Quantitative analysis of the effect of synchronous online discussions on oral and written language development for efl university students in Vietnam

Hue University Journal of Sciences:Social Science and Humanities
ISSN 2588–1213
Vol. 128, No. 6B, 2019, Tr. 16–26, DOI: 10.26459/hueuni-jssh.v128i6B.4880
* Corresponding: anhpham.dumtp@gmail.com 
Submitted:18–07–2018; Revised: 06–12–2018; Accepted: 18–12–2018. 
QUANTITATIVE ANALYSIS OF THE EFFECT OF 
SYNCHRONOUS ONLINE DISCUSSIONS ON ORAL AND 
WRITTEN LANGUAGE DEVELOPMENT FOR EFL 
UNIVERSITY STUDENTS IN VIETNAM 
Pham Dang Tram Anh* 
Universityof Foreign Languages, Hue University, 57 Nguyen KhoaChiem St., Hue, Vietnam 
Abstract. The study investigates the effects of synchronous online discussions on oral and written skills 
development for EFL university students in Vietnam after one semester of instruction, using a pretest-
posttest quasi-experimental approach. One 15-student class of B1-level English participated in oral role 
plays, whereas another 15-student class engaged in synchronous online discussions after comparable 
preparation. The results show no statistically significant differences in oral and written proficiency 
between the two treatment groups at the end of the semester. However, students in both groups achieved 
great progress in speaking and writing skills, especially speaking skill after one semester. The fact that the 
group which used written chat during the treatment had better scores in the oral test at the end of the 
semester strengthens the belief that online discussions support the development of the same processes that 
underlie oral speech. Synchronous online discussions are thus shown to be a valuable addition to the 
foreign language classroom in terms of second language acquisition and learners’ language development. 
Keywords.synchronous online discussions, oral role plays, oral skills, written skills, foreign language 
classroom 
1. Introduction 
During the last decades, research on computer-mediated communication (CMC) has 
shown an outstanding development in foreign language teaching and learning methodology 
using technology.Countless benefits have been proved through numerous world-wide research 
studies. 
Computer-mediated communication has two modes: asynchronous computer-mediated 
communication (ASCMC) and synchronous computer-mediated communication (SCMC). 
Generally, asynchronous computer-mediated communication takes place in time and place 
independence. Typically, the instructor and students discuss, give and get information or 
deliver materials through the course web pages, posting of forum messages, and listserv or 
Jos.hueuni.edu.vn Vol. 128, No. 6B, 2019
17 
email messages. In contrast, synchronous computer-mediated communication represents real-
time interactions. Usually, this means that the instructor and students meet in a text-chat room 
simultaneously at a particular location in cyberspace or from any location that has Internet 
access and interact through typed statements or questions [10]. In this study, SCMC is used as a 
synchronous online discussion ortextchat or written chat interchangeably. 
Previous research studies show that there is a link between oral communication and 
online synchronous discussions. Online synchronous discussions not only enhancewritten 
proficiency but also indirectly improve oral proficiency since writing in a computer-mediated 
environment resembles spoken language in terms of its interactional features [1, 4]. 
Withthe increasing use of technology in today's language classrooms in Vietnam, the 
current research is needed to evaluate its effects on EFL learning of Vietnamese students. The 
objective of the present study is to determine the different levels in the oral and written 
proficiency of EFL students between the role-play group and the synchronous online-discussion 
group after one semester of instruction. 
2. Literature review 
Computer-mediated communicationand language development 
The contributionof computer-mediated communication to language development has 
constantly been proved since the early period in CMC’s history. Various meta-linguistic aspects 
of language development have been explored, including negotiation of meaning, sociolinguistic 
environment, and intercultural competence. Furthermore, numerous studies on the 
interactionist approach as well as a sociocultural viewpoint to second language acquisitionhave 
been conducted on the negotiation of meaning and CMC [5]. In addition, Kitade’s findings 
show that CMC is a profitable environment for sociolinguistic development [9]. The author 
concludes that interactional features of CMC facilitate encouraging conditions for developing 
positive attitudes towards language learning: no turn-taking competition, text-based 
interaction, and a lack of nonverbal cues. 
What is more, increasing in linguistic competence, both quality and quantity, among 
learners has also been stated by Kern [7]. He reveals that CMC-supported learners create more 
language production than their friends in the face-to-face group. Kern findsthat SCMC 
discussions produce between two and four times more turns, more sentences, and more words 
than the oral discussions. The research also shows that learners’ language production is at a 
greater level of sophistication regarding grammatical accuracy and complexity. 
Finally, in terms of language skills, both written and spoken skills are enhanced through 
various modes of CMC in language learning projects. In fact, there is a common tendency to 
Pham Dang Tram Anh Vol. 128, No. 6B, 2019
18 
associate CMC with the development of specific language skills [11]. The hypothesis that SCMC 
may improve speaking proficiency has been tested by Payne and Whitney, who find that 
participants in a text-chat room have a significantly higher oral proficiency than those just 
spending time in traditional oral classes [13]. 
In short, it is obvious that CMC has been used widely in developing most language areas 
and skills, except for listening skills which are normally supported and developed through 
other forms of visual and audio technology. In general, the text-based nature of CMC brings 
about many meaningful applications in language education. This medium not only amplifies 
students’ attention to linguistic forms [14], but it also stimulates increased written production of 
the target language as well as creating a less stressful and more equitable environment for 
discussion [12]. 
Benefits of synchronous online discussion 
A wide range of potential strengths of SCMC mode hasbeen found in previous studies. 
According to Berge [2], SCMC enhances learning opportunities for students. Additionally, 
SCMC not only provides more in-depth interactions on topics, but it is also convenient and 
useful for students living long distances from the university, travelling frequently, having 
family obligations (childcare responsibility) or disabilities that affect mobility as in the study by 
Lavooy & Newlin [10]. Then, Candace claims that SCMC creates more spontaneous 
communication going back and forth, enhances learning interests and interpersonal relationship 
[3]. In addition, she adds that students are more engaged and have more equal distribution in 
asking and answering questions during the communication processes. Moreover, Candace 
proves there is a higher percentage of socio-emotional interactions; learners ask more personal 
questions and reveal more about their frustration or need for help with less hesitation. The 
students have a stronger sense of immediacy to respond to peers’ questions and spend less time 
to send messages [3].Furthermore, Khamis states that text scripts can be always retrieved for 
later reference via SCMC which is a backbone of distance learning programmes. His research 
proves that SCMC facilitates instructor-student and student-student interactions as well as 
provides natural by-products of authentic communicative activities [8]. Finally, in Watt’s study, 
SCMC’s benefit is that it is easier to provide immediate feedback to information seekers [15]. 
Furthermore, the most common benefits related tothe effects of CMC are as follows: 
reduced anxiety, increased and equalized participation among language students, and 
increased language output. Often, researchers report that these factors have an impact on one 
another, i.e., the reduced anxiety level that is prevalent in a CMC environment leads to 
increased and equalized participation among students, which, in turn, leads to more language 
production. The question that arises is why learners experience a discussion in a computer lab 
so differently from an oral discussion in the classroom. 
Jos.hueuni.edu.vn Vol. 128, No. 6B, 2019
19 
In order to capitalize these SCMC’s advantages and explore its possible influences on 
classroom language learning, the present study investigates the effects of synchronous online 
discussions on the development of oral and written language skills for EFL university students 
in Vietnam. 
3. Methods and material 
Research design 
In order to achieve the aim of this study, a quasi-experimental approach was chosen.The 
researcher also collected numeric details from quantitative data of the pre-treatment 
questionnaire to investigate students’ bio-data and background. The following hypotheses 
guide this research: 
 The chat group would write at a higher proficiency level at the end of the semester 
because they havemore opportunity to write. 
 The role-play group would demonstrate higher oral proficiency at the end of the 
semester because they havemore opportunity to speak. 
 The chat group would demonstrate better oral performance at the end of the 
semester because the language used in chatting is more similar to oral speech. 
Participants 
After fully getting information about the purpose of the study, 49 students voluntarily 
participated in the project. However, 4 students withdrew from the study during the semester, 
15 more students either dropped the class or were not present on the day of taking the post-
tests at the end of the semester. Only data from participants who were present on all data 
collection days were analyzed in this study, which made the total number of 30 participants, 
with 15 from each class, available for research data collection and analysis. One of the two 
classes was designated as the control class and the other, the experimental or SCMC class. 
Table 1 shows the selected background information of participants. There is clearly no 
statistically significant difference regarding gender, distribution of demographic background, 
and the starting point of learning English between the two treatment groups. 
Table 1. Selected background information 
Total 
n= 30 
Role-play group 
n= 15 
Text-chat group 
n= 15 
No. % No. % No. % p* 
Gender Female 20 66.7 11 73.3 9 60.0 0.700 
Pham Dang Tram Anh Vol. 128, No. 6B, 2019
20 
Male 10 33.3 4 26.7 6 40.0 
Demographic 
background 
Urban 11 36.7 7 46.7 4 26.7 
0.601 
Rural 19 63.3 8 53.3 11 73.3 
Starting point 
of learning 
English at 
school 
Grade 3 16 53.3 11 73.3 5 33.3 
0.072 Grade 6 12 40.0 3 20.0 9 60.0 
Other 2 6.7 1 6.7 1 6.7 
*Significant at p< 0.05 
In addition, the researcher measured the pre-treatment mean scores of the oral test and 
the written test between the two treatment groups. 
Table2. Mean scores of pre-treatment oral and written tests 
Groups Means SD t p* 
Role play 
n = 15 
Pre-oral 5.10 1.137 
0.295 0.770 
Pre-write 5.07 1.624 
Text-chat 
n = 15 
Pre-oral 4.97 1.329 
-0.435 0.667 
Pre-write 5.37 2.117 
*Significant at p<0.05 
Table 2 shows that there is clearly no statistically significant difference between the two 
treatment groups in mean scores of the pre-treatment oral test and the written test with p = 0.770 
and 0.667 (>0.05). 
In summary, there is no statistically significant difference between the two treatment 
groups about participants’ background and pre-treatment mean scores. This shows that the two 
groups practically have the same input before treatment. 
Instruments 
In order to establish participants’ levels of oral and written proficiency at the beginning 
and at the end of the semester, the researcher used pre- and post-treatment oral tests and 
written tests. 
Furthermore, a pre-treatment questionnaire was distributed to the two selected classes 
after the participants had been briefed on the purpose of the study and had an opportunity to 
ask questions about what would be involved. 
Research procedures 
The research procedures are described in Table 3. 
Jos.hueuni.edu.vn Vol. 128, No. 6B, 2019
21 
Table 3. Overview of the research procedures 
Pre-treatment measurement instruments 
1. Oral test 
2. Writing test 
3. Pre-questionnaire 
While-treatment 
Role-play group in the lab 
Conducted by the researcher: 
– Provide topics and links 
– Explain and answer questions 
– Provide assistance during the roleplays 
– Feedback on mistakes 
Performed by students: 
– Do web-based activities 
– Do oral roleplays 
– Correct mistakes 
– Present topic discussion in the whole group 
Text-chat group in the lab 
Conducted by the researcher: 
– Provide topics and links 
– Explain and answer questions 
– Provide assistance during the textchat 
– Feedback on mistakes 
Performed by students: 
– Do web-based activities 
– Write text messages 
– Correct mistakes 
– Present topic discussion in the whole group 
Post-treatment measurement instruments 
1. Oral test 
2. Writing test 
4. Findings and discussion 
First, to achieve the aim of the research, descriptive statistics (means and standard 
deviations) were computed. Table 4 shows the means and standard deviations for the pre- and 
post-treatment scores in the oral test and the written test according to treatment (role play, text 
chat). 
Table 4. Means and standard deviation for oral test and written test 
Groups Pre-oral Post-oral Diff. Pre-write Post-write Diff. 
Role 
play 
n = 15 
Means 5.10 6.20 1.10 5.07 5.93 0.86 
SD 1.137 1.099 1.624 1.321 
t 0.295 –0.435 
p* 0.770 0.667 
Text 
chat 
n = 15 
Means 4.97 6.57 1.60 5.37 6.40 1.03 
SD 1.329 1.100 2.117 1.242 
t –0.914 –0.997 
p* 0.369 0.327 
Pham Dang Tram Anh Vol. 128, No. 6B, 2019
22 
*Significant at p<0.05 
Both groups receive higher means in the post-treatment tests at the end of the semester, 
which indicates increased oral and written proficiency (Table 4). The standard deviations for 
both groups seem quite large, suggesting rather variable levels of proficiency within the 
different groups at the beginning and at the end of the semester. 
Usually, educators expect instruction to have a positive effect on students' learning, 
resulting in higher means in a post-test. This is the case here, too, with the two groups showing 
gains in measurements of oral and written proficiency levels. Furthermore, educators also 
expect that instruction would result in students' similar performance in thepost-test, indicated 
by a smaller standard deviation, which is also the case here. From the researcher’s viewpoint, 
several factors might have contributed to the current picture. First, the treatment using 
synchronous online discussions might have had a more positive effect on students' performance 
in an oral test than the treatment using face-to-face discussions (the difference between oral pre-
tests and post-tests in the text-chat group is 1.60 and that in the role-play group is 1.10); a 
possible reason for this effect might be that the language which was practiced duringtext-chat 
sessions is more similar to oral speech than to written language after all.Another possible 
reason is that the students monitor the use of grammar and vocabulary better in text-chat 
messages. The third reason might be that the text-chat discussion generates plenty of ideas. 
Second, the text-chat group hasbetter scores in oral tests than in written tests in spite of writing 
text-chat messages as treatment during the semester (1.60 in the oral test compared with 1.03 in 
the writing test). The oral test of the text-chat group achieves the best scores among the four 
tests both in oral and written tests at the beginning and at the end of the semester. This confirms 
the assumption that synchronous online discussions might have a positive effect on speaking 
skills. Third, oral role-playing as a treatment does not seem to have any leveling effect on 
students' performance in either oral test or written test. 
In general, language learners in their English course of one semester might develop so 
differently from one another that no matter what kind of treatment they receive, they show very 
different profiles concerning their proficiency levels at the end of that semester. 
Table 5.Means and standard deviation for pre- and post-oral and written tests 
Treatment Means SD t p* 
Role play (n = 15) 
Oral test 
1.1333 0.91548 
–1.826 0.078 
Text chat (n = 15) 1.7333 0.88372 
Role play (n = 15) 
Written test 
0.8000 1.56753 
–0.445 0.660 
Text chat (n = 15) 1.06667 1.70992 
*Significant at p< 0.05 
Jos.hueuni.edu.vn Vol. 128, No. 6B, 2019
23 
As can be seen in Table 5, the mean scores of oral and written tests between pre- and 
post-treatment in the text-chat group are higher than those in the role-play group. Thisshows 
that students achieve great progress in speaking and writing skills, especially speaking skill 
after one semester of using synchronous online discussion although there is no statistically 
significant difference between the two treatment groups with p = 0.078 and 0.660 (>0.05). The 
fact that the chat group achieves a better gain in oral proficiency at the end of the semester 
indicates that having students engage in synchronous online discussions is not counter-
productive to the development of their oral language skills. This finding further supports the 
hypothesis of the study that learners might improve their oral skills using synchronous online 
discussion. 
Since the two different treatment groups have different mean gain scores, especially 
regarding their oral proficiency level, the researcher decided to run paired samples (matched) t-
tests to compare the groups' performance before and after the treatment. Table 6 indicates 
thatthe two groups independent of treatment have a significant gain in oral and written 
proficiency between the beginning and the end of the semester. Significance levels are reached 
by the oral role-play and the text-chat groups for both oral and written proficiency gain. 
Table 6.t-test: paired samples test 
Treatment Pair 
Mean Gain 
Score 
SD df t 
sig. 
(2-tailed) 
Role play 
Pre-oral 
Post-oral 
1.1333 
0.842 29 –8.779 0.000 
Pre-write 
Post-write 
0.8000 
Text chat 
Pre-oral 
Post-oral 
1.7333 
1.694 29 –3.072 0.005 
Pre-write 
Post-write 
1.0667 
*Significant p<0.05 
Although the specific treatment does not seem to have different effects on the 
development of students' oral and written proficiency, the instruction that they receive during 
the semester results in a significant increase in their proficiency levels. It is important to note 
that the data also show that having students engage in synchronous online discussions is not 
counter-productive to the development of their oral language skills, a result which further 
supports Payne and Whitney's hypothesis that SCMC develops the same cognitive mechanisms 
that underlie oral speech. 
Pedagogical implications 
Pham Dang Tram Anh Vol. 128, No. 6B, 2019
24 
Given the very positive reactions of the participating students to the integration of web-
based search activities in their English lessons, followed by oral role plays or online discussions, 
the increased use of technology in the foreign language c