Assessing the Relationship between English as a Foreign Language (EFL) Teachers’ Self-Efficacy and Their Acceptance of Online Teaching in the Chinese Context

Abstract

This quantitative study investigated the relationship between EFL teachers’ self-efficacy and their acceptance of online teaching in the Chinese context. One online questionnaire, including a self-efficacy scale and behavioral intention scale, was administrated to 293 university-level EFL teachers in China. The descriptive results indicated that EFL teachers at Chinese universities possessed a positive intention to adopt online teaching and relatively high self-efficacy in embracing online teaching. Correlational analyses showed that all four aspects of the self-efficacy of EFL teachers—student engagement, instructional strategies, classroom management, and computer ability—had significant correlations with their acceptance of online teaching. Meanwhile, based on the results of multiple linear regression analyses, EFL teachers’ efficacy in student engagement and instructional strategies were predictors of their intention to teach online. Self-efficacy in classroom management and computer ability did not influence EFL teachers’ acceptance of online teaching. An in-depth understanding of the relationship between self-efficacy and EFL teachers’ behavior and intentions can allow teachers and educators to improve their self-confidence and engagement in online teaching at a pedagogical level and identify patterns in teacher efficiency with respect to language at a theoretical level.

1. Introduction

Online education contributed significantly to efforts to address educational disruption during COVID-19 [1] by maintaining the sustainability of teaching activities through secure and timely means [2]. During the epidemic period, ICT-based online teaching was not an option for providing excellent instruction, but rather a necessity [3]. However, this abrupt switch from conventional, in-person instruction to online instruction has impacted university teachers’ beliefs and their technical ability, since they perceive obstacles and challenges with respect to the internet and technology-based online instruction [4]. This is particularly true in the case of the second language teaching, where creating an interactive learning environment is critical for students’ language learning [5]. College teachers of English as a foreign language (EFL) in China are being encouraged by national policies to make use of advanced technology so as to create authentic, language-rich learning environments in their teaching practice [6]. However, insufficient technical proficiency, as well as the lack of confidence in technology use, has negatively affected EFL teachers’ efforts to embrace technology [7]. Previous literature has shown that Chinese EFL teachers were not enthusiastic about using technology in their classrooms [8]. Although technology has been used widely in various contexts with the appropriate resources and has been increasingly incorporated into instruction in China, the pandemic period was the first time that the internet and online platforms were suddenly transformed into the only teaching tools available to Chinese EFL teachers [9,10]. As a result of inadequate preparation for online teaching and training, these Chinese language instructors were faced with new challenges in developing their instruction skills in line with those of regular online courses [11]. Professional development improving teachers’ ability for online teaching is important. However, understanding and evaluating how these university language teachers perceive their abilities to teach in such a mandated online environment and how their perceptions affect the acceptance of online teaching are also equally important [12].

Self-efficacy is an individual’s belief in their competence when carrying out particular tasks [13]. Teacher self-efficacy is typically understood in the context of teacher education as a teacher’s assessment of their own ability to improve students’ academic outcomes, engage students, complete instructional tasks, and fulfill teaching objectives [14,15]. Teachers’ involvement and high self-efficacy support sustainable teaching and learning, both of which are the foundations of their long-term commitment and perseverance in carrying out educational activities [16]. Previous research has found that individuals who have higher teacher self-efficacy show a greater commitment to teaching and tend to adopt advanced technologies in order to facilitate their teaching [17]. The necessity of such a quality is even more pressing in the context where online teaching, as the mainstream method of delivery, is becoming increasingly popular [18].

Online teaching situates EFL teachers in more challenging settings, in which both teaching and learning depend heavily on technology. Therefore, the self-efficacy of EFL teachers may differ from that of traditional classroom-based instructors. Though teachers’ self-efficacy and its relationship with technology acceptance in various learning environments have been extensively studied in the past [19,20], most of the previous research was generally conducted in the traditional face-to-face teaching context. Considering that self-efficacy varies as the tasks and contexts change [21], the empirical literature on self-efficacy and its relationship with technology acceptance in online contexts is scarce. In particular, with respect to language education, there is a lack of recent literature on how these language teachers evaluate their capability in online teaching [12]. Such a gap is highlighted by the fact that online English language teaching and learning has increased due to COVID-19 [22].

Revealing the underlying structure of online teachers’ self-efficacy can help us to comprehend how it relates to other psychological factors. Furthermore, research on self-efficacy in online language settings can improve the efficiency of online instruction and advance the long-underutilized field of teacher psychology [15,23]. Therefore, the current study aims to assess whether language teachers’ self-efficacy is related to their acceptance of online teaching by examining how Chinese EFL teachers perceive their ability to teach online. Specifically, in this study, Chinese EFL teachers’ intention to teach online and four aspects of self-efficacy are measured, namely, students’ engagement, instructional strategies, classroom management, and computer skills. Accordingly, the study will address the following research questions:

• Q1: To what extent do EFL teachers experience online self-efficacy?
• Q2: To what extent do they accept online teaching?
• Q3: Is EFL teachers’ self-efficacy associated with their acceptance of online teaching?
• Q4: Which is the best predictor of EFL teachers’ acceptance of online teaching in relation to the specific aspects of self-efficacy: students’ engagement, instructional strategies, classroom management, or computer skills?

2. Literature Review

2.1. Self-Efficacy and Teacher Self-Efficacy

Self-efficacy, first originating from Bandura’s “social cognitive theory”, is defined as an individual’s belief in his or her capability to carry out given tasks [13]. Self-efficacy was found to be significant because those who possess it are more motivated to persevere when facing negative outcomes or expectations and to try to improve the working environment [24,25]. On the other hand, those with low self-efficacy are less likely to persist in comparable situations and tend to feel gloomy and experience pessimistic feelings. Self-efficacy primarily influences people’s ideas, feelings, motivation, and behaviors [26] owing to its capacity to predict behavior [27]. Performance experience, verbal persuasion, vicarious experience, and physiological states are the four main sources of self-efficacy [24]. These four categories are acknowledged as the essential components in the development of self-efficacy, and it is the interaction between these variables that significantly influences whether one believes in others, in general, or not.

In the context of education, a teacher’s self-efficacy refers to their perceived confidence in their ability to successfully conduct activities related to teaching and learning, such as enhancing students’ academic performance, managing the classroom, and engaging students [14,28]. High-self-efficacious teachers are more likely to develop positive teacher–student relationships, be adept at handling problems in the classroom [29], offer more professional commitment [16], or actively integrate technology into their teaching practices [30]. Teacher self-efficacy is acknowledged as a mechanism that activates the process of teaching and learning [18]. However, teacher self-efficacy varies based on the specific tasks and domains, as a multidimensional construct [15]. For instance, language teachers’ self-efficacy ranges from language error correction to language teaching skills [31]. It was found that language teachers possessed higher self-efficacy in instructional strategies than in motivating students to learn English [32].

2.2. Teacher Self-Efficacy and Technology Acceptance

Technology-enhanced learning differs from traditional, physical classroom instruction in fostering cooperation, improving interactive communication, and increasing access to educational materials [33]. Teachers need to possess the skills and knowledge necessary to successfully incorporate advanced technologies into their teaching practice in order to take advantage of the benefits of information and communication technology [34]. However, research has revealed that teachers lack the confidence or perceived ability to instruct by utilizing new technology in classrooms and do not hold a positive view of their own efficacy as online teachers [35]. Teachers with little experience in online teaching, particularly during the COVID-19 epidemic, lacked confidence in their ability to utilize cutting-edge technologies and encountered difficulties during the unanticipated switch to online instruction [36]. The growing use of online teaching emphasizes the significance of teachers’ self-efficacy [37], and this appears to be a foundation for effective online instruction [38].

Prior research has demonstrated a relationship between self-efficacy and technology adoption [39,40] (Sun and Chen, 2016; Corry and Stella, 2018). Teacher self-efficacy, in particular, is positively correlated with technology and internet-enhanced teaching, and it can greatly drive the adoption of new technologies [41,42]. For instance, a statistical association between the use of information and communication technology and the self-efficacy of EFL teachers was discovered in the study of [43].

2.3. Online Teaching Self-Efficacy

Based on Bandura’s work [13], Tschannen-Moran and Woolfolk Hoy [28] developed the Teacher Self-Efficacy Scale (TSE) to examine teacher efficacy with acceptable internal consistency. The scale consists of three dimensions, namely, efficacy in student engagement, efficacy in instructional strategies, and efficacy in classroom management. By rewording the original 24 items and including a set of 8 new items that capture the characteristics of online teaching, Robinia and Anderson modified the TSE in 2010 to measure online TSE. The modified scale, “The Michigan Nurse Educators Sense of Efficacy for Online Teaching (MNESEOT)” [44], was designed to explore the factors that influence nurse faculties’ self-efficacy and engagement in online settings. By including self-efficacy in computer ability in the previous TSE, the modified scale, with 32 items, eventually included four aspects, namely, efficacy in student engagement, efficacy in instructional strategies, efficacy in classroom management, and efficacy in computer ability.

Self-efficacy in student engagement refers to the teachers’ beliefs in their ability to engage students in learning. Factors such as teaching experience and online teaching skills were demonstrated to have an influence on teachers’ self-efficacy in student engagement [45,46]. The instructional strategies refer to tactics and techniques that teachers employ to deliver their courses. Some instructional practices, such as enhanced student involvement and flexible teaching and assessment policies, have been documented as effective in online modes of teaching in the literature [47]. Teachers who use online teaching methodologies effectively delivered high-quality instruction, resulting in a thriving classroom environment [48]. Teachers’ belief in their capacity to arrange and conduct activities and behaviors that lead to a positive learning environment are referred to as self-efficacy in classroom management. At the level of classroom management, it is widely accepted that successful teaching and learning cannot take place in a poorly managed classroom, emphasizing the significance of classroom management techniques as a prerequisite for academic achievement [49,50]. Several studies have found that teachers with stronger self-efficacy can better manage their classrooms [50], while few researchers have attempted to investigate the same factors in an online setting [51]. Self-efficacy in computer ability mainly refers to teachers’ confidence in their capacity to use computers and technology. As an emerging concept, the self-perception of teachers in regard to their ability to use the internet and other types of educational technologies is worth noting [52]. Improving teachers’ digital competencies can help us to respond to the growing requirement for technologically adept teachers [40,53].

Based on the statements mentioned above regarding self-efficacy, self-efficacy, in the context of this study, refers to the EFL teachers’ self-assessed confidence in their capabilities in online teaching, specifically in student engagement, instructional strategies, classroom management, and computer ability in online contexts. Given that the current study is designed to foster a better understanding of EFL teachers’ self-efficacy and their online teaching acceptance, as well as their relationships, the following framework (Figure 1) was proposed to guide the study:

3. Methodology

3.1. Research Approach

The current study was conducted by a coefficient correlation analysis. This approach was suitable, since it enabled us to investigate whether any significant correlation existed between the four aspects of self-efficacy and the acceptance of online instruction. Then, to examine specific EFL teachers’ self-efficacy as a predictor of EFL teachers’ acceptance of online teaching, a multiple linear regression analysis was carried out. The study was part of a research project exploring the predictors of EFL teachers’ acceptance of online teaching in China, and only the findings regarding teachers’ self-efficacy in online teaching were reported.

3.2. Participants

The target population of the study was a total of 2235 EFL teachers registered in the academic year of 2021–2022 at public universities in Henan province, China. The formula of Cochran [54] was adopted to estimate the sample size, and the minimum sample size of 239 from the population of 2235 was considered sufficient for the study. A proportional stratified cluster sampling technique was employed to ensure that the sample represented the target population to the maximum extent. After the sample totals were calculated for the respective types of universities according to the size of each stratum or subgroup, a random cluster sampling technique was carried out to obtain participant samples. Random cluster sampling is a means of selecting samples from clusters rather than single-unit elements. The cluster unit was each university within the stratified types of universities. Then, the cluster universities were randomly selected from among all the universities using the fishbowl method. A total of 307 online questionnaires were returned, which met the minimum requirement of a sample size of 239 based on Cochran’s suggestion [54]. A total of 14 questionnaires were deleted due to the answering time being less than 200 s or incomplete information, thus reducing the final number of samples to 293.

3.3. Questionnaire

For this study, a two-part online questionnaire was developed. The first section dealt with the teachers’ self-reported demographic data (gender, age, years of teaching, academic title, and highest degree). As shown in

Table 1. The demographic information on the participants.

		Frequency	Percent (%)
Gender	1 Male	61	20.8
	2 Female	232	79.2
Age	1 25–35	68	23.2
	2 36–45	141	48.1
	3 46–55	58	19.8
	4 Over 56	26	8.9
Teaching Year	1 Under 5 years	42	14.3
	2 6–10 years	41	14.0
	3 11–15 years	68	23.2
	4 16–20 years	54	18.4
	5 Over 20 years	88	30.1
Title	1 Teaching assistant	32	10.9
	2 Lecturer	133	45.4
	3 Associate Professor	96	32.8
	4 Professor	32	10.9
Highest Degree	1 Bachelor’s	26	8.9
	2 Master’s	228	77.8
	3 PhD	39	13.3

, 20.8% (n = 61) male teachers and 79.2% (n = 232) female EFL teachers were involved in this study, which indicated a higher percentage of female teachers. The ages of most of the respondents fell in the 36–45 group, accounting for 48.1% (n = 141) of the total participants. Regarding the number of teaching years, most respondents were in the group with over 20 years’ experience, which accounted for 30.1% (n = 88) of the whole population. In terms of the academic title, about 11.0% (n = 32) of the respondents were teaching assistants and professors, 45.4% (n = 133) of the respondents were lecturers, and 32.8% (n = 96) of the respondents were associate professors. Regarding the highest received educational degree, 77.8% (n = 228) of teachers were master’s degree holders, which was the group with the highest percentage among all the participants.

The second part consisted of two instruments, namely, the scale of Behavioral Intention in Online Teaching and the scale of EFL teachers’ Self-Efficacy in Online Teaching. Though these two instruments were adapted from two previous instruments with an adequate reliability and validity, the content validity of the scales was still examined by a panel of content experts before we administered the questionnaires. This verification by two educational technology experts allowed us, the researchers, to evaluate the clarity of the items offered in the survey instrument, determine whether the instrument was capturing the required phenomena, and ensure that no important components had been overlooked. The feedback from the experts was used to adjust, refine, and improve the experimental scales. Some scales were slightly modified, accordingly, to gain a better understanding of the selected samples. A detailed description of the two instruments is presented below.

3.3.1. The Scale of Behavioral Intention in Online Teaching

In this study, the scale of Behavioral Intention in Online Teaching was adopted from Venkatesh et al. [55] in order to measure the degree of EFL teachers’ willingness to conduct online teaching. The score of a 5-point Likert-type scale ranged from “strongly disagree”, with a score of 1, to “strongly agree”, with a score of 5. The respondents needed to rate the extent to which they agreed with each statement. The total score of a respondent was obtained by summing all the responses. The scores on the scale demonstrate the extent of the willingness of an individual to conduct online teaching. Therefore, the higher the scores are, the more willing the teachers are to conduct online teaching.

The Cronbach’s alpha value of the Behavioral Intention scale was 0.90 in the adopted study [55], which indicated the good internal consistency of the instrument based on the suggestion of Hair et al. [56]. To ensure the internal consistency of the factors in the actual study, Cronbach’s alpha was tested, giving the value of 0.91 (shown in

Table 2. Reliability statistics of the instruments.

		Cronbach’s Alpha
	Subscales	Adopted	Tested
Behavioral Intention		0.90	0.91
Self-Efficacy		0.93	0.70
	Student Engagement	0.93	0.92
	Instructional Strategies	0.94	0.79
	Classroom Management	0.93	0.71
	Computer Ability	0.86	0.83

), which indicated that the scale was highly reliable.

3.3.2. The Scale of Online Teachers’ Self-Efficacy

The Michigan Nurse Educators Sense of Efficacy for Online Teaching Scale (MNESEOT) [45], originally developed by Tschannen-Moran and Hoy [28], was used to assess EFL teachers’ self-efficacy in online teaching. There were a total of 32 items for the instrument. Participants were instructed to indicate their degree of self-efficacy for each question by marking their answers, ranging from “nothing”, with a score of 1, to “a great deal”, with a score of 5. The higher the score was for each item, the greater the EFL teachers’ self-efficacy was for each statement.

The reliability coefficients of the MNESEOT [45] were high for each sub-scale, including online student engagement (0.93), online instructional strategies (0.94), classroom management (0.93), computer ability (0.86), and overall self-efficacy (0.93). In this study, the actual values of Cronbach’s alpha ranged between 0.70 and 0.92 for each sub-scale and the overall scale, which indicated that all the scales in the study possessed adequate internal consistency based on the criteria of Hair et al. [56]. The reliability coefficients of the original and actual instruments for each subscale are summarized in Table 2.

3.4. Data Collection

Online questionnaires were administrated to EFL teachers in order to collect quantitative data due to the influence of COVID-19, as well as the advantages of e-questionnaires. Before the questionnaires were administrated, our application for ethical clearance for the research was approved by the Ethics Committee for Research Involving Human Subjects of the University. Additionally, permission for data collection was obtained by the relevant personnel of the target universities in China.

An e-questionnaire was sent to teachers via a link or a quick response (QR) code. The respondents were informed about the purpose of the questionnaire, the necessity of the respondents’ honest participation, and the submission of the questionnaire. The respondents could fill in the questionnaire at any time and in any way that they preferred, such as by phone, laptop, or computer, as long as it was completed within two weeks. One submission criterion was established to ensure the objectivity of the data: the same respondent could not submit a questionnaire from the same IP address twice. A reminder was expressed to the participants that their participation was voluntary and could be withdrawn from the study at any time. In addition, another extremely important aspect expressed to all the teachers was that their responses were anonymous, as we assigned numbers to the returned instruments, and all the data were confidential and used only for the research study. At the same time, the teachers were informed that the findings of the research would be offered to them if they were interested in the study. To encourage participation, any participant who submitted the questionnaire would receive a small gift for participating.

The entire data collection lasted for about three weeks in early 2022. A total of 307 completed online surveys were submitted, which met the minimum requirement of the sample size of 239. Because 14 questionnaires were invalid due to the answering time being less than 200 s and incomplete information, a total of 293 final questionnaires were included in the preliminary analysis.

3.5. Data Analysis

In this study, all the statistical analyses were completed using the IBM Statistical Package for the Social Sciences (SPSS) 25. After entering all the data into SPSS 25, the exploratory data analysis was carried out, firstly, in order to identify outliers, the normal distribution, coding issues, or missing values, and to determine whether the assumptions regarding the statistics were met for the current research. Descriptive statistical analyses of the EFL teachers’ self-efficacy and behavioral intention were performed to address research questions 1 and 2 after all the statistical assumptions were met. In terms of research question 3, Pearson’s product–moment correlation was used to assess the relationship between the EFL teachers’ self-efficacy and their acceptance of online teaching. As for research question 4, multiple linear regression was performed to determine the best predictors of the EFL teachers’ acceptance of online teaching using the self-efficacy subscales.

4. Findings

4.1. Descriptive Statistical Analysis of Teachers’ Acceptance of Online Teaching (Q1)

Table 3. Descriptive statistical analysis of EFL teachers’ acceptance of online teaching.

Items	Percent (%)					M	S.D.
	(f)	(f)	(f)	(f)	(f)
	SD	D	N	A	SA
I intend to conduct online teaching in my teaching practice in the future.	2.0 (6)	6.1 (18)	32.1 (94)	48.5 (142)	11.3 (33)	3.61	0.84
I plan to conduct online teaching as much as possible in my teaching practice in the future.	3.4 (10)	18.4 (54)	45.4 (133)	26.3 (77)	6.5 (19)	3.13	0.91
I will conduct online teaching regularly in my teaching practice in the future.	2.7 (8)	10.6 (31)	28.0 (82)	50.2 (147)	8.5 (25)	3.51	0.89
I will conduct online teaching frequently in my teaching practice in the future.	3.8 (11)	24.6 (72)	41.6 (122)	23.9 (70)	6.1 (18)	3.04	0.94
I will suggest that my colleagues conduct online teaching in their teaching practice.	3.1 (9)	18.4 (54)	44.4 (130)	28.3 (83)	5.8 (17)	3.15	0.90
I would like to conduct online teaching rather than face-to-face teaching in my teaching practice.	9.6 (28)	40.6 (119)	31.7 (93)	14.0 (41)	4.1 (12)	2.62	0.98
Overall						3.18	0.76

SD = strongly disagree; D = disagree; N = neutral; A = agree; SA = strongly agree; S.D. = standard deviation; M = mean; f = frequency.

presents the frequencies and percentages, along with the means and standard deviations, of the participant’s responses concerning their acceptance of online teaching. The overall mean score of the teachers’ intention to teach online was 3.18, which was slightly above the scale’s midpoint (3.0). It indicated that the EFL teachers possessed a moderate intention to accept online teaching.

More than half of the participant EFL teachers strongly agreed or agreed that they intended to conduct online teaching in their teaching practice (59.8%, n = 175) and conduct online teaching regularly (58.7%, n = 172). About one-third of the teachers agreed or strongly agreed on the items stating their intention to conduct online teaching as much as possible in their teaching (32.8%, n = 96), carry out online teaching frequently (30%, n = 88), and suggest that their colleagues perform online teaching (34.1%, n = 100). It was interesting to find that a small portion of the teachers (18.1%, n = 53) preferred to conduct online teaching, while more than a half of the participant teachers (51.2%, n = 147) disagreed or strongly disagreed that they would conduct online teaching rather than face-to-face teaching.

The mean score of each item of the variables is presented in Table 3. With the highest mean of 3.6 (S.D. = 0.84), the item “I intend to conduct online teaching in my teaching” suggested a positive intention of the teachers to teach online in their teaching. The second-highest mean score (M = 3.5, S.D. = 0.89) was attributed to the item indicating that teachers preferred to have a regular online class, followed by the item expressing an intention to “conduct online teaching as much as possible” (M = 3.1, S.D. = 0.91), “conduct online teaching frequently” (M = 3.0, S.D. = 0.94) and “suggest online teaching to colleagues” (M = 3.2, S.D. = 0.90). The item expressing an intention to “conduct online teaching rather than face-to-face teaching” obtained the lowest mean score of 2.6 (S.D. = 0.98), indicating that the teachers preferred offline teaching compared with online teaching.

4.2. Descriptive Statistical Analysis of EFL Teachers’ Self-Efficacy (Q2)

There were four subscales of teachers’ self-efficacy, including self-efficacy in student engagement, self-efficacy in instructional strategies, self-efficacy in classroom management, and self-efficacy in computer ability. The teachers’ responses to the questions regarding their self-efficacy are presented in

Table 4. Descriptive statistical analysis of EFL teachers’ self-efficacy in online teaching.

Items	Percent (%)					Mean	S.D.
	(f)	(f)	(f)	(f)	(f)
	Nothing	A little	Some	Quite a Bit	A Great Deal
(SE11) How much can you do to help your students think critically in an online class?	1.7 (5)	15.7 (46)	66.2 (194)	14.7 (43)	1.7 (5)	2.99	0.66
(SE12) How much can you do to motivate students who show little interest in online work?	1.4 (4)	17.4 (51)	64.2 (188)	15.4 (45)	1.7 (5)	2.99	0.67
(SE13) How well can you structure an online course that facilitates collaborative learning?	3.8 (11)	14.3 (42)	56.3 (165)	21.5 (63)	4.1 (12)	3.08	0.82
(SE14) How much can you do to get students to believe that they can do well in an online class?	1.7 (5)	16.7 (49)	58.7 (172)	20.1 (59)	2.7 (8)	3.05	0.74
(SE15) How much can you do to control disruptive behavior in an online environment?	1.4 (4)	18.1 (53)	59.7 (175)	17.4 (51)	3.4 (10)	3.03	0.74
(SE16) How much can you do to help students value learning online?	1.4 (4)	12.3 (36)	61.1 (179)	23.5 (69)	1.7 (5)	3.12	0.68
(SE17) How much can you do to foster individual student creativity in an online course?	1.7 (5)	17.7 (52)	62.5 (183)	16.7 (49)	1.4 (4)	2.98	0.68
(SE18) How much can you do to improve the understanding of a student who is failing an online class?	1.4 (4)	8.9 (26)	59.4 (174)	27.6 (81)	2.7 (8)	3.22	0.70
(SE21) How much can you gauge students’ comprehension of what you have taught in an online course?	17.4 (51)	19.5 (57)	21.2 (62)	19.8 (58)	22.2 (65)	3.10	1.40
(SE22) To what extent can you provide an alternative explanation when students in an online class seem to be confused?	0.7 (2)	1.4 (4)	14 (41)	51.9 (152)	32.1 (94)	4.13	0.75
(SE23) How well can you structure an online course that provides good learning experiences for students?	0.3 (1)	2.4 (7)	15.7 (46)	54.9 (161)	26.6 (78)	4.05	0.74
(SE24) How much can you adjust your online lessons for different learning styles?	0.7 (2)	1.7 (5)	13.3 (39)	47.4 (139)	36.9 (108)	4.18	0.78
(SE25) How well can you respond to difficult questions from online students?	2.4 (7)	10.6 (31)	21.8 (64)	47.4 (139)	17.7 (52)	3.68	0.97
(SE26) How well can you craft questions or assignments that require students to think by relating ideas to previous knowledge and experience?	0.3 (1)	0.7 (2)	18.1 (53)	48.1 (141)	32.8 (96)	4.12	0.74
(SE27) How much can you do to use a variety of assessment strategies for an online course?	0.7 (2)	1.0 (3)	15.7 (46)	50.2 (147)	32.4 (95)	4.13	0.75
(SE28) How well can you provide appropriate challenges for very capable students in an online environment?	0.3 (1)	2.7 (8)	19.1 (56)	48.1 (141)	29.7 (87)	4.04	0.79
(SE31) How well can you develop an online course that facilitates student responsibility for online learning?	2.4 (7)	19.8 (58)	60.4 (177)	15.7 (46)	1.7 (5)	2.95	0.72
(SE32) To what extent can you make your expectations clear about student behavior in an online class?	1.0 (3)	21.5 (63)	56.0 (164)	19.8 (58)	1.7 (5)	3.00	0.72
(SE33) How well can you respond to defiant students in an online setting?	9.2 (27)	18.4 (54)	35.5 (104)	24.6 (72)	12.3 (36)	3.12	1.13
(SE34) How well can you establish routines (facilitate or moderate student participation) in coursework so as to keep online activities running smoothly?)	7.2 (21)	23.5 (69)	27.6 (81)	27.6 (81)	14 (41)	3.18	1.15
(SE35) How much can you do to get through to disengaged students in an online class?	11.6 (34)	19.8 (58)	30.4 (89)	22.9 (67)	15.4 (45)	3.10	1.22
(SE36) How much can you do to get students to follow the established rules for assignments and deadlines during an online class?	0.3 (1)	2.7 (8)	46.8 (137)	43.3 (127)	6.8 (20)	3.54	0.68
(SE37) How much can you control students dominating online discussions?	1.0 (3)	5.8 (17)	61.1 (179)	28.7 (84)	3.4 (10)	3.28	0.67
(SE38) How well can you establish an online course (e.g., convey expectations, standards, course rules) for each group of students?	1.0 (3)	5.5 (16)	61.1 (179)	29.4 (86)	3.1 (9)	3.28	0.66
(SE41) How well can you use the computer for word processing, internet searching, and e-mail communication?	0.7 (2)	3.4 (10)	35.5 (104)	46.1 (135)	14.3 (42)	3.70	0.78
(SE42) To what extent does your comfort level with computers facilitate participation in online teaching?	0.7 (2)	2.7 (8)	42.7 (125)	44.7 (131)	9.2 (27)	3.59	0.72
(SE43) How well can you navigate the internet to provide links and resources to students in an online course?	1.4 (4)	1.7 (5)	37.9 (111)	44.7 (131)	14.3 (42)	3.69	0.79
(SE44) How well can you navigate the technical infrastructure of your institution to successfully create an online course?	2.0 (6)	10.9 (32)	48.1 (141)	32.8 (96)	6.1 (18)	3.30	0.82
(SE45) How well can you navigate the technical infrastructure of your institution to successfully teach an established online course?	1.4 (4)	7.5 (22)	52.2 (153)	33.8 (99)	5.1 (15)	3.34	0.75
(SE46) To what extent can you use asynchronous discussions to maximize interactions between students in an online course?	5.1 (15)	24.2 (71)	52.9 (155)	14.3 (42)	3.4 (10)	2.87	0.84
(SE47) To what extent can you use synchronous discussion to maximize interactions between students in an online course?	4.1 (12)	19.5 (57)	56.0 (164)	16.7 (49)	3.8 (11)	2.97	0.82
(SE48) To what extent can you use knowledge of copyright law to provide resources for online students?	7.8 (23)	26.3 (77)	49.5 (145)	12.6 (37)	3.8 (11)	2.78	0.90
Overall						3.19	0.49

SE = self-efficacy; S.D. = standard deviation; f = frequency.

. In terms of the subscale of student engagement (SE11-SE18), about one-third of the teachers reported that they could do quite a lot or a great deal to help a student who was failing an online class in understanding the material (30.3%, n = 89). About 20% of the teachers reported that they could do quite a lot or a great deal to design an online course in a way that encourages collaborative learning (25.6%, n = 75), convince students that they could succeed in an online course (22.8%, n = 67), control disruptive behavior in an online environment (20.8%, n = 61), and help students to value learning online (25.2%, n = 74). Less than 20% of the teachers reported that they could do quite a lot or a great deal to help students think critically (16.4%, n = 48), encourage students who are not interested in doing their online work (17.1%, n = 50), and cultivate each student’s unique originality in an online course (18.1%, n = 53). At the same time, less than 20% of the teachers reported that they could do nothing or little in regard to all items in an online environment. Generally, the majority of the teachers reported that they could do something in regard to all the statements. The item “How much can you do to improve the understanding of a student who is failing in an online class?” received the highest mean (M = 3.22, S.D. = 0.70) among all the items, which indicated that the teachers could do a lot to help students to understand online classes.

For the subscale of the teachers’ self-efficacy in instructional strategies (SE21-SE28), it was indicated that these EFL participants were very positive in reacting to this domain. A very large proportion of the teachers (over 80%) thought that they could do quite a lot or a great deal in providing an alternative explanation to students in an online class, designing an online course such that the students have positive learning experiences, modifying online classes for various learning preferences, creating tasks or questions that require students to reflect by linking concepts to prior knowledge and experience, and using a range of evaluation techniques for an online course. With regard to the difficult questions posed by online students, more than half of the teachers (65.1%, n = 191) reported that they could deal with them well. At the same time, about one-third of teachers (36.9%, n = 108) showed that they could do nothing or little in evaluating the degree to which students understood what the teachers taught in an online course. All of the items’ average scores were greater than the scale’s middle point of 3.00. The greatest mean score of 4.18 (S.D. = 0.78) indicated that EFL teachers had adequate confidence in adjusting their online lessons for different learning styles, while the lowest mean score indicated that evaluating students’ understanding of the content was a problem.

For the subscale of classroom management (SE31-SE38), a very small percentage of the teachers responded that they could do quite a lot or a great deal to create an online course that encourages students to take initiative in their learning (17.4%, n = 51). Slightly more than 20% of the teachers reported that they could deal well with disengaged students in an online class and make clear what they expected of students in regard to their conduct in an online course. About one-third of the teachers reported that they could establish an online course for each group of students (33.5%, n = 95) and deal with students dominating online discussions (32.1%, n = 94). Furthermore, 43% (n = 126) of the teachers reported that they could set up routines in their courses in order to keep online activities operating well. In an online class, about half of the teachers stated that they could influence students to adhere to the established guidelines for assignments and due dates (50.1%, n = 147), and this item also obtained the highest mean score among all the items of this domain (M = 3.54, S.D. = 0.68), indicating that the teachers were confident in assignment management. The item with the lowest mean score (M = 2.95, S.D. = 0.72) indicated that the teachers had less confidence in creating an online course that could encourages student initiative in online learning.

For the last subscale of computer ability, the same percentage of the teachers (38.9%, n = 114) reported that they could work quite effectively in navigating the institution’s technical infrastructure so as to successfully develop an online course and successfully teach a course that is already established online. More than half of the teachers reported that they could do quite a lot or a great a deal with computers in order to facilitate participation online (53.9%, n = 158) and in browsing the internet to provide links and resources to online course students (59%, n = 173) and using a computer for word processing, internet searching, and e-mail communication (60.4%, n = 177). Meanwhile, at the same time, a very small percentage of the teachers reported that they could effectively use asynchronous discussions (17.7%, n = 52) and synchronous discussions (20.5%, n = 60) to maximize students’ interactions online and provide online students with resources by applying their knowledge of copyright law (16.4%, n = 48). The item with the highest mean score was “using a computer for word processing, internet searching, and e-mail communication (M = 3.70, S.D. = 0.78), followed by “navigating the internet to provide links and resources to students in an online course” (M = 3.69, S.D. = 0.79). Meanwhile, the items with the lowest mean score were “using knowledge of copyright law to provide resources for online students” (M = 2.78, S.D. = 0.90), followed by the ability to facilitate interactions between students in asynchronous discussions (M = 2.87, S.D. = 0.84) and synchronous discussions (M = 2.97, S.D. = 0.82).

4.3. The Relationship between EFL Teachers’ Self-Efficacy and the Acceptance of Online Teaching (Q3)

The association between self-efficacy and the acceptance of online teaching among EFL teachers was examined using the Pearson product–moment correlation coefficient, namely, the relationship between the sub-dimensions of self-efficacy in students’ engagement, instructional strategies, classroom management, and computer capability and the acceptance of online teaching. Cohen’s criteria for the correlation coefficient [57] were referred to in order to present the statistical results of the relationship.

Based on the statistical results (shown in

Table 5. Relationship between self-efficacy and intention to teach online.

	Behavioral Intention
Behavioral Intention	1
Student Engagement	0.60 **
Instructional Strategies	0.51 **
Classroom Management	0.32 **
Computer Capability	0.47 **

** Correlation is significant at the 0.01 level (two-tailed).

), all four sub-dimensions of self-efficacy had a significant correlation with online teaching acceptance among EFL teachers. The strongest correlation existed between the teachers’ self-efficacy in students’ engagement and the EFL teachers’ intention to teach online (r = 0.60, and p < 0.01), followed by self-efficacy in instructional strategies (r = 0.51, p < 0.01). These were close correlations based on the guidelines of Cohen [57]. Self-efficacy in computer capability was also positively correlated with EFL teachers’ intention to teach online (r = 0.47, p < 0.01), which was considered a medium correlation based on the guidelines of Cohen (1988). The lowest positive correlation with EFL teachers’ intention to teach online was EFL teachers’ self-efficacy in classroom management online teaching, with r = 0.32 and p < 0.01. According to Cohen [57], there was a medium correlation between these variables. Because of the significantly positive relations between all the sub-dimensions of self-efficacy and teachers’ intention to teach online, it could be suggested that EFL teachers who had relatively high levels of self-efficacy in students’ engagement, classroom management, computer capability, and instructional strategies in online teaching tended to have a greater intention to teach online, and vice versa.

4.4. Multiple Linear Regression Analysis (Q4)

Multiple linear regression (MLR) can be used to predict a dependent variable based on a combination of several normally distributed predictor variables [58]. Therefore, in this study, an MLR analysis was carried out to address research question 4: which is the best predictor of EFL teachers’ acceptance of online teaching: student engagement, instructional strategies, classroom management, or computer skills?

All the assumptions for an MLR, such as the normal distribution, independence of errors, linearity, and lack of perfect multicollinearity, were met. Using the enter method, all four sub-scales of self-efficacy were entered into the regression model. As

Table 6. Model summary.

Model	R	R Square	Adjusted R Square	F	Sig.
	0.61 a	0.37	0.36	42.30	0.000 b

^a. Predictors: (constant), student engagement, instructional strategies, classroom management, computer skills. ^b. Dependent variable: behavioral intention.

shows, a significant model emerged from the ANOVA result: F (4, 292) = 42.30 and p = 0.000 < 0.05. The multiple correlation coefficient (R) was 0.61 and the R² was 0.37, meaning that 37% (adjusted R² = 0.36) of the variance in the EFL teachers’ intention to teach online could be explained by the four sub-scales of self-efficacy.

Table 7. Multiple regression analysis summarised.

Construct	Unstandardized Coefficients		Standardized Coefficients	t	Sig.
	B	Std. Error	Beta
(Constant)	0.173	0.321		0.537	0.592
Student Engagement	0.689	0.093	0.519	7.427	0.000
Instructional Strategies	0.128	0.064	0.096	1.985	0.048
Classroom Management	0.016	0.080	0.011	0.199	0.843
Computer Capability	0.107	0.093	0.078	1.150	0.251

shows the regression results of the model. The results show that self-efficacy in student engagement (t = 7.427, p = 0.000) and self-efficacy in instructional strategies (t = 1.985, p = 0.048) were the predictors of EFL teachers’ intention to teach online, while classroom management (t = 0.199, p = 0.843) and computer capability (t = 1.150, p = 0.251) were not predictors of the EFL teachers’ adoption of online teaching. The results indicated that EFL teachers with higher self-efficacy in student engagement and instructional strategies were more likely to possess a greater intention to accept online teaching.

5. Discussion

The study mainly focused on examining EFL teachers’ self-efficacy and its relationship with online teaching acceptance in the Chinese context. The overall mean score of the teachers’ intention to teach online indicated that EFL teachers possessed a high positive behavioral intention to accept online teaching. The results of the study indicated that nearly half of the teachers were very active in carrying out online teaching in their practice. However, it was interesting to find that, although many teachers intended to perform online teaching, the frequency varied. Teachers preferred to hold online classes regularly rather than frequently or too often. Another interesting finding was that a large portion of the teachers wanted to recommend online teaching modes to their colleagues; however, half of the teachers still preferred traditional teaching rather than online teaching. As seen in Table 4, there were great differences between the teachers’ answers. There were teachers, although they represented a rather small proportion of the participants, who marked “nothing” or “a little” in response to some items, while at the other end of the scale, other teachers felt the opposite.

These interesting responses of the participants generally reflect the current conditions of online teaching. On the one hand, teachers are self-efficacious in online teaching or positive and willing with respect to the integration of advanced educational technologies into their language teaching [12,59]. With educational technologies becoming increasingly prevalent, more language teachers must become familiar with using the technologies to facilitate high-quality teaching. EFL teachers may not consider performing teaching online difficult or complicated due to the rise of professional development training and frequent exchange of ideas with their peers. Teachers’ commitment to teaching, especially during COVID-19, may be partly explained by this [8]. To minimize the epidemic disruption of education, teachers had to spend a certain amount of time and make great efforts to equip themselves with the necessary skills and knowledge required to conduct online teaching [60], which improved the teachers’ capability to teach online in a short amount of time. On the other hand, the findings also revealed that physical teaching is favored by EFL teachers. These results are aligned with previous studies, demonstrating that some teachers prefer traditional classroom teaching or even reject teaching online [61,62]. New obstacles are encountered when teaching online compared with traditional face-to-face teaching, such as the need to build interesting online learning environments for students to participate in [63]. The workload, quality of online teaching, and dishonesty or plagiarism are also common concerns of online teachers [64]. All these factors have an influence on teachers’ efficacy in adopting online teaching. Body language is essential to language teaching and cannot be performed in online settings.

Overall, self-efficacy was statistically associated with the intention of EFL teachers, which indicated that EFL teachers who had higher confidence in their capabilities to conduct online teaching were more likely to accept online teaching in their EFL teaching practice. The findings were supportive of prior research [65,66]. Recent research on the technology-related self-efficacy of preservice teachers showed that teachers are more willing to incorporate instructional technology into their lessons when they have more confidence in their technical abilities [67].

Specifically, EFL teachers’ self-efficacy in student engagement and self-efficacy in instructional strategies were found to have a strong correlation with, and significantly predict, EFL teachers’ intention to teach online. Student online engagement was one of the greatest challenges for teachers. Though students registered their attendance or could even be seen online, it was unclear whether they were engaged or not. After all, an online class cannot allow teachers to observe all students in the same way as a face-to-face class. Given this limitation, the teachers made great efforts and spent a lot of time on course design for the purpose of improving students’ engagement in an online environment. Consequently, they had confidence in their capacity for student engagement and instructional strategies, which were two important factors related to teachers’ acceptance of teaching online.

However, the sub-dimensions of self-efficacy did not show significant effects on teachers’ intention to teach online. Classroom management and computer capability were not observed to have such a significant relationship with the teachers’ intentions. The findings indicated that classroom management and computer capability were not predictors of the intention to teach online. In terms of computer capability, many teachers expressed their worries about the computer ability required to conduct online teaching in the earlier stage of the implementation of online teaching. However, continuous learning through teachers’ professional development provided by administrators, technological support from technical staff, and peers’ help through exchanging ideas about online teaching and self-study could contribute to an increase in their confidence in computer operations [60]. On the other hand, in the study, more than 70% of the survey participants were under 45, and 90% of respondents were master’s and/or PhD holders. These demographic factors can partially explain the teachers’ relatively high levels of computer ability, which is consistent with previous studies. Teachers, in the digital and internet age, have frequent access to modern technologies and more online resources and, therefore, are more familiar with operations related to computers by default. Moreover, English teachers may consider online teaching a commonplace teaching mode nowadays, and this may have been especially true during the COVID-19 period. As a result, it is possible to explain the lack of correlation between instructors’ self-efficacy in computer skills and EFL teachers’ acceptance of online instruction.

6. Limitations

This research has its limitations. First, it should be acknowledged that the participants were EFL teachers who were selected from one region in China. The results might be more generalizable if more teachers from diverse regions had been included in the study. Additionally, because all the information was obtained through questionnaires, the results of the research significantly depended on the respondents’ honesty and how they subjectively evaluated their self-efficacy and their willingness to accept online teaching. Based on Rodarte-Luna and Sherry [68], these responses tended to be influenced by response sets, response bias, and reactivity. Therefore, it is advised that future researchers supplement these data with other types of data, such as interview data for triangulation. Additionally, it must be pointed out that the relationship between the participants’ self-efficacy and their acceptance of online teaching cannot be taken to be causal, since the correlational analysis reveals only the correlation between variables, instead of disclosing the cause effect [69]. Another limitation lies in the use of the scale of self-efficacy, which was originally developed for the assessment of nurse online self-efficacy. Though the scale has been used widely to measure user online efficacy, it lacks focus on the characteristics of English language teaching, such as EFL teachers’ confidence in teaching four basic language skills: listening, speaking, reading, and writing.

7. Conclusions and Implications

In the Chinese setting, this quantitative study examined the potential relationship between the self-efficacy of EFL teachers and their acceptance of online instruction. EFL teachers at Chinese universities in the study showed a positive confidence in themselves and a willingness to embrace online teaching. The empirical findings indicated that self-efficacy had a statistically significant relationship with EFL teachers’ acceptance of online teaching. Specifically, the self-efficacy of Chinese EFL teachers in both student engagement and instructional strategies were positively related to EFL teachers’ intention to accept online teaching. A thorough understanding of this subject can enable teachers and educators to boost their confidence and commitment to online teaching in a practical sense and search for patterns in language teacher effectiveness in a theoretical sense.

Despite some limitations, this study also has implications. The results indicate that teachers generally have a favorable intention to teach online and a high level of self-efficacy in adopting online teaching. Therefore, it is essential to take measures to maintain and improve these instructors’ positive attitudes toward technology integration in online teaching. Furthermore, this study shows that four aspects of self-efficacy, namely, students’ engagement, instructional strategies, classroom management, and computer capability, have a significant correlation with EFL teachers’ acceptance of online teaching. Therefore, it is also important for the personnel who are responsible for the teachers’ professional development to create training plans in order to help EFL teachers to develop the aforementioned four aspects of self-efficacy, especially those designed to improve teachers’ confidence in student engagement and instructional strategies. On the other hand, teachers must be made aware of the value of self-efficacy and know how to apply it to their online teaching practices. Last but not least, the concept of self-efficacy was further validated in this study by examining its role in the context of Chinese language teaching and learning.