TEACHING
WRITING: STUDENT RESPONSE TO TEACHERS’ WRITTEN COMMENTS
ABSTRACT
This
paper discusses the effects of teacher comments on students’ revisions when
students revise their compositions. The
study involved three writing teachers and three students. The students differed in their language
ability, ranging from advanced to low level of proficiency. Teacher comments on form enabled the
advanced and intermediate students to correct their language errors
successfully. However, for comments on
content, short teacher comments written in-between sentences were of little
help to all three students. Comments in
the form of long and complete sentences written in-between paragraphs were
found to be useful by all three students when revising. The findings suggest that a teacher’s
responding style could influence a student’s revision style.
When
teaching writing, the most important thing that we do for our students is
responding to their written work. In
the absence of a face-to-face verbal writing conference between teacher and
student, written response is the only way in which the teacher can respond to
the individual writing needs of the student. Teacher comments are therefore, essential,
if not indispensable to a student revising and rewriting his or her
composition. Given that teacher
response is such an important aspect in the teaching of writing, I find it
somewhat surprising when most writing teachers I speak to on the matter report
that when they were trainees, they hardly received any instruction on how to
respond effectively to student writing.
While most teacher trainers may not
consider teacher response as an essential component in the training module of
teachers, researchers have long recognised its role as being an important one
in the teaching of writing. There have
been numerous studies conducted in this area.
Two early notable articles written by
Sommers (1982) Zamel (1985) ,
were based on research that studied and attempted to describe teacher comments.
Sommers found that teachers of writing were preoccupied with surface level
errors (errors of form) in students’ written work when giving written
feedback. Sommers is of the opinion that the effect of
such comments is that they encouraged the students to view their compositions
as fixed pieces, which consequently gave them no inherent reason to revise the
structure and meaning of texts. To make
matters worse, comments on content,
already few in number, usually take the form of vague and abstract descriptions
that students find difficult to interpret.
A similar observation was made by Zamel (1985).
This preoccupation with language
accuracy in student essays has long been a trait of teachers who taught English
as a second language (ESL or L2 teachers).
Therefore, it is not surprising that writing achievement is often
defined as mastery of the surface level skills required for accurately written
papers. Lau (1990) reported that most
ESL teachers responding to the written work of high school students, focused on
form (grammar) and paid very little attention to content. Keh (1990) stated that one reason why
teachers tend to focus on form more than content is that, endless hours of red
pen correction is often equated to hard work and has great face-validity to
teachers and headmasters. It remains to
be seen of course, whether or not these corrections are of any help to the
student writers when they revise their compositions. This tradition continues to persist in spite of various studies
reporting that error correction has little effect on the overall writing
quality of students. For example, Robb et al. (1986) investigated the
hypothesis that more salient or more detailed feedback on errors would have a
significant effect in improving the overall writing quality of student
writers. The results of the study
showed that detailed feedback on form is not really worth the teacher’s time
and effort. In fact, many earlier
researchers like Searle & Dillon (1980); Lamberg (1980); and Knoblauch
& Brannon (1982) had voiced similar concerns. Kepner, (1991), explains this trait among teachers by saying that
teachers, especially L2 teachers, fear the fossilisation of errors, and thus
feel morally obligated to correct all mistakes in students’ written work. Kepner calls for feedback that would attend
more to content.
Besides calling on teachers to focus
more on content, another concern among researchers is that, teacher response
should be more facilitative and less directive in nature. Dunn
et al., (1989) and Hyland
(1990) contend that teachers should not just “tell” students what to do, but
instead provide a platform from which students themselves can reassess and
redraft their work. Hyland calls for
detailed and informative feedback on content, which is deemed more helpful than
short questioning remarks. Such long
comments, almost conversational in nature, would respond more to the students
themselves and not just to their writing.
Mahili (1994) expressed her view
that a teacher’s primary role should be that of a guide, shares this opinion.
This paper, which is based on a study conducted in a centre for
pre-university studies in Malaysia, seeks to address the concerns stated
above. Three ESL teachers and three students
(of different proficiency levels) were involved in the study. Specifically, I wanted to find out:
1. If these teachers paid more attention to
form than to content,
2. If they responded differently to students of
different levels of language ability, i.e.by giving more salient feedback to
the weak student and less salient feedback to the good student,
3. What their respective responding styles
were, with style referring to how clearly
the teachers phrased their comments and
where they placed their comments on
the students’ essays.
4. Which types of teacher response were
effective, i.e. able to aid the students when they
revised their essays and which types were not.
Design
Three ESL teachers, Paula, Farah, and Jenny (not their
real names) agreed to participate in the study. All three were involved in teaching a writing course at the
centre. Paula was the most experienced
with ten years of teaching experience. Jenny had been teaching for almost four
years while Farah had taught for only two years. Farah was also the only
writing teacher who did not have any formal training in the teaching of English
as a second language (TESL).
With the teachers’ help, I was able to
select three students whose language abilities ranged from the high intermediate
level to the beginner level (based on scores from an English placement test,
grades of ongoing assessments, and the ability of the students to verbalise
their thoughts). For ease of reference,
I shall refer to them as the Advanced Learner, the Intermediate Learner, and
the Beginner.
I did not prescribe any topics for the
students to write on. Instead, the
three essays used in the study were part of a number of ongoing writing
tasks. This suited my intentions of
keeping the tasks in the study as real as possible. All three essays were in the expository mode. The length of the
essays ranged from 300 to 350 words.
Once the students had written the first drafts, they were then
distributed to the three teachers for feedback.
Each teacher responded to three essays
of the same title. This means that
Paula responded to the first essay, written by all three students, while Farah
responded to the second essay and Jenny to the third. I decided on such a distribution as I wanted to find out how the
teachers would differ in their response to students of differing language
ability. Besides, getting each teacher
to respond to all three essay titles would have meant each teacher would have
had to respond to all the essays written by the three students. Consequently, each of the three students
would have had to revise each essay three times - once for each teacher! Needless to say, both the teachers and the
students were not agreeable to this idea.
The teachers were required to verbalise
their thoughts (verbal protocol) as they responded to the essays. The essays, with the teacher comments, were then
returned to the students for revision.
The students were also required to verbalise their thoughts as they
revised and rewrote the essays. The student protocol transcripts proved
valuable when determining which teacher’s comments were helpful and which were
not. Four sets of data were obtained from study. These were:
1 Essays
written by the three students (each student wrote three essays),
2 Protocol
(think aloud transcripts of teachers responding to the students’ essays,
3 Rewrites
of all the essays (each student revised and rewrote all three essays using
teaching feedback, and
4 Protocol
transcripts of the students revising and rewriting their essays.
I divided my analysis of the data into two parts. For the first part, I looked at the
teachers’ responses to form (grammar, vocabulary and spellings). I then studied how the teacher’s comments on
form affected the students’ revision on form.
In the second part, I studied the teachers’ comments on content
(development and organisation) and how their comments affected the student’s
revisions on content. Whenever the
students’ were able to make corrections of error and textual revisions by using
teacher comments, I recorded them as instances when the teacher response had
been successful or effective.
Going through the students’ essays and transcripts, I
could identify several types of students’ responses. There were three types of response with regard to form and four
for content. The three types of
responses that I noted as the students attended to form were: ’
1. Teacher comments are understood and the
suggested changes to the text are acted
upon by the student when writing the second draft
2. Suggestions are ignored and changes are
not made to the text in the second draft
though teacher comments are
understood,
3. Teacher comments are not understood by
student, therefore no change is made to the text in the subsequent draft, and
4. Teacher comment are not understood by
student, but changes are made anyway to the text in the second draft.
A) Teacher
comments on form
When responding to form in students’ essays, teachers
normally make one of three types of corrections.
1. Firstly, they might indicate only the
location of an error in the students’ essays.
2. Secondly, they may choose to indicate both
the location of the error and the type of error that the student has made (for
example, by writing “tense” to indicate that
the wrong tense had been used, or “SVA” to indicate a subject verb agreement error).
3. The teacher response gets even more salient
in the third type of response, where they opt to indicate not only the location
and type of an error, but also provide a model of the correct version.
All three teachers in this study used the second
method, that is, they indicated the location of an error and what type of error
it was. The following is a sample of a
teacher’s response to form (teacher comments in italics):
Example:
SVA
‘……..every
intention have both effects………’ write in full
‘……..because
it’s too many…..’ sp.
‘……. Perfomance……’
Paula and Farah marked a higher percentage of errors
than Jenny did. Surprisingly, it was
the Intermediate Learner and not the Beginner who received the highest number
of corrections from Paula and Farah.
Since the teachers’ protocol transcripts did not reveal the reasons for
this, I could only make a calculated guess.
That is, since both teachers had also made numerous comments on the
content of the Beginner’s essays, the fewer comments on form could be seen as
reluctance by the teachers to inundate the student’s essay with their
remarks. When interviewed later, the
teachers agreed that their many comments on the content of the Beginner’s
essays could have influenced them to reduce the number of corrections of errors
in the student’s essays.
Jenny,
however, showed a marked difference from Paula and Farah in terms of the amount
of error correction. Though she did
correct a higher number of errors for the Beginner, proportionately, the number
was similar to the amount of corrections done for the other two students (she
corrected approximately 28% of the total number of errors in each essay she
responded to for all three students).
This indicates that Jenny had a predetermined idea of how much feedback
she was going to give on form. This is
a necessary decision if we are trying to focus the student’s attention on
content.
Of the three teachers, Farah corrected
the most number of errors. She was also
the only teacher who did not have formal training in TESL, as she was a
Management degree holder. Though this
training variable warrants further investigation, it is an indication that
training could be a factor that influences teacher attitudes with regard to
responding to students’ essays. In this
case, Farah’s preoccupation with error correction seems to suggest that she was
less aware of the overall objectives of the writing course in the
pre-university programme, which emphasises
the developmental aspects of essay writing.
B) On content
There was no significant difference in the number of
comments made on content among the three teachers. All three teachers wrote more comments on content in the
Beginner’s essays than they did in the other two students’ essays.
Paula and Farah again shared
similarities in their responses to content.
They had an identical style of responding. As they read the student essays, the teachers progressed from
sentence to sentence and stopped to write comments whenever they felt it was
necessary (this is indicated in the protocol transcripts). Most of their comments were in the form of
short and vague phrases. This trait has
been the predominant method of responding to students’ essays, even in L1 classrooms
– an observation made in several studies (including in Sommer 1982; and Zamel,
1985). An example of such a method of
commenting is as follows (Paula’s comments in italics as found in the Advanced
Learner’s essay, which was titled “The Causes of Schizophrenia”):
be specific stress? only facts?
“When the brain is always
suffering overloading of facts then it is justified that the situation
will worsen and stress will catch up.”-
weak
concluding sent. – not
sensitive
As you can see in the example, Paula has pointed out
that the student’s concluding sentence is weak and she explains that it is so
because the sentence is not sensitive, but she fails to explain what it is not
sensitive to.
Though writing in-between sentences
focuses the student’s attention on the exact part of the essay that needs revision,
teacher comments that require the student to make changes to a particular
sentence are sometimes at odds with error corrections which have also to be
done in-between sentences. There is
little point in correcting errors in a sentence that the teacher wants the
student to rephrase as the structures containing the errors will probably
disappear in the rephrased sentence.
On the other hand, Jenny’s protocol
transcripts revealed that she progressed from paragraph to paragraph as she
responded to the students’ essays and only occasionally stopped at the sentence
level to correct errors. She read whole
paragraphs before writing her comments.
Her comments were mostly in the form of near complete sentences, written
in the margins or in-between paragraphs, not in-between sentences.
Some examples of her comments (as given in her response to the
third essay written by all three students, titled “The Causes and Effects of
Smoking” are given below (in italics):
• In a response given to a particular paragraph, Jenny
writes,
‘You seem to have missed the link to
the title –- smoking’
• Comment given by Jenny to a transitional paragraph
(leadingonfromcausesofsmoking to the effects of smoking),
‘Transition
should list the effects as well as not
be too general’
• Comment given by Jenny to a new paragraph,
‘
Avoid Q at the beginning of paragraphs’
Jenny, after reading the essays, also deliberated on
each of them at length and wrote some general comments below the concluding
paragraphs. Jenny’s style of responding
generated more holistic and meaningful comments on content. (This style of
commenting approximates the type called for by researchers like Dunn, et.al.,1989; Hyland, 1990; Mahili,
1994).
A) On form
The
correction of error by indicating the location of the error only, or by
indicating location and type of error, as practised by the teachers, seems to
be adequate for the Advanced Learner and the Intermediate Learner, but not for
the Beginner. The Advanced Learner and
the Intermediate Learner were able to correct most of the errors picked out by
the teachers. The Beginner was only
able to correct 47.8% of the errors marked by the teacher although many of
these corrections were for spelling errors.
Though the Beginner appears to require more salient feedback on form, it
may not be a good idea for the teacher to provide a model of the correct
structure for each corrected error as this will place the responsibility of
correction solely on the teacher.
B) On content
As for
feedback on content, short phrases were generally found to be not helpful. At the most, they helped students to rephrase sentences, which
means revisions were mostly done at the sentence level and not at the text level. Of the three students, the Advanced Learner
was able to make the most number of revisions.
She was able to utilise 16 out of the total number of 18 comments
(88.8%) she received for all three essays.
In contrast, the Intermediate Learner only utilised 16 out of the 29
comments (55.2%) that she received for all three essays in her revisions. Interestingly, she chose to ignore 8
comments. The following is an extract
from the Intermediate Learner’s think aloud protocol transcript (transcription
of the recording made when she verbalised her thoughts as she revised and
rewrote her first essay, “The Causes of Schizophrenia”),
“…Tremendous stress could
also lead to schizophrenia. Paragraph
is too short? (referring to a teacher
comment) … No, can … it doesn’t matter … it doesn’t matter if my paragraph
is short or not … I don’t care … no, I mean … there’s nothing wrong if my
paragraph is short … as long as it is more than one sentence … I have few
supporting details … but I think this short paragraph is packed with …things …
that the reader can understand … I don’t care …I want it to remain a paragraph
… whether it is short or long.”
The above
example demonstrates a point made by Sommers (1982) that too many comments
might result in the student ignoring them.
This student also indicated that she could not understand five (17.2%)
of the teacher’s comments. The
following extract shows the Intermediate Learner struggling with Paula’s
comments on the same essay:
“Person who takes drugs such
as heroin, speed and marijuana can inherently be predisposed to the disease …
meaning … what is this? (referring to a
teacher comment) … the handwriting is terrible … I cannot see properly … I
know there is something wrong with the sentence … the lecturer doesn’t like …
okay, I’ll change it …but for sure, the way she writes … it’s not clear … I
cannot detect my … the correction she wants me to make.
While the
problem here appears to be the teacher’s illegible handwriting, The Intermediate Learner
would have found more detailed feedback helpful.
The Beginner, who received a total of 41 comments on content, responded
to 27 of them. However, his revisions
were only meaningful and the essay content only showed improvement when the
feedback he was responding to was clear and detailed. He did not understand 8 of the comments and chose to ignore the
rest.
From the analysis of student revisions, I found that Jenny’s long and
clear comments were most helpful for all three students. In fact, it was only her comments that
prompted the Beginner to make some meaningful revisions when rewriting. This supports Hyland’s (1990) claim that
detailed and informative feedback on content can help students to reassess and
redraft their work.
As for the placement (location) of teacher comments in the students’
essays, comments in-between sentences (as evident in the feedback given by
Paula and Farah) helped the students to identify the parts that needed
improvement. However, the sentential
revisions did little to
improve the overall quality of
the essays. Jenny’s detailed comments
in-between paragraphs enabled the
students to rethink paragraphs, though the comments did not really prompt them
to make major changes. Similarly,
Jenny’s long comments at the end of each essay did prompt all three students to
think about the overall effectiveness of their respective essays (as evident in
the students’ protocol transcripts), but again, the comments did not seem to
prompt actual revisions, except in the case of the Advanced Learner, who made
some textual changes. This finding
suggests that there are other concerns
that the teacher of writing should be aware of when responding to students’
essays, namely, whether or not the student’s linguistic ability would permit
him/her to make the necessary changes, and his/her attitude towards teacher
response.
Conclusion
It was
difficult to determine if the teachers were paying more attention to form than
content by just counting the numbers of comments made for the respective
categories, except in the case of Jenny.
Jenny clearly showed more concern for content when responding to the
students’ essays. In the case of Paula
and Farah, it would be fair to say that they paid as much attention to form as
they did to content (clarity of comments notwithstanding). Considering that these are ESL teachers, the
findings appear promising as these teachers of writing seemed to consider
textual development to be just as important as accuracy in second language
writing.
The teachers did not make any clear
attempts to vary their responses when responding to students of different
language ability. While the difference
was not significant between the total number of comments made on the Advanced
Learner’s essays and those of the Intermediate Learner’s essays, all three
teachers did give a significantly higher number of comments on content when
responding to the Beginner’s essays.
The obvious reason for this is that the Beginner composed a poorly
developed essay and needed more revisions.
But it could also mean that the three teachers were aware that he
required more help than his peers did.
As for the effectiveness of the
teachers’ responses, when giving feedback on form, or what we generally call
error correction, indicating just the location of the error, or the location
and the type of error, appears to be effective for students of high and average
levels of proficiency, but not for weaker students. The findings suggest that students of low proficiency levels
require more detailed feedback, like having the model of the correct version or
perhaps even a grammar rule to explain the error. This is effective only if we want to help the student to edit his
or her errors. But doing this too often
could render more important feedback ineffective. If the teacher appears to focus more of her attention on error
correction, the student will be inclined to follow suit.
As far as feedback on form is concerned,
I would recommend that teachers offer their feedback sparingly. Picking out certain repetitive major errors
and giving detailed explanations (depending on the level of the student) would
probably be more effective than making random corrections like what the
teachers in the study did.
For the feedback on content, comments in
the form of short phrases are definitely not very effective. Though the Advanced Learner was able to
utilise most of the comments given on her essays when revising, the revisions
were mostly the rephrasing of sentences and not revisions centred on the development
of the content. Placing the comments
in-between sentences is also not advisable as the students in the study did not
consider such comments in relation to the whole paragraph. Comments written in-between paragraphs were
found to be more helpful. These prompted the students to consider a point in
relation to the whole paragraph. Too
many comments, repeatedly asking the student to elaborate or clarify, also do
not appear to be effective in prompting meaningful revisions. Students in the study responded by either
paraphrasing sentences or by completely ignoring the comments. Finally, retrospective comments written at
the end of the essay can be useful in aiding students to review the overall
effect of their essays. Teachers should
also pay attention to the nature and content of the comments, that is, what the
teachers actually say in their responses.
Ideally, when responding to content, teacher comments should prompt the
students to revise more at the textual level than at the sentential level (an
aspect that was largely missing in the present study).
After looking at the various aspects of
teacher feedback, on form and on content, effective or otherwise, I am of the
opinion that students, when revising, view their essays in a manner that is
similar to the way their teachers view them.
If teachers read essays and write comments at the word and sentence
levels, this “fragmented view” is transferred to the students through their
comments. Consequently, the students
revise in a corresponding manner. If
teachers want students to adopt a more global perspective when revising, they have to start providing more holistic
comments (or comments which enable students to rethink their essays in a larger
context). A point to note is that a
student’s ability to revise is also subject to factors like writing experience
and command of the language. In the
process approach writing class at least, I would recommend that teachers pay as
much attention to students’ revisions as they do to the written products.
References
Dunn, S., Florio-Ruane, S.
and Clark, C. 1989. The Teacher as Respondent to the High
School Writer. In Freedman, S. (ed.) The
Acquisition of Written Language: Response
and Revision: pp33 – 50, 2nd
edn Norwood: Ablex Publishing
Corp.
Hyland, K. 1990.
Providing productive feedback . ELT Journal, 44 (4): 279 – 285.
Keh, C. 1990.
Feedback in the writing Process.
ELT Journal, 44 (4):
294 – 304.
Kepner, C. 1991.
An experiment in the relationship of
types of written feedback to the
development of second
language writing skills. The Modern Language Journal, 75 (3):
305 – 312.
Lamberg, W. 1980.
Self-provided and peer-provided feedback. College Composition and Communication, 31 (1): 63 – 69.
Lau, L.Y. 1990.
A descriptive study of teacher responses in the English compositions
of
form four students in some
selected Malaysian schools. Unpublished
BA Thesis. Universiti Kebangsaan Malaysia.
Mahili, I. 1994.
Responding to student writing. English Teaching Forum, 24 – 27.
Robb, T., Ross, S. &
Shortreed, I. 1986. Salience of feedback on error and Its effect
on EFL writing quality.
TESOL Quarterly, 20 (1):
83 – 95.
Searle, D. & Dillon, D.
1980a. The message of marking: teacher
written responses to student writing at intermediate grade levels. Research
in the Teaching of English, 14 : 233 – 242.
Searle, D. & Dillon, D.
1980b. Responding to student
writing: What is said or how it is
said.
Language Arts, 57 (7): 773 – 781.
Sommers, N. 1982. Responding to student writing. College
Composition and Communication,
33 : 160 – 169.
Zamel, V. 1985.
Responding to student writing. TESOL Quarterly, 19 (1): 79:101.