Language Teachers’ Corrective Feedback Practice: Impact of a Teacher Education Course on Feedback Types, Uptake, and Interactional Moves

August 2025 – Volume 29, Number 2

https://doi.org/10.55593/ej.29114a1

Zia Tajeddin
Tarbiat Modares University, Iran
<tajeddinzmodares.ac.ir>

Servat Shirkhani
Dept. of English Language Teaching, Khor. C., Islamic Azad University, Khorramabad, Iran
<servatshirkhanigmail.com>

Abstract

Corrective feedback (CF) is a central issue in language education, and, undoubtedly, teachers have a key role in boosting the potential outcomes of CF for learners. This study aimed to explore the effect of a teacher education course on the types of errors the teachers treated through oral corrective feedback (OCF), the types of OCF, and OCF-driven interactional moves. Twenty teachers were observed for two consecutive sessions before and after attending a course on OCF. Analyses of OCF in 131.5 hours of pre- and post-course recorded sessions using the Wilcoxon signed rank test indicated that the course did not significantly influence phonological and grammatical errors treated, but led to a significant increase in addressing lexical errors. Moreover, significant changes were observed in the employment of OCF types, with a significant decrease in confirmation checks, recast, metalinguistic feedback, explicit correction, and multiple feedback, and a significant increase in clarification request, repetition, and negotiated feedback. Concerning uptake, although neither +Uptake nor -Uptake conditions changed significantly, a significant decrease was observed in the number of conditions with no evidence of uptake. In addition, while before the course, only 32.9% of OCF episodes consisted of more than three interactional moves, this percentage increased to 63.9% after the course. The results prove that a short-term course can reshape teachers’ OCF practice.

Keywords: corrective feedback episodes; error types; interactional move; oral corrective feedback; teacher education course; uptake

Corrective feedback (CF) has received considerable attention in language education. Accordingly, a myriad of studies have addressed various dimensions of CF in language teaching. Leeman (2007) classifies studies on CF into two broad categories of descriptive and developmental research. Whereas descriptive studies have centered on the types of CF provided and learner responses to the feedback (e.g., Llinares & Lyster, 2014), developmental studies have examined the effectiveness of different CF types (e.g., Li & Iwashita, 2021; Khezrlou, 2019; Nassaji & Kartchava, 2020; Sato & Loewen, 2018; Sippel, 2019). These studies have considered immediate uptake, test performance, or noticing error correction as an indicator of CF efficacy. Nonetheless, as stressed by Chong (2022), the focus of CF should shift from an emphasis on immediate uptake, or in his terms “feedback literacy”, to teachers’ and learners’ engagement with CF, which he calls “feedback ecology.”

Some of the developmental studies on CF have shown that teacher variables play a major role in the treatment of errors (e.g., Fu & Nassaji, 2016; Junqueira & Kim, 2013; Soruç et al., 2025). Teacher education, as Junqueira and Kim (2013) showed, is one of the factors related to teachers’ practice of CF. However, despite the key role of teacher education in developing teachers’ practice (Weekly & Pollard, 2024), its influence on their CF is under-researched. Recently, some studies (e.g., Ha, 2022; Ha & Murray, 2021; Mirzaei Shojakhanlou & Saeedian, 2023) have examined the role of teacher education on teachers’ cognition and/or practice of CF. However, scant research has explored the effect of teacher education courses based on sociocultural theory (SCT) on teachers’ CF practices. Thus, this study aims to investigate the role that a socioculturally informed teacher education course can play in teachers’ selection of OCF types and error types addressed in OCF-driven interactional moves, as well as in learner uptake. Since the objective of the teacher education course was to give teachers more awareness about the importance of providing OCF based on SCT principles, qualitative analysis of the interactional moves was carried out to see whether teacher education could lead to more effective mediation and more attention to learner reciprocity in the practice of OCF.

Background of the Study

Conceptions of Corrective Feedback

CF can be either written or oral. As posited by Li and Vuono (2019), while written CF involves the visual presentation of delayed CF, OCF, which is the focus of the current study, refers to the aural indication during speech production that an element of the learner’s output is erroneous. CF is reflected in numerous strategies classified in various ways. Depending on whether the error is directly corrected or the learner is encouraged to self-correct, OCF strategies have been divided into input-providing and output-prompting (Ellis, 2009) or have been placed on a continuum from explicit to implicit CF (Li, 2018). Moreover, CF strategies have been discussed as being dynamic or non-dynamic based on whether they fit the individual learner’s zone of proximal development (ZPD) or are fixed corrective moves given regardless of the learner’s level of ZPD (Rassaei, 2019). Additionally, based on the timing of feedback, i.e., given either during or after a task, they can be categorized as immediate or delayed. Lyster and Ranta’s (1997) study is the often-cited source for the categorization of OCF types, which lists seven types: clarification request, elicitation, recast, repetition, metalinguistic feedback, explicit correction, and multiple feedback. A clarification request entails indicating to learners that what they have said is erroneous or misunderstood and requires them to reformulate it. Elicitation involves the teacher’s efforts to prompt the learner to produce the correct form. Recast refers to the situation in which the teacher rephrases the learner’s utterance, correcting the erroneous part(s) without changing the central meaning. Repetition is when the teacher uses intonation while repeating the learner’s utterance to highlight the error. Metalinguistic feedback requires providing comments or asking questions about the form of the learner’s production (generally using grammatical metalanguage) to assist the learner in correcting the error. Explicit correction involves the teacher providing the correct form and clearly stating that an error has occurred. Multiple feedback, according to Lyster and Ranta, refers to the teacher’s use of more than one feedback strategy in a single instance to correct the learner.

The facilitative role of CF in second language (L2) development has been emphasized by both cognitive theories of L2 acquisition (e.g., Li & Iwashita, 2021) and the SCT (e.g., Aljaafreh & Lantolf, 1994; Rashidi & Majdeddin, 2023; Rassaei, 2014, 2019; Zhang & Zhang, 2023). From the cognitive perspective, a substantial body of research (e.g., Khezrlou, 2019; Li & Iwashita, 2021; Sato & Loewen, 2018; Zhao & Ellis, 2020) focuses on either the distribution or the effectiveness of various CF strategies. Based on this perspective, the effectiveness of CF can be measured by considering uptake, learners’ performance on tests, or learners’ awareness of errors based on their perceptions (Sheen, 2004). The SCT, however, is more focused on the degree to which CF is “graduated” so that it can help the learner transition from needing corrections from others to possessing the ability to self-correct. The SCT emphasizes the necessity of graduated CF to ensure that, at any given time, the learner receives the minimum assistance required to self-correct. It suggests paying attention to the whole interaction, or the feedback ecology in Chong’s (2022) terms, to determine whether teacher mediation in the form of CF has fostered learner development. The teacher as mediator may employ more than one form of CF in successive turns, and the learner may grasp the concept after the first or later CF turns. Thus, within the SCT, it is the type and amount of mediation provided during CF that best indicates CF effectiveness.

Research on Corrective Feedback

The types of CF described can be utilized to correct learners’ phonological, lexical, grammatical, and pragmatic errors. Previous studies related to the targets or foci of CF have addressed the extent to which different error types have been corrected either in general (e.g., Brown, 2016; Mackey et al., 2000) or concerning the proportion of occurring errors (Lyster, 2001; Shirkhani, 2019). Brown (2016) synthesized descriptive observational research on types and linguistic targets of CF and found that the most frequently corrected errors were grammatical in nature. Mackey et al. (2000) analyzed interactional CF episodes provided and stimulated-recall comments from the learners to compare the targets of CF and to examine the learners’ recognition of CF provided on different error types. They found that morphosyntactic and phonological errors were the most corrected. However, they reported that while learners noticed the majority of CF for lexical and phonological errors, they accurately perceived only 13 percent of CF concerning morphosyntactic errors. Lyster (2001) identified grammar as the error type receiving the highest percentage of corrections; however, considering the proportion of corrected errors to the total number of errors in each category, Lyster concluded that lexical errors were the most noticed ones. Similarly, Shirkhani (2019) demonstrated that although the frequency of corrected phonological errors was higher than that of lexical and grammatical errors, the three error types were treated almost equally concerning their frequency of occurrence.

Some other studies have explored the effectiveness of CF concerning its targets or the implicit/explicit and input-providing/output-prompting nature of CF (e.g., Khezrlou, 2019; Mackey & Goo, 2007; Sato & Loewen, 2018; Tan et al., 2024; Zarei et al., 2018). Mackey and Goo (2007) conducted a meta-analysis on interaction research and concluded that the highest rate of CF efficacy was for CF on lexical errors. Sato and Loewen (2018) reported that, in the four experimental groups they studied, only the group that received input-providing CF did not achieve significant gains after the treatment. Zarei et al. (2018) compared the effects of implicit, explicit, and emergent CF on learners’ speaking accuracy. They found a significant positive effect for emergent CF, through which the teacher provided implicit feedback and moved to explicit feedback only if the learner showed it was necessary. They also reported that learners had positive attitudes only toward emergent CF, which allowed them to correct themselves. Khezrlou (2019) examined the effects of CF types—namely, input-providing and output-prompting CF—and the types of task repetition (i.e., exact task repetition and procedural repetition) on the linguistic development of four groups of EFL learners. She found that, regardless of the type of task repetition, the output-providing groups significantly outperformed the input-providing groups.

Despite the significant potential of SCT in enhancing the practice and effectiveness of CF, research in this area is limited. Besides Aljaafreh and Lantolf (1994), a few recent studies have discussed or examined the impact of SCT-informed interventions on teachers’ practices (e.g., Rashidi & Majdeddin, 2023) and learners’ internalization of CF (e.g., Nassaji & Swain, 2000; Rashidi & Majdeddin, 2023; Rassaei, 2014, 2019; Zhang & Zhang, 2023). In their foundational study, Aljaafreh and Lantolf observed interactions between a tutor and three L2 learners. They found that, for one of the learners, the explicit CF initially required became more implicit over time; that is, the degree of scaffolding provided by the teacher decreased as the learner gained greater control over the L2. This change in the level of scaffolding occurred due to the teacher’s attention to the learner’s reciprocity. From their findings, Aljaafreh and Lantolf identified three principles that they believed govern the effectiveness of CF. They proposed that CF must be graduated (at no time should more scaffolding be provided than necessary), contingent (the provision of CF must cease when the learner demonstrates the ability to operate without scaffolding), and dialogic (the scaffolding must be based on the dynamic assessment of the learner’s ZPD). Furthermore, Nassaji and Swain (2000), in a case study of two adult Korean learners of English, compared the effectiveness of CF within the learner’s ZPD with CF provided regardless of the learner’s ZPD. The results supported the Vygotskian perspective on scaffolding, indicating greater effectiveness of CF within the learner’s ZPD than CF given randomly. Similar studies by Rassaei (2014, 2019) demonstrated that feedback tailored to learners’ needs is more beneficial than CF provided without consideration of their ZPD. Additionally, Rashidi and Majdeddin (2023) revealed that negotiated feedback enhanced learners’ internalization of the CF received from their teacher. Regarding the role of SCT-based teacher education in teachers’ CF provision, Rashidi and Majdeddin explored the impact of mediational discourse between a coach and an English language teacher during and after observations of the teacher’s CF delivery. The SCT-based study found mediation effective in increasing the teacher’s understanding of his conceptual thinking about CF.

As the preceding review shows, several CF-related studies have focused on the distribution of CF strategies used by language teachers. Although these studies have revealed varying distributions of CF types, most of them (e.g., Lyster & Ranta, 1997; Shirkhani & Tajeddin, 2016) have found input-providing strategies to be the most frequently employed. Moreover, only a few studies (e.g., Brown, 2016; Mackey & Goo, 2007) have examined CF targets in language classrooms, arriving at contrasting findings. Furthermore, several studies have investigated the amount of CF uptake (e.g., Fu & Nassaji, 2016; Gholami & Gholami, 2020) and reported differential effects for various types of given CF (e.g., Llinares & Lyster, 2014; Nassaji & Kartchava, 2020; Zhao & Ellis, 2020). The findings from these studies suggest a need for changes in certain aspects of CF provision that can be addressed through language teacher education. However, studies on the role of teacher education in the use of CF in classrooms are scarce (e.g., Ha, 2022; Junqueira & Kim, 2013; Rashidi & Majdeddin, 2023; To et al., 2023). To the best of our knowledge, limited research has investigated the effects of a teacher education program based on SCT principles on various aspects of CF provision. Therefore, the present study focused on the extent to which teachers’ OCF practices could be influenced by raising their awareness of how SCT principles might manifest in OCF provision. Specifically, it sought to examine whether an SCT-informed, feedback-focused teacher education program could influence the types of errors addressed, the types of OCF provided, and the interactional moves in OCF episodes. In line with the study’s purpose, three questions were posed:

Does an SCT-informed feedback-focused teacher education course have any significant impact on the following?

1. The type of errors corrected by language teachers
2. The type of OCF used by language teachers
3. OCF episodes in terms of the number of moves and uptake

Method

Participants

The participants, selected through convenience sampling, were 20 teachers instructing English at the intermediate level in two language institutes in Tehran, Iran. They voluntarily attended an OCF-focused teacher education crash course, and their classes were observed for two sessions before and two after the course. To encourage attendance, the institutes marketed the course as part of in-service teacher education. The course was conducted separately in each institute so that all teachers had ample opportunity to participate in the discussion sessions. In one institute, 8 teachers (6 female and 2 male) attended, while in the other, 12 (all female) participated. The teaching experience of the participating teachers ranged from 1 to 10 years, and their ages varied from 20 to 35. Eleven of the teachers held bachelor’s degrees (5 English majors and 6 non-English majors), and nine had master’s degrees (5 English majors and 4 non-English majors). Teachers in these language institutes, like many others in the country, are hired based on their language proficiency and knowledge of language teaching skills. Consequently, non-English majors with intermediate and advanced levels of English proficiency can gain qualifications through short-term teacher education courses. Both institutes under study have their own teacher education programs aimed at familiarizing their teachers with the fundamentals of language teaching and equipping them with various techniques for instructing different aspects of the language included in the course books they are assigned to teach.

Data Sources

The data for the study were collected through pre-course observations, the teacher education course, and post-course observations.

Pre-course observations. Observations made during two class sessions for each teacher before the teacher education course were conducted and compared with those after the course. Overall, 68.5 hours of classroom instruction were observed before the course. The session lengths ranged from 74 minutes to 121 minutes, and the average session duration was about 103 minutes. The session lengths varied as some of the teachers did not turn on the recorder exactly at the beginning of the class or turned it off a short time before the end of the session or even during the class; these teachers preferred to avoid recording some short informal parts of classes. The recorded materials were analyzed for the frequency and types of errors treated, the frequency and types of OCF, and the number of interactional moves.

An OCF-focused teacher education course was developed, featuring a comprehensive plan for a three-hour crash course as a guide for implementation. This plan provided details on the timing and topics covered in each session, as shown in Table 1. The course, instructed by the second author, comprised two 90-minute sessions. The first session introduced OCF, its significance in language teaching, and how its effectiveness is often assessed. Additionally, this session explored the basics of SCT and its role in providing OCF. The second session involved participating teachers in discussions about patterns of mediation and learner reciprocity in delivering negotiated feedback. Emphasizing the importance of identifying learners’ ZPD for effective OCF, this session required participants to apply the discussed principles to several examples. Each example presented a scenario where a learner made an error, and the teacher was expected to provide contingent dialogic OCF. Finally, the teachers were encouraged to share their thoughts on appropriate OCF practices with one another and with the course trainer.

Table 1. Teacher education course plan: ZPD-driven OCF practice

Content	Time (minutes)
Session One (90 minutes)
1.       Introduction to the course	5
2.       Introduction to OCF	15
3.       Significance of OCF	15
4.       Measurement of OCF effectiveness	10
5.       SCT: An introduction	30
6.       OCF within the SCT: Phases and measurement of its effectiveness	15
Session Two (90 minutes)
7.       Discussion of patterns of mediation and reciprocity in the practice of OCF in a few examples	25
8.       Phases of ZPD-driven OCF	15
9.       Application of ZPD-driven OCF phases to a few examples	10
10.     Discussion of the participants’ practice of ZPD-driven OCF	30
11.     Summary and conclusion	10

Post-course observations. The procedure for these observations was parallel to the one for pre-course observations and included 63 hours of observing the same participant teachers after the course. The session lengths ranged from 70 to 122 minutes, and the average session duration was about 96 minutes.

Data Collection and Analysis

Data collection was carried out over seven weeks. The 20 teachers who had agreed to participate in the study were informed that the data from their classes would be used for research purposes without their identities being revealed. They were told that the recordings of observations would be used for classroom interaction analysis without giving them a detailed explanation about the focal purpose of the study so that their regular feedback provision would not change. Afterward, the data collection was carried out based on the timeframe below:

Weeks 1-3: pre-course observations
Week 4: teacher education course
Weeks 5-7: post-course observations

For data analysis, detailed descriptions were made of OCF-related interactional moves in the recorded sessions. In making the descriptions, the moves provided by the teachers and learners were identified. Afterward, for each of those moves, the types of errors being corrected, the types of OCF given, the number of moves by teachers and learners, and the outcome akin to immediate uptake were determined. In categorizing the data, three aspects of language (i.e., grammatical, lexical, and phonological) were identified as the probable focus of any OCF move (i.e., as error types) based on Lyster and Ranta’s (1997) study. The original categorization of OCF types was taken from Lyster and Ranta, which included clarification request, elicitation, recast, repetition, metalinguistic feedback, explicit correction, and multiple feedback. In addition, confirmation check was adopted from Ellis (2008), and negotiated feedback emerged during data coding. Ellis defines confirmation check as the utterance used to confirm that the first speaker’s previous utterance has been correctly interpreted. By negotiated feedback, we mean those instances where the teacher, tailoring the feedback to the learner’s needs, provided more than one OCF type on a single error each given in one separate move. Therefore, in general, there were nine categories for OCF types. Table 2 presents examples from the data for the nine OCF types to clarify what is meant by each. Furthermore, three conditions were defined for uptake: (1) +Uptake: where the correct form by the learner followed the OCF move; (2) No evidence of uptake: where there was neither immediate uptake nor was there error repetition by the same learner; and (3) -Uptake: where the moves following the corrective move contained either the same error or similar errors showing that the corrected feature had not been taken by the learner. It should be noted that some instances of uptake immediately following OCF might not necessarily indicate learning. Thus, if a corrected case, followed by immediate feedback, was later used erroneously by the same learner, it was considered -Uptake based on the definition of this uptake condition. Also, to verify the correct coding of uptake conditions, the utterance following the last move in each OCF episode was examined carefully to make sure that the episode had ended and the next utterance was on another topic.

Table 2. Examples of OCF types with identified error types and uptake conditions

CF Type	Error Type	Uptake Condition	Interactant	Example
Clarification request	Lexicon	No evidence	Learner	We saw many exhibitions in that show.
			Teacher	What do you mean? What are exhibitions?
Confirmation check	Lexicon	No evidence	Learner	Because I always work hard, I got a rise last week.
			Teacher	Do you mean a raise?
Recast	Grammar	No evidence	Learner 1	What are you going to do?
			Learner 2	I’m going to camping.
			Teacher	I’m going to go camping.
Repetition	Phonology	No evidence	Learner	He is a chef /tòef/.
			Teacher	a chef /tòef/?!
Metalinguistic feedback	Grammar	No evidence	Learner	Which one is the largest: Iran or Japan?
			Teacher	Oh, no, you are comparing two things.
Elicitation	Grammar	+Uptake	Learner	I’ll just stay in home.
			Teacher	Stay …?
			Learner	Stay home
Explicit correction	Grammar	No evidence	Learner	Because he has a children,
			Teacher	He has a child, not CHILDREN
Multiple feedback	Grammar	No evidence	Learner	I have took some medicine.
			Teacher	Have took? No, have taken, I have taken some medicine.
Negotiated feedback	Grammar	+Uptake	Learner	If I will going to another country,
			Teacher	If I …?
			Learner	If I going to
			Teacher	If I GO to
			Learner	If I go to another country, I can speak with them.
			Teacher	Yes. Right.

The study employed a pre-test and post-test design. The pre- and post-course data were subjected to two types of non-parametric analyses because of the categorical nature of the variables. First, frequencies and percentages were calculated for the linguistic aspects being corrected, OCF types, the number of OCF moves, and uptake conditions. These computations were carried out separately for pre- and post-course data. Next, the Wilcoxon signed rank test was conducted for comparison purposes to explore the changes in teachers’ practice of OCF.

Results

Types of Errors Treated

To examine whether the feedback-focused teacher education course produced effects on OCF targets, descriptive analyses were carried out for the types of errors addressed before and after the course. As Figure 1 displays, the highest proportion of OCFs targeted phonological errors, and the lowest was directed at lexical errors. Before the course, phonological and lexical errors constituted nearly 48% and 16% of all errors addressed, respectively. After the course, phonological and lexical errors accounted for around 41% and 27% of all errors treated. In addition, the frequency of lexical errors addressed increased while the frequencies of grammatical and phonological errors decreased after the course.

Figure 1. Errors treated before and after the course

Next, the Wilcoxon signed rank test was performed to examine the significance of the differences between the error types. The findings, as shown in Table 3, indicated no significant difference between the total number of errors addressed before and after the education course. Furthermore, no significant differences were detected between pre- and post-course frequencies of grammatical and phonological errors. However, the analysis revealed a statistically significant increase in the frequency of lexical errors corrected after the course, z = 2.16, p < .05. Nonetheless, the effect size for this test was small (r = .01), indicating a weak difference between the two frequencies.

Table 3. Wilcoxon signed rank test for types of errors before and after the course

	TCF	G	L	P
Z	-.448a	-1.089a	-2.158b	-1.588a
Asymp. Sig. (2-tailed)	.654	.276	.031	.112

^aBased on negative ranks; ^bBased on positive ranks. Note.  TCF = total OCF provided, G = grammatical, L = lexical, P = phonological.

Types of OCF

As part of the analysis, descriptive statistics were run for types of OCF to see how the frequencies of OCF types changed after the teacher education course. The types of OCF studied are based on the categorization in Table 2.

The results for types of OCF used before the course (Figure 2) demonstrated that explicit correction was the most frequently used (47.8%) and that recast was the second mostly used (30.3%). All the other OCF types accounted for only 21.9% of the total OCF, with repetition as the least frequent (.5%). On the other hand, as shown in Figure 2, after the course, negotiated feedback was provided in 51.4% of the cases. Explicit correction and recast were second and third (18.6% and 10.6%, respectively). Confirmation check and metalinguistic feedback were not used at all. The other four categories, namely clarification request, repetition, elicitation, and multiple feedback constituted 19.4% of all the OCF. Among these four strategies, multiple feedback occurred the least, accounting for 1.6% of all the OCF turns.

Figure 2. Types of OCF given before and after the teacher education course

Figure 2 shows that after the course, the frequency of clarification requests, repetition, elicitation and negotiated feedback increased, while the use of confirmation checks, recasts, metalinguistic feedback, explicit corrections, and multiple feedback decreased. The frequencies of confirmation checks and metalinguistic feedback dropped to zero, while the frequencies of recasts, explicit corrections, and multiple feedback fell to 33%, 38%, and 31% compared with their rates before the course. In contrast, clarification requests and elicitation were employed 1.78 and 1.32 times more than their rates before the course. Furthermore, the post-course frequencies of repetition and negotiated feedback reached 10.5 and 15.74 times their pre-course frequencies.

To see whether the differences observed in Figure 2 were statistically significant, the Wilcoxon signed rank test was run for the nine OCF categories. The results, shown in Table 4, revealed significant differences between the frequencies before and after the course for eight OCF categories; elicitation was the only category with no significant change. The effect size was then calculated for all the comparisons with significant differences. Based on Cohen’s (1988) criteria, the analysis indicated large effect sizes for confirmation check (r = .74) and metalinguistic feedback (r = .54); medium effect sizes for clarification request (r = .28), repetition (r = .4), and multiple feedback (r = .27); and small effect sizes for recast (r = .16), explicit correction (r = .12), and negotiated feedback (r = .15). The results, therefore, demonstrate that despite being statistically significant, the changes in the frequencies of recast, explicit correction, and negotiated feedback after the course were weak.

Table 4. Wilcoxon signed rank test for pre- and post-course use of OCF types

	CR	CC	Rec	Rep	ML	El	EC	MF	NF
Z	-2.225a	-2.879b	-3.549b	-3.328a	-3.550b	-.648a	-3.362b	-2.404b	-3.921a
Asymp. Sig. (2-tailed)	.026	.004	.000	.001	.000	.517	.001	.016	.000
r (effect size)	.28	.74	.16	.40	.54	–	.12	.27	.15

^aBased on negative ranks; ^bBased on positive ranks. Note. CR = clarification request, CC = confirmation check, Rec = recast, Rep = repetition, ML = metalinguistic feedback, El = elicitation, EC = explicit correction, MF = multiple feedback, NF = negotiated feedback.

Excerpts of explicit correction and negotiated feedback, as the most frequently used strategies before and after the course, respectively, are provided in this section to show why the significant changes observed in the results are considered positive in line with the objectives of the teacher education program.

Excerpt 1. Explicit correction of a grammatical error:
Learner: I think Leila is well at writing paragraphs, so …
Teacher: “Well” is wrong. Leila is good at writing.

Excerpt 2. Negotiated feedback on a lexical error:
Learner: Practice helps us remind the information.
Teacher: REMIND the information?
Learner: Yes. Ummm
Teacher: Remind or remember?
Learner: They are synonyms, I think.
Teacher: No. Look at these examples (The teacher explains the difference through examples she writes on the board.)

In Excerpt 1, the OCF episode involves only two moves (i.e., the erroneous utterance by the learner and the explicit correction by the teacher). The teacher does not seem to check whether the learner has noticed the correction. However, in Excerpt 2, the negotiated feedback consists of six moves, reflecting the teacher’s effort to provide contingent feedback. When the learner demonstrates a lack of knowledge about the lexical item in question in move 5, the teacher begins to correct the error explicitly. In this excerpt, the interactive collaborative negotiation prompts the teacher to implement graduated dialogic feedback.

OCF Episodes in Terms of Moves and Uptake

To assess the impact of the teacher education course, OCF episodes were analyzed in terms of the number of interactional moves and uptake. First, the frequency and percentage of different moves before and after the course were calculated. A total of 1,204 OCF instances were recorded before the course and 1,163 after, comprising 3,800 and 5,756 moves, respectively. Thus, the average number of moves per OCF increased from 3.2 to 4.9, indicating a significant rise in the number of moves following the course. The results in Figure 3 show that the percentage of OCF episodes containing more than three moves grew after the course, while episodes with two and three moves saw a marked decrease. Prior to the course, the number of moves ranged from 2 to 12. Overall, 77.1% of all the moves belonged to two- and three-move OCFs, which accounted for 31.3% and 45.8% of the OCFs, respectively. Therefore, only 32.9% of OCF episodes exhibited more than three moves, and 8-to-12-move OCFs made up just 1.2% of all the moves. As anticipated, the number of moves noticeably increased after the course, ranging from 2 to 20 moves. Similar to the pre-course phase, the highest percentage was three-move OCFs. Although this type constituted only 24.6% of all the moves, 63.9% of the corrections included more than three moves.

Figure 3. Number of moves before and after the teacher education course

Next, descriptive statistics for the uptake conditions before and after the education course were conducted to examine changes in uptake. As noted earlier, uptake was investigated in terms of three conditions: +Uptake, no evidence of uptake, and -Uptake. As evident from Figure 4, the frequency of +Uptake increased by 1.23 times, while no evidence of uptake dropped to half, and -Uptake fell to 0.17 times.

Figure 4. Uptake conditions before and after the teacher education course

To analyze the significance of the differences between pre- and post-course frequencies for each of the three uptake conditions, the Wilcoxon signed ranks test was performed. The results (Table 5) indicated a significant difference at the .05 level only for the second condition, meaning instances where there was no evidence of uptake. However, the effect size for this significant difference (r = .11) demonstrated that the difference was weak.

Table 5. Wilcoxon signed rank test for uptake conditions before and after the course

	+Uptake	No evidence of uptake	-Uptake
Z	-.825a	-2.840b	-1.890b
Asymp. Sig. (2-tailed)	.409	.005	.059

^aBased on negative ranks; ^bBased on positive ranks.

Excerpt 3. OCF episode with no evidence of uptake:
Learner: My sister /ˈestʌdiz/ math.
Teacher: /sˈtudiz/. Your sister /sˈtudiz/ math.

This excerpt shows the teacher’s explicit correction of a learner’s phonological error. Similar to Excerpt 1, this excerpt does not involve any move showing the learner’s uptake or the teacher’s concern about uptake.

The following excerpts of OCF episodes with -Uptake and +Uptake conditions show how the +Uptake condition is different from the other two conditions in terms of negotiation of feedback.

Excerpt 4. OCF episode with -Uptake:
Learner: My brother take my little sister to school and …
Teacher: Takes, not take.
Learner (later in the same session): She work with her friend.

Excerpt 5. OCF episode with +Uptake:
Learner: Between the three, choose one.
Teacher: Between …?
Learner: Between the three, … mmm ….
Teacher: We use between for two things, right?
Learner: Oh, among, among the three. Sara, you can choose one among the three books.

The OCF episode in Excerpt 4, like Excerpt 1 and Excerpt 3, involves an erroneous move by the learner and explicit correction from the teacher. Later, error repetition illustrates the ineffectiveness of explicit correction. Conversely, Excerpt 5 consists of five interactional moves, including the teacher’s use of two OCF strategies. First, the teacher uses a confirmation check and then provides metalinguistic feedback as the learner repeats the error. This metalinguistic feedback is contingent on the learner’s level, as evident from the learner’s reciprocity. The contingent OCF leads to the learner’s uptake and, consequently, the production of the correct form.

Discussion

This study examined the effect of a feedback-focused teacher education course on different aspects of OCF in language classrooms. Concerning the types of errors addressed, the results indicated that, both before and after the course, the highest rate of feedback was directed at phonological errors. This finding suggests that the teachers’ CF practice does not align with efforts to enhance the effectiveness of OCF, as the results of other studies (e.g., Mackey & Goo, 2007) have demonstrated greater effectiveness of CF in correcting lexical errors. The finding related to error types does not conform to other studies (e.g., Brown, 2016; Lyster, 2001; Mackey et al., 2000). For instance, Brown (2016) and Lyster (2001) identified grammar as the most frequently corrected error type, while Mackey et al. (2000) found grammatical and lexical errors to be the most corrected. These contrasting results may relate to variations in the distribution of types of errors occurring in different contexts, as confirmed by Lyster (2001) and Shirkhani (2019).

To account for the results related to OCF types in light of previous research, the taxonomy of CF proposed by Ellis (2009) is quite illuminating. In this taxonomy, CF strategies are categorized as implicit versus explicit and as input-providing versus output-prompting. The current study shows evidence that a socioculturally informed, feedback-focused teacher education course positively influenced the reduction in the use of input-providing OCF (i.e., recast, confirmation check, and explicit correction). Furthermore, among the two explicit output-prompting strategies, metalinguistic feedback was not used at all, and elicitation showed no significant change. Additionally, the provision of multiple feedback decreased to 31% of its usage before the course. This notable decline in the use of this strategy indicates the teachers’ heightened awareness of the need to create space for learners to correct themselves.

As the findings indicate, the use of the two implicit output-prompting strategies increased significantly. Repetition was utilized 10.5 times more than before the course, and clarification requests rose to 1.8. Implicit output-prompting OCF can enhance interaction between the teacher and the learner, thereby giving the teacher the opportunity to identify the learner’s ZPD and align OCF with that ZPD. The changes in OCF practice are positive considering the results of similar empirical studies (e.g., Khezrlou, 2019; Lyster & Ranta, 1997; Sato & Loewen, 2018), which suggest that implicit output-prompting CF types are more effective than explicit input-providing. Lyster and Ranta (1997) reported recast as the least effective CF type, with elicitation and metalinguistic being the most successful, arguing that output-prompting strategies are more effective because they provide learners with opportunities to engage in the error correction process. The results further align with Zarei et al.’s (2018) finding that emergent CF moving from implicit to explicit based on the learner’s needs has a more significant impact on grammatical accuracy than implicit and explicit CF. According to Zarei et al., learners preferred emergent CF over implicit and explicit CF because they enjoyed taking an active role in error correction. However, the findings of this study do not align with some others (e.g., Zhao & Ellis, 2020) that found no significant difference between the effects of explicit and implicit CF on learners’ accuracy and reported that explicit CF led to more uptake-with-repair (the same as +Uptake in the present study) than implicit CF and no feedback.

As to OCF moves and uptake, the results revealed that the post-course moves were 1.5 times more than the pre-course ones. Another positive impact of the course was an increase in negotiated feedback (15.74 times), which was implicit in the course material and discussions, as the teachers were guided to navigate through the regulatory scale while considering the learner’s reciprocity. Empirical studies (e.g., Aljaafreh & Lantolf, 1994; Nassaji & Swain, 2000; Rashidi & Majdeddin, 2023; Rassaei, 2014, 2019; Zhang & Zhang, 2023) emphasize that dynamic feedback within a learner’s ZPD is more effective than non-dynamic feedback that overlooks ZPD. In addition to indicating a teacher’s effort to apply SCT-informed CF strategies, negotiated feedback helps learners focus more on their errors and ensures they do not recur.

Regarding uptake, it was found that +Uptake and -Uptake conditions did not change significantly after the teacher education course. However, a significant decrease was evidenced in post-course frequencies for the second condition, that is, no evidence of uptake. Based on post-course results, the percentage of cases with no evidence of uptake dropped by 50%, which is a noticeable decrease in line with the aim of the study. The increase in the number of moves used in treating errors is suggestive of the impact of the course on teachers’ treatment of errors. This finding aligns with the study by Rashidi and Majdeddin (2023), which confirmed the effectiveness of SCT-informed teacher education on CF episodes. Additionally, the descriptive findings of the study, indicating +Uptake as the most frequent, resonate with previous studies on learner uptake (e.g., Gholami & Gholami, 2020; Lyster & Ranta, 1997; Nassaji & Kartchava, 2020). Gholami and Gholami reported uptake for 47% of the feedback. Furthermore, Nassaji and Kartchava demonstrated that 58.2% of corrections led to uptake. Naturally, several studies have shown that learner uptake may vary depending on other factors, including CF types (Nassaji & Kartchava, 2020) and error types (Gholami & Gholami, 2020; Kartchava & Ammar, 2014).

Overall, the findings related to OCF types, uptake, and OCF moves suggest that the teacher education course was effective in reshaping teachers’ OCF practices. Multiple dimensions of OCF provision by the teachers changed significantly after the course. Taken together, all these dimensions illustrated that OCF underwent profound changes in line with SCT principles. The notable increase in the number of moves in OCF episodes resulted from the teachers’ use of more output-prompting implicit OCF and negotiated feedback, which was, in turn, due to their intolerance of OCF that brings no evidence of uptake. All these changes appear to have occurred as a consequence of the teachers’ awareness of the need to provide OCF that is dialogic, contingent, and graduated.

Conclusion

OCF is found in most language classrooms. However, the results of this study show that the way teachers deliver OCF is not necessarily aligned with SLA theories. While theoretical perspectives on CF provision view learner reciprocity as crucial to OCF effectiveness, this study reveals that teachers are not attuned to learner engagement with OCF. Providing OCF demands that teachers pay attention to the learner’s ongoing progress so they can dynamically adapt their support to the learner’s level. Therefore, OCF provision requires more skillful teachers who can identify the learner’s ZPD at any moment and observe the learner’s reciprocity to offer contingent OCF. The results indicate the effectiveness of teacher education in assisting language teachers in applying SLA findings to their OCF practice. While there were more input-providing and explicit strategies prior to the course, the use of output-prompting and implicit strategies increased significantly afterward. This aligned with the course’s goal to familiarize teachers with the implications of SCT for OCF treatment. As a result of applying the SCT framework, more strategies were utilized in addressing errors, leading to greater attentiveness to learners’ reciprocity.

The study holds implications for teachers and teacher educators. The findings suggest that teachers should stay updated with SLA research and apply it to their practices, particularly in OCF provision. Thus, they are encouraged to reflect on their OCF practices to evaluate, using the SCT framework, the degree to which they attend to learner reciprocity, allowing them to offer OCF that meets learner needs. Furthermore, the study indicates that teacher educators can improve L2 teachers’ OCF practices by creating courses that showcase the latest theoretical findings and help them apply these findings in their teaching methods.

This research does have some limitations that should be acknowledged. First, since the study explored the influence of a three-hour crash course on the teachers’ practice of OCF, future research should focus on the effects of longer-duration courses on teachers’ OCF practices. Second, due to the large spoken corpus analyzed for this study, it was not feasible to examine all the corrective interactions in terms of mediation and reciprocity patterns, as practiced in some SCT-based studies. Therefore, further research with narrower focuses addressing teacher mediation and learner reciprocity moves in providing OCF is recommended. Third, to connect SLA research with language teaching practices, additional studies can examine teachers’ knowledge of SLA findings related to OCF, their perceptions of the necessity to apply this knowledge to language pedagogy, and the degree to which this research-based knowledge is reflected in their classroom practices.

About the Authors

Zia Tajeddin is Professor of Applied Linguistics at Tarbiat Modares University, Iran. He is the co-editor of Applied Pragmatics (John Benjamins) and Second Language Teacher Education (University of Toronto Press). His research interests center on teacher education, L2 pragmatics, and EIL/ELF pedagogy. He has published papers in many journals, including System, Language Teaching Research, The Language Learning Journal, Journal of Intercultural Communication Research, RELC Journal, Australian Journal of Teacher Education, The Asia-Pacific Education Researcher, TESL-EJ, and Journal of Language, Identity & Education. His most recent publication is Handbook of Language Teacher Education: Critical Review and Research Synthesis (with Thomas S. C. Farrell, Springer, 2025). ORCID ID: 0000-0002-0430-6408

Servat Shirkhani holds a Ph.D. in Applied Linguistics and is an assistant professor at the Department of English Language Teaching, Khor. C., Islamic Azad University, Khorramabad, Iran. Her areas of interest include corrective feedback in language instruction, teacher education, interlanguage pragmatics, psychology of language learning, and technology-assisted language learning. ORCID ID: 0000-0003-2114-5391

To Cite this Article

Tajeddin, Z., & Shirkhani, S. (2025). Language teachers’ corrective feedback practice: Impact of a teacher education course on feedback types, uptake, and interactional moves. Teaching English as a Second Language Electronic Journal (TESL-EJ), 29(2). https://doi.org/10.55593/ej.29114a1

References

Aljaafreh, A., & Lantolf, J. P. (1994). Negative feedback as regulation and second language learning in the zone of proximal development. The Modern Language Journal, 78(4), 465-483. https://doi.org/10.1111/j.1540-4781.1994.tb02064.x

Brown, D. (2016). The type and linguistic foci of oral corrective feedback in the L2 classroom: A meta-analysis. Language Teaching Research, 20(4), 436-458. https://doi.org/10.1177/1362168814563200

Chong, S. W. (2022). The role of feedback literacy in written corrective feedback research: From feedback information to feedback ecology. Cogent Education, 9(1), 2082120. https://doi.org/10.1080/2331186X.2022.2082120

Cohen, J. W. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum.

Ellis, R. (2008). The study of second language acquisition (2^nd ed.). Oxford University Press.

Ellis, R. (2009). Corrective feedback and teacher development. L2 Journal, 1(1), 3-18. https://doi.org/10.5070/l2.v1i1.9054

Fu, T., & Nassaji, H. (2016). Corrective feedback, learner uptake, and feedback perception in a Chinese as a foreign language classroom. Studies in Second Language Learning and Teaching, 6(1), 159-181.  https://doi.org/10.14746/ssllt.2016.6.1.8

Gholami, L., & Gholami, J. (2020). Uptake in incidental focus-on-form episodes concerning formulaic language in advanced adult EFL classes. Language Teaching Research, 24(2), 189-219. https://doi.org/10.1177/1362168818783442

Ha, X. V. (2022). Effects of a professional development program on teachers’ oral corrective feedback practices. System, 110, 102917. https://doi.org/10.1016/j.system.2022.102917

Ha, X., & Murray, J. C. (2021). The impact of a professional development program on EFL teachers’ beliefs about corrective feedback. System, 96, 102405. https://doi.org/10.1016/j.system.2020.102405

Junqueira, L., & Kim, Y. (2013). Exploring the relationship between training, beliefs, and teachers’ corrective feedback practices: A case study of a novice and an experienced ESL teacher. Canadian Modern Language Review, 69(2), 181-206. https://doi.org/10.3138/cmlr.1536

Kartchava, E., & Ammar, A. (2014). The noticeability and effectiveness of corrective feedback in relation to target type. Language Teaching Research, 18(4), 428-452. https://doi.org/10.1177/1362168813519373

Khezrlou, S. (2019). Task repetition and corrective feedback: The role of feedback types and structure saliency. English Teaching & Learning, 43(2), 213-233. https://doi.org/10.1007/s42321-019-00025-2

Leeman, J. (2007). Feedback in L2 learning: Responding to errors during practice. In R. DeKeyser (Ed.), Practice in a second language: Perspectives from applied linguistics and cognitive psy­chology (pp. 111-137). Cambridge University Press.

Li, H., & Iwashita, N. (2021). The role of recasts and negotiated prompts in an FL learning context in China with non-English major university students. Language Teaching Research, 25(2), 209-233. https://doi.org/10.1177/1362168819839727

Li, S. (2018). Corrective feedback in L2 speech production. In J. I. Liontas (Ed.), The TESOL encyclopedia of English language teaching (pp. 1-9). Wiley.

Li, S., & Vuono, A. (2019). Twenty-five years of research on oral and written corrective feedback in System. System, 84, 93-109. https://doi.org/10.1016/j.system.2019.05.006

Llinares, A., & Lyster, R. (2014). The influence of context on patterns of corrective feedback and learner uptake: A comparison of CLIL and immersion classrooms. The Language Learning Journal, 42(2), 181-194. https://doi.org/10.1080/09571736.2014.889509

Lyster, R. (2001). Negotiation of form, recasts, and explicit correction in relation to error types and learner repair in immersion classrooms. Language Learning, 51(1), 265-301. https://doi.org/10.1111/j.1467-1770.2001.tb00019.x

Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in communicative classrooms. Studies in Second Language Acquisition, 19(1), 37-66. https://doi.org/10.1017/S0272263197001034

Mackey, A., Gass, S., & McDonough, K. (2000). How do learners perceive interactional feedback? Studies in Second Language Acquisition, 22(4), 471-497. https://doi.org/10.1017/S0272263100004010

Mackey, A., & Goo, J. (2007). Interaction research in SLA: A meta-analysis and research synthesis. In A. Mackey (Ed.), Conversational interaction in second language acquisition: A collection of empirical studies (pp. 407-452). Oxford University Press.

Mirzaei Shojakhanlou, M., & Saeedian, A. (2023). Developing EFL teachers’ cognition of corrective feedback using a task-based approach. Journal of Modern Research in English Language Studies, 10(2), 69-90. https://doi.org/10.30479/jmrels.2022.17313.2075

Nassaji, H., & Kartchava, E. (2020). Corrective feedback and good language teachers. In C. Griffiths & Z. Tajeddin (Eds.), Lessons from good language teachers (pp. 151-164). Cambridge University Press.

Nassaji, H., & Swain, M. (2000). A Vygotskian perspective on corrective feedback in L2: The effect of random versus negotiated help on the learning of English articles. Language Awareness, 9(1), 34-51. https://doi.org/10.1080/09658410008667135

Rashidi, N., & Majdeddin, M. (2023). The Role of mediational discourse in developing an EFL teacher’s oral corrective feedback strategies: A sociocultural perspective. Qualitative Report, 28(3), 784-815. https://doi.org/10.46743/2160-3715/2023.5463

Rassaei, E. (2014). Scaffolded feedback, recasts, and L2 development: A sociocultural perspective. The Modern Language Journal, 98(1), 417-31. https://doi.org/10.1111/j.1540-4781.2014.12060.x

Rassaei, E. (2019). Tailoring mediation to learners’ ZPD: Effects of dynamic and non-dynamic corrective feedback on L2 development. The Language Learning Journal, 47(5), 591-607. https://doi.org/10.1080/09571736.2017.1343863

Sato, S., & Loewen, S. (2018). Metacognitive instruction enhances the effectiveness of corrective feedback: Variable effects of feedback types and linguistic targets. Language Learning, 68(2), 507-545. https://doi.org/10.1111/lang.12283

Sheen, Y. (2004). Corrective feedback and learner uptake in communicative classrooms across instructional settings. Language Teaching Research, 8(3), 263-300. https://doi.org/10.1191/1362168804lr146oa

Shirkhani, S. (2019). Exploring the role of occurring errors distribution in the distribution of corrective feedback targets. Journal of Language and Translation, 9(3), 71-84. https://ttlt.stb.iau.ir/article_668980_1854dfab2fa69f95557ada54861a851d.pdf

Shirkhani, S., & Tajeddin, Z. (2016). L2 teachers’ explicit and implicit corrective feedback and its linguistic focus. Iranian Journal of Applied Linguistics, 19(1), 181-206. https://doi.org/10.18869/acadpub.ijal.19.1.181

Sippel, L. (2019). The impact of peer corrective feedback on vocabulary development. Foreign Language Annals, 52(3), 1-17. https://doi.org/10.1111/flan.12416

Soruç, A., Yuksel, D., McKinley, J., & Grimshaw, T. (2025). Factors influencing EFL teachers’ provision of oral corrective feedback: The role of teaching experience. The Language Learning Journal, 53 (1), 98-113. https://doi.org/10.1080/09571736.2024.2338346

Tan, X., Reynolds, B. L., & Ha, X. V. (2024). Oral corrective feedback on lexical errors: A systematic review. Applied Linguistics Review, 15(3), 1177-1221. https://doi.org/10.1515/applirev-2022-0053

To, J., Tan, K., & Lim, M. (2023). From error-focused to learner-centred feedback practices: Unpacking the development of teacher feedback literacy. Teaching and Teacher Education, 131, 104185. https://doi.org/10.1016/j.tate.2023.104185

Weekly, R., & Pollard, A. (2024). Repositioning corrective feedback to a meaning-orientated approach in the English language classroom. TESL-EJ, 27(4), n4. https://doi.org/10.55593/ej.27108a10

Zarei, M., Ahour, T., & Seifoori, Z. (2018). The effects of implicit, explicit, and emergent oral feedback on Iranian EFL learners’ accuracy, fluency, and attitude. Language Horizons, 2(2), 75-102. https://doi.org/10.22051/lghor.2019.26029.1122

Zhang, X., & Zhang, R. (2023). Feedback, response, and learner development: A sociocultural approach to corrective feedback in second language writing. SAGE Open, 13(1), 21582440231157680. https://doi.org/10.1177/21582440231157680

Zhao, Y., & Ellis, R. (2020). The relative effects of implicit and explicit corrective feedback on the acquisition of 3rd person -s by Chinese university students: A classroom-based study.  Language Teaching Research, 26(3), 361-381. https://doi.org/10.1177/1362168820903343

Copyright of articles rests with the authors. Please cite TESL-EJ appropriately. Editor’s Note: The HTML version contains no page numbers. Please use the PDF version of this article for citations.