Impact of Self-Correction on Extrovert and Introvert Students in EFL Writing Progress

To investigate the impact of self-correction method as an alternative to the traditional teacher-correction method, on the one side, and to eva lua te the impac t o f per sona l i ty t r a i t s o f Extroversion/Introversion, on the other side, on the writing progress of the pre-intermediate learners three null-hypotheses were proposed. In spite of students absolutely different personalities some teachers probably expect all the students to react to their incompetent feedback in an indistinguishable way. To put it in other words, these teachers pay no attention to the fact that each student is completely a different human being with his/her exclusive personality. These instructors pursue the traditional correction method and correct every error and even do not agonize to present work for different correction methods let alone thinking of making learners engage in their own learning. Therefore, 120 pre-intermediate Iranian female EFL students were selected through employing of Nelson English Language Test (NELT) and the Eysenck Personality Questionnaire (EPQ). They were assigned to four different groups-two groups include extroverts and two groups include introverts. Subsequently, they were given five expository topics to write about in a five-week period. While two groups, one extroverted and one introverted enjoyed self-correction, the teacher corrected the writings of the other two groups. The obtained results showed that personality type had no significant effect on learners' progress in writing. Nevertheless, self-correction method showed to be significantly effective at 0.5.


Introduction
There has been a steady argument among scholars and teachers all through the history of teaching writing to second language (L2) learners with regard to the role of error feedback in helping students learn how to write (Fathman and Whalley, 1990;Ferris, 1999a;Lalande, 1982;Semke, 1984;Truscott, 1996).Despite studies that have been conducted to scrutinize this matter, a lot of uncertainty remains regarding what type of error correction help which learners to progress which parts of the writing process.So, many English as a Second/ Foreign Language (ESL/EFL) writing teachers are often confused about how to help their students.
It has been observed that students are often slow in recognizing errors of their own making but are more able to spot errors in sentences written by others (Bartlett, 1982).A feasible clarification is that when a student reads his own composition, he does not read it with the eyes and mind of a reader who does not as yet know what the writer tends to imply.In its place, he reads with the advantage of the background information which he, as writer, has and which enables him to provide missing links that has not been overtly articulated.Since correcting learners' errors is perhaps the most responsive part of teacher's job and it is barely amazing that it makes so much consideration and deliberations.Correcting learners' errors is one of the main tasks of a teacher.For language teachers error correction of students' written work is a time consuming duty.
Based on some studies, the students feel the need to be corrected; on the other hand, they lose their self-confidence if they are corrected repeatedly (Ferris, 2002).Among the several self-correction techniques, merely attaining the correct answer is the most commonly used and preferential techniques.However, there are several other self-correction techniques which would guarantee longer time retention in the memories of learners (Ferris, 2002).As a matter of fact, for attaining a logical level of retention through some cognitive processes which are extremely different from an easy disruption and correction and in order for a successful correction to take place, some procedures should be taken into considerations.For the meantime, a main significance to the efficiency of each and every error correction technique is the personal features of the learner.Learners with different cognitive features would come up to the techniques of error correction in various ways to deal with their differences in the properties of cognitive styles.
Personality is one of the individual differences which is broadly established to have an outcome on learning generally and second language acquisition especially.It sounds that personality traits have types of result on the learners' language learning.In addition, many research projects have shown that corrective feedback in the classroom situation is a real need (Bitchener, Young, & Cameron, 2005).
Though, little research has been achieved to discover if definite error correction techniques are more effective with regard to the different personality traits of the language learners.It looks that the feedback to a type of error correction technique is different with regard to the learners' cognitive styles.Many teachers who study the students' performances consider that extroverts tend to be better students and carry out much better than their introverted classmates.
Some teachers justify the principle by saying that extroverts are more sociable and better risk-takers and consequently, they would be inclined to learn faster and better than their introverted counterparts.Some early researchers such as Pritchard (1952) andPimsleur, Sunland, andMeintyre (1966) were on the side of extroversion and regarded it as the popular stereotype of good learners whereas some other researchers believed that to be sociable and unreserved could be suggested as a proper strategy to be adopted by learners, mainly in the progress of communicative skills.
To reinvestigate this trust, the researcher tried to conduct an experimental study considering the different rates of writing progress in morphological, lexical, syntactic and mechanical errors of extroverted students versus their introverted counterparts through two different methods of correction, namely self-correction and teacher-correction.
Regardless of the reality that every time teachers check the students' writing meticulously and return them to their students many EFL teachers protest that their students do not show substantial progress in their writings.This inadequacy on the part of the students may be principally because of the weak and unproductive strategies used to give feedback to the poor students on the part of these teachers.Without doubt, these teachers are not aware that the correction of writing should go after some principles, techniques and strategies (Ferris, 2007).They only want to bombard the students with the disappointing red marks on their papers.Furthermore, they may not be aware that learning will not take place in any way until cognitive processes are achieved in the students' minds (Swain & Lapkin, 1995).
Moreover, in spite of students absolutely different personalities these teachers probably expect all the students to react to their incompetent feedback in an indistinguishable way.To put it in other words, these teachers pay no attention to the fact that each student is completely a different human being with his/her exclusive personality.
From the task-based point of view, learning will take place in the best way when students get involved in learning and teaching processes personally and feel responsible for their own rate of learning.To correct the writing works of the students is certainly a significant part of any writing process and self-correction as an alternative to the traditional teacher-correction method is appropriate to get the students directly concerned in their own learning in that from the critical point of view it gives any individual student the opportunity to look at his/her writing.In addition, focusing on the students, many teachers have suggested the idea of regarding the learner factors such as personality type, motivation, attitudes and other personal factors.Thus, these factors have gained importance.One of these factors is the personality trait of Extroversion/Introversion. Considering a student as an extrovert or introvert would likely affect many teachers' attitudes toward him due to the erroneous stereotype that some teachers have against introverted students.The present study intends to investigate the impact of these two personality factors as well as correction methods on writing progress of students.
The general objective of this study is an account of the relationship between extroversion / introversion personality trait, on the one hand, and self-correction method versus teacher-correction, on the other hand, on students' writing progress.
The first specific objective of the present study is that it investigates the effectiveness of self-correction as two different correction methods of the writing task in extroversion and introversion groups.As far as writing progress is concerned, some teachers have suggested that a part of the dependability for learning be put on the students themselves.For instance, the students can be taught to read their own pieces of writings and check for the errors.That implies self-correction instead of the traditional teacher-correction method.
The other specific objective of this study is that teachers have to employ inspiring feedback methods that foster students to mull over, rework, and self-correct their drafts in extroversion and introversion groups for making marking to be efficient in the writing classroom.In this method, learners will be enthusiastically involved in the correction of their errors, rather than being passive audience of the teacher's remarks and corrective comments.
To investigate the performance of the extroverts and introverts with regard to their progress in writing the following questions are proposed: (1).Is self-correction method more effective than teacher-correction method in students' writing progress?
(2).Is there any relationship between Extroversion and students' writing progress?
(3).Is there any relationship between Introversion and students' writing progress?
Extrovert and Introvert are two significant personality factors affecting language learning.Self-correction, on the other hand, is of main significance.In point of fact, most of the teachers pay no attention to different techniques of self-correction; in addition, those who are aware of the techniques of self-correction usually don't have the smallest amount of idea on the efficacy of different self-correction techniques on different people.
The first significant of the present study is to develop learner independence and responsibility.The second significance of this study is the comparison carries out between the performances of extroverted students versus introverted students in fulfillment of the tasks (i.e.progress in writing).And, finally the third significance of this study is to investigate the role of error feedback in helping students learn how to write.

Methodology
The participants of this study have been randomly selected from among pre-intermediate EFL learners.Since all early intermediate EFL learners in Kish Air English Language Institute which is located in Tehran, Iran had an equal and independent chance of being included in the sample, the researcher assigned subjects to the experimental and control groups by random assignment; the type of sampling in this study is simple random sampling.The subjects of this study were 120 female students who (according to the chart of the institute) have been supposed to be pre-intermediate learners.These 120 students have been selected from among other students according to their scores in a pre-intermediate version of Nelson English Language Test and also on the basis of the responses to Eysenck Personality Questionnaire.
The following steps have been taken as the procedures of the study: A pre-intermediate version of Nelson English Language Test was administered to select a homogeneous group out of the available population.205 sample female students who were pre-intermediate students in Kish Air English Language Institute which is located in Tehran, Iran have been chosen from the pre-intermediate classes.
(2) According to the students' scores in NELT, 191 students were selected out of the total number of 205 students on the basis of normal distribution of scores.To be more accurate, the students whose scores fell one standard deviation above and below the mean were selected as the homogeneous subjects in the first phase (i.e.scores between 30 and 43), and the other 14 students whose scores did not fall within this range were crossed out.
(3) On the basis of the students' responses to the 57 questions, the degrees of extroversion of all the 191 students were measured.It is to be noted that according to EPQ standards, more than three unreliable answers to the 57 items would mean that the given student has not been honest enough to be included in the study.Therefore, to be on the safe side, 19 students who had more than three 'unreliable' responses were crossed out in this phase.Also, 52 students were randomly crossed out, 31 students were extroverts and 21 were introverts.The ultimate number of the subjects of the study was 120.
(4) After the subjects have been selected, they have been randomly divided into four groups.Group 'A' included extroverted students whose writings were supposed to be corrected by the teacher.Group 'B' included extroverted students who were supposed to correct their own writings.Group 'C' included introverted students whose writings have been supposed to be corrected by the teacher.And finally, group 'D' included introverted students who have been supposed to correct their own writings.According to the subjects' scores in Nelson English Language Test, a one-way ANOVA has been conducted to ensure the homogeneity of the four groups.
(5) Prior to the administration of the pre-test, for two sessions, the participants in all groups have been given detailed identical instruction and practice, all by one teacher, on how to write well-formed paragraphs on expository topics.This maintains both the reliability and the validity of the scoring of writings.
(6) After some extroverts and introverts have been identified and assigned into the four main groups, the correction-code sheets have been distributed among the students of the two experimental groups (i.e.groups 'A' and 'C') who were supposed to undertake self-correction during the treatment weeks.
(7) During a 5-week period, five different expository topics which were selected by the Test of Written English (TWE) have been given to the students in all the groups.Throughout this study, all the participants were assigned to write expository paragraphs.On the whole, there were five topics assigned.Two topics were utilized as the pre-test and the post-test topics with a time limit of 20 minutes, and the remaining three were the writing topics on which the participants composed their paragraphs and received the relevant feedback in the course of the treatment phase.Each week the teachers (the researcher and one of his colleagues) corrected the writing of the students in groups 'A' and 'C' (i.e. the control groups).Meanwhile, the writings of the students in groups 'A' and 'C' (i.e. the two experimental groups) have first been checked and commented on by the teachers giving some clues on the error types and then corrected by the subjects themselves using those clues and the correction-codes.
(8) The next step was to design a method to evaluate and score the writings.The method for correction is a formula as a practical measure of composition rating, especially for those teachers who are quite busy to develop their own scale "Total number of errors/total number of words 100".
(9) A rater training session has been held before the raters begin to score the written texts in order to minimize an individual rater's variability and to enhance inter-rater reliability.
(10) An inter rater reliability coefficient has been used to show the agreement between the two raters.The relationship between rater 1 and rater 2 in first writing was investigated using Pearson product-moment correlation coefficient.Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity.There was a strong, positive correlation between the two variables [r=.98, n=120, p<.05].Also, the relationship between rater 1 and rater 2 in fifth writing was investigated using Pearson product-moment correlation coefficient.Preliminary analyses were performed to ensure no violation of the assumptions of normality, linearity and homoscedasticity.There was also a strong, positive correlation between the two variables [r=.99, n=120, p<.05].
(11) In the fifth week, a topic has been given to all the subjects to write a composition about.This time, according to the design rating scale, all the writings checked and scored by two raters.For each individual piece of writing, the two obtained scores summed up and then averaged.The final scores in this phase have been considered as the main data of the study.
(12) Multivariate Analysis of Variance was conducted to compare scores on the confidence in coping with statistics test at self-correction and teacher-correction.

Data Analysis
As it was stated before, to select and homogenize the subjects, a Nelson English Language Test was administered at the onset of the study.Table 1 represents the descriptive statistics for the NELT.
A pre-intermediate version of Nelson English Language Test was administered to select a homogeneous group out of the available population.According to the students' scores in NELT, 191 students were selected out of the total number of 205 students on the basis of normal distribution of scores.The students whose scores fell one standard deviation above and below the mean were selected as the homogeneous subjects in the first phase (i.e.scores between 30 and 43).So the 14 students whose scores did not fall within this range were crossed out.
On the basis of the students' responses to the 57 questions, the degrees of extroversion of all the 191 students were measured.It is to be noted again that according to EPQ standards, more than three unreliable answers to the 57 items would mean that the given student has not been honest enough to be included in the study.Therefore, to be on the safe side, 19 students who had more than three 'unreliable' responses were crossed out in this phase.Also, 52 students were randomly crossed out, 31 students were extroverts and 21 were introverts.The ultimate number of the subjects of the study was 120.
After the 120 subjects were assigned to the four groups, a one-way ANOVA was carried out to ensure the homogeneity of the four groups regarding their general English proficiency.The general descriptive statistics related to the distribution of the subjects in the four groups according to their performance in NELT are presented in Table 2.
Compared with the critical F value of 2.68, the low observed value of .072depicts that there is no significant difference between the four groups as far as the students' performances in NELT are concerned.The test is not significant (i.e..975>.05).This implies that the subjects have been assigned to the four homogeneous groups.
After the 120 subjects were selected they were randomly divided into four groups who were given personality traits questionnaire.Group 'A' included extroverted students who were supposed to correct their own writings.Group 'B' included extroverted students who were supposed to be corrected by the teacher.Group 'C' included introverted students who were supposed to correct their own writings.And finally, group 'D' included introverted students whose writings were supposed to be corrected by the teacher (Table 3).Then, In order to ensure the homogeneity of the first writing, a one way ANOVA was carried out (Table 4).
Compared with the critical F value of 2.68, the low observed value of 1.06 depicts that there is no significant difference among students in writing one (i.e.p=.365 >.05).
To find the answer to the first question, the researcher obtained the main data from the final compositions of the four groups after the fifth week of treatment.
To analyze the data, Multivariate Analysis of Variance was employed to investigate self-correction versus teacher-correction method.
Multivariate Analysis of Variance was conducted to compare scores on the confidence in coping with statistics test at self-correction and teacher-correction.There is a significant effect for self-correction [Wilks' Lambda=.081,F (4,113) =320.220,p<.05, multivariate partial eta squared=.919],table 5.
Table 6 depicts that self-correction (in extrovert and introvert) subjects is statistically significant in comparison with teacher-correction in all four kinds of errors.Thus, the first null hypothesis is rejected.
In morphology the significance of Self-correction Extrovert in comparison with Teacher-correction Extrovert is .004,at .05 levels of significance; (P< .05);so Self-correction Extrovert is more significant and effective than Teacher-correction Extrovert.
In morphology the significance of Self-correction Introvert in comparison with Teacher-correction Introvert is .011,at .05 levels of significance; (P< .05);so Self-correction Introvert is more significant and effective than Teacher-correction Introvert.

c) In lexical the significance of Self-correction Extrovert in comparison with
Teacher-correction Extrovert is .000,at .05 levels of significance; (P< .05);so Self-correction Extrovert is more significant and effective than Teacher-correction Extrovert.d) In lexical the significance of Self-correction Introvert in comparison with Teacher-correction Introvert is .000,at .05 levels of significance; (P< .05);so Self-correction Introvert is more significant and effective than Teacher-correction Introvert.e) In syntactic the significance of Self-correction Extrovert in comparison with Teacher-correction Extrovert is .000,at .05 levels of significance; (P< .05);so Self-correction Extrovert is more significant and effective than Teacher-correction Extrovert.f) In syntactic the significance of Self-correction Introvert in comparison with Teacher-correction Introvert is .000,at .05 levels of significance; (P< .05);so Self-correction Introvert is more significant and effective than Teacher-correction Introvert.g) In mechanical the significance of Self-correction Extrovert in comparison with Teacher-correction Extrovert is .003,at .05 levels of significance; (P< .05);so Self-correction Extrovert is more significant and effective than Teacher-correction Extrovert.h) In mechanical the significance of Self-correction Introvert in comparison with Teacher-correction Introvert is .000,at .05 levels of significance; (P< .05);so Self-correction Introvert is more significant and effective than Teacher-correction Introvert.
Considering a student as an extrovert or introvert would likely affect many teachers' attitudes toward him due to the erroneous stereotypes that some teachers have against introverted students.Thus, to analyze the data for research questions 2 and 3, Multivariate Analysis of Variance was employed to investigate extroversion/introversion as personality traits and students' writing progress in all four types of errors (Morphological, Lexical, Syntactic and Mechanical), (table 7) As table 7 shows in morphology, self-correction extrovert subjects are not statistically significant in comparison with self-correction introvert subjects (p=.990>.05)and, also, teacher-correction extrovert subjects are not statistically significant in comparison with teacher-correction introvert subjects (p=1.000>.05);also, in lexical, self-correction extrovert subjects are not statistically significant in comparison with self-correction introvert subjects (p=.979>.05),and teacher-correction extrovert subjects are not statistically significant in comparison with teacher-correction introvert subjects (p=.551>.05); in addition, in syntactic, self-correction extrovert subjects are not statistically significant in comparison with self-correction introvert subjects (p=.975>.05),and teacher-correction extrovert subjects are not statistically significant in comparison with teacher-correction introvert subjects (p=.156>.05);and finally in mechanical self-correction extrovert subjects are not statistically significant in comparison with self-correction introvert subjects (p=.206>.05),and teacher-correction extrovert subjects are not statistically significant in comparison with teacher-correction introvert subjects (p=.991>.05).So hypotheses two and three are accepted.This depicts that being extroverted or introverted alone, that is personality trait regardless of the correction method, has no significant effect on students' progress in writing.

Results and Discussions
According to the first null-hypothesis, correction method would have no effect on students' progress in writings.However, the results depict that self-correction method has a greater effect on students' progress in writing than the traditional teacher-correction method.Therefore, the first null-hypothesis could be rejected in this study.
The second and third null-hypotheses point out that there are no relationship between extroversion and introversion with learners' writing progress.Personality traits demonstrate no significant effect on learners' writing progress in this study.Therefore, the second and third null-hypotheses were supported in the present study.

Conclusion
In the present study, it has been challenged to provide a historical account of frequent trends in the teaching of writing.After that, the main approaches of giving feedback to learners' written works, that is to say error-correction was involved on.Introducing two different methods of error-correction (self-correction and teacher-correction) in writing, it was attempted to examine the efficacy of the self-correction method as an alternative method to the traditional teacher-correction method in teaching of writing to learners.And therefore, the effectiveness of the traditional method was questioned.
The present research is an investigation of the impact of two different correction methods, on the one hand, and personality traits of extroversion/introversion, on the other hand, on pre-intermediate EFL students' progress in writing.
The first null-hypothesis, proposing that 'the two methods of self-correction and teacher-correction are not significantly different in affecting the students' writing progress' was rejected at 0.05, implying that there is a significant difference between the two correction techniques, with self-correction being more effective in writing progress than the teacher-correction method.
However, the second null-hypothesis proposing that 'there is no relationship between extroverts and their writings progress' was supported.

Table 1 .
Descriptive statistics for Nelson English Language Test (NELT)

Table 2 .
Statistics related to the homogeneity of the four groups

Table 3 .
The distribution of the students in the four main groups according to their personality trait and the correction Methods

Table 4 .
Statistics related to the homogeneity of the first writing

Table 6 .
Multivariate Analysis of Variance.Effect of Self-correction method and Teacher-correction method

Table 7 .
Multivariate Analysis of Variance The effect of Self-correction method and Personality traits on writing progress