ESL students’ perceptions of error correction techniques in oral production: a level-based approach

Размещено на http://www.Allbest.Ru/

Khmelnytskyi National University,

ESL students' perceptions of error correction techniques in oral production: a level-based approach

M. Zembytska

Yu. Romanova

N. Chumak

Khmelnytskyi, Kyiv - Ukraine

Abstract

The article reveals the findings from a survey examining students' perceptions of corrective feedback involving six groups of the first- through fifth-year ESL students enrolled in Bachelor's and Master's programs in English Language and Literature, Applied Linguistics, International Law, and International Communication and Global Media. The study attempts to reveal students' perceptions of corrective feedback, as well as its perceived effectiveness and psychological relevance, which are analyzed on the basis of students' answers. The aim of this survey-based research was to explore ESL students' preferences for the amount and type of corrective feedback in speaking/reading and develop a method to help educators effectively choose the types of corrective feedback on the basis of their students' level of English. In order to reinforce the study with substantial theoretical evidence, each type of corrective feedback was characterized on the basis of a rigorous review of related evidence-focused literature. The survey, which was administered to 78 ESL students at Taras Shevchenko National University of Kyiv and Khmelnytskyi National University, Ukraine, demonstrated a number of discrepancies in students' preferences and attitudes. Their responses constituted grounds for assessing and ranking the prevalent verbal correction techniques in ESL teaching according to their perceived relevance. The implications of the current study could be taken into consideration by ESL teachers for determining an optimal set of error correction techniques in their own classrooms on the basis of their students' level of English.

Keywords: ESL students, perceptions, error correction, corrective feedback, error treatment, oral production, level-based approach, learner uptake, the Bayes' formula/theorem.

Анотація

Зембицька М., Романова Ю., Чумак Н. Сприйняття методів виправлення помилок в усному мовленні студентами, які вивчають англійську мову як другу іноземну: підхід на основі рівня володіння мовою

У статті викладено результати дослідження, проведеного на основі опитування студентів у шістьох групах з першого по п'ятий курси, які вивчають англійську мову як другу іноземну в рамках освітнього кваліфікаційного рівня “Бакалавр” та “Магістр” за спеціальностями “Англійська мова та література”, “Прикладна лінгвістика”, “Міжнародне право” та “Міжнародні комунікації”.

У дослідженні зроблено спробу з'ясувати особливості сприйняття студентами коригувального зворотного зв'язку за допомогою аналізу його передбачуваної ефективності та психологічної комфортності на основі індивідуальних відповідей студентів.

Метою дослідження було на основі опитування з'ясувати, який тип та обсяг коригувального зворотного зв'язку під час говоріння/читання є бажаним з погляду студентів, які вивчають англійську мову як іноземну, та розробити метод, який допоможе вчителям ефективно обирати типи коригувального зворотного зв'язку на основі рівня володіння англійської мовою їхніми студентами. Кожен тип коригувального зворотного зв'язку був теоретично обґрунтований на основі вивчення наукової літератури, що спирається на відповідні емпіричні дослідження. Опитування, проведене у двох університетах України (Київському національному університеті імені Тараса Шевченка і Хмельницькому національному університеті) серед 78 студентів, які вивчають англійську мову як першу іноземну, продемонструвало низку розбіжностей в уподобаннях та ставленні студентів до різних методів виправлення помилок.

Ключові слова: студенти, які вивчають англійську мову як іноземну; сприйняття, виправлення помилок, коригувальний зворотній зв 'язок, виправлення помилок, усне мовлення, підхід на основі рівня володіння мовою, засвоєння, теорема/формула Баєса.

Introduction

In the course of learning a second/foreign language, students regularly produce oral and written utterances which do not conform to the standards of the language being acquired, i.e. erroneous or ill-formed utterances (Corder, 2008). The traditional perception of errors as a sign of learners' incompetence has given way to viewing errors as a marker of the difficulties that learners are having with particular aspects of the language, which could be explained by the persistence of the habits imposed by their mother tongue. Numerous studies, taking the stance that corrective feedback constitutes an integral part of L2 instruction, testify to the acute need of providing more coherence between theory, research and pedagogy with respect to this domain. Since error making is a natural and inevitable part of SLA process, error treatment could reasonably be considered a significant component of L2 teaching which directly affects improvement. Hence, the question arises, which strategies should be chosen to facilitate better student learning on the one hand, and to avoid discouraging, demotivating or simply ineffective instruction on the other hand.

There is a vast body of research on students' perceptions regarding error treatment in foreign language teaching (Chen, 2005; Lasagabaster & Sierra, 2005; Lyster et al., 2013; Perdomo, 2018; Quinn, 2014; Sheen, 2010), greater attention being paid to corrective feedback in L2 written activities (Bitchener, 2008; Ferris & Roberts, 2001; Lee, 2004; Rezaei et al., 2017). Numerous studies have examined error correction strategies implemented in teaching ESL students worldwide. However, a substantially smaller number of studies have focused on Eastern European ESL learners' attitudes and preferences in this domain, with Ukraine kept away from the spotlight. Therefore, this research was carried out to reveal Ukrainian students' perceptions and attitudes towards the oral correction techniques currently used in ESL teaching. Although researchers and ESL educators generally maintain that discovering and matching the expectations of teachers and students is beneficial for language learning, they have not demonstrated consistent approaches to error correction, disagreeing about the extent to which errors should be corrected, the types of errors that need correction, and the techniques to be used.

Therefore, the present research aims to explore ESL students' attitude towards the corrective feedback received while speaking/reading and identify the correlation between the students' L2 proficiency level and their preferences for corrective feedback, which could help educators develop the most effective error correction model for their ESL classroom corresponding to the students' expectations for the type and the amount of the corrective feedback provided.

The research questions posed are as follows:

1. What are the most effective oral correction techniques according to ESL students' perceptions?

2. What types of corrective feedback are more psychologically relevant than others?

3. How do students' preferences correlate with their level of English?

There are different views on the nature of linguistic errors, such as Lennon's error analysis (Lennon, 2008), which has indicated that certain errors recurring among language learners result from the intrinsic difficulty of the subsystem involved, rather than due to the cross-lingual influence. The author maintains that `No matter what your first language is, whether it has prepositions or not, you will almost certainly find it very difficult to make no mistakes in English prepositions' (Lennon, 2008). As explained by the researcher, even if the ESL learner's first language has the category of verbal aspect, the choice between simple and progressive verb forms in English are likely to be erroneous. These `developmental errors' are usually made regardless of their first language backgroun d and can hardly be predicted by contrastive analysis (Lennon, 2008). The assumption that error correction constitutes an essential aspect of language teaching also resonates in studies of Bartram and Walton (1991), Purwati (2012), Touchie (1986), and others.

To have a clear understanding of what types of errors are speculated on, it is important to differentiate between true errors of competence, revealing the extent of the learner's proficiency in the second language, and mistakes - errors of performance, having a circumstantial origin similar to slips of the tongue in the native language (Corder, 1967). Mistakes are made by both native speakers and foreign language learners, and can be self-corrected when pointed to the speaker. Errors in SLA are traditionally treated as a deviation from the rule, primarily due to the lack of knowledge. As defined by Lennon (1991), an error is `a linguistic form or combination of forms which in the same context and under similar conditions of production would, in all likelihood, not be produced by native speakers' counterparts' (Lennon, 1991, p. 182).

Method. The study involved six groups of EFL students in their first, second, third, fourth and fifth year of studies earning their Bachelor's and Master's degrees at Taras Shevchenko National University of Kyiv (Ukraine) and Khmelnytskyi National University (Ukraine) majoring in English Language and Literature, Applied Linguistics, International Law, International Communication and Global Media. The data were collected through the use of an online survey (Zembytska et al., 2019a; 2019b), classroom observations, and follow-up interviews with students. To eliminate a possibility of misinterpretation in students' responses, the terms `error' and `mistake' were clarified. Also, the questions selected for the survey were pilot-tested with a small group of similar candidates to avoid ambiguity. All the participants (78 students) were instructed to ensure their understanding of the terminology used for different types of corrective feedback.

Results and Discussion

When asked to reflect on each type of corrective feedback during the follow -up interviews, students typically characterized explicit correction as a method `easier for comprehension', `conventional and therefore understandable', and `faster'. While other students sustained the benefits of repetition because it was `more likely to prevent errors in the future', `quite understandable', `easy to remember', much fewer students preferred clarification and recast as techniques which `stimulated thinking', `worked on a larger scale' (contributed to a deeper revision of rules, patterns etc.) were `challenging in a good way' and `helpful for memorizing the errors'.

Figure 1. Students' responses to the question addressing the range of error correction techniques used by their teachers

While analyzing students' responses in the online survey (Zembytska et al., 2019a) (Fig. 1), we found that the prevailing error correction technique used by their teachers was explicit correction. This type of response was given by the majority of students (46.7%). This result supports the findings from a qualitative study by Tsang (2004), which demonstrated that teachers mostly used explicit correction and recast in ESL error correction. However, according to Tsang (2004), these techniques did not guarantee student-generated repair, while repetition was the most frequent type of feedback resulting in student repairs. Still, their study suggests that explicit correction and recast appeared to be helpful in treating phonological errors. Surprisingly different are findings from a study by Lee (2013), in which the most frequent type of corrective feedback was recast, which generated 92.09% learner repair. Other studies (Bitchener & Knoch, 2010; Chandler, 2003; Lee, 2005) suggest that explicit correction is preferable in writing, since its precision minimizes students' confusion over the corrective feedback received.

The second most common type of corrective feedback that students attributed to teachers within the current study was recast (16.7 % of the students). However, the study by Lyster and Ranta (1997) proved that `recast, the most popular feedback technique, is the least likely to lead to uptake of any kind' (Lyster & Ranta, 1997: 54). Similarly, in Housen and Pierrard (2005), no student-generated repairs were observed using explicit correction and recasts since both techniques provide learners with correct forms. A study by Mackey et al. (2000) demonstrates that recasts were more preferable in treating morphosyntactic errors, while correction of pronunciation errors implied a more extensive use of clarification requests. In addition, Lyster (2001) has found recasts to be clearly beneficial for error treatment in the case of phonological errors. At the same time, Havranek and Cesnik (2001) emphasize the importance of reinforcing recasts with further comments or repetition by the corrected learner, with `unfortified' recast being viewed as `the least successful format of corrective feedback' (Havranek & Cesnik, 2001, p. 99). Analyzing this technique, Perdomo (2018) refers to it as one of the most common types of negative feedback in native- and nonnative speakers' interaction in different settings; however, the same study admits that there is no sound evidence to properly claim that it is also the most effective corrective feedback for all teaching situations and language learners.

Clarification request ranked third among the most extensively used techniques in ESL classroom within our study, as specified by 10.3% of students. This technique, which involves asking a comprehension question to make students repeat what they said, is believed to help them reconsider and change the form of the utterance produced, especially if the students are given the awareness that comprehension check is used for confirmation of form, rather than meaning (Nicholas et al., 2001). Lyster (2005) has found negotiation of form, which includes elicitation, metalinguistic clues, clarification request, and repetition, to be more effective at leading to immediate repair than recasts or explicit correction, particularly in the case of lexical errors, as well as grammar errors and unsolicited uses of L1.

Repetition, which is recognized as an effective type of corrective feedback, along with metalinguistic feedback, elicitation and clarification requests (Heift, 2004), was reported to be the fourth extensively used error correction strategy by teachers in the current study, as reported by 11.5% of the respondents. This is at variance with a study by Havranek & Cesnik (2001), in which repetition was found the most frequent type of error treatment during language practice, in which the correct form was provided by the teacher and repeated by the learner.

A mere 7.7% of the respondents recognized metalinguistic feedback as a strategy commonly used by their teachers. However, according to Lyster & Ranta (1997), metalinguistic feedback, clarification request and repetition are similarly effective at eliciting student uptake, although metalinguistic feedback is more likely to elicit studentgenerated repair than clarification request or repetition.

Elicitation, specified by 5.1% of the respondents, was the most uncommon technique in students' responses. By contrast, it was proved to be the most successful technique for eliciting uptake by Lyster & Ranta (1997). Similarly, in Housen & Pierrard (2005) metalinguistic feedback and elicitation were leading to learner uptake 98% of the time. This result is consistent with Lyster & Ranta's study (1997) in which 100% of learner utterances following elicitation involved learner uptake.

Before classroom observations, teachers participating in the study were encouraged to diversify their corrective feedback by using a wide range of error correction techniques. The results of classroom observations carried out within this study support those obtained by Lee (2013), where recasts appeared to be the most frequent type of corrective feedback used by ESL teachers. While observed, teachers in this study tended to react to students' erroneous utterances through recasts and explicit correction.

The online student survey revealed an alarming fact that as much as one third of the respondents admitted that error correction by the teacher in the classroom could be psychologically uncomfortable for them or other students (Fig. 2). This finding could serve a stimulus for ESL teachers to reflect on the types of positive corrective feedback that would reduce the levels of students' anxiety and therefore contribute to oral production fluency.

Figure 2. Students' perceptions of oral corrective feedback in terms of the psychological discomfort it may cause

psychological corrective feedback error еnglish student

Furthermore, to support the previous statement, another question regarding the psychological relevance of error treatment was posed, in which the respondents reported that they were quite often (32 %) or very often (5%) discouraged by the teacher interrupting their oral production to correct the errors (Fig. 3).

In order to identify the types of errors that are recognized by university-level ESL learners as necessary for treatment, students were offered to choose one type of errors that needs corrective feedback more than others (Fig. 4).

Figure 3. Students' assessment of the frequency of discouragement associated with intrusive corrective feedback

In a similar study by Zhang et al. (2010) ESL students regarded lexical errors as the most critical and deserving the greatest teacher's attention; while grammatical errors ranked the second and phonological errors - the last (Zhang et al., 2010, p. 307).

Figure 4. Students' vision of the type of errors in SLA which should necessarily be corrected

A controversial issue is whether corrective feedback should be provided simultaneously with the learner's oral production or after a while - as soon as a learner has finished speaking, after a particular activity is completed, or at the end of the lesson. Simultaneous feedback on the accuracy of ESL learners' oral production is associated with high rates of success and reduced likelihood of practicing errors (Archer & Hughes, 2011). However, analyzing its relevance, Park (2010) maintains that even though delay between an error and its treatment may undermine the effectiveness of corrective feedback, `immediate error treatment can interrupt the flow of communication and even make students feel embarrassed or afraid of making errors' (Park, 2010: 61). To highlight the point, Park (2010) provides the evidence that 60% of the students in the high anxiety group preferred their errors to be treated after the activities are completed (Park, 2010, p. 49), as opposed to the students in the low anxiety group, who generally did not favour delayed error correction.

There was no major divergence in students' attitudes to simultaneous and delayed corrective feedback within the current study. The difference in response distribution constituted 5.2% in favour of simultaneous error correction (Fig. 5).

Figure 5. Preferences of the respondents towards the desired speed of corrective feedback received

We strongly agree with Touchie (1986) that frequent correction of oral errors may disrupt the process of language learning and discourage shy students from communicating in the target language. According to Touchie (1986), teachers should correct errors global errors which affect the general comprehension of utterances, putting more emphasis on high frequency and generality errors and errors produced by a large percentage of their students. As explained by Radecki & Swales (1988), students' need for error correction does not necessarily imply the usefulness of such feedback. A similar idea is expressed by Hedge (2000), who maintains that teachers should reduce their intrusion by responding to errors, which are systematic and result from the lack of knowledge, rather than mistakes, which are non-systematic and accidental. Teachers in a study by Lee (2013) strongly disagreed that they should correct all of the students' errors and mistakes, although they recognized the benefits of teachers' corrective feedback and the efficacy of immediate correction of the students' errors to enhance their oral proficiency.

In teacher interviews held within the current study, 5 out of 14 instructors disapproved of the idea of correcting all errors, referring to the impediment it creates to the natural flow of communication in ESL classroom. However, both teachers and students completely opposed to the idea that teachers should not correct any errors at all. The majority of students admitted that corrective feedback raises their awareness of the existing problems and prevents them from making repetitive errors. Moreover, 56.4% of students strongly supported the idea of receiving regular corrective feedback, suggesting that teachers should correct all students' errors.

Since positive reinforcement during language acquisition has been associated with increases in correct responding (Miltenberger, 2003; Worsdell et al., 2005), ESL teachers are generally in favour of correcting students' errors. The results of a similar study (Park, 2010) revealed that although teachers and students had significantly different opinions about types of errors that need correction, as well as the timing, method, and delivering agents of corrective feedback, students wished to get more corrective feedback than their teachers expected. In our survey, both teachers and students agreed on the necessity of error correction (Fig. 6). The majority of teachers expressed a belief that either `all students' errors should be corrected' (64.3% of the teachers), or `all major errors but not the minor ones' (21.5%). The students generally reacted to this question in a similar way, 56.4% of them supporting treatment of `all students' errors' and 15.4% of the respondents expecting a corrective feedback on `all major errors'. However, there is a clear distinction in terms of students' and teachers' attitude towards reducing corrective feedback to treating global errors leading to misunderstanding, hindering communication, and interfering with clarity of expression. While 9% of students in our survey were in favour of limited corrective feedback (6.4% of students supporting the idea of correction focused on errors leading to misunderstanding and 2.6% of students discarding any error correction as long as the expressed ideas are clear), teachers completely disregarded these options in the survey (0%) (Fig.6).

Figure 6. Comparison of teachers' and learners' perceptions of the extent to which errors should be corrected in SLA

According to Havranek and Cesnik (2001), corrective feedback is considered successful if it elicits a corrected version of the original utterance from the learner it was generated by, even if it is a mere repetition of the corrected version of the utterance. Speculations on the optimal type of corrective feedback resulting in student-generated repairs generally lead researchers to viewing self-correction as a highly productive and psychologically relevant strategy. Self-correction may `enable students to have more autonomy in learning, requiring that they take responsibility in the process of monitoring their own errors' (Hong, 2004, p. 14).

Whether foreign language learners are capable of self-correction or teacher generated corrective feedback is often dependent on their L2 level, which is also the case with particular error correction techniques they are more responsive to. In this respect, Perdomo (2018) rightly points that the level of L2 proficiency is an important factor influencing the decisions on the type of corrective feedback chosen. We strongly support Nicholas et al. (2001), who state that the effectiveness of corrective feedback depends on `the overall developmental level of proficiency or interlanguage e variety of the learner' (Nicholas et al., 2001, p. 752). Therefore, in order to answer the key question in our survey (Fig. 7) students were to specify their English proficiency level before choosing the error correction technique which they considered to have the strongest preventative effect.

Figure 7. Distribution of students' choices of the error correction technique perceived as having the strongest preventative effect

Interpretation of the data obtained was a complex task. We formed an interim table in which the sorted responses were distributed according to the ESL learners' proficiency level (Table 1).

Table 1