A Comparative Study of the Effectiveness of Direct Feedback and Indirect Feedback Methods for Urdu EFL Learners’ Writings

The current study is designed to find the comparati ve effects of direct feedback method and indirect feedback method of error treatm nt on Urdu EFL learners’ writings at intermediate level. To find out the comparative res ults about these two treatment methods, 46 students having Urdu as their mother tongue were di vided into two groups. Errors in writings of one group were treated by direct feedback method wh ile indirect feedback method was adopted to treat the errors of the other group. Repeated me asure experimental research design was used to find the accuracy in students’ writings. The stu dy revealed that both methods helped the students to decrease the errors and created accurac y over the time but direct feedback method yielded better results in comparison to indirect fe edback group. Comparative effects of two methods on error categories showed that for verb er ors, article errors and sentence structure errors direct feedback method was more helpful. Whi le for noun ending errors, wrong word errors, indirect feedback method of error treatment was more useful.


INTRODUCTION
Second language is always learned due to some political, economic, social or educational purposes therefore proficiency in Second language is of great importance.Feedback by teacher plays an important role in a language learning process.It is a tool through which teachers can gauge that learning process is actually being taking place.Without feedback, language learning process cannot be completed (Antokhin, 2000).
Second language is always learned due to some political, economic, social or educational purposes therefore proficiency in Second language is of great importance.Feedback by teacher plays an important role in a language learning process.It is a tool through which teachers can gauge that learning process is actually being taking place.Without feedback, language learning process cannot be completed (Antokhin, 2000).
Error correction also termed as error treatment, is a typical and widely used form of feedback.For giving response to learner's error, different terms are being used.These are 'feedback', 'repair', 'error correction', 'error treatment' 'negative feedback' and 'negative evidence'.Longman Dictionary of Language Teaching and Applied Linguistics (2002) defines feedback as "… refers to comments or other information that learners receive concerning their success on learning tasks or tests, either from the teacher or other persons " (p. 199).For any kind of response to learner's errors, Dulay et al. (1982) used the term 'feedback' in their work.According to Chaudron (1988) feedback is "various types of classroom interactions" used by the teachers to response the learners' errors.He adopts the term 'error treatment' and describes it as "any teacher behaviour following an error that minimally attempts to inform the learner of the fact of error" (p.150).For whole process of error treatment, Hendrickson's (1980) in his article used the only term error correction.
In fact the terms like error treatment, error correction, corrective feedback and repair are used with same meanings in the area of L2 learning even these are used interchangeably by different researchers in different studies or even in the same study.However, the final objective of error treatment is to enable the learner to identify the error, correct it and modify the underlying rule so that proficiency and accuracy may be created in their learning.
In second language learning, when error corrective feedback is provided for students' L2 writings, it becomes a difficult job for language teacher to construct an effective corrective feedback strategy.There are controversies over methods adopted for written corrective feedback (WCF) strategies.Regarding the issue that how to correct the errors, Ferris (2004) asserts that "there is a variety of options for error feedback-from direct correction of error to some fairly indirect and less informative approaches-from which the teacher must choose, again bearing in mind the needs of the students and goals of the writing course and task" (Ferris, 2003, p. 124).
Written corrective feedback is usually divided in two method by the researchers, that is direct feedback (explicit feedback or overt feedback) method and indirect feedback (implicit feedback or covert feedback) method (Hendrickson, 1980;Laland, 1982;Ferris, 1995aFerris, , 1995c;;Sheen, 2004;Erel and Bulut, 2007).In direct feedback method of error treatment, errors are located by the teacher and proper feedback is provided for these errors.Ferris (2001) describes that "direct feedback is when teacher provides the correct form for the student writer; if the student revises the text, and he / she needs only to transcribe the correction into the final version".Whereas indirect feedback refers to the situation where a teacher only identifies the error but does not provide a correct form of it, thereby leaving the student to diagnose and correct it.Ferris (2001) defines it as, "Indirect feedback occurs when the teacher indicates in some way that an error exists but does not provide the correction, thus letting the writer know that there is a problem but leaving it to the student to solve it".In this method teacher points out student's errors by underlining, circling, highlighting it or by providing codes (Lee, 2004).
When providing feedback to L2 writing there are several issues.One of the major issues is that whether the error should be corrected explicitly or implicitly.Research work done in the area of effectiveness of written corrective feedback method is less as compared to other areas of corrective feedback.Most studies, so far done, have focused on large scale instruction programs as well as on specific differences in explicit presentation and explanation of rules and implicit practice of rules (Chaudron, 1988).
For the effective use of error corrective methods Ferris (2004) asserts that feedback provided by teacher may vary from explicit type to less informative type depending upon the need of task assigned in a lesson but an instructor should keep in mind that adopted method of CF should not causes embarrassment to the students.Further CF method should be as per need of the students.Grammatical categories should be kept in mind before selecting the types of error correction method.
About the effectiveness of error correction there are studies in which different aspects of feedback methods have been discussed (Hendrickson, 1978(Hendrickson, , 1980;;Lalande, 1982;Robb et al., 1986;Ferris, 1995a, b;Ferris and Hedgcock, 1998;Ferris, 2002) but still there is controversy to claim which method of WCF is better to create accuracy in L2 writings.
Results of a longitudinal study by Lalande (1982) showed that students who received indirect CF outperformed as compared to direct CF group.Erel and Bulut (2007) conducted a research to identify the longitudinal effects of the two kinds of error feedback strategies in Turkish EFL context.The results of Erel and Bulut (2007) study show that the indirect coded feedback group performed better than the direct feedback group.While the indirect coded feedback group had an average of 6.04 errors for every hundred words, the direct feedback group had an average of 9.51 errors.However, at the initial stage the numbers of errors were less for the direct feedback group but at the end of second semester indirect coded feedback group outperformed the direct feedback group.However Chandler (2003) found that direct CF resulted to help more to create accuracy.For creating accuracy in L2 writing corrective feedback is of great importance rather it is always considered an important variable in variable in L2 learning (Lier, 1988).In spite of its importance less importance is given to it in teachers training (Hendrickson, 1984).Although L2 instructors daily spent time for correction still they do not have any model to follow to correct the errors in students L2 writings (woods, 1989).As teachers have no knowledge and training for about WCF strategies so they adopt WCF strategies according to their own will.
In Urdu EFL context too, teachers are without proper knowledge about WCF methods and their effectiveness.They give response to students' writings just by considering that as a language teacher it is their duty to correct the errors.Some foreign researchers (Laland, 1982;Erel andBulut, 2007 andEllis 2009) have investigated to know which method of WCF is more effective but contrary in Urdu EFL context there is no study to find the effects of WCF methods especially direct feedback versus indirect feedback.This piece of research is an attempt to contribute into the area of error treatment by providing a great opportunity for language teachers to review and to reconsider effective ways of teacher responses to various writings of L2 learners.

Objectives of the study
The following are the objectives of this study: 1.
To explore the possible effects of direct and indirect method of WCF in Urdu EFL context with regard to accuracy in L2 writings.2.
To evaluate which method of WCF among direct feedback and indirect feedback is more effective to decrease the errors in L2 writings of Urdu EFL learners.

3.
To find out the possible effects of direct feedback method and indirect feedback method on different error categories.

RESEARCH TOOLS
In the present study two instruments were used to collect the data i.e. questionnaire and test.Test is described as "gateway" to find out the progress of an individual and evidence of potential for learners abilities.In this study, data about students' views regarding the effectiveness of error treatment methods was collected through questionnaire and a test as a research tool was used to empirically find the comparative effectiveness of error treatment methods.

Error taxonomy adopted for current study
Error taxonomy for treatment of error as suggested by Ferris and Hedgcock (2005) is adopted for current studies.Further the effects of error treatment groups are analyzed in the light of these seven types of error categories (Table 1).

Procedure of the study
For the present study, parallel group repeated measure research design was used for this experimental study as this model is helpful for long term studies as well as for short term practices.This research design allows the researcher to observe the changes in the performance of subjects over time, due to the effects of some treatment.
This study was conducted at Government Post Graduate College of district Vehari, Pakistan.All the students at intermediate level constitute the population of this study.To find the beliefs of students about effectiveness of error correction methods, eighty two students having Urdu as their mother tongue were selected.They were asked to respond a questionnaire.After investigating the beliefs of students about the error correction methods, 46 students among eighty two were selected randomly.Further they were divided into two groups (direct feedback group and indirect feedback The writings of the students in direct feedback group were corrected by direct feedback method of error correction whereas for writings of second group indirect feedback of WCF was adopted.All the tests were conducted after a period of one week as this time of one week was convenient for teacher to provide the error treatment to both groups according to their feedback method group.All the feedback provided writings were properly documented with reference to number of error made and corrected.These corrected errors are also logged error category wise.Since length wise writings of different student were different, although they were with the same time to write.So, in order to avoid this difference, numbers of treated errors in all writings were considered till maximum length of 200 words.During writing tasks of learners and teacher's treatment of errors, no instructions were given to students or teacher by researcher, except for teacher to be specific for type of error correction method for each group.The correction codes used by the teacher were also made clear to students of indirect feedback group in the beginning.

QUESTIONNAIRE FINDINGS
Questionnaire was used before the experimental study to know the beliefs of students about the study.The second purpose of use of this questionnaire was to support the result of main experimental study.The responded results of the questionnaire are given in Table 2. 1.
For the first entity of questionnaire, all students want their teachers to correct their errors.

2.
For second question, 62% students responded that error correction should be for all errors.23% students expressed that the error correction should be for some errors and 15% of students claimed that error correction should be only for errors that are important to correct.

3.
For questionnaire third entity 39% of students claimed that their teachers use direct feedback for their errors.Percentage of students to claim that their teacher use indirect feedback is 19.51% and 41.5% students responded that their teacher use both direct and indirect feedback method for correction of their errors.4.
For Questionnaire 4 th entity, 57% of students responded that direct WCF is more effective to decrease errors in writings.About 25% of students' were of the view that indirect WCF is more effective.Also there were 17%.Students who were of the view that teachers should use both methods for WCF.

Comparative Results of Direct WCF Group and Indirect WCF Group
Below are given the comparative results of both direct and direct WCF group's results.This comparison is made on two bases (Table 3). 1.Comparison of errors appeared from 1 st to last test of both groups.2. Comparison of total errors of all the test of both groups.
The graphical presentation of above results is given in Figure 1.It presents the comparative graph of error decrease due to dose therapy of WCF for direct feedback group and indirect feedback group.The black colored line in this figure demonstrates the decrease in errors of direct feedback group from first to last test.The white  colored lined presents the decrease in errors of indirect feedback group.So, from Figure 1 it is clear that direct WCF method is more helpful to decrease the errors of the learners as compared to indirect WCF method.

Comparison of total errors of both groups
The total number of errors made by students in both groups in all the tests is given in Table 4.The total numbers of error of indirect feedback group are greater than direct feedback group.The comparison of both groups on the basis of total errors shows that indirect feedback group made 88 errors more as compared to direct feedback group.The significant difference in total errors as described by the data reveals that direct feedback method is more helpful to decrease the errors as compared to indirect feedback method (Figure 2).The graphical presentation of total errors in both direct feedback group and indirect feedback group, as illustrated in Figure above demonstrates that indirect feedback group made more errors and direct feedback group made less errors comparatively.The significant difference of 88 errors proves the fact that direct feedback method is more helpful to decrease the errors of students as compared to indirect feedback method.

Verb errors
To find out comparatively what effects do direct feedback method and indirect feedback method have on correction of verb errors, the percentage and mean of verb errors of both group is presented in Table 5.
The percentage of verb errors in direct feedback method is 12.04 while the percentage of verb errors under indirect feedback is 14.69.Similarly, the mean of verb error for indirect feedback is also greater than direct feedback group.Thus, comparatively direct feedback is more helpful to decrease the verb errors.

Article errors
Table 6 presents the frequency of article errors as appeared in both groups, followed by respective percen-

Noun ending errors
Data in Table 7 present the error frequency, percentage and mean of noun ending errors of direct feedback group and indirect feedback group to find the relative effects of treatment methods on noun ending errors.
The percentage and mean of direct feedback group for noun ending errors is more as compared to indirect feedback group.Thus indirect treatment method decreased

Sentence structure errors
To find the effectiveness of error treatment methods data of sentence structure errors is listed in Table 8.Table 8 shows that the mean and percentage for sentence structure errors is less for direct feedback group relatively to indirect feedback method.So, direct feedback method of error treatment is valuable to decrease the sentence structure errors as compared to indirect feedback method.

Wrong word errors
Percentage and means of indirect feedback group are less than direct feedback group.So, this significant difference in mean and percentage of both groups reveals that indirect feedback method is comparatively more helpful for wrong word errors (Table 9).

Spelling errors
To find out the effects of direct treatment and indirect treatment of error correction on spelling errors the error frequency, percentage and mean of both groups is illustrated in Table 10.
The difference in percentage shows that spelling errors in direct feedback group are 0.35% more as compared to indirect feedback group.So, comparatively indirect feedback method is more helpful to decrease the spelling errors.The difference between two groups for this error category is not much to signify.Similarly, there is not much difference for means of both groups.Thus, direct and indirect feedback methods are equally helpful to decrease the errors of other errors category.

DISCUSSION
Data obtained from these two groups to find out the answers of research questions, revealed the following results.
1.The study results reveal that accuracy of students in direct feedback group increased over the time and students' errors were gradually decreased.This significant decrease in total errors and average errors as described by the data proves that direct feedback method is helpful to reduce the errors in Urdu EFL students' writings.2. Data results about the second research question explored that under the indirect feedback, errors of indirect feedback group were gradually decreased.The significant decrease in total errors and average errors of students as described by the data proves that indirect feedback method is helpful to decrease the errors in Urdu EFL students' writings.3. Regarding the third research question, results of study describe that direct feedback method is more helpful to reduce the errors as compared to indirect feedback method.In this study, the comparison of direct feedback and indirect feedback was made.Data show that total errors of direct feedback group are 1851 and total errors made by students in all the tests of indirect feedback group are 1939.It is a significant improvement by direct feedback method to decrease more errors compare to indirect feedback.So results of data prove that comparatively direct feedback method is more helpful to decrease the errors in writings of Urdu EFL learners.4. The results of study, for fourth research question explored that the direct feedback method and indirect feedback method have different effects on different errors categories.a.As described by the data, the percentage of verb errors under the direct feedback error is 12.4 % and percentage of verb errors under the indirect feedback is 14.69 %.Errors percentage of indirect feedback group is 2.65 % more as compared to direct feedback group.2.65% less errors of direct feedback group reveals that direct feedback method is more helpful to decrease the verb errors.b.The percentage of article errors under direct feedback is 11.45 and the percentage of article errors under indirect feedback is 14.18.Data reveals that article errors made by indirect feedback group are 2.73 % more as compared to direct feedback group.Thus, this difference in percentage of errors shows that direct feedback method is more useful to decrease the article errors.c.The data about comparative effects of direct feedback and indirect feedback on noun ending error category revealed that percentage of noun ending errors under the direct feedback is 14.20 and under indirect feedback it is 11.60 %.Percentage wise the noun ending errors are 2.6 % more under the direct feedback group.This significant difference in percentage of noun ending errors reveals the fact that indirect feedback method is more successful to decrease the noun ending errors.d.The data about sentence structure errors described that the percentage of sentence structure errors under direct feedback is 11.88 and under indirect feedback these errors are 13.61%.Sentence structure errors of direct feedback group are 1.73 % less as compared to indirect feedback method.This difference of 1.73 in percentage as described by the data indicates that direct feedback is more useful to decrease the sentence structure errors.e.The data results about wrong word errors category explored that under direct feedback noun ending errors are 17.50 % and under indirect feedback this error category is 13.40 %.The percentage of wrong word errors is 4.1 % more under direct feedback group as compared to indirect feedback group.This significant difference in wrong word errors verifies the fact that indirect feedback is comparatively more helpful to decrease the wrong word errors.f.For spelling errors data revealed that percentage of spelling errors in direct;8 feedback group is 20.20 %, while under indirect feedback the percentage of this error category is 19.85.Difference in percentage errors of these two groups for spelling errors is 0.35.This significant difference in spelling errors as described by the data reveals that indirect feedback is more effective to decrease the spelling errors.g.Comparison of other errors category for direct feedback and indirect feedback specified that under direct feedback students made 12.69 % errors and under indirect feedback method students made 12.63 % errors of other errors category.Difference in percentage of both groups for this error category is 0.07 %.This significantly less difference indicates that both feedback methods of error correction are equally helpful to decrease the other errors category.
The comparative results of both groups reveal the fact that both the methods of error correction are helpful in reducing the errors in students' written scripts but direct feedback method of error treatment is more useful as compared to indirect feedback.These results of experimental study also supported from questionnaire results where 57.3 % of students responded that direct feedback method was more helpful for them to decrease the errors.

Pedagogical implications
As summarized above, the current research has addressed several fascinating issues about error treatment methods for writings of L2 learners.These findings may contribute to a new implication to second language acquisition, particularly, in the area of error treatment, by helping the teacher to find right method of error treatment while treating the errors in students writing.These pedagogical implications suggested through this study are as under: 1.There should be no hesitation to correct learners' errors in their L2 writings because this study proves that error treatment is helpful to create accuracy in L2 writings.Considering the learners' proficiency level and giving them written error feedback on some selected types of linguistic errors enables them to improve accuracy right after they receive written feedback.2. In relation to giving feedback to learners' errors at intermediate level it is suggested to provide direct feedback in preference to indirect feedback.Learners at this level do not have much ability to correct the errors that are just highlighted by teachers through indirect feedback method.Learners in this study were at intermediate level, and it was found in research outcomes that learners who received direct feedback performed better than those who received indirect feedback.This proves that learners have less ability to self-correct their errors at this level.3. Indirect feedback method of error treatment is also useful to decrease the errors in L2 writings but it is less useful as compared to direct feedback method of ET.Teachers should use this method only if they are sure that students are well aware of the rules related to that specific error and just giving a hint about that error will enable them to correct that error.4. Before treating an error a teacher should diagnose the type of error.Then he should adopt the method of error correction according to type of error. 5. Types of errors that are governed by some grammatical rules (such as verb, article and sentence structure errors), better accuracy for such types of errors can be gained by direct feedback method.6. Types of errors that are not rule governed, such as spelling errors, noun ending errors, wrong word errors and capitalization errors, can be better treated by indirect feedback method because indirect feedback makes the learner more responsible by creating their interest in error correction process.7.At different level of learning, different methods of error treatment should be used.If learners are well aware of the grammatical rules then indirect feedback method can be useful but if student are not well aware of rules then direct feedback correction is more helpful.

Figure 1 .
Figure 1.Comparative presentation of both groups' errors.

Figure 2 .
Figure 2. Bar chart to present the comparison of total errors of both groups.

Table 1 .
Types of errors.
group) with 23 students in each group.A series of tests were conducted to gauge the effects of experimental factors.The topics of essay were selected from syllabus of intermediate level and to create convenience, students were provided with brief outlines of the essay.

Table 3 .
Comparison of Errors in all tests.

Table 4 .
Comparison of total errors of both groups.

Table 5 .
Comparison of verb errors.

Table 6 .
Comparison of article errors.

Table 7 .
Comparison of noun ending errors.

Table 8 .
Comparison of sentence structure errors.

Table 9 .
Comparison of wrong word errors.
Table 11 presents the error frequency, percentage and mean of direct feedback and indirect feedback group.

Table 10 .
Comparison of spelling errors.

Table 11 .
Comparison of other errors category.