Lợi ích từ việc giảng viên nhận xét tương tác vào bài viết tiếng Anh của sinh viên

84 KHOA HỌC NGOẠI NGỮ QUÂN SỰSố 08 - 7/2017
v NGHIÊN CỨU - TRAO ĐỔI
LỢI ÍCH TỪ VIỆC GIẢNG VIÊN NHẬN XÉT 
TƯƠNG TÁC VÀO BÀI VIẾT TIẾNG ANH 
CỦA SINH VIÊN
1. INTRODUCTION
Teachers’ responses to student writing has been 
acknowledged as central to teaching composition 
(Freedman, Greenleaf, & Sperling, 1987). In 
fact, since the early twentieth century, Carpenter 
et al (1913) considered the role of response or 
“criticism” to the teaching and learning of writing 
as “one of the most important in the whole problem 
of teaching English, upon which the value of the 
criticism success in teaching composition finally 
depends” (Carpenter, Baker, & Scott, 1913, p. 142). 
Responding to students’ writing is arguably 
a most widely adopted method; yet it is time 
consuming and “the least understood” (Sommers, 
1982, p. 170). The questions of how to write helpful 
comments, to what extent teacher written response 
is supportive to student revision, and whether 
student successful revision is the result of teacher 
comments, are never simple to answer. 
A growing body of research has attempted to 
answer these tricky questions. Teacher written 
response has been examined in both first language 
TÓM TẮT
Trong quá trình dạy và học viết tiếng Anh, giáo viên thường phản hồi trực tiếp vào bài viết của 
sinh viên, làm cơ sở để người học chỉnh sửa trước khi hoàn thiện bài viết. Việc này được coi là 
tốn thời gian, công sức của giáo viên, nhưng giới nghiên cứu vẫn đang tranh luận về hiệu quả của 
nó đối với chất lượng bài viết. Trong nghiên cứu này, chúng tôi phân tích tác động của phản hồi 
tương tác của giảng viên đối với chất lượng bài viết tiếng Anh của sinh viên Việt Nam học tiếng 
Anh như một ngôn ngữ thứ 2. Chúng tôi thu thập trên 30 bài viết về 15 chủ đề của 03 sinh viên 
đại học người Việt trong 24 tuần. Tác động của phản hồi tương tác được phân tích theo chuẩn 
của Ferris, chất lượng bài viết được phân tích định tính theo chuẩn Viết Phân tích của Hoa Kỳ, so 
sánh kết quả sử dụng phương pháp ANOVA (định lượng). Kết quả cho thấy, người học tiếp thu, 
sử dụng gần 70% góp ý nhận xét của giảng viên, và có cơ sở thống kê để nhận định chất lượng 
bài viết lần cuối cao hơn lần đầu, đặc biệt về nội dung, bố cục, văn phong (không cải thiện về sử 
dụng từ và ngữ pháp). Kết quả nghiên cứu giúp cải thiện quy trình dạy và học viết tiếng Anh trình 
độ đại học tại Việt Nam.
Từ khóa: nhận xét của giáo viên, phản hồi, phản hồi tương tác, viết tiếng Anh.
TRƯƠNG ANH TUẤN*; LANNIN AMY**; NGÔ QUÝ CHUNG***
*Trung tâm gìn giữ hòa bình Việt Nam - BQP, ✉ tuanpkc@yahoo.com
**Đại học Tổng hợp Missouri, Hoa Kỳ
***Học viện Khoa học Quân sự, ✉ cuaquychung@yahoo.com
85KHOA HỌC NGOẠI NGỮ QUÂN SỰSố 08 - 7/2017
NGHIÊN CỨU - TRAO ĐỔI v
(L1) and second language (L2) writing classes. 
Teacher response, as agreed upon by most teachers 
and researchers, has evolved into more than just 
written marginal or end comments. Responses may 
include all types of interaction with student drafts. 
They could be formal, informal, in written, or oral 
forms to a series of drafts, or to one polished final 
paper. Responses may be used in formal mainstream 
classrooms, or in an informal, casual interaction 
between teacher and student (Freedman et al., 1987).
Teacher response might be explicit, implicit, or 
a combination of both. A teacher might comment 
as explicitly as “I’m interested in your idea here,” 
“I like your voice in this paragraph,” or “I think 
this sentence needs a verb.” Teachers might also 
engage indirectly, such as “What do you think this 
paragraph lacks?” or “I’m lost here!” Reflective 
response might also be used, such as “I’m just 
curious to see what is happening here,” or “as a 
reader, I like to see more details in this scene.” 
In this study, we attempted to explore the effects 
of reflective response on student revision as defined 
by Anson (Anson, 1989). The study was a pilot 
study for a future research with greater sample. We 
examined 15 papers, including 30 drafts produced 
by three college students who studied English as 
a second language over a period of two academic 
semesters (24 weeks). These papers were written as 
an additional writing exercise, out of the students’ 
normal class time, and not for credit or grading. No 
pressure was placed on the students with regard to 
what they wrote, when they wrote, and where. By 
doing this, we intended to give more freedom to 
the students, and avoid imposing the concepts of 
teacherly “ideal text” on the students (Sommers, 
1982). The students would revise their drafts only 
because they wanted to do so, not because of 
meeting any requirements by the teacher for the 
purpose of grading. 
The effects of reflective response were analyzed 
using a rating scale developed by Ferris (1997). We 
assessed if the students’ subsequent revisions were 
the result of the teacher response, and if the changes 
in drafts improved the overall writing quality as 
evaluated using a version of the National Writing 
Project’s analytical writing continuum (NWP, 
2009). Improvement in a student’s paper was 
determined by two procedures: (a) holistic scoring 
of the first and final drafts on a six-point scale, and 
(b) analytical scoring centered on six traits: content, 
structure, stance, sentence fluency, word choice, 
and conventions. 
2. REVIEW OF RELATED LITERATURE
2.1. L1 response research and theory
Written teacher response has been a topic 
drawing concern from a large number of researchers 
and educators, resulting in a growing body of 
research in the field. As early as 1913, Walter 
Barnes wrote: 
I believe that children in the grades live, so 
far as the composition work is concerned, in an 
absolute monarchy, in which they are the subjects, 
the teacher the king (more often, the queen), and the 
red-ink pen the royal scepter...In our efforts to train 
our children, we ... earcher categorized written teacher response 
styles into three groups of dualistic, relativistic, 
and reflective. Dualistic responders tend to focus 
their attention on surface errors and mechanics. 
Teacher responders clearly prescribed what is right 
from what is wrong, and that students should make 
changes in their revision. “The tone of the responses 
implied that there were standards for correct and 
incorrect ways to complete the assignment, and that 
a teacher’s job was to act as a judge by applying the 
standards to the student’s writing,” or “[the tone] 
was highly authoritative and teacherly” (Anson, 
1989, pp. 344, 348). Grammatical issues seem to be 
the focus of dualistic comments. Dualistic response 
emphasized narrowly prescriptive comments 
(Straub & Lunsford, 1995). Dualistic response 
tends to focus on spelling out issues, not to offer 
options for revision. The following example is a 
typical dualistic response: 
There are some serious problems with this 
paper. For one thing it is far too short, and the ideas 
in it, if any, are at the moment barely articulated 
88 KHOA HỌC NGOẠI NGỮ QUÂN SỰSố 08 - 7/2017
v NGHIÊN CỨU - TRAO ĐỔI
one obvious reason why you did not write more, 
is that you have very serious deficiencies in your 
knowledge of the mechanics of writing. I am 
referring here to tense, spelling, punctuation, and 
sentence structure (Anson, 1989, p. 344).
The second type of responders, relativistic, 
commented almost nothing. They wrote minimum 
comments on the margins of student papers as well as 
in the summary statements. Relativistic responders 
seemed to avoid focusing on the student’s text, 
and to be “entirely unconcerned with giving the 
students anything more than a casual reaction
the text seems ‘owned’ by the writer” and teachers 
did not want to intrude into the text. Relativistic 
responder provides “no options for revision,” just 
“idiosyncratic response of a single reader” (Anson, 
1989, pp. 349-350).
The third approach examined was reflective 
response. Reflective responders tend to make 
suggestions and possibilities for future revision. 
This type of comment expresses concerns for 
student writers in “ideas, textual decisions, 
personal reactions.” Reflective responders acted 
as “representative readers” of student text, not 
authoritative teachers. Final choices of whether or 
not making any changes to the drafts will be decided 
by the students themselves. Reflective response also 
implies that the student writing was “in-process 
drafts” which serves as “tools for further learning.” 
Reflective responders often phase “maybe you could 
think about”, “what if you”, “and how about 
seeing if there’s a way to” The tone of reflective 
response tended to be collaborating, suggesting, 
guiding, and modeling. The reflective responder 
seems to be “rhetorically sitting next to the writer” 
(Anson, 1989, pp. 351, 353). Below is an example 
of a reflective commentary to the student’s writing:
Hi Bobby. The first thing that strikes me before 
I even read your story is that it’s very short I’m 
wondering if it’s short for a good reason, or it’s 
short because you just couldn’t think of things to 
say. It’s possible for a piece of writing that’s very 
short to be very good. Poetry is that way, certainly. 
On the other hand, the more you put in, the more 
chances are that your reader is going to be able to 
get into your story. Stories generally- and this essay 
is a story- are fairly well-detailed if you just keep 
it short and don’t put in many details then we never 
really get into your story at all (Anson, 1989, p. 351). 
Reflective responses tend to “place more 
responsibility on the writer  not just in the style or 
form of the response, but in its focus and content.” 
By challenging the students to rethink their essays, 
reflective response appeared to “challenge the 
students to rethink their ways of viewing the world” 
(Anson, 1989, p. 352).
2.4. L2 written feedback research
Research in L2 written feedback has been 
growing, with attention being paid to the 
effectiveness of teacher’s written comments to 
student writing and in the ways feedback is given 
(e.g., Ashwell, 2000; Bitchener, 2008; Bitchener 
& Ferris, 2012; Bruton, 2009a, 2009b, 2010; 
Chandler, 2003; Conrad & Goldstein, 1999; 
Fathman & Whalley, 1990; Ferris, 1995b, 1997, 
2001, 2003, 2004; Ferris, 2010; Ferris, Brown, Liu, 
& Stine, 2011; Ferris, Pezone, Tade, & Tinti, 1997; 
Guénette, 2007; Hartshorn et al., 2010; F. Hyland 
& Hyland, 2001; K. Hyland & Hyland, 2006; Leki, 
1990; Storch & Wigglesworth, 2010; Truscott, 
1996, 1999, 2007; Van Beuningen, De Jong, & 
Kuiken, 2012; Zamel, 1985). Earlier L2 written 
feedback research yielded similar findings to L1 
research. Teacher comments were reported to be 
vague and form-related. They focused on language 
errors rather than on global issues such as ideas and 
organization (Zamel, 1985).
Research in the 1990s tended to focus on what 
to respond to (either on form, content, or both), 
and reported mixed findings. Focus on form was 
believed to be helpful for student writing (Ashwell, 
2000; Chandler, 2003; Cohen & Cavalcanti, 1990; 
Fathman & Whalley, 1990; Ferris, 1997; Leki, 
1990). In an empirical study with 72 college 
89KHOA HỌC NGOẠI NGỮ QUÂN SỰSố 08 - 7/2017
NGHIÊN CỨU - TRAO ĐỔI v
students from mixed backgrounds, Fathman & 
Whalley (1990) reported that specific comments 
on grammatical errors have greater effect on the 
improvement of grammatical accuracy than general 
comments on content do. The researchers also 
noted that both grammar and content response 
might be provided either separately or at the same 
time “without overburdening the students” (p. 187). 
This claim was further supported by later studies 
(e.g. Ashwell, 2000; Ferris, 1997). Feedback on 
some selective patterns of errors was helpful to 
student writing (Ferris, 1995b). Chandler (2003) 
reported, for example, that error correction helped 
students gain greater accuracy than when they did 
not receive error feedback. Form-related comments 
(on grammatical errors) led to better grammatical 
accuracy than content-related feedback did (Fathman 
& Whalley, 1990; Ferris & Roberts, 2001). Error 
correction helped prevent error fossilization (i.e. a 
tendency to resist to change errors so that the errors 
become fixed) in L2 learners (Higgs & Clifford, 
1982; Lalande, 1982).
However, earlier studies in L2 written feedback 
also revealed that error correction was ineffective, 
even harmful to students’ fluency, and led to no 
improvement in long-term progress (Fazio, 2001; 
Kepner, 1991; Polio, Fleck, & Leder, 1998; Robb, 
Ross, & Shortreed, 1986; Semke, 1984; Sheppard, 
1992; Truscott, 1996). Truscott (1996), for 
example, claimed that error correction was harmful 
to student fluency and led to no improvement in 
long-term progress and that students might not gain 
anything from error feedback. Zamel (1985) and 
Lunsford and Connors (1993) reported that teacher 
feedback was often vague, form-related, and 
inaccurate. Truscott (1999) suggested that teachers 
should adopt a correction-free approach in teaching 
writing, and teachers should focus on extra writing 
practice rather than spending time handling errors.
In recent reviews, Ferris summarizes a number 
of issues in response research: i) teachers often rely 
on marginal or end of paper notes whose purpose is 
to request, suggest, give information, encourage, and 
provide positive feedback. A number of techniques 
have been utilized to respond: questioning, making 
statements and imperatives, recommending, etc., ii) 
teachers adjust their responses to types of writing 
task and student writing proficiency; and iii) some 
response styles tend to be more effective to revision 
than the others. Comments about information, 
grammar, or mechanics are more likely to lead to 
successful revision than comments about such issue 
as thinking or argumentation (Bitchener & Ferris, 
2012; Ferris, 2003).
One of the main concerns in L2 response 
scholarship is how to determine if teacher response 
affects student revision. Several taxonomies have 
been developed to trace revision changes. Faigley 
& Witte (1981) proposed a system that traces 
revision by classifying changes into surface changes 
(changes that do not result in new information) 
and text-based changes (changes that lead to new 
content or deletion of old content) (Faigley & Witte, 
1981). Storch (2010) and Ferris (2003) argues that 
this revision scheme tends to be misleading because 
i) students tend to make by far greater number of 
surface formal changes than text-based changes 
within a writing, and ii) the scheme does not deal 
with how such a change affects the general quality 
of the draft. 
Another procedure monitoring teacher response 
and student revision is proposed by Ferris (1997). 
This rating scale traced the students’ drafts and 
the teacher’s response to see how students utilized 
the comments in their revision. Students’ revision 
was coded as not revised, successful revision, and 
unsuccessful revision. These changes were also 
determined if they improve quality of the paper, 
have mixed effects, or have negative effect. This 
method “more directly addresses the influence of 
teacher feedback and its effects not only on the 
types of revisions students make but on whether 
those changes actually improve the quality of the 
students’ texts” (Ferris, 2003, p. 36). A number of 
studies have applied this analytic model in working 
with teacher’s comments and the effect on student’s 
90 KHOA HỌC NGOẠI NGỮ QUÂN SỰSố 08 - 7/2017
v NGHIÊN CỨU - TRAO ĐỔI
revision (i.e. Conrad & Goldstein, 1999; Ferris, 
1997, 2001; F. Hyland, 1998). 
Many researchers, such as Bitchener and Ferris 
(2012), Storch (2010), Guénette (2007); K. Hyland 
and Hyland (2006), and Ferris (2003, 2004), 
suggests that future studies in L2 response should 
consider student background and motivation level 
for L2 learning. These include the amount of time 
students commit to spend on writing (in-class and 
out-of-class). The relationship between students and 
teacher should also be noted. The types of writing 
student compose, the ways teacher constructs 
responses (linguistic, pragmatic, etc.) might also 
count in the relationship between response and 
revision. Whether or not teacher’s written feedback 
is harmful to student’s writing as Truscott (1996, 
1999) claims or whether teacher’s feedback is 
helpful to students’ immediate revision are also 
issues that merit further explorations.
There has been a debate about whether or 
not teacher’s written feedback is helpful to non-
native students of English (e.g., Chandler, 2003; 
Ferris, 1997, 1999, 2004, 2011; Ferris, 2006; A. 
Lunsford & Connors, 1993; Truscott, 1996, 1999, 
2007; Zamel, 1985). Among many types of written 
feedback, the current study only explores only one 
type, written reflective feedback, and to examine 
if written reflective response has any effect on 
ESL students’ revision. Given the fact that written 
feedback is still the most widely adopted method 
by writing teachers and is time consuming and 
yet appropriately examined, it is necessary to 
investigate whether or not teachers’ feedback make 
a difference to students’ writing progress. The study 
was designed to answer the two following research 
questions: 
1. To what extent does teacher’s written 
reflective response influence ESL learners’ revision 
process?
2. To what extent does ESL students’ revised 
draft improve after receiving teacher’s written 
reflective response? 
3. THE STUDY
3.1. The participants
The study was conducted on three female 
college students. They were Vietnamese first year 
students (mean age is 20). They were pursuing 
different degrees in different majors, at different 
universities. For the purpose of ethics, their names 
are coded as Queen, Pie and Tea. Queen was 
studying English and commerce in Singapore; Pie 
was following a business program at a university 
in Wellington, New Zealand; and Tea was studying 
finance in Russia.
The length of the participants’ experience 
with English varies. Queen has been learning 
English since she was at her secondary school in 
Vietnam (for about seven years) and she is now 
learning English in Singapore. Pie and Tea have 
acquired Russian as their second language. Pie 
learned Russian for six years before switching to 
English when she began her business program in 
Wellington in 2009. By the time data for the study 
was collected, Pie has been learning English in New 
Zealand for roughly a year. Tea, interestingly, still 
used Russian as a means for her accounting program 
since Russian was a language of instruction at her 
university. Tea, however, wanted to learn English 
since she was considering a Master’s degree in an 
English speaking university. By the date of the data 
collection process, Tea had been learning English 
for almost two years. To fully examine the effect 
of teachers’ responses (if any), it is appropriate to 
select the participants with diverse backgrounds of 
English learning.
All of the participants were former students 
at the universities in Vietnam where two of the 
researchers used to teach but were not current 
students at the time of this study. Following the 
university’s approval, an email was sent out to 
recruit the participants. These three students were 
the ones who agreed to join the study. They were 
female students who appeared to have a clear 
commitment and plan to their studies, which might 
give credit to their motivation in learning English. 
All of them were eager to participate in the study