(Mis) Alignment of Iranian EFL Teacher’s Written Corrective Feedback beliefs and practices from an Activity Theory Perspective

Abstract

While recent research on written corrective feedback (WCF) has elaborated on the connection between teachers’ beliefs and practices, little research has investigated EFL teachers’ WCF as a multifaceted notion. Drawing on activity theory (AT), in this case, study, we aimed at presenting a picture of EFL teachers’ beliefs about WCF and how their beliefs interrelate with their WCF practices. To this end, we collected data from three EFL teachers working at a private language center in Iran. The data were collected from multiple resources including semi-structured interviews, teachers’ WCF on sample essays, and stimulated recall interviews. The results showed that teachers’ beliefs did not align with their practices as far as the scope and type of WCF are concerned. Teachers’ beliefs, on the other hand, corresponded to their practices regarding the types of errors they targeted. Teachers’ reasoning for (mis)alignments of their beliefs and practices were analyzed from an AT perspective and were then matched against an activity system diagram.

Keywords:

• EFL teachers
• written corrective feedback
• activity theory
• perception
• practices

PUBLIC INTEREST STATEMENT

Teachers’ beliefs are assumed to be playing a filtering role in their practices and give direction to their performance in different areas. The way through which beliefs are translated into practices, however, may not always be linear and depending on diverse factors, misalignments show up. For example, contextual factors, personal attitudes, and working conditions might be the major reasons for the mismatch of beliefs and practices. In this paper, using activity theory, the connection between beliefs and practice regarding written corrective feedback was examined. The results of this study highlight the particularity of each teacher’s written corrective feedback activity system shaped and reshaped through individual factors and external values and expectations. It also attempts to raise attention that teachers should reassess their practices in accordance with specific needs of the context and avoid using a one-fits-all size strategy to address errors in writing.

1. Introduction

Studies into L2 writing have shown a remarkable surge of focus on WFC in second language acquisition research over the last few decades (Bitchener & Ferris, 2012; Ene & Kosobucki, 2016; Lee, 2016). Extant studies have geared toward the effectiveness of WCF (e.g., Ashwell, 2000; Chandler, 2003; Truscott, 1996; Van Beuningen et al., 2012), learners’ engagement with and reactions to WCF (e.g., Lee, 2005; Radecki & Swales, 1988; Zheng & Yu, 2018), self and peer feedback (e.g., Lundstrom & Baker, 2009; Min, 2006; Rahimi, 2013; Sullivan & Lindgren, 2002), and teachers’ WCF performance and beliefs (e.g., Al Shahrani & Storch 2014; R. D. Ferris et al., 2013; Ferris, 2014; Ferris et al., 2011a; Junqueira & Payant, 2015; Lee, 2009; Mao & Crosthwaite, 2019). However, these studies have been criticized for approaching WCF as a decontextualized, cognitively inspired, and linear activity, providing no details on teachers’ purposes and reasoning, classroom contexts, learners’ goals and responsiveness to WCF, and how and why teachers’ WCF practices are projected in certain ways (Bitchener & Storch, 2016; Lee, 2016; Storch, 2018).

Recent studies highlighted the role of context in language teachers’ performance and its potential to affect the relationship between teachers’ beliefs and practices (Borg, 2003; Cross, 2010; Freeman, 2004; Johnson, 2006). As Ortega (2012) argues, teachers’ WCF provision, like other pedagogical practices, does not occur in a social vacuum and, depending on individual and contextual features, as well as educational and social backgrounds, the nature and quality of WCF are not the same. Hence, gaining a contextualized knowledge of WCF might upgrade teachers’ feedback practices (Lee, 2018).

Lee (2016) contends that previous studies were in a broad-brush fashion and revealed a limited snapshot of teachers’, particularly EFL teachers’, WCF performances. Storch (2018) added that previous research lacked ecological validity because they have failed to address the impact of classroom realities, contextual elements contributing to teachers’ beliefs and practices, and the reasons accounting for the possible mismatches between WCF beliefs and practices (Storch, 2018).

Recent calls to explore WCF motivate researchers to adopt an AT lens to capture a comprehensive picture of WCF as a situated activity rather than a stand-alone phenomenon (Bitchener & Storch, 2016; Lee, 2014, 2016; Storch, 2018). To heed these calls, this study is set up to examine EFL teachers’ WCF beliefs and practices from an AT perspective. In this study, we adopted AT to draw EFL teachers’ beliefs about WCF, demonstrate the (mis)alignments between WCF beliefs and practices, and show how individual and social factors shape and guide their beliefs and practices in the context of Iran.

2. Review of literature

2.1. An overview of teachers’ WCF

The body of research on WCF has mainly focused on whether WCF leads to improving L2 learners’ written accuracy (Ashwell, 2000; Truscott, 1996; Truscott & Hsu, 2008; Van Beuningen et al., 2012) and what feedback type is more effective (Bitchener & Knoch, 2009; Chandler, 2003; Ferris & Hedgcock, 2005). Earlier studies dealt with the question of explicit/implicit dichotomy (Bitchener & Knoch, 2009 ; Ferris & Roberts, 2001; Van Beuningen et al., 2008), focused/comprehensive WCF (D. Ferris, 2003; Sheen, 2007), degree of explicitness of WCF (Carroll & Swain, 1993; Ellis, 2009; Loewen & Philp, 2006), and types of errors targeted in writing (Ferris, 2014, Kang & Han, 2015).

Recently, there has been an interest in how teachers’ beliefs about CF affect their feedback practices. More specifically, studies have focused on the (in)consistency between teachers’ beliefs and practices. For instance, Montgomery and Baker (2007) found that while teachers opted for global and content-related aspects of writing, their actual performances showed that their WCF mainly addressed local language errors. The paradox between EFL teachers’ WCF beliefs and actual practices was also reflected in studies by Lee (2008b) and Lee (2009). She found that teachers, despite preferring selective and implicit WCF, practiced comprehensive and explicit WCF.

In a case study, Junqueira and Payant (2015) used multiple resources of data collection (i.e., interviews, reflective journals, comments on learners’ essays) to discover mismatches between a pre-service L2 teacher’s belief and real practices. Findings of the study showed that, although the teacher believed in commenting on global issues of writing, she practiced giving explicit WCF and focused on local issues of writing. Being aware of the mismatches between WCF beliefs and practices, she expressed she could not incorporate her beliefs and practices all the time.

In another study, Mao and Crosthwaite (2019) investigated the extent to which Chinese teachers’ WCF beliefs and practices aligned from Borg’s (2006) cognitive theory point of view. They showed three areas of mismatch between beliefs and practices: provision of explicit and implicit WCF, different forms of implicit WCF, and distribution of local vs. global errors. They found the lack of training for WCF and contextual factors as the main sources of mismatch between teachers’ beliefs and practice. This study, however, did not provide any data on teachers’ arguments for their practices.

Although previous research has focused on the sources of mismatch between teachers’ WCF beliefs and practices, the findings are not monolithic and consistent yet (Storch, 2018). Bitchener and Storch (2016) claimed that this problem might be due to studying teachers’ beliefs and practices from a cognitive theory perspective. Similarly, Cross (2010) leveled criticism against using cognitive theory to inform the relationship between beliefs and practices as it is not a unified and comprehensive conceptual framework to support the changes and development of beliefs and practices. The power of teachers’ agency, social positioning within the context, and the relationship between thinking and doing an activity seems to have been ignored in cognitive theory. Furthermore, the malleability of teachers’ beliefs and its dynamic nature across time and context might not be traceable by cognitive theory. Storch (2018) suggested using activity theory as a framework to provide a more comprehensive and covering picture of the relationship between teachers’ WCF beliefs and practices.

3. Theoretical framework: activity theory

This study is grounded in activity theory (AT), an extension of the sociocultural theory that considers all human activities as a socio-culturally rooted phenomenon. From an AT perspective, language learning is a goal-directed activity mediated by a myriad of factors (Engeström, 2000, 2001). These factors might be physical tools (e.g., computer, pen, paper, and book) and/or symbolic artifacts (e.g., language, moral codes, religious and geopolitical ideas) mediating the quality of the relationship between the doer of the activity and the surrounding world (Lantolf & Thorne, 2006; Vygotsky, 1978).

AT was first proposed by Vygotsky (1978) and, later, Engeström (1987, 2001) proposed a second and third versions for AT. In all the versions of AT, an activity is explained in a network of connections between six main components. The first three components that correspond to stimulus, response, and the outcome are tools (instruments through which an action is done), subject (doer of an action), and the object (a particular achievement or outcome of an action), respectively. The fourth component of an activity is the community of stakeholders and/or those who are involved with and share a goal or interest. The communications and interactions between the community of practice and the subject of the activity are controlled by the “rules” of the activity system, the fifth component. Rules regulate the quality of interconnection between components of the activity system. The division of labor, the sixth component, shows the position of the people and the power relations in the activity system. It indicates the division and quality of access to the resources when an activity is done.

The third version of AT or G3, which serves as the theoretical framework of the present study, attempts to show the inter-subjectivity and shared voices of the participants involved in an activity (Engeström, 2001). Figure 1 illustrates how the components of AT are interconnected.

Figure 1. Engeström’s expended activity theory model (Engeström, 2001)

As mentioned above, most of the studies focusing on teachers’ WCF beliefs and practices are informed by cognitive theory. Cognitively informed studies have not allowed monolithic and consistent findings; nonetheless, they all indicated that individuals’ cognition impacts WCF. On the plus side, a few of these studies elaborated on the complexity and malleability of beliefs and cognition and offered pedagogical implications for L2 education (Junqueira & Payant, 2015). However, these studies seem to present a unilateral perspective lacking the contributions of other community members (e.g., supervisors, teacher trainers, curriculum developers, etc.) to teachers’ beliefs and practices. Lee (2014, 2016), Bitchener and Storch (2016) and Storch (2018) recommended using activity theory as the theoretical framework to bridge the gaps in WCF research because this theory interprets an activity from multiple perspectives and takes individual and contextual factors into account to explain it. As discussed earlier, this theory enables the researchers to have multilateral outlooks toward the teachers’ WCF and explain their WCF behaviors.

3.1. Research on WCF from an AT perspective

A quick review of the literature shows that most of the recent studies informed by AT dealt with WCF from the learner's standpoint. These studies focused on EFL learners’ peer feedback strategies (Yu & Lee, 2016), assessing language learning in L2 writing from an AT perspective (Mak & Lee, 2014), comparing less skilled and skilled learners’ WCF strategies (Lei, 2016), ESL learners’ policy paper assignment (Li, 2013), and learners’ beliefs about engaging with WCF and receiving mediation (Han, 2017). All of these studies used qualitative designs and multiple sources of data collection (e.g., semi-structured interviews, stimulated recalls, and learners’ drafts of writing), to show feedback as a socially mediated activity that facilitates group interactions through using L1, using L2 writing norms (rule), group activities (rule), seeking help from the teachers (community), and playing different roles (division of labor).

The analytical and theoretical potentials of AT to examine teachers’ WCF beliefs and practices has inspired a few studies. Lee (2014) used AT to inform her study and criticize conventional WCF in an EFL context. She described her experience as a teacher educator in Hong Kong and introduced mediating learning experience (MLE) as an innovative approach to WCF. Unlike conventional approaches, which ignore teacher–learner interaction during and after WCF, MLE focused on the human role, mediated and dialogic process of feedback provision, giving rise to purposeful and effective WCF.

In another study to explain the issues and challenges of WCF, Lee (2016) suggested practicing new trends of WCF, running teacher training programs for innovative WCF strategies, and drawing on AT to track possible changes in teachers’ WCF approaches. The study argued that more investigations are needed to yield new insights into how meaningful, purposeful, and contextualized WCF are provided in a real classroom.

To address this call for further research and to fill the gap in the literature on teachers’ WCF, the present case study, using AT as its theoretical framework, aims at investigating the factors shaping three Iranian EFL teachers’ beliefs about and practices of WCF, as well as the sources of (mis)match between their beliefs and practices. In particular, the study attempts to answer the following question:

1. What are the teachers’ beliefs about WCF?

2. What are the teachers’ beliefs about the most appropriate type, focus, and scope of WCF?

3. Are there any mismatches between the teachers’ WCF beliefs and their real practices? If so, how can they be explained from an AT point of view?

4. How can the teachers’ WCF activity be explained from an AT perspective?

4. Method

4.1. Context

This multiple case study was conducted at an Iranian language center. The language center offers general English courses to the students with different levels of proficiency focusing on all language skills (listening, speaking, writing, and reading). Each level is offered over 10 weeks (two one-hour-and-forty-five-minute sessions per week). In Iran, English is mainly taught through Grammar-Translation Method at schools whereby no chance is provided for the Iranian language learners to use English for communication; to compensate for this inadequacy, they attend private language learning centers in which English is normally taught through communicative approach. Writing in the context of private language centers of Iran is not attached as much importance as reading and oral skills due to a multitude of factors including high-class size, lack of enough time, students’ demands, and teacher’s workload (Naghdipour, 2016). However, since teachers are normally required to cover all the textbooks’ tasks and activities, they assign the writing tasks to the students and may provide some feedback on their writings, although the students are not normally required to apply their teachers’ feedback and revise their writings.

4.2. Participants

The participants of the study (n = 3) were selected on a voluntary basis. We contacted the teachers who were teaching one of the B2 levels at the language center (N = 8), informing them of the aim of the study and if they were willing to contribute to our study, among whom three teachers agreed to participate. The reason for targeting the teachers at B2 levels was that, based on the book’s content (Topnotch Series by Saslow & Ascher, 2014) and the language center curriculum, at B2 levels, the students start writing extended texts, including at least three or four paragraphs; at lower levels, writing instruction focuses on controlled and guided writing, which is limited to writing short paragraphs that are highly controlled in terms of sentence structure, tense, word choice, etc.; this did not allow us to observe students’ natural writing errors and have a good sample of teachers’ feedback. All teachers were native speakers of Persian holding an MA degree in TEFL (teaching English as a foreign language). They all had more than 10 years of teaching experience but had not received any training on teaching writing and WCF before. Table 1 provides more detailed information about the participants.

Table 1. Participants’ profile

*Teachers	Age	Degree	Teaching experience	Level of course
Amin	34	MA	11 years	Upper-intermediate
Goli	37	MA	15 years	Upper-Intermediate
Zeinab	39	MA	13 years	Upper-Intermediate

*Please note that the names are pseudonyms.

At the time of the experiment, each teacher had two or three classes each with around 10 students. They were all university students aged 18–35. As mentioned above, they all had the experience of writing controlled and guided paragraphs before this study. They attended general English classes to improve their communication abilities and develop their language proficiency to fulfill their academic needs (e.g., reading research articles in English and passing English proficiency test such as IELTS or TOEFL).

4.3. Instruments

4.3.1. Semi-structured interview

One-on-one semi-structured interviews were conducted to gain information about teachers’ perceptions of WCF, how they provide WCF, and the factors affecting their WCF practices. Mak and Lee’s (2014) interview questions inspired the interview items of our study.

Accordingly, the following interview questions were raised:

1. What is your attitude towards WCF? What factors shape your attitude towards WCF? Explain.

2. What kind of errors should be addressed in WCF: local or global errors? Why?

3. How implicit/explicit should WCF be?

4. Should WCF address all the student errors or some specific ones? Why?

4.3.2. Stimulated recall interview

We ran stimulated recall interviews with the participants to find the reasons for their comments. This helped us provide justifications for the (mis)match between their beliefs stated at the outset of the study and their real feedback practices. The stimulus recall interviews took around 45 minutes for each teacher. (see the data collection section for more details about the interviews)

4.4. Materials

4.4.1. Writing tasks

The writing tasks were used to elicit samples of teachers’ real practices of WCF. The tasks were written by the students of the teachers who participated in the study in response to a prompt that appeared in the second unit of the book the teachers were teaching at the time of the study (Summit 1, by Saslow & Ascher, 2014). Normally, in each book, consisting of five units, the students must accomplish five writing tasks. The topic of the writing was violence. The students were required to write three paragraphs to address the issue raised in the prompt. The number of words ranged from 200 to 250 words (the average length was 247 words).

The teachers provided feedback on these writings. All the teachers had already been notified to submit a copy of their students’ writings on unit two of the book they were teaching with their feedback to the office of the language center before returning the writings to the students. Then, for each teacher, we sorted out the papers and identified the ones which included a large number of comments and then randomly selected five papers. This would provide us with enough data for our study. We used a mixture of purposive and random sampling because the aim of the study was to examine teachers’ feedback practices (comprehensive vs. focused; implicit vs. explicit; and form vs. content) and the papers with a low number of comments, belonging to the students with a relatively good writing ability, did not let us have a representative sample of teachers’ feedback.

4.5. Data collection

To access the teachers and conduct the research, we obtained the institute manager’s consent. We started with a semi-structured interview in the second week of a 10-week-long semester. Each interview was done in Persian and took around 90 minutes . At the end of week seven, when all the teachers had submitted a copy of their students’ writings on which they had provided feedback, we collected the three teachers’ papers from the office of the institute. Then, in week eight, within 3 days after the papers had been collected, the first author conducted the stimulated recall interviews. Each teacher was given around 15 minutes to review all the five papers to recall what he/she had done with respect to his/her feedback. They were asked to think about the reasons for their choice of errors they had targeted as well as the feedback method (explicit/implicit) they had used.

4.6. Data analysis

4.6.1. WCF coding

Coding of local and global feedback was done based on Montgomery and Baker's (2007) categorization of global and local errors; based on this categorization, local issues refer to sentence level or errors of form (e.g., grammar, punctuation, spelling and mechanics of writing) and global errors are related to organization of ideas and content of writing.

Feedback type was coded based on Ferris’ (1995) categorization of explicit/implicit feedback. Explicit WCF referred to correct answers provided by the teachers while implicit WCF included circling, underlying, and using symbols or codes. The writing assignments were also classified in terms of whether WCF was comprehensive (addressing all the errors) or focused (focusing on a limited number of errors).

The two authors coded a random 20% of feedback units (n = 74) for global/local as well as explicit/implicit feedback. There were 83%, and 94% agreement on global/local and explicit/implicit errors, respectively, between the two authors. The discrepancies between the two codings were discussed and resolved. Then, the first author coded the remaining 80% of the teachers’ comments.

4.6.2. Interview analysis

Interviews were analyzed based on Creswell and Poth's (2018) data analysis spiral. Based on this procedure, data analysis goes through four main loops; that is, reading and memoing emergent ideas, describing and classifying codes into themes, developing and assessing interpretations, and representing and visualizing the data. The first loop, reading and memoing, involves synthesizing the data into higher level analytic meanings. In this stage, we made notes of the interview transcripts and tried to see what meanings the transcripts convey. In the next stage, forming codes or categories, we developed themes based on the most frequently happening reasons the teachers provided for their feedback and then, in the third stage, we provided an interpretation for the emerged themes in the light of literature on WCF. We came up with such themes as 'teachers' negative feeling regarding WCF’ or 'learners' expectations as reasons for WCF provision.' To support each theme, we selected one example from each teacher’s statements. Next, we crosschecked the teachers’ statements about WCF with their real practices to find the match or the mismatch between them. We followed the same process of coding for stimulated recalls. Finally, we visualized the themes through tables as well as through the AT diagram (see the results section).

5. Results

5.1. RQ 1: What are the teachers’ beliefs about WCF?

Table 2 illustrates the themes that emerged from the interviews showing teachers’ beliefs about WCF, their reasons and examples of their statements.

Table 2. Teachers’ attitudes about WCF

Beliefs about WCF	Reasons	Example of interview
WCF is unworthy	1. Time & energy consuming 2. Learners’ unwillingness and reluctance to review the WCF and revise essays.	It is a very time-consuming job especially when I see that my learners have too many mistakes in their writing. In addition, the money I get does not pay off the labor (Goli) “I give my learners WCF as much as possible, but, most of time I do not see any development whatsoever in their writing or language ability in general, which questions the effectiveness of WCF” (Amin). “most of the time, they (the learners) leave the writings on the table and pay no attention to the WCF. This is really disappointing!” (Zeinab).

Participants showed negative feelings about providing feedback in general. They described WCF task as a pointless and exhausting job bringing no real benefits. The reasons they provided for their negative attitudes included low pay, the tediousness of the job, students’ lack of appreciation, and no improvement observed in students’ writings. Despite this negative feeling, however, they did provide WCF. Table 3 summarizes their reasons for the provision of WCF.

Table 3. Reasons for WCF provision

Justifications for WCF provision

Excerpts of interviews

1.Sense of obligation arising from teaching responsibilities and textbook 2. Students’ expectation

I do not really enjoy providing WCF and do it just because it is part of my duties and not because I want to give WCF (Amin). “Well, my colleagues are covering writing tasks and provide WCF. So, I think to be consistent with others in the institute I have to provide WCF too (Goli). “In each chapter, writing is the final task to be covered and the learners expect me to cover the whole chapter, so I have to assign writing and give WCF afterward”(Zeinab). “I assume that if I do not give WCF, my students label me as an irresponsible teacher “ (Amin).

As Table 3 reveals, teachers believed that their responsibilities for teaching and learners’ expectations were two main reasons for giving feedback. Learners expected the teachers to cover all tasks including the writing tasks in the book. This sense of obligation made them provide WCF on the students’ writings.

5.2. RQ2: What are the teachers’ beliefs about the most appropriate WCF type, focus, and scope?

As for the focus of WCF, the errors targeted by WCF, Table 4 reveals the themes emerging from the interviews.

Table 4. Teachers’ beliefs about the errors targeted by WCF

	Beliefs about WCF	Justification	Excerpts of interviews
Goli	Priority must be assigned to local errors (or form of writing) (if any) rather than global errors (content).	writing is a chance to practice new grammatical and structural concepts and help learners eliminate their oral errors	by checking the most frequent mistakes and local errors, I can see what grammatical point needs more attention in teaching. Students need to learn about the mechanics and primary principles of writing to be able to develop well-organized writings.
Zeinab	both global and local errors must be targeted	Writing is not just grammar	Writing is not just grammar. Although it is very important, we need to pay attention to the coherence and cohesion, too.
Amin	both global and local errors must be targeted	Following his teacher	At a high level, we should also consider the organization of the writing and how they present their ideas. This is what my teacher did at university

The data showed variations in teachers’ beliefs regarding errors to be addressed. While Goli believed that the primary focus should be on local errors in writing, Zeinab and Amin considered global issues equally important.

Table 5 illustrates the themes that emerged from the interviews with regard to feedback scope.

Table 5. Scope of WCF: comprehensive vs. selective

Teachers’ Justifications	Excerpts of interviews
Focused WCF offers learners the chance of deeper understanding of their errors and a sustained learning experience.	Selective WCF is like a gradual learning. It makes them focused and minimizes the possibility of repeating the same errors (Amin). “I think it is easier for learners to focus on fewer errors and learning will be easier in this case” (Goli) Comparing to unfocused approach whereby a wide range of errors need to be addressed and remembered, focused WCF impose a much lighter cognitive load on  learners.
Unfocused WCF might be demotivating and embarrassing for some learners	If I were a student, getting a paper marked in red would be disappointing to me (Zeinab). I think addressing all errors might decrease the confidence in learners and discourage them from writing(Amin)

As the table reveals, the teachers believed that focusing on a limited number of errors is more effective, helps learners experience less cognitive load and learn more effectively. However, one of them also pointed out comprehensive WCF, offers the learners a chance of having a complete picture of errors and impede fossilization of errors.

Regarding WCF type, as Table 6 shows, participants preferred implicit WCF. They argued that implicit WCF develops a sense of autonomy and self-learning in learners. They also believed that implicit WCF would enhance learners’ confidence and self-learning, develop curiosity and autonomy; on the other hand, they also believed that explicit WCF is more time and energy saving than implicit feedback.

Table 6. Teacher’ beliefs about explicitness of WCF

Teachers’ Justifications

Excerpts of interviews

1. developing a sense of autonomy and self-learning by implicit WCF 2. creativity and expanding knowledge

Providing WCF implicitly results in learners’ independent learning and critical thinking (Amin). Implicit WCF enhances discovery learning experience and leads into a sustainable learning outcome. Implicit WCF helps learners memorize their mistakes (Goli). Implicit WCF fosters creativity in learners as when they are looking for the correct answers, they might face some other points and expand the range of knowledge (Zeinab). “Implicit feedback helps learners look at the writing from a different perspective and develop papers in a fixed and already shaped frame” (Goli). .

5.3. RQ 3: Are there any mismatches between the teachers’ WCF beliefs and their real practices? If so, how can they be explained from an AT point of view?

5.3.1. Results of WCF analyses

In total, there were 371 feedback items in 15 pieces of writing and they all addressed local issues of the writings rather than global problems. The higher-order issues, such as content and organization problems, were ignored in teachers’ feedback practices. The number and percentage of WCF for each teacher have been illustrated in Table 7.

Table 7. Type of errors targeted by WCF

	No. of feedback points	% of WCF type and focus
		Global	Local	Explicit	Implicit	Comprehensive/focused
Goli	158	0%	100%	87.97%	12.03%	Comprehensive
Zeinab	105	0%	100%	68.95%	39.05%	Comprehensive
Amin	108	0%	100%	69.44%	30.56%	Comprehensive
Total no. of feedback marks:	371

*Please note that all types of WCF that indicated that there is an error without explicitly providing the correct form were marked implicit (e.g., underlining, coding, metalinguistic explanation).

As shown in Table 7, all comments addressed local errors and no feedback was given on global issues such as content and organization in writing. Also, all teachers addressed all the existing language errors in the texts (comprehensive feedback). As for feedback type, the table indicates that the common practice seems to be explicit feedback. More specifically, about 70% of the errors have been treated explicitly. The highest percentage of explicit feedback (87.97%) belonged to Goli and the lowest percentage (68.95%) to Zeinab.

Table 8 compares the teachers’ beliefs about feedback type, focus, and scope to their real practices.

Table 8. Teachers’ WCF beliefs vs. practice

Teacher	Feedback Focus		Feedback Scope		Feedback type
	Belief	Practice	Belief	Practice	Belief	Practice
Goli	Focused	Comprehensive	Local	Local	Implicit	Explicit
Zeinab	Focused	Comprehensive	Local/Global	Local	Implicit	Explicit
Amin	Focused	Comprehensive	Local/Global	Local	Implicit	Explicit

As the table shows, there is an almost complete mismatch between the teachers’ beliefs and practices. As far as the feedback scope (focused vs. comprehensive) is concerned, none of the three teachers’ real feedback aligned with their beliefs. While the teachers believed that WCF must be focused, they provided feedback on almost all existing errors of the students. The teachers also stated earlier in the interview that feedback should be implicit, but the majority of their comments on the students’ errors were explicit. Around 90% of Goli’s and 70% of Amin’s and Zeinab’s comments were explicit (See Table 7). Two teachers, Zeinab and Amin, believed that feedback must be provided on both global and local issues, but later they provided feedback on local issues only.

Since the comparison between the teachers’ beliefs and practices did not show an alignment between the two, we conducted a stimulus recall interview to find the reasons for this mismatch. This helped us contextualize their feedback activity system within the AT framework. Regarding the scope of feedback, although the teachers (subject component of AT) believed that writing errors (the object component of AT) must be addressed selectively (selective WCF as a tool to give WCF), in practice they used comprehensive WCF. The reason for the paradox between beliefs and practices regarding the tool component of AT, was attributed to the community component of AT. They complained that they had no training for WCF, and so they felt they should address all the students’ errors. For instance, Goli stated:

I think selective WCF requires training and time to see what errors should be selected. Knowing about which errors to focus needs expertise.

Amin and Goli referred to apprenticeship of observation or following colleagues, community of practice component of AT, as another reason for providing comprehensive feedback. For instance, Amin stated,

I remember that my writing professor did the same when I was a student and he, similarly, gave us comprehensive WCF.

Teachers pointed to the learners’ expectation as another factor mediating comprehensive WCF practice. They argued that in an EFL context, teachers are assumed to be the mere source of knowledge. Here, cultural-historical artifacts could justify the inconsistencies and mismatches between teachers’ WCF beliefs and practices. Goli expressed,

I believe in efficacy of selective WCF, but we, Iranians, expect teachers to do everything for the learners. This expectation unconsciously affects the way I practice WCF.

As for the focus of feedback (targeting local or global errors), all teachers, unlike what they had stated in the interview, focused just on local errors. The reason for this mismatch was time limitation and main goals of teaching which restricted the teachers to deal with teaching the process of writing and focusing their feedback on the global writing issues. Amin said,

Here at the language center, the main goal of language teaching seems to be improving speaking skill rather than writing. In addition, we do not have time to teach writing to our students; so I do not think it is useful at all to give feedback on macro aspects of writing. This makes the students more confused.

Zeinab believed that their students’ proficiency level was not high enough to go beyond accuracy to writing quality. She said,

our students are not even able to correct their local errors let alone their errors of writing organization.

They also believed that the students should first improve the accuracy of their writing and then focus on global errors. Amin stated,

I did not address the students’ global errors because they were still struggling with form. How can we expect them to improve writing coherence or content?

Focusing on accuracy, and the mismatch between believing in global issues and practicing local WCF, could be explained in terms of outcome component of AT because the final goal of writing in the context of our research is improving linguistics knowledge than the writing ability.

As for feedback type, they majorly used explicit feedback although they had favored implicit feedback in their interviews. Teachers assumed that their students were reluctant to revise papers and so implicit feedback would not be helpful to them. As an example, Goli said,

My learners do not show interest in writing and I think using implicit WCF makes them more demotivated.

Another reason the teachers provided for providing explicit WCF was the teacher-centered approach to pedagogy in their EFL context. This corresponds to the labor division component of AT model as teachers are expected to transfer knowledge to the students and the students, in turn, are considered passive recipients. To support this argument, Amin stated,

My learners expect the teacher to be the main source of knowledge and give all correct forms in writing. Sometimes they say, if they had known the correct form, they could have used it while writing. They do not find implicit feedback helpful.

Zeinab also added,

Although using codes and symbols might trigger a sense of curiosity and self-learning, given the level of my students and the fact that they do not normally revise their writings, I have to provide the correct form for my students.

5.4. RQ4: How can the teachers’ WCF activity be explained from an AT perspective?

Teachers’ WCF activity system was analyzed in the light of the findings of this study and was then matched against the components of AT diagram (see Figure 2). As Figure 2 illustrates, in the feedback activity system, the subjects are EFL teachers working at the language center. The object is the learners’ writing errors. The rule based on which the teachers were providing WCF was single draft approach whereby the students submitted their writings and the teacher provided feedback without requiring the students to revise their writing based on the feedback. Teachers’ feedback type, focus, and scope were mediated by their community of practice, that is, colleagues, manager, students and their expectations, and what they had learnt as a student from their teachers (apprenticeship of observation). Due to the teacher-centered approach, which is dominant in the EFL context, the division of labor is limited to teacher being the source of knowledge and responsible for students’ learning. So, students do submit their papers and receive feedback on all their errors from the teachers, and they are not even responsible for revising their writings. Finally, the outcome or goal of teachers’ WCF practices (the outcome component of AT model) was learners’ improvement of written accuracy.

short-legendFigure 2.

6. Discussion

6.1. Teachers’ WCF beliefs and practices

Drawing on AT, we used a qualitative design to study EFL teachers’ WCF beliefs, real practices, and reasons behind these practices to provide a more comprehensive picture of Iranian EFL teachers’ WCF activity. We also showed that the (mis)matches between EFL teachers’ WCF beliefs and practices feature a collective act depending on an ongoing, multilayered, and socially distributed system whereby multiple internal and external artifacts are mediated. Teachers believed that WCF might not come into fruition and they described negative attitude towards providing WCF. However, they practiced WCF due to the teaching obligations.

Regarding WCF practices, as far as the method and scope of WCF are concerned, the findings of this study are in line with Lee (2008b). She showed that, in Hong Kong, 70% of EFL teachers practiced explicit WCF, as they believed finding the meaning of codes for learners would be confusing. Likewise, Al-Jarrah (2016), Al Shahrani and Storch (2013) found explicit WCF was common practice in EFL contexts as the issue of responsibility in teaching influenced teachers’ real WCF practice. However, Hamouda (2011) and Mao and Crosthwaite (2019) found the majority of WCF were implicit. In these studies, teachers emphasized the efficacy of explicit WCF but they practiced implicit WCF and gave coded (selective) WCF. Hamouda (2011) argued that leaners must be assigned the responsibility of finding correct answers. Teachers in Mao and Crosthwaite's (2019) study also asserted that implicit and selective WCF saves time and labor for teachers. Lee (2008b) regarded learners’ level of proficiency were among variables mediating teachers’ WCF practices in reality.

Teachers’ beliefs and practices both indicated the primacy of local issues in WCF. The focus on grammatical accuracy and using explicit feedback as the tool of WCF is rooted in the deep-seated belief that in the EFL context (in particular, the context of Iran), the lion's share of learning a language is grammar and there is no separate systematic program to teach writing skills at private language learning centers (Naghdipour, 2016).

The traditional approach to teaching English at both university and school level has also been reported in a study by Farhady and Hedayati (2009). They mentioned that “the way English is taught at the universities is often translation oriented, because the main objective is to enable students to read and understand materials written in English in their majors. Since the language of instruction in the universities is Persian, there is no strong feeling toward the need for English, especially at the undergraduate level” (p. 139). In such a context, communicative English has become gradually overshadowed and dealing with productive skills like speaking and writing is barely of concern (Dahmardeh, 2009).

Naghdipour (2016) also reported that, in Iran, writing skill is not valued in language learning policies and there is a lack of agenda about what to do and how to proceed with teaching writing. It is this history and traditional teaching background that mediate teachers’ lack of rules to address writing errors and teach writing. Lack of agenda for WCF and the following a product approach to WCF were earlier confirmed in the EFL context (Lee, 2009; Naghdipour, 2016). To escape from the state of perplexity, teachers resort to their “apprenticeship of observation” (Lortie, 1975) or their recollection of teaching as learners as the source of idea to practice WCF.

Lee (2014) clarified that the reason for following traditional comprehensive WCF is the maxim of the more the better, which indicates that teachers feel responsible about all learners’ errors and shoulder the major burden of teaching. This practice, from the AT viewpoint, is an example of contradictions arising from components of systems. Considering grammar teaching as the outcome of practicing WCF and focusing on grammatical accuracy (outcome within AT), considering teachers as a sage on the stage and treating learners as passive recipients of knowledge (labor division within AT) hinders practicing implicit WCF. Participants in this study believed that, in the EFL context, learners expect their teachers to help them understand all errors and spotting certain types of errors might raise misunderstanding for learners that teachers shrink their responsibility. Such a belief is also due to non-reciprocal whole class work that reflects teachers’ heavy responsibility to present the knowledge and be responsible for providing input. Naturally, this traditional non-reciprocal culture of teaching and learning in Iran and teacher-fronted pedagogy has resulted in learners’ undeniably heavy reliance on teachers as mere source of knowledge (Ataee & Mazloom, 2013).

6.2. Misalignment of teacher WCF beliefs and practices

The study also showed a misalignment between beliefs and practices concerning the type and scope of WCF. Teachers believed that implicit methods of WCF would work better and errors should be tapped selectively. In reality, however, they used explicit WCF and addressed all errors. From the AT viewpoint, the mismatches between thoughts and practice are inevitable, as classroom practices do not take place simply on the basis of the immediate setting (Cross, 2010). The activity system of WCF continually draws on and develops through the unstable and dynamic context in which internal and external contradictions work (Engeström, 2001). More specifically, the history of teachers (subject within the activity system), along with cultural-historic expectations, the object and the outcome of the activity are distributed through the system of activity and shape and reshape the quality of practices in real context. In this study, learners’ expectation of the obligations imposed by the administration (community component of AT) mediated teachers’ WCF practices and led to the disequilibrium between WCF beliefs and practices. Given that, as Lantolf (2004b) points out, “activities, whether in the workplace, classrooms, or other settings, do not always unfold smoothly” (p. 339) and a unit of analysis is “an irreducible [emphasis made in the original] theoretical entity that cannot be broken down into elements [emphasis made in the original]” (Roth, 2014, p. 143).

The conflict between beliefs and real practices is, however, regarded as “sources of development; activities are virtually always in the process of working through contradictions” (Kuutti, 1996, p. 34). Studying about these mismatches might assist teachers in becoming more “principled, less random, perhaps more experimental and innovative, more connected to the learning of particular students, and more subject to our own critical evaluation of techniques, methods, successes, and failures” (Casanave, 2004, p. 45). To homogenize teachers’ practices, enhance motivation, and attain effective WCF strategies leading to more promoted language-learning programs, however, it is important to refer to Fullan’s (2005) theory of change and capacity and improve teachers’ collective effectiveness. Teacher-training workshops, formulating systematic writing instruction, and evaluation programs, placing a more serious emphasis on significance of writing in teaching and learning English might be assumed decent for overshadowed stance of WCF in language learning system of Iran and minimize the distance between beliefs and practices.

6.3. Limitations and suggestion for further research

Although the study attempted to suggest a model of Iranian EFL teachers’ WCF activity system, due to a few limitations, the model may not present a completely precise picture of teachers’ feedback system in this context. For starters, we used a small number of teachers from a single institute, which certainly affects the generalizability of the findings to the whole context; the results should, hence, be interpreted cautiously. Future research should employ a larger number of teachers teaching in different institutional settings such as universities, schools, or different institutes with different language learning policies and approaches.

Another limitation of the study that may affect the accuracy of the suggested model of teachers’ WCF activity system is that we did not ask the learners (as one of the main groups of community component of the activity system) to express their beliefs and expectations of the type, focus, and scope of their teachers’ WCF. We only relied on what the teachers stated about their learners’ expectations. Future research should include the students’ ideas as well as how they would respond to the WCF they receive. This would help present a more accurate and comprehensive model of feedback activity system. In other words, as AT is claimed to present a comprehensive portrayal of a problem, it is necessary to delve into learners’ ideas and reactions to EFL teachers’ WCF practices and see if teachers’ perceptions and visions about WCF match learners’ expectations. In particular, leaners’ stances in WCF activity system might yield fruitful insights into how WCF works best in the EFL context and what contextual and cultural factors should be adjusted to bring about a promoted WCF system.

In addition, we did not observe the real context to take note of what really happens in a real classroom context. A follow-up study might provide a better picture of this reality using observation of teachers’ and learners’ reactions and examining their expectations to see how these layers are interrelated.

Author details

Neda Soleimani is a PhD candidate in teaching English as a foreign language teaching at Shiraz University. She has been doing studies related to teachers’ education, activity theory, and epistemological beliefs.

Mohammad Rahimi is an associate professor of TEFL at Shiraz University. His research focuses on writing evaluation, written corrective feedback, and dynamic assessment of writing. He has published in Language Teaching Research, Language and Sociocultural Theory, System, Journal of Response to Writing, CALL, and Reading and Writing.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

Mohammad Rahimi supervised and designed the method of the study, contributed to data collection and analyses, and read and revised the paper.

Neda Soleimani collected the data, analyzed them, and wrote the preliminary draft of the paper.

Additional information

Funding

This study did not receive any funding.

Notes on contributors

Mohammad Rahimi

Neda Soleimani is a PhD Candidate in TESOL in the department of foreign languages and linguistics, Shiraz University, Iran. She has been doing research in teacher education, classroom management, epistemological beliefs and discourse studies.

Mohammad Rahimi is an associate professor of TEFL in the Department of Foreign Languages and Linguistics, Shiraz University, Iran. His current research interests include L2 writing evaluation, written corrective feedback, and dynamic assessment of writing.