Teacher Belief of Corrective Feedback in Grammar Teaching: Current Status and Future Prospects

Having been generally believed to impact on teachers’ practice, teachers’ belief has been studied for several decades, most of which concentrates on grammar teaching and learning. Moreover, the role of corrective feedback (CF) in second language acquisition has received considerable attention over the past decades. The studies concentrated on the overall effect of feedback on second and foreign language development, including the occurrence and effects of different types of feedback, leaners’ perception of CF, etc. In general, having been divided into three major categories (grammar, vocabulary and pronunciation), CF has been proved to have a positive effect on overall language learning. However, despite numerous studies on CF from various perspectives, the researches on teachers’ CF beliefs are fewer and that on CF beliefs in relation to practices are even scantier. As both teachers’ belief and CF are drawing greater attention, some researchers suggested making a combination between them. But up to now, few studies have focused on senior high school teachers’ CF belief on practices in EFL contexts, making it hard to figure out the significance of the relationship between CF belief and practice, and the factors that impact the relationship. Hence, this paper aims at exploring the history of the development of teachers’ belief and instructional practices of CF in grammar teaching at home and abroad, aiming at deepening the understanding of teachers’ underlying thoughts and providing suggestions for teacher education and future study orientation.


Introduction
Over the past decades, researchers have initiated different definitions of teachers' belief. Kagan proposed that teachers' beliefs were a kind of special and personal implicit presumptions about teaching practice [1]. Pajares thought that teachers' beliefs referred to teachers' firm views on teaching work, the role of the teacher, students, course, learning, which cover teachers' practice and experience to guide teachers' thoughts and behaviors [2]. Teacher beliefs are essential as emphasized by Richards and Lockhart who stated that what teachers believe and know as well as teacher-knowledge and teacher-thinking provide the underlying framework and schemata which guides the teachers' classroom actions [3]. Basturkmen et al. emphasized that teachers' stated beliefs are statements that teachers expressed about their ideas, thoughts, and knowledge as evaluations of what 'should be done', 'should be the case', and 'is preferable' [4]. Later, Borg argued that although the definitions of teachers' beliefs vary, the term is usually used to refer to evaluative propositions which teachers hold consciously or unconsciously and which they accept as true while recognizing that other teachers may hold alternative beliefs on the same issue [5]. Accordingly, it is commonly agreed that teachers' beliefs have impact on their practical teaching practices. But it is difficult for researchers to arrive at a consensus about the relationship between teacher beliefs and teaching behaviors.
Moreover, CF has long been a significant focus of attention in research into language learning and teaching, especially since it became accepted that an explicit focus on form could contribute to learning within a communicative language teaching framework. According to Ellis et al., corrective feedback takes the form of responses to learner utterances that contain an error, which consist of (a) an indication that an error has been committed, (b) provision of the correct target language form, or (c) metalinguistic information about the nature of the error, or any combination of these [6]. Lyster and Ranta introduced a refined taxonomy for coding incidental CF in meaning-focused L2 classroom instruction that included six prominent types of oral CF teachers provide, which continues to be widely adopted in coding CF types in observational reports [7]. With the increasing interest in both corrective feedback and teachers' belief, there is a tendency that researchers commence investigating teachers' CF beliefs. Despite the fact that providing CF is an essential task in foreign language acquisition, teachers' responses to students' errors and the pedagogical beliefs behind have not been thoroughly investigated, especially in Chinese EFL (English as a Foreign Language) context. Hence, it is necessary to investigate teachers' CF beliefs and the way different types of CF are put into practice in different contexts in order to further both teachers and researches' understanding of the relationship between CF beliefs and practices in Chinese EFL contexts, improve teachers' teaching by making more conscious and informed decisions when dealing with their learners' errors and provide corresponding suggestions for teacher training and education. Therefore, this paper reviews former researches on both teacher belief and corrective feedback in order to present a holistic picture and offer feasible suggestions for relevant future research.

Teacher Belief About Grammar Teaching
Ever since the 1970s, teacher belief has been scrutinized in a plethora of studies that have been conducted to understand what teachers do in operational teaching settings. Research into teacher belief has long been acknowledged as a major area of research in the field of second and foreign language teaching, especially in grammar teaching. There is a consensus that teachers' beliefs play a significant role in teaching practices, by informing, motivating, guiding, or shaping teachers' decision-making process and pedagogical behaviors. In terms of grammar, Brown believed that it is "a system of rules governing the conventional arrangement and relationship of words in a sentence", which can facilitate the acquisition of a foreign language and is conducive for cultivating comprehensive language competence [8]. The grammar teaching beliefs held by teachers can affect their practical teaching behaviors in class, thus can have different teaching results in the end. Research on teacher belief about grammar teaching has centered on several aspects, ranging from the correspondence between teacher belief and practice as regard to grammar teaching, factors that influence what teachers believe in and their practices, to the comparison between teacher belief and student belief, as well as the comparison between experienced and novice teachers, etc.

The relationship Between Teacher Belief and Practice in Grammar Teaching
Tracing back to the 1990s, numerous studies contribute to a holistic picture of the relationship between teacher belief and grammar teaching, pointing to different versions of conclusion, ranging from consistencies to inconsistencies. Johnson investigated and observed how three secondary ESL teachers' theoretical beliefs correspond with their instructional practices, and found that their literacy instruction was consistent with theoretical orientation [9]. Borg conducted classroom observations and interview with teachers and presented two case studies regarding teachers' self-perception and practice in teaching grammar, aiming at gaining insight into the factors that influenced teachers' instructional decisions in teaching grammar. It was concluded that the way teachers perceive their knowledge about grammar clearly does influence what they do in the classroom, although how this actually occurs may vary across teachers, and teachers' self-perception is but one of a range of other factors which shape their practices in teaching grammar [10]. Phipps and Borg examined the divergences between what English teachers say and do in teaching grammar, and, by exploring the reasons for these, also provide insight into deeper tensions among competing beliefs that teachers hold [11]. Deng and Lin proposed that teachers' beliefs can influence their practical teaching and there are numerous studies on teaching beliefs and teaching behaviors. However, it can be difficult for all the researchers to hold a consistent opinion on the relationship between teaching beliefs and teaching behaviors [12].

Factors Shaping Teacher Belief and Actual Practice in Grammar Teaching
The former researches witness a variety of factors that may shape teachers' beliefs and actual practice in grammar teaching. To begin with, Richards listed two different types of knowledge that may influence teachers' understanding and practice of teaching. The first one concerns the curricular goals, lesson plans, instructional activities, materials, tasks, and teaching techniques. The other relates to teachers' personal and subjective philosophies and their understanding of what constitute good teaching [13]. In his review of teacher cognition in grammar teaching, Borg pointed out that teachers referred to various factors shaping their views in articulating their rationales, including student wants and syllabus expectations, etc. However, teachers' experience as teachers and learners had a particularly powerful influence on their views about grammar teaching [14]. Exploring teachers' tensions between their grammar teaching beliefs and practices, Phipps and Borg pointed out that not only student expectations and preferences, but also classroom management concerns led teachers to take decisions and actions that may be contrary to their stated beliefs [11]. Borg examined the impact of in-service teacher education on teachers' beliefs and concluded that through teacher education, teachers' beliefs can be strengthened, extended, and made more apparent to themselves. And teachers can learn how to put their beliefs into practice and also develop links between their beliefs and theories [15]. Sanchez and Borg explored the role of cognitive and contextual factors in defining the selection and use of teachers' classroom decisions and practices, including teachers' perceptions of their knowledge of grammar, their beliefs about the value of grammar in L2 learning, their interpretations of their context (particularly of their learners), as well as their personal experience [16]. In investigating the relationship between teachers' CF beliefs and practices, Bao concluded several factors that may account for the inconsistencies in the relationship, including the complexities of classroom life, teachers' varying teaching experience and individual learner differences [17]. All in all, there is a large amount of studies that dig deep into the factors that influence teachers' beliefs and practices, which offers significant enlightenment for further research into teachers' beliefs.

Comparative Studies About Teacher Belief in Grammar Teaching
Comparative studies regarding teacher belief in grammar teaching has also been abundant in recent decades. On the one hand, researchers shed light on the comparison between teacher belief and student belief in which the high consistency is believed to lead to effective teaching [18]. Borg concluded in his review that teachers' and students' views about aspects of grammar teaching may differ considerably and suggested that such differences may be detrimental to the effectiveness of teachers' formal instruction [14]. Deng and Lin compared grammar teaching beliefs between teachers and students and pointed out two main reasons for the inconsistencies, including the tension between communicative teaching and examination-oriented learning, and students' preference for the traditional teacher-centered way of grammar teaching. Several corresponding suggestions for both teachers and students are provided. On the other hand, it was proposed that most English language teachers have their own set of beliefs of how grammar should be taught and the years of teaching experience can easily shape such beliefs [12]. To illustrate, through an experimental research centering on teachers' beliefs in grammar teaching, Nurusus et al. came to the conclusion that the years of teaching experience can have an influence on a teachers' beliefs regarding the teaching of grammar, and suggested that teacher educators and other stakeholders in education must therefore pay great attention to continuous professional development and training in order to ensure that teachers continue to hold correct and appropriate beliefs in the teaching of grammar and consequently teach grammar in their classes accordingly [19].

Corrective Feedback in Language Teaching
Schegloff et al., as cited by Kirgoz and Agcam, defined the term corrective feedback (CF) as "the replacement of error or mistake by what is correct" [20]. CF, according to Chaudron, is "any teacher behavior that minimally attempts to inform the learner of the fact of error". The purpose of feedback is to aid the learners with correcting mistakes and improving their understanding, fluency, and accuracy in an L2 [21]. In 1978, Hendrickson proposed five questions concerning error correction, which incorporate "Should learners' errors be corrected? When should learners' errors be corrected? Which errors should be corrected? How should errors be corrected? Who should do the correction?" [22]. Henceforth, a considerable amount of researches have been carried out to study CF taxonomy, to testify the effectiveness of different types of CF, and to find out teachers' and learners' attitudes towards error correction, etc. This part puts greater emphasis only on these aspects, which lay the foundation for further illustration.
To begin with, CF can generally be divided into two types-oral and written CF. Sheen contended that there are a number of ways in which oral and written CF potentially differ. First, oral CF is more likely to be noticed by the leaner as correction. Second, oral CF is provided online while written CF is delayed. Third, oral CF is typically directed at individual leaners but is available to the rest of the class as hearers [23]. In terms of written CF, Ellis identified six basic strategies for providing written CF, including direct CF, indirect CF, metalinguistic CF, focused versus unfocused CF, electronic feedback and reformulation [24]. Junqueira and Payant investigated teacher feedback beliefs and practices of a pre-service L2 writing teacher over one academic semester by coding the types of written CF as "direct, direct with explanation, indirect, indirect with explanation" [25]. Sheen also divided written CF into written direct correction and written metalinguistic correction, giving detailed definition at the same time [23]. As for oral corrective feedback, Lyster and Ranta's seminal study identified six types of CF that teachers use in response to students' oral errors (recast, explicit correction, elicitation, clarification request, metalinguistic cue, and repetition) and linguistic foci (lexical, phonological, and grammatical errors) [7]. These taxonomies have remained dominant in observational studies conducted in a growing range of second language teaching contexts [26]. Ellis et al. pointed out that CF differs in terms of how implicit or explicit it is. In the case of implicit feedback, there is no overt indicator that an error has been committed, whereas in explicit feedback types, there is, lending theoretical relevance regarding the importance of noticing in language learning [6]. Besides, in Li's meta-analysis, regarding the implicit and explicit type of CF, it was stated that implicit feedback included recasts, negotiation (clarification requests, elicitation, and repetition), and any type of feedback that was not intended to overtly draw the learner's attention to his/her erroneous production. And explicit feedback included metalinguistic feedback, explicit correction, and any feedback type that overtly indicated that the learner's L2 output was not acceptable [27]. Moreover, according to Li, Loewen and Nabei pointed out that recasts and explicit correction could be labeled "other repair" and prompts "self-repair" [27]. Regarding the timing of CF, Li et al. argued that when feedback is directed at specific errors and not the process of task completion, immediate feedback is more effective. In contrast, delayed feedback is more helpful when it addresses process issues and how a task is carried out [28]. Furthermore, Fallah and Nazari interpreted that CF only impacts declarative/explicit knowledge and not the procedural/implicit knowledge [29]. In addition, research has also attested to the effect of peer CF. Sippel and Jackson compared the effect of oral peer and teacher feedback on the acquisition of the German present perfect tense in an experimental study, with the result showing that both teacher and peer feedback groups improved significantly, and peer group also outperformed the teacher feedback group. The findings suggested that peer CF raises students' awareness of linguistic forms and they also benefit from not only receiving feedback, but also from providing it [30]. All in all, the taxonomy of CF is diversified and research with different focuses can apply different kinds of taxonomy accordingly.
The past decades have also witnessed a rapid increase in the research on the effectiveness of CF. According to Brown, corrective feedback in second language acquisition (SLA) refers to the responses to a learner's non-target-like L2 production [27]. As cited by Pawlak, Ramirez and Stromquist found a positive correlation between error correction and individual gains in second language proficiency. And Lightbown and Spada also contended that CF delivered during communicative lessons in an intensive EFL program helped to reduce frequency of some types of errors [31]. Norris and Ortega meta-analyzed the empirical studies published between 1980 and 1998 on the effectiveness of second language instructional treatments, and found that the average effect size for focus-on-form treatments was slightly larger than that for focus-on-forms treatments and that explicit instruction had substantially larger effect size than implicit instruction [32]. In comparing the effectiveness of implicit and explicit feedback, Ellis et al. contended that both types of feedback have a positive impact on second language acquisition [6]. Li's meta-analysis on the effectiveness of CF in SLA also concluded that there was a medium overall effect for CF and the effect was maintained over time [27]. Yoshida contended that researchers focused on how different types of CF are effective for learners' second language/foreign language acquisition [33]. Roothooft reported that there is a general consensus within language teaching circles that some form of feedback is positive and even necessary. And several other meta-analyses have been published, which all indicate a positive role for CF for the acquisition of second language grammar [34]. Brown also stated that recent years have witnessed a growing number of meta-analyses that synthesized various domains of CF research with general findings that lend substantial support to the efficacy of CF in second language learning [26].
Besides the bulk of studies exploring the effectiveness of CF, research has also dealt with the perceptions of those involved in the act of providing and receiving feedback, with a major focus on the latter. For instance, Mackey et al. explored learners' perception of CF and found that learners were relatively accurate in their perceptions about lexical, semantic and phonological feedback [35]. Yoshida investigated the CF preferences of seven learners of Japanese in stimulated recall sessions and found that most of them preferred having time for self-correction [36]. Yoshida examined Japanese language teachers' and learners' perceptions of CF and found that their perceptions of CF and the learners' responses to CF are associated with teachers' perception of individual learners, learners' perception of classroom interactions, as well as the various types of CF [33]. Fu and Nassaji stated that a number of studies have looked in particular at leaners' perception of recasts, one of which examined perception of recasts and its relationship with error types and feedback characteristics and the results indicated that leaners are fairly accurate in perceiving the target of recasts [37]. Furthermore, comparative studies of teachers' and students' views of corrective feedback have also emerged. Schulz investigated teachers' and students' views about CF, suggesting that while students tend to prefer explicit forms of error correction, most teachers have a preference for implicit types. And it was proposed that given the detrimental effects brought about by discrepancies between teacher and students' belief systems, teachers should pay great attention to explore students' perceptions regarding the factors believed to enhance the learning of a new language and make effort to deal with potential conflicts between students' beliefs and instructional practices [38]. Lyster et al., as cited by Kirgoz and Agcam, also advocated that learner and teacher preference for CF have been investigated for two main reasons: learner preferences can influence learning behaviors and mismatches may result in negative effects on learning [20].

Teacher Belief About Corrective Feedback in Grammar Teaching
Basturkmen reviewed the researches into the correspondence between language teachers' stated beliefs and practices and suggested that further research is needed on teachers' beliefs in relation to incidental aspects of teaching practice, such as error correction [39]. Roothooft presented a comparison between the observation of 10 EFL teachers and their sated beliefs about oral corrective feedback and pointed out that the question of teachers' beliefs about oral corrective feedback has received little attention, especially in comparison to the studies investigating feedback effectiveness [34]. According to Fallah and Nazari, "although research on CF is well visited in second language acquisition, the exploration of teachers' cognitions of CF has not moved in tandem with this thick literature" [29].
Beyond reporting on the types and foci of CF provided in language classroom, earlier research also examined how teachers conceptualize CF and compared teachers' perceptions to their students' expectations. These studies turned out that there was a mismatch between teachers' and students' preference for CF, which was triggered by teachers' perception that CF was not necessary. According to Junqueira and Kim, the dichotomy was seen as an important factor in understanding teachers' beliefs toward their students and the CF practices in their classroom, promoting researchers to further examine the correlation between teachers' beliefs and their actual use of CF [40]. Some studies have reported consistency between teachers' beliefs and actual practices [17], while others have provided evidence to the contrary. For example, Basturkmen et al. conducted a case study with 3 ESL teachers teaching the same courses, examining their stated beliefs about and the use of CF. The findings revealed that there were discrepancies between teachers' stated beliefs and their own practices [4]. Dilāns studied with L2 Latvian teachers and found that teachers' beliefs on CF were in line with their classroom practices regarding CF techniques but were inconsistent in the frequency with which they used these techniques [41].
Previous studies have also analyzed possible reasons for the inconsistencies between teachers' stated beliefs and actual practices. Some identified reasons were related to teachers' varying learning and teaching experience [4]. For instance, Junqueira and Kim pointed out that as an important factor in the interaction between teacher beliefs and provision of CF, teaching experience and previous training needs more attention [40]. In a relatively recent study, Rahimi and Zhang explored novice teachers' and experienced teachers' cognition of CF through questionnaire and interview, concluding that the teachers' personal experience had an impact on their choice of CF, its timing and their overall view on CF effectiveness [42]. Fallah and Nazari examined the mediating role of experience in L2 teachers' beliefs about corrective feedback, and pointed out that although research on CF is well-visited in second language acquisition, the exploration of teachers' cognitions of CF has not been conducted enough. It was proposed that by raising teachers' awareness of their own beliefs and also the empirical results of the effectiveness of CF, teachers may be able to scrutinize their entrenched beliefs and aid them with (re)constructing their conceptions through launching teacher education initiatives [29]. Some studies attributed these discrepancies to the complexity of classroom life that may constrain teachers in their attempts to align their instructional practices with their beliefs. To illustrate, Sheen examined the contextual variables that influence CF, which revealed that proportions of CF types vary considerably across contexts [43]. Based on qualitative data, Mori explored how the knowledge and beliefs of two EFL professionals shaped their beliefs and proposed that both instructional focus and time constraints can influence teachers' actual practice regarding CF [44]. Moreover, the inconsistencies are also attributed to individual learner differences such as personality, level of communication ability etc. Han suggested that teachers' perception of leaners' problems that are causing errors can influence their choice of CF [45]. In analyzing teachers' explanations to shed light on their concept of pedagogical content knowledge in an English as a second language context, Jonhnston and Goettsch's study, as cited by Borg, found that "teachers' beliefs about how learners learn and what they know affect their pedagogical strategies" [14].
Junqueira and Kim also concluded that teachers' beliefs about what types of learner-error CF should focus on, as well as their beliefs about their own students' expectation, appear to heavily influence their actual CF practices [40]. Bao proposed that reasons for the inconsistencies between teachers' CF beliefs and practices may be attributed to the instructional focus, the target language and the class size [17]. Figuring out the influencing factors of the inconsistencies between teacher beliefs and practices about CF contributes to teachers' further understanding of their teaching practices, which can help to strengthen or reconstruct their CF beliefs, or to moderate their CF practices.

Future Direction
Overall, it can be concluded from the above-reviewed studies that teachers' stated beliefs and actual practices on CF represent a complex interplay of various factors. While substantial empirical research is found in western contexts, there is little information of this kind from China. Brown's meta-analysis found that most CF focused on grammar (43%), followed by lexis (28%) and pronunciation (22%). However, many investigations have not distinguished between CF on grammar, lexis and pronunciation, making the extent to which research findings apply to grammar CF vague and obscure [26].
According to Roothooft, even if further research on CF manages to answer all the questions posed by Hendrickson as to whether, when and how leaners errors should be corrected, it will clearly still be necessary to gather information about teachers' beliefs and feedback practices if we intend second language research to have an impact on classroom practice [34]. Furthermore, as grammar teaching plays an essential role in foreign language acquisition, which influences learning outcomes to a large extent, it is quite necessary and essential to conduct further research on teacher belief about corrective feedback in grammar teaching in EFL context to figure out a holistic picture of topic. Finding out the relationship between teacher beliefs and behavior and analyzing the factors that influence teachers' choice of corrective feedback can not only contribute to teachers' further understanding of their own stated beliefs and behaviors, but also have a significant enlightenment on the improvement of both pre-service and in-service English teacher education. In addition, research on teachers' beliefs on CF would provide teacher educator with important information to help determine curricula and program direction.

Conclusion
In conclusion, the current research offers a general review of studies that concentrate on teachers' beliefs in grammar teaching, elaborating on prior research on the relationship between teachers' beliefs and practice, factors that shape the beliefs, as well as the comparative studies in this field. Furthermore, as corrective feedback is gaining increasing interest in the field of second language acquisition, this study also gives an overview of relevant research on corrective feedback, which mainly focuses on the classification and effectiveness of corrective feedback, and the perceptions of those involved in the process of corrective feedback. And it turned out that there is a dearth of research to investigate teachers' beliefs about corrective feedback in grammar teaching. Therefore, for the enrichment of this field, further studies that focus on the relationship between teachers' beliefs and practices of corrective feedback in grammar teaching and factors that influence the beliefs are needed.