Scholarly article on topic 'Teachers’ Response to Serbian EFL Learners’ Pronunciation Errors'

Teachers’ Response to Serbian EFL Learners’ Pronunciation Errors Academic research paper on "Educational sciences"

0
0
Share paper
Academic journal
Romanian Journal of English Studies
OECD Field of science
Keywords
{""}

Academic research paper on topic "Teachers’ Response to Serbian EFL Learners’ Pronunciation Errors"

VERSITA

DOI: 10.2478/rjes-2013-0008

TEACHERS' RESPONSE TO SERBIAN EFL LEARNERS' PRONUNCIATION ERRORS

DANICA JEROTIJEVIC

University of Kragujevac Email: danicajerotijevic@gmail.com

Abstract: The present paper aims at discovering what type of feedback Serbian teachers resort to when correcting their students ' mispronounced words or utterances. To accomplish the previously stated aim of the study, we conducted a survey investigating teachers' preferences for specific types of corrective feedback and the results indicate that the most frequently employed type of feedback among Serbian EFL teachers is recast, whereas the least preferred one is direct or explicit correction. The total of 55 teachers from primary, secondary schools and colleges participated in the survey. Keywords: corrective feedback, EFL, pronunciation errors

1. Introduction

Although systematic research in SLA has expanded for the last three or four decades, the field of interlanguage phonology remains slightly disregarded (Baptista 2002) compared to other linguistic areas, such as grammar and vocabulary, for example. We dare say that in Serbian EFL context, as well, research in interlanguage phonology is still in its infancy. However, with the predominant use of English as a medium of international communication, acquiring the capacity to correctly perceive and produce foreign, i.e. English language sounds has become an essential step towards successful foreign language acquisition. Defective pronunciation can lead to communication breakdowns and sometimes result in unnecessary tension and conflict. Hence, there is a renewed interest in the relevance of pronunciation teaching in recent years with the number of studies increasing (Derwing and Munro 2005).

The aim of the present study is to discover the strategies Serbian EFL teachers employ when teaching pronunciation, more precisely, to determine what type of feedback they provide for the incorrect production of English sounds. After several introductory sections presenting relevant theoretical considerations, the results of the conducted research are presented and discussed.

2. Effects of Corrective Feedback

Corrective feedback remains a matter of controversy among scholars, especially concerning the following issues: (a) whether CF contributes to L2 acquisition, (b) which errors are to be corrected, (c) who should do the correcting (the teacher or the learner him/herself), (d) which type of CF is the most effective, and (e) what is the best timing for CF (i.e. immediate or delayed) (Ellis 2007). Feedback may be helpful for it enables learners to distinguish whether their production is accurate or not (Panova and Lyster 2002), and, when their performance is not correct, further feedback incites them to take some remedial action. Another considerable role of corrective feedback is in rejecting false hypotheses that may arise from various misleading sources as well as in preventing overgeneralisation (Nassaji and Swain 2000). Under the right conditions and salient feedback, learners may cognitively restructure current learning techniques. Frequent and thoughtful corrections may have a metalinguistic or reflective function. Through consistent feedback teachers encourage students to view learning situations form a different standpoint and to observe their own progress. In the absence of feedback students may feel disappointed and lose interest and motivation (Richards and Lockhart 1994). Moreover, lack of feedback causes repetition of errors eventually resulting in fossilization.

To ensure the effectiveness of corrective feedback, one must take the following factors into consideration (Brookhart 2008): (1) Timing, (2) Amount, (3) Mode, and (4) Audience. El Tatawy (2002) adds that feedback should be clear enough in order to be perceived as such and the learner's readiness to process it should also be carefully considered. The learner should actively participate in pronunciation monitoring and correction, promoting thus his/her communicative self-confidence.

Nevertheless, if feedback is not systematically applied, it may trigger negative outcomes. Occasionally, avoidance of error correction is recommendable. A learner may produce a slip or mistake, not an actual error, which likewise naturally occurs in mother tongue, and this is when corrective feedback is undesirable. Affective filter must also be taken into account, since frequent corrections may lower learners' self-confidence and decrease motivation or even result in disappointment, frustration and ultimately, reluctance and rejection (Murphy 1991). Possible problems in applying corrective feedback may be ambiguity and inconsistency. Teachers sometimes adopt ambiguous feedback strategies, i.e. they neglect errors for the sake of not interrupting continuous flow of conversation or correct the same errors. Furthermore, students often pay scarce attention to feedback they receive and frequently do not know how to benefit from it, either because their strategies are limited or because the feedback they expect does not match the one they are provided with. Additional negative effects of feedback may refer to the amount of time necessary for processing information since numerous corrections cannot be processed simultaneously thus causing confusion and overload of cognitive capacities (Richards and Lockhart 1994).

3. Types of Corrective Feedback

Some authors suggest that feedback be classified into a) implicit or explicit and b) input or output based (Ellis 2007).

Implicit feedback is covert and inconspicuous incorporating:

a) A recast - a reformulation and correction of a deviant utterance while the meaning is preserved;

b) A request for clarification - urges the students to reformulate and repeat the utterance since pronunciation or meaning was unclear;

c) Repetition of errors - different from recasts because the teacher repeats the student's utterance with the error using intonation to indicate that it was incorrect.

Explicit feedback signifies that the teacher provides the correct form and clearly indicates that the student has made a mistake, and it includes:

a) Direct / explicit correction - telling students they made a mistake and correcting it;

b) Explanation or metalinguistic feedback - comments about the error intended to make students reflect upon and correct errors by themselves;

c) Elicitation - encourages students to pronounce correctly by open-ended questions or fill-in-the-gap utterances.

Input-based feedback designs the correct form for learners (e.g. a recast), while output-based corrections demand the correct form from them (e.g. a request for clarification).

There is yet another division of feedback into external and internal according to the source it emerges from (Ellis 1991). External feedback derives from teachers and peers, whereas internal feedback is a self-regulating and monitoring cognitive process by which a learner assesses the current state of affairs and adopts appropriate techniques for future improvement. Teachers' corrections provide a sense of direction, identification of errors and advice, whilst peer feedback enables interaction without anxiety and may have more powerful results. Finally, regarding the time span during which it occurs, feedback can be immediate, instantly following the incorrect production, and delayed, with a time lapse between the performance and feedback (Opitz et al. 2011).

4. Methodology

4.1. The Aim of the Research

The essential goal of the current research is to provide an insight into the feedback strategies Serbian EFL teachers employ when correcting students' pronunciation errors. We are primarily interested in ranking the types of feedback according to the level of frequency and relevance as well as in determining what variables affect the teachers' preference for any of the suggested feedback types.

4.2. Research Questions

Having considered relevant theoretical concerns, we formulated the most significant questions for the present research. We are especially concerned with:

1. What is the most frequently used type of teacher feedback for correcting pronunciation errors in Serbian EFL classroom?

2. Is there a correlation between independent variables, such as work experience, age of students, teaching method employed, number of classes per week and the teachers' pronunciation variant, and the ranking of preferred types of feedback?

4.3. Participants

A total of 55 teachers from primary (25), secondary schools (18) and colleges (12) in Jagodina, Cuprija, Paracin, Kragujevac, Nis and Belgrade participated in the study. Our aim was to discover the types of feedback teachers most frequently use in Serbian EFL classroom in general, so we found it suitable to question teachers working at different levels of education.

4.4. Instruments

In order to examine the teachers' preference regarding which type of feedback they resort to when correcting students' pronunciation errors, we conducted a questionnaire containing three parts. The first set of questions was related to general information concerning the teachers' work experience, gender, age of students, preferred English variant in terms of pronunciation (British or American English), number of classes per week and the predominant teaching method they employed. The second set was a three-point Likert scale and the questions were concerned with the teachers' general viewpoints on feedback and pronunciation errors. The final set in the questionnaire represented an ordinal scale, more precisely a rank-ordering scale, in which the participants had the task to rank the statements regarding the types of feedback from the most frequently used (1) to the least frequently used (6). The third set of the survey is considered the most relevant for the present research, yet the previous sets are not disregarded, either, since the first set was essential for establishing the variables affecting the results in the final section of the questionnaire, whereas the second part enabled us to obtain a wider image regarding teachers' attitudes to feedback in general.

4.5. Pilot Study

The reliability and validity of the questionnaire was initially tested through a pilot study which was conducted on a sample of 15 teachers, and the results of the pilot study led to the reformulation of several questions and the exclusion of two, because they were vague and irrelevant to the issue.

4.6. Procedure

The survey was conducted from November to April in the 2010/2011 academic year with teachers of primary, secondary schools and colleges in Jagodina, Cuprija, Paracin, Kragujevac, Nis and Belgrade. The questionnaire was distributed either personally or via the Internet. The results were later coded and analysed by providing frequency counts, i.e. the percentage scores for each variable and the results were compared to establish whether these particular variables affected the outcomes or not. The percentage values were displayed for the total number of participants, as well. The types of feedback considered the most frequently used were the ones ranked 1 and 2 (the percentage scores were added to determine the cumulative value), whereas the least preferred ones were those with ranks 5 and 6. Ranks 3 and 4 represented medium points.

5. Results

5.1. The Second Part of the Questionnaire - Teachers' General Attitudes

The majority of teachers participating in the present survey believe that pronunciation is important (67.27 %), whereas 21.82 % think it is extremely important and 10.9 % of teachers are of the opinion that pronunciation is unimportant for successful interaction in L2. The second part of the questionnaire contained nine statements regarding teachers' general views on pronunciation instruction, errors and corrective feedback. Although the statements were not directly related to the actual types of feedback, we believe the answers may be beneficial since they represent the teachers' general notions and attitudes. The following table represents the percentage scores of teachers' answers:

Statements Always (%) Occasions lh~ (%) Never (%)

I disregard my students' pronunciation mistakes. 9.09 41.82 i9M

I emphasize the importance of pronunciation to my students. 51.18 34.55 25.46

My students have trouble pronouncing English consonants that are different from Serbian. 36.36 63.64 0

My students have trouble pronouncing English vowels that are different from Serbian. 40 60 0

My students have trouble pronouncing English diphthongs. 36.36 60 3.64

I devote special time to practise pronunciation with iT WMV/MVA -1 my students. 27.27 61.82 10.91

My students are bored when practising IWvVvWvVvWP pronunciation. 27.27 65.45 7.27

I correct my students' pronunciation mistakes. 61.82 34.55 3.64

My students respond to my feedback by immediately correcting themselves. 40 40 20

5.2. The Third Part of the Questionnaire - Preferred Types of Feedback

For the sake of conciseness we present the results of the third part of the questionnaire in tables, yet it is important to clarify the acronyms: Direct/Explicit Correction (D/E.C.), Repetition of Errors (R.O.E.), Elicitation (E.), Metalinguistic Feedback (M.F.), Request for Clarification

(R.F.C.), Recast (R.). It is likewise essential to note that the rank of the results starts at 1 (the most frequently used) and finishes with 6 (the least frequently used).

5.3. General Results

According to the results obtained in the current research, the majority of teachers rank recast as the most frequently used type of feedback; the second most preferred are repetition of errors and elicitation. Regarding the least frequently used feedback strategy, teachers rarely use request for clarification and direct/explicit correction:

Rank 1 2 3 4 5 6

D/E.C. 5 11 7 6 3 23

(9.09 %) (20%) (12.73%) (10.9%) (5.45%) (41.82%)

ROE. 6 14 25 2 5 3

(10.9%) (25.45%) (45.45%) (3.64%) (9.09%) (5.45%)

E. 12 S 11 16 3 5

(21.82%) (14.55%) (20%) (29.09%) (5.45%) (9.09%)

M.F. 4 4 1 21 21 5

(7.27%) (7.27%) (1.82%) (38.18%) (38.18%) (9.09%)

R.F.C. 3 5 2 5 21 IS

(5.45%) (9.09%) (3.64%) (9.09%) (3S.1S%) (32.73%)

R. 25 14 9 5 2 0

(45.45%) (25.45%) (16.36%) (9.09%) (3.64%) (0%)

5.4 Results according to variables

The following results in the tables represent the rank order of feedback types in percentage scores according to the independent variables we found potentially significant for the overall outcome. As a reminder, the highest percentage scores for ranks 1 and 2 represent feedback types used most often, whereas the highest percentage scores for ranks 5 and 6 represent the least frequently employed types of feedback by Serbian EFL teachers.

5.4. a) Gender

The first variable that we would like to consider is gender. Females were predominant in the questionnaire:

Apparently, there are no significant differences between male and female teachers regarding the most rarely used type of feedback since it is request for clarification as well as direct/explicit correction. The most frequently used type for both men and women is recast, yet there is a slight difference concerning the second most preferred type of feedback, since it is repetition of errors for male and elicitation for female teachers.

Male (%) Female (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 5.SS 23.53 17.65 5.88 5.8 S 41.18 10.53 18.42 10.53 13.16 5.26 42.1

R.O.E. 17.65 29.41 29.41 5.88 17.65 5.88 7.89 23.68 52.63 2.63 7.89 5.26

E. 17.65 11.76 17.65 41.IS 0 17.65 23.68 15.79 21.05 23.68 7.89 7.89

M.F. 5.SS 5.88 0 47.06 47.06 0 7.89 7.89 2.63 34.21 34.21 13.16

R.F.C. 5.SS 5.88 17.65 0 35.29 35.29 5.26 10.53 0 13.16 39.47 31.58

R. 47.06 23.53 23.53 5.88 0 0 44.74 26.32 13.16 10.53 5.26 0

5.4. b) Work Experience

The majority of teachers in the survey had less than 10 years of experience and we further investigated whether it had any effects on the choice of corrective feedback type.

The participants were sorted into four groups according to their work experience, yet merely six teachers had more than 20 years of experience, which must be recognized as an impediment to more valid research results. The first group (0-5 years of work experience) uses recasts most often, followed by elicitation and repetition of errors, which have similar percentage scores. The least preferred types of feedback are direct/explicit direction. Interestingly, teachers chose request for clarification as one of the most frequently as well as the least frequently used feedback types. The second group (5-10 years) ranked recasts as most frequent, and elicitation as the second, while the least employed are again direct/explicit correction and request for clarification.

Recasts are the most preferred feedback type for the third group (10-20 years), as well, yet there is an alteration regarding the second ranked feedback type, since it is direct/explicit correction. The least preferred are request for clarification and metalinguistic feedback. The last group (over 20 years) will not be taken as relevant, since the number of participants was deficient to make any valid conclusions. Intriguingly though, absolutely all the teachers chose recasts as the most frequently used feedback type.

0-5 years (%) 5-10 years (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 14.26 14.26 9.52 14.26 4.76 33.1 0 17.65 0 5.SS 11.76 64.71

R.O.E. 4.76 33.33 28.57 4.76 19.05 4.76 23.53 11.76 52.94 5.88 5.88 0

E. 28.57 9.52 23.S 23.S 0 14.26 29.41 23.53 23.53 17.65 0 5.88

M.F. 14.26 14.26 4.76 28.57 28.57 9.52 5.8S 0 0 58.82 35.29 0

R.F.C. 28.57 9.52 4.76 14.26 33.1 23.8 0 5.SS 5.88 11.76 47.06 29.41

R. 28.57 19.05 23.57 14.26 9.52 0 41.18 41.IS 17.65 0 0 0

10-21} years (%) Ch er 20 years (%

D/E.C. 13.13 27.27 27.27 0 0 27.27 0 33.33 33.33 16.67 0 16.67

R.O.E. 9.09 IS.IS 63.64 0 0 9.09 0 50 50 0 0 0

E. 9.09 9.09 9.09 54.55 IS.IS 0 0 16.67 16.67 33.33 16.67 16.67

M.F. 0 9.09 0 18.18 45.45 27.27 0 0 0 50 50 0

R.F.C. 9.09 9.09 0 9.09 36.36 36.36 0 0 0 0 33.33 66.61

R. 54.55 27.27 0 18.18 0 0 100 0 0 0 0 0

5.4. c) Students' Age

Another factor that we shall consider is the age of students taught by the participants in our survey.

Primarv School Secondare School Co Lege

1-4Ù Grade 1-4Ù Year

l-4a Grade 5-8 ù Grade IS (32.73 %) 12 (21.S2 %)

13 (23.64 %) 12 (21.82 %)

1-4th grade primary school teachers prefer recast and explicit correction and avoid using request for clarification and metalinguistic feedback. 5-8& grade primary school teachers prefer recasts, as well, along with elicitation, repetition of errors and direct/explicit correction, while they do not use request for clarification and some of them direct/explicit correction. Regarding the preferred feedback types, the situation is identical with secondary school teachers, whereas the least frequently used ones are request for clarification, metalinguistic feedback and direct/explicit correction. College teachers prefer recasts and elicitation, and rarely use direct/explicit correction and request for clarification.

1-4° Grade Primary School (%) S-S01 Grade Primary School (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 23.03 23.03 0 23.08 7.69 23.08 0 33.33 0 8.33 3.33 50

R.O.E. 0 23.08 53.85 7.69 15.38 0 3.33 25 50 0 8.33 0

E. 15.38 7.69 23.08 38.46 0 7.69 25 8.33 25 33.33 0 8.33

M.F. 0 15.38 0 15.38 53.35 15.33 S.33 8.33 0 41.67 33.33 S.33

R.F.C. 7.69 0 7.69 7.69 23.03 53.35 0 8.33 0 16.67 41.67 33.33

R. 46.15 30.77 15.38 7.69 0 0 53.33 8.33 25 0 8.33 0

1-4° Grade Secondaiy School (%) 1-41" College

D/E.C. 0 11.11 38.89 11.11 0 38.89 16.67 16.67 0 0 3.33 53.33

R.O.E. 11.11 38.89 27.28 0 5.56 16.67 25 0 58.33 8.33 8.33 0

E. 27.73 11.11 16.67 22.22 16.67 5.56 3.33 33.33 16.67 25 0 16.67

M.F. 5.56 0 0 50 33.33 11.11 16.67 8.33 8.33 41.67 25 0

R.F.C. 11.11 5.56 5.56 5.56 44.44 27.23 0 16.67 0 8.33 50 25

R. 44.44 33.33 11.11 11.11 0 0 33.33 25 16.67 16.67 8.33 0

5.4. d) Number of Classes per Week

Since lack of time may be a problem for Serbian EFL teachers, perhaps the number of classes may affect the options for corrective feedback, which is why we included this question in the survey and discovered that the majority of teachers we surveyed taught 2 classes per week (per one class only).

Similarly to the situation with work experience, the number of teachers who teach the same group of students more than two classes a week is deficient, so the results we obtained are insufficient to lead to reliable conclusions. Regardless, we will mention that the most frequently used feedback types are recasts, elicitation and repetition of errors for teachers who teach two classes per week, whereas the least used ones are request for clarification, metalinguistic feedback and direct/explicit correction. The situation is similar when it comes to the group of teachers who teach 4 classes per week, as can be seen from the relevant tables:

2 Classes (%) 4 Classes (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 9.09 20.45 11.36 13.64 6.S2 38.64 12.5 12 5 0 0 0 75

R.O.E. 6.S2 29.55 43.18 4.55 11.36 4.55 37.5 0 62.5 0 0 0

E. 25 11.36 20.45 27.27 4.55 11.36 12.5 37.5 25 25 0 0

M.F. 6.82 6.82 2.27 36.36 38.64 9.09 0 12.5 0 62.5 25 0

R.F.C. 4.55 6.82 4.55 11.36 36.36 36.36 0 0 0 0 75 25

R. 47.73 25 18.18 6.82 9.09 0 37.5 37.5 12.5 12.5 0 0

5 Classes (%)

D/E.C. 0 33.33 66.67 0 0 0

R.O.E. 0 33.33 33.33 0 0 33.33

E. 0 0 0 66.67 33.33 0

M.F. 33.33 0 0 0 33.33 33.33

R.F.C. 33.33 33.33 0 0 0 33.33

R. 33.33 0 0 33.33 33.33 0

5.4. e) Teaching Approach

In the theoretical sections concerning feedback and pronunciation instruction in general we mentioned the different views various teaching approaches had with regard to pronunciation instruction, which is why we consider the teachers' preferred teaching approach may affect the choice of feedback strategy.

Among teachers who apply the Communicative approach in their classroom, the majority prefers elicitation and recasts, whereas direct/explicit direction is the most disliked type along with request for clarification. Grammar-translation supporters prefer recasts and direct/explicit

correction, however, while metalinguistic feedback is the least used and so is request for clarification. Teachers who combine approaches say that they most frequently use recast and repetition of errors, and avoid using direct/explicit correction and request for clarification.

Communicative (%) Grammar-translation (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 5.8S 5.88 0 17.65 11.76 58.82 6.67 53.33 20 13.33 0 6.67

R.O.E. 0 29.41 52.94 5.88 11.76 0 0 26.67 66.67 0 6.67 0

E. 64.71 17.65 11.76 0 0 5.88 0 6.67 6.67 53.33 20 13.33

M.F. 5.88 5.88 0 47.06 29.42 5.88 0 6.67 0 20 46.67 26.67

R.F.C. 0 4.88 5.88 17.65 35.29 29.41 6.67 0 6.67 6.67 26.67 53.33

R. 23.53 35.29 29.41 5.88 5.88 0 S6.67 6.67 0 6.67 0 0

Combined (%)

D/E.C. 13.04 8.7 17.39 4.35 4.35 52.17

R.O.E. 26.09 21.74 26.09 4.35 8.7 13.04

E. 4.35 17.39 34.78 34.78 0 8.7

M.F. 13.04 8.7 4.35 39.13 34.78 0

R.F.C. 8.7 13.04 0 4.35 47.33 26.09

R. 34.78 30.43 17.39 13.04 4.35 0

5.4. f) Teachers' Preferred Pronunciation Variant

The American English variant is prevailing in films, TV shows and music, whereas British English is less frequently heard, which is why we believe that teachers who opt for British English pronunciation are more careful and meticulous about the production of target language sounds.

British English

American English

28 (50.91 %)

27 (49.09 %)

Teachers who adopted British English pronunciation use recasts and direct/explicit correction to correct their students' pronunciation errors, and rarely use request for clarification and metalinguistic feedback. Those teachers who speak American English prefer elicitation, recast and repetition of errors, yet dislike direct/explicit correction and request for clarification:

British English (%) American English (%)

Rank 1 2 3 4 5 6 1 2 3 4 5 6

D/E.C. 7.14 32.14 21.43 7.14 0 32.14 11.11 7.41 3.7 14.81 li.ii 51.85

R.O.E. 10.71 21.43 50 0 10.71 7.14 11.11 29.63 40.74 7.41 7.41 3.7

E. 7.14 10.71 17.86 46.43 10.71 7.14 37.04 18.52 22.22 11.11 0 11.11

M.F. 0 7.14 0 32.14 46.43 14.29 14.81 7.41 3.7 44.44 25.93 3.7

R.F.C. 10.71 3.57 7.14 7.14 32.14 39.29 0 11.11 0 11.11 48.15 29.63

R. 64.29 25 3.57 7.14 0 0 25.93 25.93 29.63 11.11 7.41 0

6. Discussion

The fact that the majority of Serbian teachers who participated in our study believe that pronunciation is important for successful interaction seems encouraging for students who intend to perfect their production of English sounds, since teachers are willing to invest their efforts and knowledge to help them. One of the ways in which teachers can help their students is by not disregarding their errors and by emphasizing the importance of pronunciation, which is how teachers predominantly behave, according to the results of our questionnaire. The prevailing number of teachers said that their students occasionally had problems with the pronunciation of English vowels, consonants and diphthongs, yet with a careful approach and consistent attention, these problems can be overcome successfully. The discouraging fact, however, is that teachers believe that their students are occasionally bored with pronunciation practice, which can be improved by including more interesting exercises and involving students in projects they would feel comfortable with. The majority of teachers likewise said that they only occasionally devoted special time to practising pronunciation, which is probably the result of the curriculum requirements that do not put significant emphasis on pronunciation practice. Most of the teachers reported that they always corrected the students' pronunciation errors, which goes in line with stressing the importance of pronunciation. An equal number of teachers said that their students responded to corrective feedback always or occasionally, which may be considered positive since it serves as evidence that teachers and their corrective feedback affect their students and provoke reactions.

The results obtained regarding the general Serbian EFL teachers' attitudes towards corrective feedback and pronunciation instruction in general may be beneficial because of their pedagogical implications for applied linguistics research and English language teaching in particular.

The results of the questionnaire showed that the most frequently used type of feedback by Serbian teachers was recast, followed by elicitation and repetition of errors. Our findings thus confirm the ones previously obtained by various scholars about which we reported earlier (Lyster 1998; Lyster and Ranta 1997). However, although recasts are the most frequent type, they may be confusing for students and they have less effect on the students' production than, for example, metalinguistic feedback or other types of implicit feedback because these enhance the students' involvement and motivate them to rely on their own resources. The most unpopular feedback types are request for clarification and direct/explicit feedback. The possible explanation perhaps derives from the fact that pronunciation errors are different from other types of error and when a student produces an inaccurately pronounced word, the teacher understands what type of error was made and needs no explanation from the student regarding what he/she was trying to say. Direct/explicit correction may be used most rarely because teachers are reluctant to discourage their students by explicitly indicating that they made a mistake, since direct correction may lead to the increase of affective filtering and result in the decrease of self-confidence and motivation. However, it is irrefutable that explicit correction may be useful occasionally, and it is the teachers' task to discover what type of error and classroom situation it is suitable for.

Regarding the factors, i.e. independent variables that may affect the teachers' choice of corrective feedback type, we may conclude that gender is not a relevant factor, since both men and women have the same preference for recast and avoid request for clarification and explicit correction. However, it may be significant to mention that male and female teachers differ in terms of the second best ranked type of feedback, since males prefer to repeat errors, which is implicit, whereas females tend to elicit the correct production from their students, which is an explicit type of corrective feedback. Work experience may seem a significant factor, yet we are not allowed to make any steadfast conclusions due to insufficient evidence, as we mentioned earlier. However, we may underscore the fact that as work experience is greater, teachers prefer direct/explicit correction as second-ranked, but recasts remain the most frequently used type. Further research on a larger number of participants may provide more substantial evidence for the relevance of work experience for the type of feedback teachers choose. The age of students may have a certain influence on the type of feedback teachers use, yet not regarding the first-ranked choice, but the second one and especially with younger learners. Teachers who teach lower grades in primary schools reported direct/explicit correction as the second most preferred choice (after recasts), whereas teachers in older primary school grades and secondary schools use elicitation and repetition of errors most frequently, along with recasts. This difference may arise from the fact that younger learners need a more explicit indication that they made an error in pronunciation and also from the fact that the older learners' curriculum is more diverse and the students' proficiency is hopefully at a higher level, so teachers can use various materials and adjust their approach to correspond to their students' needs. The same argument may account for the college teachers' preference for elicitation since they may want to involve their students more and stimulate them to rely on their own intuition and knowledge.

The deficiency of evidence disallows us to draw any viable conclusions regarding the correlation between the number of classes taught and the types of feedback teachers use. Bearing in mind that pronunciation was approached differently depending on the teaching approach, we assumed that the teaching approaches applied by the participants in the survey might have an influence over the type of feedback they choose. Our assumption was correct to a certain extent, since the teachers who use the Communicative approach predominantly use elicitation, probably seeking to involve students more, yet recasts also remain the most frequently used type. Furthermore, the teachers who apply the Grammar-translation method prefer recasts; however, they resort to explicit correction to a significant extent, which is comprehensible enough since the teaching approach in question is somewhat traditional and so is direct/explicit correction. There is likewise a difference in the least preferred type of feedback, and that is direct/explicit correction for Communicative-approach teachers, and metalinguistic feedback for Grammar-translation teachers, while request for clarification is avoided by both groups of teachers. Teachers who combine several approaches tend to use recasts and, interestingly, repetition of errors, and avoid direct correction and request for clarification.

The results showed a difference between teachers with British and American English variants, meaning that the teachers who speak British English regard direct/explicit correction

more favourably than those who speak American English, who prefer elicitation and repetition of errors, yet recasts are common with both groups. The possible explanation may be that the teachers who prefer British English tend to emphasize the importance of pronunciation and thus explicitly draw the students' attention to erroneous production, whereas teachers with American English variant choose to correct their students somewhat more implicitly.

7. Conclusion

The goal of the previous study was to explore the types of feedback Serbian EFL teachers use to correct their students' pronunciation errors.

The results of the questionnaire designed to measure the most frequently employed type of feedback indicated that the predominantly used corrective feedback type was recast, and the most rarely used one was direct/explicit correction. We drew attention to certain possibly relevant factors influencing the teachers' choice of feedback and we arrived at the conclusion that the most significant factor is the teaching approach, along with the age of students and the teachers' pronunciation variant. However, further research involving a greater number of participants seems necessary to explore the aforementioned suggestions.

References

Baptista, Barbara Oughton. 2002. "Languages in contact: Brazilian English interlanguage phonology." Paper presented at the I Congresso Internacional das Linguagens. June 30 - August 02, 2002. Erechin: URI: Campus de Erechin.

Brookhart, Susan M. 2008. How to give effective feedback to your students. Association for Supervision and

Curriculum Development (ASCD). USA. Carroll, Susanne and Swain, Merrill. 1993. "Explicit and implicit negative feedback: An empirical study of the

learning of linguistic generalizations." Studies in Second Language Acquisition 15(3):357-386. Derwing, Tracey M. and Munro, Murray J. 2005. "Second Language Accent and Pronunciation Teaching: A

Research-Based Approach." TESOL Quarterly 39(3):379-397. El Tatawy, Mounira. 2002. "Corrective feedback in second language acquisition." Working Papers in TESOL & Applied Linguistics Teacher's College, Columbia University [Online]. Available: http://www.tc.columbia. edu/academic/tesol/WebJournal/El-Tatawy.pdf [Accessed 2011, 14 May]. Ellis, Rod. 1991. Instructed second langage acquisition. Oxford: Blackwell Publishers.

Ellis, Rod. 2007. "Corrective Feedback in Theory, Research and Practice" as a keynote speech in The 5th Annual International Conference on ELT in Beijing [Online]. Available: http://www.celea.org.cn/2007/edefault.asp, [Accessed 2011, 25 July].

Lyster, Roy and Ranta, Leila. 1997. "Corrective feedback and learner uptake." Studies in Second Language Acquisition 19(1):37-66.

Lyster, Roy. 1998. "Recasts, repetition and ambiguity in L2 classroom discourse." Studies in Second Language Acquisition 20:51-81.

Morley, Joan. 1991. "The Pronunciation Component in Teaching English to Speakers of Other Languages." TESOL Quarterly 25(3):491-520.

Murphy, John M. 1991. "Oral Communication in TESOL: Integrating Speaking, Listening, and Pronunciation."

TESOL Quarterly 25(1):51-75. Nassaji, Hossein and Swain, Merrill. 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. Opitz, Bertram, Ferdinand, Nicola K., and Mecklinger, Axel. 2011. "Timing Matters: The Impact of Immediate and

Delayed Feedback on Artificial Language Learning." Frontiers in Human Neuroscience (5):1-9. Panova, Iliana and Lyster, Roy. 2002. "Patterns of corrective feedback and uptake in an adult ESL classroom."

TESOL Quarterly 36:573-595. Richards, Jack C. and Lockhart, Charles. 1994. Reflective Teaching in Second Language Classrooms. Cambridge: Cambridge University Press.

Note on the author: Danica Jerotijevic is currently working as a language instructor at the Faculty of Philology and Arts, University of Kragujevac, Serbia, where she is also engaged in the PhD studies, linguistics module. She has so far presented and published several papers mainly related to phonetics and phonology, applied linguistics, pragmatics and sociolinguistics.