in front of the whole class may damage his confidence.
In the case of the best time for error correction, however, teachers are faced with the choice of either correcting immediately following the learner’s erroneous utterance or delaying the correction until later. Long (1977) shed light on the importance of time for the efficacy of corrective feedback. He claimed that an increase in the time between the performance of the skill and the feedback made the feedback become less effective.
The significance of examining immediate and delayed correction derives from the fact that a teacher should identify the best time for error correction. A teacher can focus students’ attention to form reactively, telling them about their errors at the moment the generalization is made, or proactively, telling them in advance about linguistic regularity plus its exception. In their Garden Path technique, Tomasello and Herron (1988, 1989) made a comparison between these two methods for correcting students’ erroneous utterances in the language classroom. They mentioned that if learners’ transfer errors received immediate correction reactively by form-based cognitive comparisons, their performance would be better.
Lightbown (1998) held a different view. She believed that although immediate correction is in order, it has a negative effect on meaning focused activities and it may be safer to leave it until a later time after the commission of the error by the learner. Doughty (2001) underlined the need for exploring when the best time for error correction is, to intrude upon the ordinary language processing. She suggested that immediate correction is more effective than delayed correction since there is a need for the learner to be interrupted in the middle of an act of communication.
Dabaghi (2006) conducted a study to compare the effects of timing of correction — immediate versus delayed correction — and manner of correction — explicit versus implicit correction — on students’ oral production. 56 intermediate level students of English as a Foreign Language the university and private language school settings in Iran participated in this study. The participants were divided into two groups, the immediate correction group and the delayed correction group. The treatment was done during or following the interviews. Students’ discussions were recorded and then transcribed. Subsequently, Individualised tests focusing on the errors that had been corrected were constructed for each participant and administered. Statistical analyses were conducted on the scores the participants received on their individualised tests. The findings indicated no significant differences for timing of correction.
More recently, Rahimi and Dastjerdi (2012) carried out a study to explore an effective error correction method (in this case immediate and/or delayed) in developing learners’ complexity, fluency and accuracy in speech. They also tried to measure the level of anxiety that students experience in the class while the teacher corrects their errors immediately or with some delay. The participants in this study were 20 female intermediate English as a Foreign Language (EFL) learners aged 15 to 20, at one of the English language institutes in Isfahan, Iran. The participants were divided into 2 groups of 10. For group1 errors were corrected immediately and for group2 with some delay, i.e. after finishing their speech during 30 sessions. At the end of the term, each student was asked to discuss one of the topics they had discussed during the term. Students’ discussions were recorded and then transcribed. Measures of accuracy, fluency, and complexity were developed and the findings revealed that delayed error correction has positive effect on fluency and accuracy but not on complexity. For the second aim of the study, all the participants received a Foreign Language Anxiety questionnaire at the end of the term and the results revealed that group2 with delayed correction experienced less anxiety in class.
In sum, it is impossible to come to a firm conclusion pertaining to the relative efficacy of immediate and delayed corrective feedback. As Ellis et al. (2001) reported, the claim that immediate corrective feedback inevitably disrupted fluency work was probably not justified. In their study, the teachers engaged in frequent immediate corrective feedback (along with pre-emptive attention to form), but this did not appear to disrupt the overall communicative flow of the lessons. Accordingly, one of the main arguments for delaying correction would seem to be invalid. Furthermore, there is no proof to show that immediate correction is any more effective than delayed. Despite that, little research has been done on the timing of error correction. It has not been clear, whether, or to what extent the error correction should be carried within the interactional context.
Teacher-, Peer-, or Self-Correction
The choice of corrector has always been a topic that has sparked off long-running debate. In the past, there was no problem since the teacher made the correction by informing the learner directly, either verbally or in writing. Some researchers are very much against the teacher’s dominant role in correction (Hendrickson, 1980). To take this idea one step further, Hedge (2000) suggested that, the students should be given the opportunity to self-correct and, if that fails, other students can perform the correction.
According to Lam (2006), self-correction occurs when students hear themselves making a mistake in pronunciation, grammar, choice of words etc., and they instantly correct it. Sultana (2009) defined ‘self correction’ as the technique which engages students to do the self-repair. A number of studies were conducted on the issue previously. Hendrickson (1980) suggested that, the higher the learners’ level of proficiency, the higher their ability to self-correct because beginners and intermediate learners have not acquired a large amount of knowledge about the target language they are learning, compared with advanced learners. The results of Lalande’s (1982) study suggested that for spelling and grammatical errors, learners should only be informed of the location and features of mistakes, and it is their duty to correct their own errors. Hammerly (1991, p.103) stated that “learning can only take place when students experience the cognitive modifications that will enable them to use each structure and element correctly”. That is, to remember the accurate language, learners should be encouraged by the teachers to find the correct forms of the language themselves. They themselves should correct their errors using teachers’ clues. The studies which have been conducted on input-pushing and output-prompting CF techniques also reveal that output-pushing techniques are superior to input-providing ones (Lyster, 2004; Ammar & Spada, 2006).Some corrective feedback strategies automatically put the burden of correction on the learner. For instance, signaling an error by means of a clarification request or by simply repeating the erroneous utterance by means of a repetition (Hinkel, 2011).
But as Ellis (2009) argued, the lack of sufficient linguistic knowledge on the part of the learners to self correct and ambiguity of output-prompting techniques in terms of being a response to form or meaning are problematic issues concerning self-repair. An alternative approach is peer-correction which has been more extensively practiced in the case of writing. Peer-correction gives emphasis to the significance of allowing learners opportunities to correct their fellow learners’ errors. In Adams’ (۲۰۰۷) study, the feedback between learners on different linguistic features (i.e., English locatives, past tense, questions, and vocabulary) during communicative tasks was tallied. Analyses on the tailor-made test revealed that the learners showed improvement on the items on which they had received feedback from other learners. Likewise, Sato and Ly
ster’s (2007) interview data revealed while working with their peers, learners thought they had more time to decide what to say and felt much more comfortable testing their linguistic hypotheses (see also Swain, Brooks, & Tocalli-Beller, 2002; Gass, Mackey, & Ross-Feldman, 2005). Sato and Lyster (2012) conducted correlational analyses and found a positive correlation between the frequency of feedback provided by peers and L2 development scores (the difference between pre- and posttests). However, Hyland and Hyland (2006) stood against “idealizing L2 peer group interactions as sites of constructive interaction, since the reality can be quite different” (p. 86). On the other hand, Sheen (2011) suggested that learners need careful training in how to conduct a peer correction.
Another alternative is Ellis’s (2009) solution. According to Ellis, CF can be conducted in two stages. First, the teacher needs to focus on and draw learners’ attention to the erroneous form for self-correction and, if that fails, it is the teacher’ responsibility to correct the errors. There are, however, several other options available today:
۱. Another student corrects the error;
۲. The student corrects himself( perhaps after a hint from the teacher);
۳. Small groups of students discuss how to correct the error;
۴. The teacher directs the learner to a reference book (Mirzaei & Saadati, 2010).
On the whole, if a teacher feels that the error is recurrent, or obscures the learner’s intended meaning, he would probably be right in choosing to handle the correction him/herself, either at the time of the error or after the lesson, according to the criteria discussed above. But there may be times when the teacher feels there is more to be gained by encouraging the students themselves or the students as a whole, to correct the error. When the teacher wants to do this, he should indicate that an error has been made and possibly indicate the type of error, and where it occurs in the sentence, at this point he would invite either the error-maker, a member of the class or small groups/pairs to consider the problem and to offer a correction. Thus, the teacher’s role as the choice of corrector is strongly influenced by a number of factors such as the degree of explicitness and implicitness of correction, the type of errors, proficiency level of learners, learning setting and purpose.
Corrective Feedback from Different Viewpoints
From the historical stand point, it is obvious that corrective feedback is a frequent practice in the field of education in general, and in language learning and teaching in particular. It typically involves a learner receiving either formal or informal feedback on his or her performance on various tasks by a teacher or peer(s).The most common terms used in the second language acquisition (SLA) field that refer to corrective feedback are negative evidence, negative feedback and corrective feedback. Owing to possible perplexity arising from the use of this terminology, a brief review of the definitions of terms by some researchers in the field is presented below.
Dulay, Burt, and Krashen (1982) defined feedback as the listener or reader’s response provided to the learners’