Database

This observational study has yielded 2.5 hours of tape recordings of English lessons in two levels, one primary 5 classroom and one primary 6 classroom. I am a female teacher who has taught English as a second language for nearly 9 years at primary level. My classes of 58 pupils include 24 in primary 5 and 34 in primary 6.

The study involves a description how teachers and students engage in error treatment during communicative interaction during English language lessons. It is because pupils at primary level do not have English subject matter lessons. As we are interested in analyzing teacher behaviour, I focused exclusively in my analyses on teacher-student interaction. Even when students were involved in pair work, I was still able to capture teacher-student interaction as I invited each pair to come out in front of the classroom when interacting with students.

My school is located nearby in a government-subsidized public housing estate in Tuen Mun. My pupils largely come from families who live in the nearby public housing estates. Their parents are manual or service workers and they are illiterate. They speak only Cantonese at home. They can not provide much help in pupils’ learning English. Thus, pupils have very limited use of English outside the classroom.

Data Analysis

The sequence begins with a learner’s utterance with at least one error. The utterance is followed either by the teacher’s corrective feedback or there is topic continuation. If corrective feedback is provided by the teacher, then it is either followed by uptake on the part of the student or not. If there is uptake, then the student’s initially utterance is either repaired or continues to need repair. If the utterance needs repair, then corrective feedback may again are provided by the teacher; if no further feedback is provided, then there is topic continuation. If and when there is repair, then it is followed either by topic continuation or by some repair-related reinforcement provided by the teacher. Following the reinforcement, there is topic continuation. The error treatment sequence emerged as the model presented in Figure 1.




 



Figure 1. Error treatment sequence.

Error

All student turns were coded as either having an error or not. Ill-formed utterances were classified as having either one or more than one error. Errors were classified as phonological, lexical, or grammatical. When more than one type of error occurred in a student turn, these were coded as ‘multiple’. Since the students are not encouraged to use L1 in English lessons, I am not interested in examining teachers’ reactions to student’s’ unsolicited use of the L1 and nonnative-like uses of English as those Lyster and Ranta did. The example of the error is as follows:

(1) a Primary

6 studnet

Teacher:

Was the medicine expensive?

Kelly:

Yes, the medicine is expensive. [Error-grammatical]

Feedback

Six different types of feedback in this study are distinguished.

1.Explicit correction refers to the explicit provision of the correct form. As the correct form is provided, what the student had said was incorrect is clearly indicated.

(2)Primary 6 students

Phoebe: My finger hurt.

Kelly: What happened?

Phoebe: I burned my finger with a cook. [Error-lexical]

Teacher: A cook is a person. A cooker is a utensil. [FB-explicit Correction]

Phoebe: I burned my finger with a cooker.

2. Recasts involve the teacher’s reformulation of all or part of a student’s utterance, minus the error. Some recasts may focus on one word only, whereas other incorporates the grammatical or lexical modification into a sustained piece of discourse.

(3) a Primary 6 student

Eric: I have a sore ‘lee’. [Error-phonological]

Teacher: I have a sore knee. [FB-recasts]

Eric: I have a sore knee.

3. Clarification requests: indicate to students either that their utterance has been misunderstood or that the utterance is ill-formed in some way and that a repetition or a reformulation is required. This is a feedback type that can refer to problems in either comprehensibility or accuracy, or both.

(4) Primary 6 students

Iva (I): I have a broken finger.

Oscar: What happened?

Iva: I fell down. [Error-grammatical]

Teacher: I fell down what? [FB-clarification requests]

Iva: I fell down the chair.

4. Metalinguistic feedback contains either comments, information, or questions related to the well formedness of the student’s utterance, without explicitly providing the correct form. Metalinguistic information generally provides either some grammatical metalanguage that refers to the nature of the error or a word definition in the case of lexical errors. Metalinguistic questions also point to the nature of the error but attempt to elicit the information from the student.

(5) Primary 5 students

Ivy: What did the robber look like?

Lai Man: He had a long hair. [Error-grammatical]

Teacher: without ‘a’. Because the hair is uncountable. [FB-metalinguistic feedback]

5. Elicitation refers to at least three techniques that are used to directly elicit the correct form from the student. First, teachers elicit completion of their own utterance by strategically pausing to allow students to ‘fill in the blank’. Such ‘elicit completion’ moves may be preceded by some metalinguistic comment. Second, teachers use questions to elicit correct forms. Third, teachers occasionally ask students to reformulate their utterance.

(6) a Primary 6 student

Dabbie: Where did he go?

Angel: He turned ‘light’ into the park. [Error-phonological]

Teacher: He turned ……. [FB-elicitation]

Angel: He turned left into the park.

6. Repetition refers to teacher’s repetition of the erroneous utterance. In most cases, teachers adjust their intonation as to highlight the error.

(7) a Primary 5 student

Dick: What did the ‘rubber’ look like? [Error-phonological]

Teacher: not ‘rubber’ [FB-repetition]

Dick: ‘rubber’

Teacher: robber

Dick: What did the robber look like?

In addition to the preceding six feedback types, there is a seventh category called

multiple feedback which referred to combinations of more than one type of feedback in one teacher turn. Because this category revealed little information as to the nature of the combinations, various combinations to determine (a) whether certain combinations tended to occur more than others and (b) whether one particular type of feedback tended to override others would be examined.

Repetition clearly occurred with all other feedback types with the exception of recasts: in clarification requests, in metalinguistic feedback, in elicitation, and in explicit correction.

(8) a Primary 5 student

May: My name’s ‘is’ May Wu. [Error-grammatical]

Teacher: My name’s May Wu. Why don’t we use ‘is’? My name’s is the short form of my name is. [FB-metalinguistics feedback]

(9) a Primary 5 student

Rose: What did he ‘wore’? [Error-grammatical]

Teacher: The past tense is wore. The present tense is wear. [FB-metalinguistic feedback]

Another combination that occurred was recast and metalinguistic feedback. Such a combination was not ‘multiple’ and necessitated instead the creation of the category ‘explicit correction’. That is, as soon as the teacher’s provision of the correct form is somehow framed metalinguistically, then the characteristics of recast no longer apply. Similarly when elicitation accompanied either a recast or an explicit correction, this was coded as ‘explicit correction’ in order to capture in the coding instances where correct forms were explicitly provided. Finally, there were a few instances of elicitation occurring with metalinguistic feedback. This was coded as ‘elicitation’ because the elicitation technique prevails in term of illocutionary force in that uptake is clearly expected.

However, multiple feedback could seldom be found in this finding because the feedback that was given by the teacher is simple and clear since the students could only understand very little English. It is hard to explain their mistakes in English. Thus, the multiple feedback occasionally occurs.

Uptake

Uptake refers to a student’s utterance that immediately follows the teacher’s feedback and that constitutes a reaction in some way to the teacher’s intention to draw attention to some aspect of the student’s initial utterance. A description of uptake, then, reveals what the student attempts to do with the teacher’s feedback. If there is no uptake, then there is topic continuation, which is initiated by either the same or another student or by the teacher. There are two types of student uptake (a) uptake that results in ‘repair’ of the error on which the feedback focused and (b) uptake that results in an utterance that still needs repair (coded as ‘needs-repair’). Four types of other-initiated repair have been distinguished.

  1. Repetition refers to a student’s repetition of the teacher’s feedback when the latter includes the correct form.
  2. (10) a Primary 6 student

    Phoebe: No, he didn’t enough money. [Error-grammatical]

    Teacher: No, he didn’t have enough money. [FB-recast]

    Phoebe: No, he didn’t have enough money. [Repair-repetition]

  3. Incorporation refers to a student’s repetition of the correct form provided by the teacher, which is then incorporated into a longer utterance produced by the student.
  4. (11) a Primary 6 student

    Dave (D): What did the ‘rubber’ look like? [Error-phonology]

    Teacher: robber [FB-recast]

    Dave: What did the robber look like? [Repair-incorporation]

  5. Self-repair refers to a self-correction, produced by the student who made the initial error, in response to the teacher’s feedback when the latter does not already provide the correct form.
  6. (12) a Primary 5 student

    Dick: What did he ‘wore’? [Error-phonology]

    Teacher: What did he …. [FB-elicitation]

    Dick: wear [Repair-self]

  7. Peer-repair refers to peer-correction provided by a student, other than the one who made the initial error, in response to the teacher’s feedback.

(13) a Primary 5 student

Angel (A): Who ‘where’ you with? [Error-phonological]

Teacher: not ‘where’ [FB-repetition]

Angel: ‘wore’

Teacher: Who can help her?

Michael: were [Repair-peer]

Teacher: Who were you with?

Angel: Who were you with?

The category of ‘need-repair’ includes the following six types of utterances. Among the six types of utterances, only one of the categories can be found.

  1. Acknowledgement generally refers to a simple ‘yes’ on the part of the student in response to the teacher’s feedback.
  2. Same error refers to uptake that includes a repetition of the student’s initial error.
  3. (14) a Primary 5 student

    Ivy: What did he wear?

    Lai Man: He ‘worn’ a white shirt. [Error-phonological]

    Teacher: not ‘worn’ [FB-repetition]

    Lai Man: He ‘worn’ [Needs Repair-same error]

  4. Different error refers to a student’s uptake that is in response to the teacher’s feedback but that neither corrects nor repeats the initial error, instead, a different error is make.
  5. Off target refers to uptake that is clearly in response to the teacher’s feedback turn but that circumvents that teacher’s linguistic focus altogether, without including any furthers errors.
  6. Hesitation refers to a student’s hesitation in response to the teacher’s feedback.
  7. Partial repair refers to uptake that includes a correction of only part of the initial error.

Results

The focus of the present paper is on different types of feedback and uptake. The effect of error type on feedback type is an important variable and will be reported on in a subsequent study.

Table 1 provides a breakdown, totals for the entire database, of the number of student turns, the number of student turns with at least one error or in need of repair, the number of teacher turns containing feedback; the number of student turns with uptake and the number of student turns with repair.

Table 1 Frequency of turns with student error, teacher feedback, and student uptake

Total Student Turns

Student Turns with Error or Needs-Repair (% of Total Student Turns)

Teacher Turns with Feedback (% of Total Errors)

Student Turns with Uptake(% of Feedback)

Student Turns with Repair (% of Feedback)

Student Turns with Repair (% of Total errors)

Total

Errors

Feedback

Uptake

Repair

Repair

Primary 5

72

36

36

33

24

24

50%

100%

92%

67%

67%

Primary 6

44

23

23

20

18

18

52%

100%

87%

78%

78%

Total

116

59

59

53

42

42

51%

100%

90%

71%

71%

The totals for the entire database are illustrated by the graph in Figure 2. Of these, 100% receive feedback from the teacher. This means that 0% of the time learners’ errors are followed by a teacher or student topic continuation move. Of all the feedback moves provided by teachers in response to learner errors, 90% lead to uptake of some kind on the part of the learner, only 71% of the feedback turns lead to student repair. From the perspective of the total number of errors produced by students, 71% of errors eventually lead to repair.

Preferences for different feedback types are displayed in Table 2 as well as the total number of teacher turns containing feedback. Same as the finding of Lyster and Ranta, across the 2 classes, the single largest category is the recast that accounts for just nearly 36% of the total number of teacher turns containing feedback. The other feedback types are distributed in decreasing frequency as follows: Repetition (27%), Elicitation (19%), Metalinguistic feedback (15%), Clarification requests (2%) and Explicit Correction (2%). The high figure for repetition is somewhat deceptive because teacher hopes to highlight student’s mistake and made them notice their wrong utterances. From this finding, it appears that recasting the learner’s ill-formed utterance for is the feedback method of choice of the second language teacher in Primary 5 & 6. Other difference is noted in the larger amount of repetition used in Primary 5 than Primary 6.

Table 2. Distribution of feedback types

Primary 5

Primary 6

Total

n =

36

n =

23

n=

59

Repetition

15

1

16

42%

4%

27%

Recasts

12

9

21

33%

39%

36%

Metalinguistic feedback

5

4

9

14%

17%

15%

Elicitation

4

7

11

11%

30%

19%

Clarification requests

0

1

1

0%

4%

2%

Explicit correction

0

1

1

0%

4%

2%

It may be asked whether all types of feedback are equally effective in leading to learner uptake. This question can be addressed by referring to the pattern of uptake following the different types of feedback, which is presented in Table 3. According to the model has been used (figure 1), every learner response that follows teacher feedback is coded according to whether or not there is evidence of uptake. Recall that uptake consists of the categories of repair (repetition, incorporation, self, or peer) and needs-repair (attempts at repair that are in some way inadequate). In many cases, feedback does not lead to uptake because there is topic continuation, provided for the most part by the teacher or by a student. The number and percentage of feedback moves that do not lead to uptake are provided in the ‘No Uptake’ column in Table 3.

Table 3. Uptake following teacher feedback

Repair

Needs Repair

No Uptake

Repetition

n =

16

8

6

2

50%

38%

13%

Recasts

n =

21

15

3

3

71%

14%

14%

Metalinguistic

n =

9

9

0

0

100%

0%

0%

Elicitation

n =

11

8

2

1

73%

18%

9%

Clarification requests

n =

1

1

0

0

100%

0%

0%

Explicit correction

n =

1

1

0

0

100%

0%

0%

The most successful techniques for eliciting uptake are metalinguistic feedback, clarification requests and explicit correction. Repetition and recast are similar in that they are effective at eliciting uptake from the student (87% and 86% respectively). All learners following repetition involve uptake with an almost even distribution between repair and needs-repair. For the elicitation and recasts, it is more than triple as likely to lead to repair than needs-repair. The results are slightly different from what Lyster and Ranta have done. As they found that, clarification requests, metalinguistic feedback, repetition are similar in that they are effective at eliciting uptake form the students.

The preceding analysis focused on the relationship between feedback type and learner uptake. The goal of the teacher should be for the learner to self-correct or to have another student correct the error. A repair in which the student simply repeats what the teacher has said does not necessarily imply that the feedback has been understood.

The breakdown has been done in two ways. The first breakdown involves the number and percentage of each feedback type leading to repair, as presented in Table 4. Thus, when student repetition is removed, the percentages do not change for the categories of elicitation, clarification request, metatlinguistic request, and teacher repetition or error – none of which can elicit student repetition because none provides the correct form. On the other hand, the percentages for recast and explicit correction are reduced to nil because these two techniques provide learners with the correct forms and thus cannot lead to student-generated repair.

Table 4. Number and percentage of feedback turns leading to repair

Number of Repair

Repairs as % of Feedback Type

Number of Student-Generated Repairs

Student-Generated Repairs as % of Feedback Type

Repetition

n=

16

8

50%

8

50%

Recasts

n=

21

15

71%

0

0%

Metalinguistic feedback

n=

9

9

100%

9

100%

Elicitation

n=

11

8

73%

8

73%

Clarification requests

n=

1

1

100%

1

100%

Explicit correction

n=

1

1

100%

0

0

The second breakdown involves the number and percentage of repairs attributed to each feedback type, as displayed in Table5. When we consider all types of repairs, recasts account for the highest percentage (25%) and the others ranging from 15% to 2%. However, if we focus on student-generated repairs alone, the picture changes dramatically. In this case, recasts do not account for any repairs, while metalinguistic feedback, elicitation and repetition are responsible for the reaming self-generated repairs: 21%, 19% and 19% respectively. From these tables, it is clear that the choice of feedback technique has an effect on the type of repair that follows.

Table 5. Number and percentage of repairs attributed to each feedback type

Repetition

Recasts

Metalinguistic feedback

Elicitation

Clarification requests

Explicit correction

All repairs

n=

59

8

15

9

8

1

1

14%

25%

15%

14%

2%

2%

Student-generated repairs

n=

43

8

0

9

8

1

0

19%

21%

19%

2%

Discussion

The purpose of this study was twofold: to develop an analytic model comprising the various moves in an error treatment sequence and to apply the model to a database of interaction in two primary L2 classrooms with a view to documenting the frequency and the distribution of corrective feedback in relation to learner uptake. The analytic model was designed in accordance with the database itself and so remains to be validated by means of coding sets of classroom data from other contexts.

Although, it is likely undesirable for teacher to provide corrective feedback more frequently than this, the result suggests that when the teacher provides feedback, I might want to consider the whole range of techniques they have at their disposal rather than relying so extensively on recasts and repetition, which comprised over 36% & 27% of all feedback moves.

It is essential to acknowledge the need for teachers to carefully take into account their student’s level of L2 proficiency when making decisions about feedback. For example, Primary 6 students have a higher degree of proficiency uses recasts considerably less than Primary 5. This allows me to draw more on other feedback types, for instance, clarification requests and explicit correction were only used in Primary 6. Thus, given my students’ higher level of proficiency, I hope to push my students more in their output and to rely less on the modeling techniques.

The transcripts of classroom interaction reveal a large number of teacher repetitions of well-formed student utterances, teachers do this consistently so as to reinforce what students have said and to build further on students’ statements. As a result, there is a great deal of ambiguity in these communicative classrooms as students are expected to sort out whether the teacher’s intention are concerned with form or meaning. Feedback types other than recasts eliminate this ambiguity by allowing student themselves to either self-correct or to correct their peers. Explicit correction also eliminates ambiguity but doesn’t allow for student-generated repair.

The student-generated repairs in the error treatment sequence may be important in L2 learning for at least two reasons. First, they allow opportunities for learner to automatize the retrieval of target language knowledge that already exists in some form. Second, when repair is generated by students, the latter draw on their own resources and thus actively confront error in ways that may lead to revision of their hypotheses about the target language.

None of the feedback types stopped the flow of classroom interaction and that uptake-that is, the student’s turn in the error treatment sequence-clearly does not break the communicative flow either. Besides, the discourse was structured in ways that allowed teacher to intervene regularly, they were able to do so by interacting with students without causing frustration because students appeared to expect such interventions. Thus, it appears that is clearly anticipated in classroom discourse and that occurs as an insertion sequence without stopping the flow of communication.

The finding indicates that the feedback-uptake sequence engages students more actively when there is negotiation of form, when the correct form is not provided to the students- as it is in recasts and explicit correction- and when signals are provided to the leaner that assist in the reformulation of the erroneous utterance.

Providing feedback as part of a negotiated sequence in this ways is of course feasible in L2 classrooms only where learners already possess an adequate level of proficiency.

In conclusion, corrective feedback does not break the flow of the classroom interaction. Among the six types of feedback that have been analyzed, recasts and repetitions are usually been given in the classrooms. They are commonly used because students do not have highly language proficiency in L2. However, students may not understand the feedback since they merely repeat what the teacher says. In fact, students would be benefited if they were involved in the negotiation of the correct form. Providing opportunities for students self-correcting and peer-correcting is the goal of second language acquisition.