Open Journal of Modern Linguistics
2012. Vol.2, No.1, 18-25
Published Online March 2012 in SciRes (http://www.SciRP.org/journal/ojml) http://dx.doi.org/10.4236/ojml.2012.21003
Copyright © 2012 SciRes.
18
Enhancing Students’ Fluency in Writing: Learning to Use
Transition Words
Yoshihito Sugita
Faculty of Nursing, Yamanashi Prefectural University, Kofu, Japan
Email: sugita@yamanashi-ken.ac.jp
Received January 6th, 2012; revised February 14th, 2012; accepted February 22nd, 2012
This study uses experimental and control group data to investigate whether learning to use transition
words results in enhancing students’ fluency in writing. Common sentence connectors, such as moreover,
however, thus, etc were chosen in order that students learn the use of transition words in text and improve
their writing fluency. 36 first-year university students were placed in an intermediate class: 18 control
group students and 18 experimental group students. Over a 12-week period, both groups received equal
amounts of writing assignments. During the first half of the period, both groups were given content and
form feedback, but the experimental group was given additional marginal comments on the use of sen-
tence connectors. After six weeks, both groups were given identical types of feedback and comments.
Fluency was measured by the number of words written and successful connections (SCs). These results
were analyzed to determine if there was a significant difference in fluency between the two groups. Find-
ings suggest that writing teachers should teach students the effectiveness of using transition words in EFL
writing classes, and this may in part help to improve students’ fluency.
Keywords: Writing Fluency; Feedback; Transition Words
Introduction
English education in Japan has traditionally emphasized teach-
ing sentence patterns, vocabulary and grammatical rules, and
thus has focused mostly on accuracy in teaching English. In the
field of writing instruction, a large survey conducted on 1,027
Japanese university students revealed that more than 90% of the
students practiced translation of Japanese into English in writ-
ing classrooms and more than 80% of the students had no ex-
periences in longer paragraph writing (Hirota et al., 1995).
Moreover, a survey conducted to examine 786 Japanese univer-
sity-level writing teachers’ view on the instruction resulted in
structure and expression practice (31.7%), translation from
Japanese to English (31.2%), free composition (17.6%), item
replacement (6.6%), and others (JACET, 1993). These survey
results indicate that most Japanese teachers of English have
considered writing instruction to be characterized by accu-
racy-centered activities in order to reinforce the teaching of
grammatical structures or vocabulary.
As globalization proceeds, it is quite likely that Japanese stu-
dents will have to learn and use English for the purpose of in-
ternational communication. In accordance with such social
needs, the importance of developing students’ practical com-
munication abilities has been emphasized, and many research-
ers and educators in Japan have proposed a paradigm shift from
accuracy-oriented to fluency-oriented writing instruction (e.g.,
Iseno, 1991; Kurihara, 1994; Oi, 2004). As a result of their
efforts, this evolution of writing pedagogy for Japanese learners
of English is clearly reflected in the government guidelines for
foreign language teaching published by the Ministry of Educa-
tion, Culture, Sports, Science and Technology (MEXT). The
objective of “Writing” is a good example to illustrate the flu-
ency-oriented instruction:
To further develop students abilities to write down infor-
mation, ideas, etc. in English in accordance with the situation
and the purpose, and to foster a positive attitude toward com-
municating by utilizing these abilities
(MEXT, 1999)
However, considering the results of 2003 and those of 1995
given above, the same tendency can be seen in the experiences
of word-by-word translation (83.9% - 77.1%) and longer para-
graph writing (31.9% - 34.6%). These figures may reflect the
fact that the shift in emphasis from accuracy to fluency in writ-
ing has not strongly reinforced fluency-oriented instruction in
writing classrooms (Hayashi et al., 2003). Consequently, re-
searchers and educators in Japan need to consider again the
practical demands of the learning situation and contribute to
this pedagogic transformation by facilitating improvement in
students’ writing fluency.
Conceptual Frameworks
The primary purpose of this research is to investigate whether
learning to use transition words results in enhancing students’
fluency in writing. The conceptual framework is based on the
work of Halliday and Hasan (1976), and Halliday (1994). The
set of Halliday and Hasan’s cohesive relations is comprised of
reference, substitutions, ellipsis, conjunctions and lexical rela-
tions. Reference is the relation between a linguistic expression
and its pronoun. Substitution is marked at representing a pre-
ceding expression by a pro-form like one, do or so. Ellipsis is
the omission of part of a sentence whose meaning will be re-
trievable from the preceding text. A conjunction is a cohesive
device which makes logical-semantic relations between linguis-
tic expressions and links paragraphs. Lexical cohesion is the
writer’s choice of particular lexical items, which are related to
Y. SUGITA
the relevant preceding expressions.
In this classification, the use of conjunctions is an effective
way for L2 writers to build connections between ideas in text.
White and McGovern (1994) call them “cohesive markers” (p.
67), and Jordan (1990) describes them as “connectives” (p.
121). These sources point out that such conjunctions lend a
framework to meaning, helping the reader to follow the devel-
opment of a text. According to Parrot (2004), learning to use
conjunctions also helps learners to show how the points they
relate to each other and to the whole text. He defines transitions
as words which indicate logical relationships and sequences as
“textual discourse markers” (p. 302), and explains the main
functions as numbering and ordering points, adding something,
linking similar things together, introducing something that con-
trasts with expectations, generalizing, exemplifying and nar-
rowing down, re-stating, and rounding off (see Textual dis-
course markers in Appendix 1). In teaching ESL or EFL stu-
dents’ writing, Fukushima and Sato (1989) investigated the
effectiveness of teaching transition words in expository dis-
course writing. Their research revealed that the teaching of
transition words helps average student writers “construct proper
organization in which their ideas are logically sequenced and
developed, and this entails writing of better quality” (p. 35).
Spycher (2007) also found that a student at lower levels of Eng-
lish proficiency could constitute logical relationships by using
newly learned conjunctions (although, such as) and guide the
reader to logical understanding.
However, there seems to be surprisingly few attempts to
examine the effectiveness of teaching transition words in terms
of writing fluency. Some potentially challenging areas might be
identified from the previous research on the use of connectives
as an effective device for improving oral fluency. For example,
Ejzenberg (1992) found that connectives were one of the facili-
tating factors to promote nonnative speakers’ fluency. Yashima
et al. (1995) also found that the high school students’ fluency
was facilitated by the use of connectives such as “and”, “but”,
“or”, “so”, “because”, “then”, etc. Kawaguchi and Kamimoto
(2000) observed distinctive features of oral production by fluent
and nonfluent EFL learners, and the results indicated that the
fluent speakers employed significantly more cases of coordina-
tion (“and”, “so”, “but”, etc.) and subordination (“because,”
“when”, “unless”, “that”, etc.) than the nonfluent speakers.
They referred to the fact that “connectives were used to com-
bine sentences successively and to develop the idea of the pre-
vious sentence” (p. 29). It follows from the previous studies
that the use of connectives facilitates speeches and that learning
to use different types of transition words increases fluency in
spoken discourse.
Although results from the studies mentioned above cannot be
used for making predictions about what results will show in the
present study, many language specialists and theorists point out
similarities between speaking and writing, and emphasize the
close relationship of oral and written language (Cook, 1989;
Halliday, 1989, Burns & Seidlhofer, 2002). Therefore, it is
presumed that the instruction of the appropriate use of transi-
tion words helps students improve fluency in written discourse
as well as in spoken discourse. Furthermore, since written dis-
course is not as time bound as spoken one, which has specific
time lag between production and reception (Brown & Yule,
1983; Nunan, 1993), it seems reasonable to suppose that stu-
dents have more time to think of connections between sen-
tences and to organize ideas logically while they are writing,
rather than while speaking. Accordingly, learning to use transi-
tion words could be more effective at facilitating students’ flu-
ency in written language than in oral language, so that the use
of connectives may enable students to write more about the
relevant information and to enhance their fluency in writing.
This paper is intended as an investigation of how students
learn the use of transition words from the teachers’ feedback of
connectives and improve their fluency in writing. Common
sentence connectors, such as moreover, however, thus, etc were
chosen in order that students learn how to use connectives ef-
fectively and try to use them in writing assignments over a 12-
week period. The results of the period were analyzed in order to
determine if there was a significant effect on fluency improve-
ment between the experimental and control groups. Thus, this
study uses experimental and control group data to investigate
whether learning to use transition words results in enhancing
students’ fluency in writing.
The Study
Research Questions
In order to examine how students learn the use of transition
words from the teachers’ feedback of connectives and improve
their writing fluency, the following questions were addressed in
this study:
1) How did the experimental group respond to intentional
feedback that explicitly focused their learning to use transition
words?
2) To what extent did the students’ learning to use transition
words enhance fluency in new pieces of writing?
Participants
Thirty-six first-year nursing students at a Japanese university
were invited to participate in the study. From April to July 2010,
they took one 90-minute General English class every week,
with the aim of developing students’ practical communication
skills in listening, speaking, reading and writing. Each class
was comprised of pre-reading activities, reading comprehension
and vocabulary learning. As a post-reading activity, writing
assignments were given to all the students and speech presenta-
tion based on the written text was made by three students at the
beginning of the following class. No specific transition-word
instruction was conducted in class during the 12-week period.
To be placed in an intermediate class, they achieved scores
between 140 and 150 on the TOEIC Bridge Test. According to
the test scores, the students were randomly divided into two
groups: the control and experimental groups. Both groups re-
ceived the same amount of writing assignments. Since this
study was conducted sequentially over 12 weeks, the data of
those who did not submit any assignments were excluded from
analysis. Thus, the final number of the participants was 30 (2
males and 28 females). Table 1 indicates more information
about students’ overall English proficiency and writing fluency
levels of each group. There was no significant difference be-
tween the two groups.
Design and Measures
The General English class in which this study was conducted
provided an appropriate setting to examine the research ques-
tions because the 12 written homework assignments were in-
Copyright © 2012 SciRes. 19
Y. SUGITA
tended to have identical writing structures: a topic sentence,
two or three reasons representing students’ opinion, and a con-
cluding sentence. Although each topic of the reading passages
in the textbook was different, the topic sentence was given in a
clear general statement, such as “I agree/disagree with Profes-
sor X’s proposal”. The concluding sentence was also provided
beforehand as in the form of “For these reasons, I agree/dis-
agree with Professor X’s proposal.” Thus, the students could
concentrate on thinking and writing about reasons to support
their opinion or choice, and the reasons they gave were sup-
posed to coherently relate to the main statement.
According to the feedback schedule shown in Table 2, dur-
ing Period 1, the control group was given the content and form
feedback, and the experimental group was given additional
marginal comments on the use of sentence connectors as well
as the content and form feedback. During Period 2, both groups
were given identical types of feedback and comments.
Many previous studies measured fluency by number of words
written (e.g., Carlson, Bridgeman, Camp & Waanders, 1985;
Fathman & Whalley, 1996; Reid & Findlay, 1986; Reid, 1996).
In this present study, fluency is defined as length of body para-
graphs coherently related to the main statement. Therefore, the
experimental and control groups were compared in terms of the
changes in the amount of body paragraphs over a 12-week pe-
riod. Another measure of fluency has been the number of con-
nectives used in a written text. As pointed out in “Conceptual
frameworks,” the use of connectives may enable students to
write more about the relevant information and thus the numbers
of connectives reveal the enhancement of their fluency in writ-
ing.
Although we know that the use of sentence connectors cre-
Table 1.
Student information.
TOEIC Bridge Writing fluency
Group Experiment Control Experiment Control
N 15 15 15 15
Mean 143.6 142.5 44.7 41.3
SD 2.75 2.20 18.2 16.2
Min. 140 140 17 19
Max. 148 146 66 67
Note: Writing fluency shown in the table is the number of words written in As-
signment 1.
Table2.
Assignment-feedback schedule.
Classes Writing Assignments Given Feedback
Week 1 Assignment 1 Feedback 1
Week 2 Assignment 2 Feedback 2
Week 3 Assignment 3 Feedback 3
Week 4 Assignment 4 Feedback 4
Week 5 Assignment 5 Feedback 5
Week 6 Assignment 6 Feedback 6
Period 1
Week 7 Assignment 7 Feedback 7
Week 8 Assignment 8 Feedback 8
Week 9 Assignment 9 Feedback 9
Week 10 Assignment 10 Feedback 10
Week 11 Assignment 11 Feedback 11
Period 2
Week 12 Assignment 12 Feedback 12
ates cohesion in a text, their overuse or misuse often becomes
obtrusive (Cooley & Lewkowicz, 2003). Aoki (1991) analyzed
the coherence of English compositions of Japanese university
students. He described the coding system, which is comprised
of the following six schemes: local connections, multi connec-
tions, local-remote connections, remote connections, unsuc-
cessful connections, and topic change (p. 104).
In order to examine the use of connectives with the concept
of coherence, one of Aoki’s schemes, “Unsuccessful connec-
tions (UC)” was chosen, which is defined as “a sentence which
has no semantic relation with any other sentence and a topic, or
causes inconsistency” (p. 104). A new scheme, “Successful
connections (SC)” was also developed for this research. The
definition is “a sentence which is semantically connected with a
previous sentence and/or a topic.” When sentence connectors
are found in a text, their use is classified into SC or UC (see
Examples in Appendix 2). Thus, connectives classified into SC
are only counted as the number of connectives which affect
writing fluency.
Procedures
The teacher-researcher evaluated all written texts of each as-
signment, giving feedback and/or comments according to the
experimental or control group conditions. I tried to provide at
least one comment on the use of connectives to each student of
the experimental group during Period 1 and to both groups
during Period 2. Then, in the new pieces of writing I counted
the number of words written, and calculated frequency on suc-
cessful connections (the actual number of SCs) as evidence for
the benefits of intentional comments.
Group type (C: feedback only, E: feedback and comments)
was treated as a between-participants factor, and time at five
levels (Period 1: Weeks 3 - 7, Period 2: Weeks 8 - 12) was
treated as a within-participants factor. In order to investigate
interactions between factors as well as the effects of individual
factors, a two-way ANOVA was chosen as an appropriate sta-
tistical procedure. A mixed between-within participants analy-
sis was used to analyze the fluency at two levels (number of
words, and number of SCs). If a test showed statistical signifi-
cance, a Scheffé test to evaluate differences among specific
means was also conducted. Additionally, in order to examine to
what extent each factor can affect fluency, the effect size was
obtained. Among some index of effect sizes (Mizumoto & Ta-
keuchi, 2008), the present study uses eta squared (η2).
Results
The 1st Period (Weeks 3 - 7)
This section presents the results of Period 1, investigating the
extent to which marginal comments on the use of transitional
words helped students improve the fluency of their writing. The
mean and standard deviation (SD) for each group are shown in
Table 3.
As an example of the information revealed in this table, it
can be seen that as for the number of words written in Assign-
ment 3 (Week 3), participants who received feedback and
comments (Experimental group) had a mean score of 51.1 and a
standard deviation of 16.3. As for the number of SCs in As-
signment 4 (Week 4), participants who received feedback only
(Control group) had a mean score of 2.3, while the mean score
of Experimental group was 3.4. Figure 1 illustrates the result
Copyright © 2012 SciRes.
20
Y. SUGITA
for the frequency of using connectives between the two groups.
The data during Period 1 (Weeks 3 - 7) were then used to
find out whether there was an effect of interactions between
factors as well as effects of individual factors. Table 4 presents
the results of the two-way ANOVA for SCs.
Although the interaction effect of feedback type and time
was not significant [F(4, 168) = 1.34, p = .25, η2 = .024], the
average successful connections (SCs) differed according to the
type of feedback provided: F(1, 28) = 19.32, p < .01. In addi-
tion, the effect size indicated that the type of feedback had a
large effect (η2 = .160, large: .14 < η2) on the frequency of us-
ing connectives. The effect of time was also significant: F(1,
112) = 3.68, p < .01, η2 = .067 (medium: .06 < η2 < .14). In
Figure 1, we can see that Experimental group began to use
more transition words after the third week, and the Scheffé test
showed a significant difference across the four writing assign-
ments: Week 4 = 38.16, Week 5 = 53.81, Week 6 = 25.19, and
Week 7 = 53.81; p < .01. Accordingly, the results confirmed
that the comments on the use of transition words had a signifi-
Table 3.
Descriptive statistics of group performance.
Word written Successful connections
Group Experimental Control Experimental Control
Time Mean SD Mean SDMean SD Mean SD
Week 1 44.7 18.2 41.3 16.21.8 1.2 1.7 1.1
Week 2 49.3 15.4 49.4 16.62.2 1.3 2.0 1.1
Week 3 51.1 16.3 49.3 22.82.5 1.2 2.3 1.4
Week 4 59.5 18.5 54.6 15.63.4 0.7 2.3 0.9
Week 5 56.3 18.1 50.9 14.24.0 1.3 2.7 1.0
Week 6 57.9 18.6 50.2 17.33.3 0.8 2.5 0.8
Week 7 61.3 15.2 52.8 11.03.6 0.9 2.3 1.2
Figure 1.
Frequency of SCs by groups.
Table 4.
Results of two-way ANOVA for SCs.
SS df MS F
Group (A) 33.61 1 33.613 19.32**
S 48.69 28 1.73
Time (B) 14.20 4 3.55 3.69**
A × B 5.16 4 1.29 1.34
S × B 107.84 112 0.96
Total 209.50 149
**p < .01.
cant effect on the frequency of the students using successful
connectives.
Table 5 shows the result of the two-way ANOVA for num-
ber of words. As for the amount of writing in each assignment,
the interaction effect of feedback type and time was not sig-
nificant, F(4, 112) = .30, p = .87, η2 = .004. The average num-
ber of words did not vary according to the types of feedback
provided: F(1, 28) = 1.57, p = .21. Taking the effect size into
account, the effect of group type is small (.027, small: .01 < η2
< .06). Although the ANOVA test did not indicate that the
number of words written over the five weeks differed signifi-
cantly according to the time in which the writing was produced
[F(4, 112) = 1.40, p = .23, η2 = .023], the Scheffé test showed a
significant difference across the two writing assignments: Week
6 = 7.61, Week 7 = 9.28; p < .01. The results are also presented
in Figure 2.
To put these results together, it can be stated that the com-
ments on the use of transition words have a significant effect on
the students using successful connectives. However, no main
effect was revealed for increased number of words written dur-
ing this period except for the differences in Weeks 6 and 7, and
thus it cannot be concluded that the type of feedback provided
have a significant effect on increasing the amount of writing. In
order to verify the effectiveness of learning and using connec-
tives on fluency performance, the subsequent research was
conducted over a period of the following five weeks.
The 2nd Period (Weeks 8 - 12)
This section presents the results of Period 2. During this pe-
riod, both groups were given identical types of feedback and
comments: the content and form feedback and the additional
marginal comments on the use of sentence connectors. The
mean and standard deviation (SD) for each group are shown in
Table 6. Figure 3 shows that Control group gradually increases
Table 5.
Results of two-way ANOVA for words.
SS df MS F
Group (A) 1187.2 1 1187.2 1.57
S 21100.3 28 753.58
Time (B) 973.77 4 243.44 1.40
A × B 209.64 4 52.41 0.30
S × B 19448.5 112 173.64
Total 42919.5 149
Figure 2.
Average number of words by group.
Copyright © 2012 SciRes. 21
Y. SUGITA
the frequency of successful connections.
The data were then used to find out whether there was an ef-
fect of interactions between factors as well as effects of indi-
vidual factors. Table 7 presents the results of the two-way
ANOVA for SCs.
Although the interaction effect of feedback type and time
was not significant [F(4, 112) = 1.22, p > .10, η2 = .0002], the
average SCs differed between Experimental and Control groups:
F(1, 28) = 6.35, p < .05. The effect size is moderately large (η2
= .085), while a larger effect (η2 = .160) was revealed during
Period 1. There was a significant difference between the two
groups across one assignment (Week 8): F(1, 105) = 10.21, p
< .01. The Scheffé test also showed a significant difference
across the four writing assignments: Week 8 = 51.08, Week 9 =
8.54, Week 10 = 15.19, Week 11 = 8.54; p < .01. No significant
difference was found in the assignment of Week 12. Judging
from these results, the difference in the frequency of the stu-
dents using successful connectives between the groups gradu-
ally decreased due to the identical types of feedback and com-
ments on the use of sentence connectors.
As Figure 4 indicates, the differences in number of words
Table 6.
Descriptive statistics of group performance.
Word written Successful connections
Group Experimental Control Experimental Control
Time Mean SD Mean SDMean SD Mean SD
Week 8 63.3 16.0 52.3 13.64.0 1.3 2.5 1.2
Week 9 57.6 15.0 51.5 14.43.6 1.5 3.0 0.8
Week 10 56.4 16.3 48.9 10.03.9 1.3 3.1 0.8
Week 11 60.4 20.5 53.9 12.94.0 1.6 3.4 0.7
Week 12 57.9 18.4 53.8 16.03.9 1.6 3.5 1.3
Figure 3.
Frequency of SCs by groups.
Table 7.
Results of two-way ANOVA for SCs.
SS df MS F
Group (A) 21.66 1 21.66 6.35*
S 95.44 28 3.40
Time (B) 5.27 4 1.32 1.17
A × B 5.51 4 1.38 1.22
S × B 125.63 112 1.12
Total 253.50 149
*p < .05.
between the groups are disappearing. Table 8 shows the result
of the two-way ANOVA for number of words. As for the
amount of writing in each assignment, the average number of
words did not vary between Experimental and Control groups:
F(1, 28) = 3.08, p > .10. Taking the effect size into account, the
size is still small (η2 = .050), but it became almost twice as
large as that of the first period (η2 = .027). Although the
ANOVA test did not indicate that the number of words written
over the five weeks differed significantly according to the time
in which the writing was produced [F(4, 112) = 0.84, p < .10,
η2 = .014], the Scheffé test showed a significant difference
across the four writing assignments: Week 8 = 18.72, Week 10
= 8.62; p < .01, Week 9 = 5.82, Week 11 = 6.47; p < .05. No
significant difference was found in the assignment of Week 12.
These results imply that the identical types of feedback and
comments had a positive effect on the number of words written
by Control and Experimental groups, respectively. The results,
therefore, suggest that the type of feedback provided during
Period 1 may have a significant effect on the number of words
written by students during this period. In addition, the effect
size indicated that different types of feedback during Period 1
result in producing a certain effect on students’ writing fluency
during this period, instead of the previous period. Furthermore,
the statistical results did not show significant differences both
in the frequency of SCs and the number of words written in the
assignment of Week 12. This can be attributed to the enhance-
ment of writing after the four opportunities of receiving feed-
back and comment (Feedback 8 - 11). Therefore, it can be pre-
sumed that students need a certain experimental period to learn
and use transition words, so that they will probably succeed in
significantly increasing the amount of writing.
Findings and Discussion
Research Question #1
How did the experimental group respond to intentional feed-
Figure 4.
Average number of words by groups.
Table 8.
Results of two-way ANOVA for Words.
SS df MS F
Group (A) 1851.52 1 1851.52 3.08
S 16829.8 28 601.06
Time (B) 517.16 4 129.29 0.84
A × B 193.90 4 48.47 0.31
S × B 17099.3 112 152.67
Total 36491.8 149
Copyright © 2012 SciRes.
22
Y. SUGITA
back that explicitly focused their learning to use transition
words?
The results of Period 1 confirmed that the intentional feed-
back that explicitly focused on transition words had a signifi-
cant effect on the frequency of the students using successful
connectives. Teachers’ comments on the use of connectives
resulted in significantly greater use of transition words. This
finding suggests that writing teachers should provide feedback
on the use of transition words and teach students the effective-
ness of using such words in EFL writing classes. In terms of
this result, there are two important points to emphasize.
First, teachers hope that their students will improve in the
next assignment while they respond to and influence students’
writing. In order to achieve this ultimate goal of feedback, the
findings of the previous studies suggest that ESL writing in-
structors should be straightforward, concrete, and fairly direc-
tive in their feedback to L2 writers (Conrad & Goldstein, 1999;
Ferris, 1997, 2003a). Bates, Lane, and Lange (1993) also sug-
gest that teachers should make text-specific comments, which
relates to the text rather than general rules. During the first
period of this study, I provided the experimental group with
fairly text-specific comments on using connectives, such as in
1), while I have chosen to express my general comments for the
control group, as in 2).
1) Try to use for example to mention specific things in or-
der to explain what you mean.
2) Try to express your ideas as specifically as possible and
give extra information.
It is important here to note that it may be more efficient for
students to improve their writing with teachers’ explicit advice
on what expression they should use, like for example. Accord-
ingly text-specific comments on using connectives can extend
to future writing behavior.
Second, the results of Period 2 indicated that the difference
in the frequency of the students using successful connectives
between the groups gradually decreased due to the intentional
feedback on the use of sentence connectors. However, until the
12th week, there were significant differences in the frequency,
implying that it takes time for students to learn and use transi-
tion words. Dekeyser (2007), for instance, points out that once
procedural knowledge has been acquired, there is still a long
way to go before the relevant behavior can be consistently dis-
played with complete fluency (p. 98). Therefore, writing teach-
ers should raise awareness of students’ sense of coherence even
in the earlier stage of writing classes, and thus they understand
the effectiveness of learning transition words so that students’
acquisition of sentence connectors may be effectively promoted
in EFL writing classes.
Research Question #2
To what extent did the students learning to use transition
words enhance fluency in new pieces of writing?
According to the review of previous studies, learning to use
transition words could be effective at increasing students’ flu-
ency in written language as well as oral language, so that the
use of connectives may enable students to write more about the
relevant information and to enhance their fluency in writing.
The findings of the present study implied that the teacher’s
marginal comments on the use of transitional words had an
effect on increasing the number of words written, and partly
confirmed that the instruction of the use of sentence connectors
enhances students’ fluency in writing. It is helpful to consider
some possible reasons for further investigation.
The first reason is that teachers’ comments on the use of
connectives meet the students’ needs for their writing assign-
ments, resulting in the increased number of words written. It
has been recognized that teacher feedback should be con-
structed according to the most critical needs of individual stu-
dent writers (Conrad & Goldstein, 1999; Ferris, 1997, 2003b;
Reid 1994). The General English class in which this study was
conducted provided the 12 written homework assignments with
identical writing structures: a topic sentence, two or three rea-
sons representing students’ opinion, and a concluding sentence.
Thus, the students could concentrate on thinking and writing
about reasons to support their opinion or choice, and the rea-
sons they gave were supposed to coherently relate to the main
statement. As the previous studies point out that using connec-
tives lends a framework to meaning (Jordan, 1990; White &
McGovern, 1994) and helps student writers to show how the
points they relate to each other and to the whole text (Parrot,
2004), receiving comments on the use of connectives was
helpful and useful feedback when the students wrote reasons
for supporting their opinions, and thus had a certain effect on
increasing the amount of writing in the subsequent assignment.
The second reason is related to the notion mentioned in the
increased number of successful connections. Text-specific com-
ments on the use of connectives may foster the tendency for
learners to add ideas in their writing. In other words, providing
comments on the use of connectives can make students think
logically and affect their attitudes toward writing. Here is an
example in point. In the sixth assignment, one student wrote, “I
think that personality is the most important for a human being.”
I provided the student with a text-specific comment, such as
“Try to describe specific reasons after the statement by using
because.” In Week 8, she submitted the assignment including
the following sentences: If I don’t get along with my email
friends, I can easily break off the friendship. This is because I
don’t keep in touch if I stop sending email to them. We can
infer that the student tried to add the reason by using the con-
nective exemplified in my previous comment.
Gordon (2008) mentioned that useful feedback inspires writ-
ers to re-plan, re-draft, or re-edit their texts so as to best convey
their intended meaning. Since connectives express a number of
logical relationships, such as addition, contrast, causation, cir-
cumstance and so on, comments on the use of connectives may
enable students to think of relevant information logically. As a
consequence, learning to use different types of transition words
helps students to expand their ideas logically and thus write
more about the logical information.
However, it is true that the subsequent research conducted
during Period 2 may in part help to confirm that teachers’
comments on the use of connectives enhance students’ fluency
in writing, but the improvement of fluency observed through
the entire period is still small, implying that they will probably
need a certain experimental period for the acquisition of transi-
tion words. Thus, it is prudent to assume that although learning
to use different types of transition words helps students to write
more about the logical information, further research is neces-
sary to examine how students use transition words to expand
their ideas logically and how teachers give more effective
comments on the use of connectives, which facilitate their greater
fluency performance.
Copyright © 2012 SciRes. 23
Y. SUGITA
Conclusion
The results of the 12-week period indicate that teachers’
comments on the use of connectives were helpful and useful
feedback, and resulted in significantly greater use of transition
words. However, the effect on the amount of writing was mar-
ginally significant and it can be presumed that students need a
certain experimental period for the acquisition of transition
words, so that they will succeed in more significantly increas-
ing the number of words. At present, the assumption that learn-
ing to use transition words enhances students’ fluency in writ-
ing is nearly verified.
The main findings also suggest that writing teachers should
provide feedback on the use of transition words and teach stu-
dents the effectiveness of using sentence connectors. This also
helps to improve students’ fluency in EFL writing classes.
However, this study only partly confirms the effectiveness of
teaching transition words in terms of writing fluency. In order
to be able to investigate the tendency for learners to add logical
information in their writing and observe patterns of consistent
improvement, there could be a need for research to longitudi-
nally examine the effects of such instruction. Therefore, a fur-
ther direction of this study will be to provide more evidence for
these findings and to improve the effectiveness of teachers’
written feedback on the use of transition words.
Acknowledgements
This paper was based upon an oral presentation by the author
at the 15th International Conference of Pan-Pacific Association
of Applied Linguistics (PAAL) held in Seoul on August 17,
2010. The author would like to express gratitude to anonymous
reviewers for their helpful comments on an earlier draft of this
paper.
REFERENCES
Aoki, N. (1991). The effects of “global consciousness” on coherence of
English compositions of Japanese students writers. Annual Review of
English Language Educa tion in Ja pan, 2, 101-109.
Bates, L., Lane, J., & Lange, E. (1993). Writing clearly: Responding
ESL composition. Boston: Heinle & Heinle.
Brown, G. A., & Yule, G. (1983). Teaching the spoken language.
Cambridge: Cambridge University Press.
Burns, A., & Seidlhofer, B. (2002). Speaking and pronunciation. In N.
Schmitt, (Ed.), An introduction to applied linguistics (pp. 211-232).
London: Arnold.
Carlson, S., Bridgeman, B., Camp, R., & Waanders, J. (1985). Rela-
tionship of admission test scores to writing performance of native
and nonnative speakers of English. TOEFL Research Report No. 19.
Princeton, NJ: Educational Testing Service.
Conrad, S. M., & Goldstein, L. M. (1999). ESL student revision after
teacher-written comments: Text, contexts, and individuals. Journal
of Second Language Writing, 8, 147-180.
doi:10.1016/S1060-3743(99)80126-X
Cook, G. (1989). Discourse. Oxford: Oxford University Press.
Cooley, L., & Lewkowicz, J. (2003). Dissertation writing in practice:
Turning ideas into text. Aberdeen: Hong Kong University Press.
Dekeyser, R. (2007). Skill acquisition theory. In B. VanPatten, & J. Wil-
liams (Eds.), Theories in second language acquisition (pp. 97-113).
Mahwah, NJ: Lawrence Erlbaum.
Ejzenberg, R. (1992). Understanding nonnative oral fluency. Ann Ar-
bor: UMI.
Enginarlar, H. (1993). Student response to teacher feedback in EFL
writing. System, 21, 193-204. doi:10.1016/0346-251X(93)90041-E
Fathman, A. K., & Whalley, E. (1996). Teacher response to student writ-
ing: Focus on form versus content. In B. Kroll (Ed.), Second lan-
guage writing (pp. 178-190). Cambridge: Cambridge University Press.
Ferris, D. R. (1997). The influence of teacher commentary on student
revision. TESOL Quarterly, 31, 315-339. doi:10.2307/3588049
Ferris, D. R. (2003a). Response to student writing. Mahwah, NJ: Law-
rence Erlbaum.
Ferris, D. R. (2003b). Responding to writing. In B. Kroll (Ed.), Explor-
ing the dynamics of second language writing (pp. 119-140). Cam-
bridge: Cambridge University Press.
Fukushima, S., & Sato, T. (1989). The effectiveness of teaching transi-
tion words in EFL composition class. KATE Journal, 3, 29-39.
Gordon, L. (2008). Writing and good language learners. In C. Griffiths
(Ed.), Lessons from good language learners (pp. 244-254). Cam-
bridge: Cambridge University Press.
Halliday, M. A. K. (1989). Spoken and written language. Oxford: Ox-
ford University Press.
Halliday, M. A. K. (1994). An introduction to functional grammar. (2nd
ed.). London: Edward Arnold.
Halliday, M. A. K., & Hassan, R. (1976). Cohesion in English. London:
Longman.
Hayashi, K., Masaki, M., & Tokioka, Y. (2003). Daigaku ni okeru
eisakubun shido no arikata. Research project on the teaching of
writing English under the auspices of the Kansai Chapter of the Jap-
anese Association of College English Teachers (JACET), Writing
Research Group.
Hirota, T., Okada, T., Okumura, K., & Tokioka, Y. (1995). Daigaku ni
okeru eisakubun shido no arikata. Research project on the teaching
of writing English under the auspices of the Kansai Chapter of the
Japanese Association of College English Teachers (JACET), Writing
Research Group.
Iseno, K. (1991). Free writing instruction. Annual Review of English
Language Education in Japan, 2, 73-82.
JACET (1993). English education toward the 21st century in Japan. The
English Teachers’ Magazine, 41, Tokyo: Taishukanshoten.
Jordan, R. R. (1990). Academic writing course (2nd ed.). Harlow:
Longman.
Kawaguchi, C., & Kamimoto, T. (2000). Distinctive features of oral
production bt fluent and nonfluent EFL learners. Language Labora-
tory, 37, 21-36.
Kurihara, Y. (1994). Which criteria should be used for assessing Eng-
lish composition? Modern English Teaching, 8, 23-25.
Mizumoto, A., & Takeuchi, O. (2008). Basics and considerations for
reporting effect sizes in research papers. English Education Research,
31, 57-66.
Nunan, D. (1993). Task-based syllabus design: Selecting, grading and
sequencing tasks. In G. Crookes, & S. M. Gass (Eds.), Tasks in a
pedagogical context (pp. 55-66). Cleveland, UK: Multilingual Matters.
Oi, K. (2004). The present studies of second language acquisition (I.
Koike, M. Terauchi, K. Kinoshita, & M. Narita (Eds.)), Chapter 11
writing, 201-218.
Parrot, M. (2004). Grammar for English language teachers (9th ed.).
Cambridge: Cambridge University Press.
Reid, J., & G. Findlay (1986). Writer’s workbench analysis of holisti-
cally scored essays. Computers and Composition, 3, 6-32.
doi:10.1016/S8755-4615(86)80014-7
Reid, J. (1994). Responding to ESL students’ texts: The myths of ap-
propriation. TESOL Quarterly, 28, 273-292. doi:10.2307/3587434
Reid, J. (1996). Responding to different topic types: A quantitative
analysis from a contrastive rhetoric perspective. In B. Kroll, (Ed.),
Second language writing (pp. 191-210). Cambridge: Cambridge Uni-
versity Press.
Spycher, K. (2007). Academic writing of adolescent English learners:
Learning to use “although”. Journal of Second Language Writing, 16,
238-254. doi:10.1016/j.jslw.2007.07.001
The Course of Study for Foreign Languages. (1999). URL (last checked
3 August 2010). http://www.mext.go.jp/english/shotou/030301.html
White, R., & McGovern, D. (1994). Writing. Hemel Hempstead: Pren-
tice Hall.
Yashima, T., Yamamoto, S., & Viswat, L. (1994). Acquisition of
communicative competence through overseas study in America: A
case study of eight English Japanese high school students. Language
Laboratory, 31, 31-44.
Copyright © 2012 SciRes.
24
Y. SUGITA
Copyright © 2012 SciRes. 25
Appendices
Appendix 1: Textual Discourse Markers
Numbering and ordering points
First, firstly, in the first place, first of all, Second, secondly/
third, thirdly/last, lastly, finally
Adding something
Moreover, and, in addition, also, furthermore, additionally,
alternatively, instead
Linking similar things together
Similarly, equally, likewise
Introducing something that contrast with expectations
However, but, nevertheless, although, though, on the other
hand, by contrast, conversely, on the contrary
Cause and results
Therefore, so, then, as a result, thus, consequently, that’s be-
cause, for, because (of), since, for this reason.
Generalizing
Generally, in general, on the whole
Exemplifying and narrowing down
Specifically, that is, for example, in fact, namely
Re-stating
In other words, in a sense, that is (to say)
Rounding-off
In summary, to summarize, in conclusion, to conclude, to
sum up
Appendix 2: Exa mples of S uc ce ss f ul and Unsuccessful
Connections (Excerpts from Chapter 2)
Successful connection: American young people like to be on
their own. In Japan, however, young people tend to stay at
home with Mom and Dad.
Unsuccessful connection: Many Japanese young women live
with their parents much longer. In addition, they have the less
chance of marrying.
Unsuccessful connection: There would be not so many para-
site singles in Japan. And, in fact, there are a lot of young peo-
ple who depend on their parents.
Unsuccessful connection: Parasite singles can have no chil-
dren. In short, her family will die out in the future.
Unsuccessful connection: We can think of how to use money
although we live with parents at home.