaviors although all participants had no online teaching experience before the pandemic. This indicates that simply moving face-to-face teaching practices online was insufficient. Teaching during such emergencies requires teachers to practice pedagogical flexibility. Thus, future teacher programs must develop teachers’ capacity to comprehend the teaching context and draw on its characteristics to generate teaching pedagogies.

5.3. LTI and practice

In this study, teachers also reconstruct their teacher identity through practice. Teachers engaged in online teaching with digital technologies such as managing teaching systems, communicating with students, uploading resources, receiving feedback, and making decisions on classroom design. This contributed to the reconstruction of the language teacher's identity as a solution seeker. Imagination is the second mode of practice through which participants recognize the role of educational technology in transforming teacher identity. The process of imagination entails participants' professional commitment and decision making through which they can recognize the challenges and difficulties of teaching in online environments. Teachers' corresponding solutions and decisions are an important step in their future development of teacher identity as positive learners. Alignment also involves transcending teaching time and space into wider teaching contexts and practices. By reflecting on the influence of online teaching, participants aligned their practices of online teaching to their future professional commitment and teachers' learning program. This alignment is essential for teacher identity because it contributes to the construction of teacher identity not only as teachers performing specific teaching tasks in a particular time and place, but also as self-assured professionals.

6. Conclusion

This study further investigated how LTI is challenged and reconstructed in an online teaching environment during a crisis. These findings have important implications for online FFL teaching and teacher education. First, this study reveals that challenges for English language teachers are not only confined to technology skills, because pedagogical creation is needed. For example, as revealed in the interview, May (9 years) pre-recorded lectures and then used them with synchronous teaching platforms, similar to a campus-based flipped class. When schools shift to emergency remote teaching due to natural disasters or public health crises, minor advance warning is provided. Teachers are required to generate pedagogies depending on their local classroom contexts, students, and teachers themselves. Face-to-face instruction will remain an essential mode of education for the foreseeable future, but it will be necessary to provide multiple channels, such as blended learning and hybrid learning, to address not only extreme contexts, such as COVID-19, but also the interruptions that occur on a daily basis. In addition to educational technology knowledge and skills, teachers must also possess pedagogical flexibility. Moreover, future teacher programs need to develop teachers’ ability to make sense of different teaching contexts and draw on the characteristics of teaching contexts to generate various teaching pedagogies.

Second, activating and communicating with students is the greatest challenge for language teachers in online classes. The COVID-19 pandemic has become an unprecedented test for teacher-student relationships. Teachers and students have to readjust their communication without daily in-person interactions. Communication becomes more difficult, but not impossible, when we rely on digital technologies. The most universal communication method is to use online chat tools. As described by Tina (3 years), students appreciated having the opportunity to ask and answer questions by typing into the chat window instead of speaking in front of other students. This finding implies that communicating online can effectively enhance students' engagement because students feel less intimidated when typing a comment into the chat window than when speaking face-to-face. In addition, other communication strategies for teaching, such as using platforms for small group discussion and various activities, are supplementary for physical classroom teaching. For teachers, communication channels with students need to be well considered to support students’ learning. Therefore, this study also calls for further exploration of communication in online teaching environments or hybrid learning processes.

Third, this study found that teachers' personal characteristics—teaching experience played an important role in how teachers behaved in online teaching environments. Novice teachers are more concerned with managing classroom discipline, establishing good relationships with learners, and mastering subject knowledge as well as teaching methods. These findings support previous research (Tsui, 2003). One important step in teachers' professional development is to engage in the teacher community by collaborating with colleagues and other staff members. For novice teachers, the exchange of ideas can encourage them to maximize their teaching experience through cooperative work and pedagogical practices. Additionally, the development of teacher identity is a teacher-learning process. Participants emphasized the necessity of life-long learning for their professional development. Education evolves quickly with new educational technology and teaching modes, such as hybrid teaching and blended teaching, so the role of a lifelong learner helps teachers’ professional development. Teachers and educators should be encouraged and provided with opportunities to constantly challenge themselves to employ new methods in various teaching contexts.

Despite its significance, our study has two limitations. One is the lack of diversity among the participants. All participants in this study were Chinese EFL teachers. Studies involving more culturally diverse English teachers would be worthwhile. In addition, although we attempted to differentiate between teacher identity and LTI in this study, it was insufficient given that the COVID-19 pandemic creates a teaching context in general, warranting further research.

Funding

This work was supported by Hebei Academy of Educational Sciences: "14th five-year plan" of Hebei educational science [2205276].

Biographies

Li Zhang, Ph.D., received her Ph.D. in TESOL from Jeonju University in South Korea and is currently a lecturer of Department of College English at Hengshui University in China.

Yohan Hwang is an assistant professor of English Language & Literature at the Jeonju University in South Korea. He holds a Ph.D. in Language and Literacy Education with a specialization of TESOL and World Language Education from the University of Georgia.

Data availability

Data will be made available on request.