Corrective Feedback in Second Language Teaching and Learning

Research, Theory, Applications, Implications

Hossein Nassaji, Eva Kartchava

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781317219934
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: English Language
• Index: Languages & Linguistics

Bringing together current research, analysis, and discussion of the role of corrective feedback in second language teaching and learning, this volume bridges the gap between research and pedagogy by identifying principles of effective feedback strategies and how to use them successfully in classroom instruction. By synthesizing recent works on a range of related themes and topics in this area and integrating them into a single volume, it provides a valuable resource for researchers, graduate students, teachers, and teacher educators in various contexts who seek to enhance their skills and to further their understanding in this key area of second language education.

PART I

Oral Corrective Feedback

1

ORAL CORRECTIVE FEEDBACK IN L2 CLASSROOMS

What We Know so Far

Rod Ellis

Introduction

Corrective feedback (CF) is an aspect of language pedagogy that is important for both teachers and second language acquisition (SLA) researchers. Teachers are concerned with whether they should correct learners’ errors and, if so, when and how. SLA researchers are interested in testing theories of second language (L2) acquisition which make differing claims about the effect that CF has on acquisition and which type is the most effective. CF, then, constitutes an ‘interface issue’ by bringing together the concerns of teachers and researchers. It is perhaps for this reason that CF has attracted enormous interest over the years, as reflected both in teacher guides such as Hedge (2000) and Scrivener (2005) and in the plethora of research articles in both theoretically oriented journals such as Studies in Second Language Acquisition and Language Learning and more teaching-oriented journals such as TESOL Quarterly and Language Teaching Research. Potentially, then, CF constitutes an issue where the insights of teachers and researchers can be mutually informing. What we know about CF requires considering both what experienced teachers have to say about it and what researchers have found out through their research.

There are some notable differences in the perspectives of teachers and researchers. One key difference is that teacher-oriented discussions of feedback consider both positive and negative (corrective) feedback, whereas researchers have focused exclusively on CF. Positive feedback is feedback that provides “an affirmation of the content or correctness of a learner utterance” (Nassaji, 2015, p. 11) and is intended to provide the learner with affective support. Teacher guides often emphasize the importance of positive feedback. Nunan (1991), for example, devotes more attention to positive than corrective feedback, noting that it serves the dual function of letting students know that they have performed correctly and increasing motivation through praise. Teachers and teacher educators see CF as potentially damaging because it can lead to defensiveness in the learner. Thus, they propose it should occur in “an atmosphere of support and warm solidarity” (Ur, 1996, p. 255). In contrast, SLA researchers are concerned only with the cognitive dimension of CF, as they are interested in whether it facilitates acquisition.¹

In the following sections of this chapter, I will draw on an early article by Hendrickson (1978) to examine what teachers and researchers ‘know’ about five key aspects of CF.

1.Should learners’ errors be corrected

2.When should learners’ errors be corrected

3.Which errors should be corrected

4.How should errors be corrected

5.Who should do the correcting

I will draw on some popular teacher guides as evidence of what experienced teacher educators have to say about these issues and on published articles that have investigated CF. As I address each issue, I will not try to resolve potential conflicts in the ‘knowledge’ that these two sources of evidence provide, but in the conclusion to the chapter I will venture some general statements about what we know about CF.

Should Learners’ Errors be Corrected

CF has always been viewed with some suspicion in language pedagogy. Some language teaching methods reject it. According to the Audiolingual Method, errors should be prevented through strict control of learner output, thus removing the need for CF, which was viewed as a form of punishment that can inhibit learning. The humanistic approaches that appeared in the 1970s also advocated against correction on the grounds that it was judgmental and would have a negative impact on the positive self-image that these approaches aimed to foster in learners. Gattegno (1972, p. 31), who developed the Silent Way, for example, proposed that students be allowed “to try their hand and to make mistakes in order to develop their own criteria of rightness, correctness, and adequacy” and that therefore “correction is seldom part of the teacher’s work.” Caring and sharing was the order of the day (Moskowitz, 1978). In Krashen and Terrell’s (1983) Natural Approach error correction is also viewed negatively, as “even in the best of circumstances (it) is likely to have a negative effect on the students’ willingness to express themselves” (p. 177). In early versions of Communicative Language Teaching, where there was an exclusive focus on meaning in order to develop fluency, there was little room for error correction. In later versions of Communicative Language Teaching and in Task-Based Language Teaching, however, CF was reinstated as important for helping learners to develop accuracy, especially if it occurred while learners were struggling to communicate.

The teacher guides, written by authors who had grown up with and experienced the use of these different methods, not surprisingly manifest some uncertainty about CF. Written in the era of post-method pedagogy (Kumaradivelu, 1994), they acknowledge the need for correction but also show concern for its dangers. Ur (1996), for example, recognized that “there is certainly a place for correction” but “we should not over-estimate this contribution” and argued that it would be better to invest time in avoiding errors rather than in correcting them. In other words, Ur favored what Lightbown (1998) called “preventive pedagogy” (p. 139).

CF is often discussed in relation to ‘accuracy’ and ‘fluency.’ Scrivener (2005), for example, proposed that error correction was necessary in accuracy work, but should be largely avoided in fluency work or delayed until the fluency activity had been completed. Bohlke (2014) also advised against correction during fluency activities, commenting that “during fluency activity, it is generally accepted that the teacher should not interrupt students to point out a grammar or vocabulary error, or to correct pronunciation” (p. 127) and going on to claim that “many teachers feel that the only appropriate time to focus on error correction is after the activity is completed” (p. 128). This view reflects the belief that in Communicative Language Teaching, communication should be unfettered. However, the authors of the guides hedge their advice. Scrivener did allow for “brief, unobtrusive, immediate correction” (p. 299) during fluency work, as did Ur (1996), who saw merit in “gentle, supportive intervention” to help the “floundering” student. These views reflect an attempt to achieve a balance among the positions adopted in the different methods. They are apparently based on the authors’ experiences of teaching and observing teachers teach and not on research, which is never mentioned.

Research suggests, however, that while Bohlke (2014) is right that teachers resist correcting during fluency work, in fact they do correct. Basturkmen, Lowen and Ellis (2004) investigated experienced teachers’ beliefs about focusing on form by means of CF during communicative lessons and their actual practice. They reported inconsistencies in the teachers’ beliefs and a tenuous relationship between their beliefs and actual practice. Even though all of the teachers believed that correction in general should be avoided during fluency work, they frequently engaged in it.

The key issue, however, is whether CF ‘works’ (i.e., assists acquisition). There has been a wealth of classroom- and laboratory-based research that has addressed this issue. The research has been synthesized in a number of meta-analyses (e.g., Russell & Spada, 2006; Mackey & Goo, 2007; Li, 2010; Lyster & Saito, 2010), which show that CF is indeed effective in assisting acquisition. Li (2010), for example, reviewed a total of 33 oral CF studies, involving 1,773 learners, concluding that “corrective feedback had a medium effect on acquisition” (p. 335), which was evident in tests that immediately followed the treatment involving CF and over time. However, he also reported that the effect was much stronger in studies carried out in a laboratory than in a classroom, perhaps because learners are more likely to pay attention to the feedback they receive in the one-on-one interactions in a laboratory context than in the teacher–class interactions typical of the classroom studies. Li also found that the effect of CF was greater in foreign language than in second language settings and suggested that this might be because learners in the former are more predisposed to pay attention to the corrections that they receive. Another finding was that CF was more effective in treatments that involved discrete item practice of grammatical structures (i.e., in accuracy work), where the feedback is intensive, than in communicative activities (i.e., in fluency work), where it is not focused on a single linguistic feature. A key issue in determining whether CF has any effect is the nature of the tests used to measure learning. Li also investigated this, reporting that the effects of CF were evident in both tests that measured controlled language use and free production. Two general conclusions can be drawn from Li’s meta-analysis: (1) CF does assist L2 acquisition and (2) it is more likely to be effective in contexts where it is salient to learners.

Sociocultural theory points to the need for the ‘graduated’ correction of errors and proposes that CF can also be examined in terms of the level of support that teachers provide when they correct learners’ errors. Aljaafreh and Lantolf (1994) and Nassaji and Swain (2000) showed that the level of correction that a teacher used to correct a specific type of error diminished over time. That is, whereas quite explicit correction was needed to enable a learner to self-correct at one point in time, more implicit correction sufficed at a later time. They suggested that this constituted evidence of acquisition taking place.² Nassaji (2011a) also reported that negotiated feedback is more effective than non-negotiated feedback or feedback with limited negotiation.

When Should Learners’ Errors be Corrected

Oral errors can be corrected immediately or shortly after they are committed. Alternatively, correction can be delayed until the pedagogic activity has been completed. We have seen that the teacher guides favor immediate correction in accuracy work but recommend delaying it in fluency work. However, Gattegno (1972) advised against interfering immediately even in accuracy-oriented work, so as to “give time to a student to make sense of ‘mistakes’” (p. 31). Discussing computer-mediated communication, where both immediate and delayed correction are possible, Bowyer and Kawaguchi (2011) argued that delayed, asynchronous correction is preferable to synchronous correction because it gives teachers time to identify and explain errors and also because learners have plenty of time to consider the corrections to their output.

To date, there has been very little research that has investigated the timing of CF. Research so far has focused on describing and investigating the effects of immediate CF. Rolin-Ianziti (2010), however, reported a descriptive study of delayed oral CF carried out by teachers of L2 French. The teachers noted the errors that learners made as they performed a role-play activity and then reviewed them later. One teacher simply provided the corrections, while another attempted to elicit corrections from the students and only provided the correction if the students failed to self-correct. Rolin-Ianzati drew on sociocultural theory to argue that the second approach was likely to be more effective. In an experimental study, Li, Zhu and Ellis (2016) investigated the comparative effects of immediate and delayed CF on the learning of the English past passive construction by Chinese junior high school students. In this study there were two feedback groups, one of which received immediate feedback and the other delayed feedback (i.e., after the learners had completed two tasks). Both immediate and delayed feedback led to gains in scores on a grammaticality judgment test, with immediate feedback also showing some advantage over delayed feedback. However, neither type of feedback had any effect on scores from an oral elicitation test. The limited effect of the CF in this study may have been because the learners were not developmentally ready to acquire the target feature (the English past passive construction).

There are theoretical grounds for preferring immediate feedback. Doughty (2001) argued that immediate feedback enables learners to map a specific form onto the meaning it conveys in a ‘window of opportunity’ (i.e., at that moment when the learner is struggling to express him/herself). The Transfer Appropriate Processing Hypothesis (Lightbown, 1998) also lends support to immediate feedback. This hypothesis posits that learning is context dependent and that learners will be better able to recall rules and forms in a communicative context if they had acquired them in such a context. In other words, immediate feedback occurring while learners communicate is more likely to result in the kind of L2 knowledge that can be accessed later for communication. From this perspective, immediate feedback is preferred because it integrates ...

Table of contents

• Cover
• Half-Title
• Series
• Title
• Copyright
• Contents
• Acknowledgements
• Introduction
• Part I: Oral Corrective Feedback
• Part II: Computer-Mediated Corrective Feedback
• Part III: Written Corrective Feedback
• Part IV: Student and Teacher Issues in Corrective Feedback
• Conclusion, reflections, and final remarks
• List of Contributors
• Index