Transcript The Effect of Different Kinds of Feedback on Second

The Impact of Written Corrective
Feedback on Student Writing
Accuracy
Preview:
Language teachers spend hours to
correct students’ writing in
different
Correcting
ways:
grammatical
errors
Marking
Grading
Commenting
Responding
The continuous debate:
There have been different approaches to error
correction, but there is not still a sense of
certainty about how best to provide such
corrective feedback on writing and the debate
has continued for several years between the
advocates and opponents of corrective feedback.
John Truscott (1996) rejected the practice for
grammar correction in his review essay.
‘ … that correction is harmful rather than simply
ineffective…[and] that no valid reasons have been
offered for continuing the practice in spite of
these overwhelming problems…thus, for the
foreseeable future my conclusion stands: Grammar
correction has no place in writing classes and
should be abandoned’ (pp. 360-367)
Truscott’s reasons for such a
strong claim:
Error correction neglects the learners
developmental sequence of acquisition
This practice suffers from a range of practical
problems such as teachers’ ability and willingness
to give and students’ motivation to receive error
correction.
It wastes time and energy while this amount of time
can be spent on more productive aspects of a
writing program.
Truscott’s claim has faced a great deal of
criticism in different reviews (e.g.
Chandler, 2003, Ferris, 1999), but he has
not withdrawn his case against grammar
correction and grammar correction is, in
general a bad idea until future research
prove that there are particular cases in
which it might not be a misguided activity
(Truscott 2007).
Direct CF
Indirect CF
Narrative
writing
Short term
Long term
Theoretical definitions:
1. Feedback:
Penny Ur (1996) defines it as ‘type of
information which is provided for the
learners about his or her performance of
a learning task, usually with the aim of
improving this performance (p.242)
Feedback
Assessment:
The learner is
informed how well
or badly has
performed.
Correction:
some specific
information is
provided on what
the learners
perform.
Direct feedback: It refers to overt
correction of errors. All the error in the
students’ written assignments are
corrected by the instructor.
Indirect feedback: It refers to prompting
students about the location or type of
errors and leaves monitoring and
correction to the students themselves.
Narrative writing: It tells a story. It
can be whether true or fictional.
Avoidance strategy: A common
communication strategy which
learners use to avoid a difficult word
or structure and use a simpler word
and structure instead.
Statement of the problem:
Although teachers give feedback, their
feedback on form and content are often
vague, contradictory, unsystematic and
inconsistent. This leads to various
reactions by students including
frustration, confusion and inattention to
comments. Many teachers still tend to
respond to their students’ written works
by using the traditional method of
correcting all grammatical errors.
Research questions:
1. Does the Indirect CF decrease the number of grammatical errors
of both total and specific error categories in Iranian EFL
narrative writing and thus contribute to their grammatical
accuracy both in the short and long-run?
2. Does the direct CF decrease the number of grammatical errors of
both total and specific error categories in Iranian EFL narrative
writing and thus contribute to their grammatical accuracy both in
the short and long-run?
3. Does CF provision encourage the learners to avoid the treated
grammatical errors?
Method:
Participants:
Number & gender:41 students (13 males, 18
females) majoring in English literature.
Age average:18.7
The writing classes met for one and half
hour once a week over 16 sessions.
They were randomly divided into three
groups with two different types of
provided feedback. The first group which
received the direct feedback was named
group A. Participants in this group
received teacher's correction.
Participants in second group which is named
group B, received indirect feedback. They
were just informed about the location of
the error by the errors being underlined.
The third group's (named group C)
participants were those who did not
receive any kinds of feedback.
Instruments:
Cloze tests
Narrative writing
tasks
Narrative Writing Topics:
Spotting and measuring the
errors:
Contributing to the fact that different
linguistic categories should not be treated
in the same way (Bitchener et al., 2005;
Ferris, 1995 a; Ferris et al, 2000; Liu,
2008; Lalande, 1982, Sheppard, 1992), the
present study dealt with six linguistic
categories to investigate the effects of
the two types of CF on them.
Linguistic errors are divided
into two groups:
Treatable
Untreatable
Verb tense and
form, Articles,
Noun endings,
sentence
fragment
Word choice,
Prepositions,
Unidiomatic
sentence
structure
To investigate the effects of the provided
feedback, the errors, considering their
linguistic category, were counted in each
individual writing task.
As the text length of the drafts varied,
the means of errors were calculated by
dividing number of errors by number of
words and multiplying by a standard which
was set at 1000.
.
Procedure:
Results & Discussion:
Three writings out of five ( the first, fourth, and
fifth) were analyzed and marked concerning the
six linguistic error categories to measure and
compare the means of errors before and after CF
provision. The fifth writing included the same
topic of the first writing task with a six week
interval to see the long term effect of corrective
feedback.
Statistical procedures used to analyze the data
included descriptive (percentages, means, and
standard deviations) and ANOVAs to analyze
error reduction across the three groups.
The control group did not show an
significant difference in error reduction
both in posttest 1 and 2. on the other
hand, the two treatment groups revealed
significant reduction but the changes were
not in the same pattern for each
grammatical category.
Verb errors
Verb errors
Noun ending errors
Noun ending errors
Wrong words
Wrong words
Sentence structure
errors
Sentence structure
error
Article errors
Article errors
Preposition error
Preposition error
Total errors
Total errors
Cloze tests analysis & results:
It has been argued that learners
tend to avoid the categories that
have been the subject of CF(e.g.
Sheppard, 1992; Truscott, 1996,
2004).
Based on the evidence of using
avoidance strategy it could be
concluded that the significant error
reduction was due to the use of
avoidance strategy and the provided
CF did not result in self correction
and accuracy in writing.
As a result cloze tests were
constructed based on students’
committed errors on the first writing
to compensate for the use of
avoidance strategy. In this way, they
could not ignore their errors any
more.
Example of the cloze test
Percentages were derived by dividing right answers (i.e. the
number of errors corrected) by the number of blanks of the cloze
test (number of committed errors of each individual on the
pretest)
The percentages revealed that
although the use of avoidance
strategy is inevitable, the effects of
CF can not be underestimated.
Conclusion:
Research objective 1: The short and long term effects of the
direct feedback on writing accuracy.
• Data analysis showed significant error reduction in the use of noun
endings, sentence structures, articles and prepositions just in the
short term. Considering the long term effect of the direct
feedback, there was not any significant error reduction after the
six week interval.
• The case was different for the verb tense and word choice. They
did not reach the statistical significance in either time periods. In
other words, the direct feedback did not have any significant
effects on these two error categories. For total errors, data
analysis showed significant error reduction in the short term and
like the separate categories, it did not last for the longer period.
Objective 2: The short and long term effects of the
indirect feedback on writing accuracy.
final analysis showed that the indirect feedback enabled students to use
the verb tense, articles, sentence structure, noun endings with
significantly greater accuracy in both short and long term periods than was
the case with their use of prepositions and word choice.
For preposition errors and wrong words, the indirect feedback showed
significant change only in the short term and the error means increased
again after the six week interval. However, for the total errors, the
reduction of error means was significant in both time periods. Considering
the short term effect of the two provided feedback, it was the indirect
feedback which outperformed the direct feedback group for the total
errors.
• It also found that both direct (only in the short
term period) and indirect feedback (in the short
and long term periods) facilitated improvement in
the more "treatable", rule-governed features
(Verb errors, Noun ending errors, Sentence
structure errors, Article errors) than in the less
"treatable" features ( Wrong words and
Preposition errors).There was not any significant
long term effect on prepositions and word choice
in the indirect group and even any short term
effect on the wrong words in the direct group.
Consequently, it is recommended that teachers
provide the learners with corrective feedbacks on
the more "treatable" types of linguistic errors on
a regular basis.
Objective 3: Use of avoidance strategy due to the provided CF
It was found that although CF provision may lead to avoidance strategy and
decrease in the length of the written work (Semke, 1984), no feedback is
not an option. Grammar correction (whether direct or indirect) generally
improve accuracy in the revised essays and it is more probable in new
writings and long term for the indirect corrective feedback.
Finally, the findings of this study have demonstrated that intermediate EFL
writers can improve the accuracy of their use of rule-governed linguistic
features due to the provision of indirect (both in the short and long term)
and direct feedback (just in the short term).
Implications of the study
The students' output in writing cannot improve if they do not receive
teacher's systematic comments to change and remove their
errors. It shows that what they have produced is incorrect and
thus helps them to 'notice the gap between their own deviant
productions and grammatically correct productions (Ellis, 1998,
p.52). This is what grammar correction hopes to achieve. Thus,
searching for the most efficient method of providing feedback is
one of the crucial factors in helping students improve in writing
skill.
Further research:
Further research would need to be undertaken to
see if this finding is also true for other linguistic
forms where rules of usage are more complex and
idiosyncratic than they are for the use of the
verb tense, article, noun ending and sentence
structure. It should also be mentioned that there
are a variety of strategies for providing CF.
Other studies could be undertaken to see the
impacts of them on other different error
categories.
Thank you for your
attention.