CHAPTER II
LITERATURE REVIEW
This chapter begins with a discussion over the effect of text-based CMC on the
development of language learning, which serves well as a pedagogical background for
investigating the effect of voice-based CMC on the enhancement of foreign language
oral proficiency, then followed by the studies in respect of the comparison between
the effect of text-based ACMC and SCMC in language learning. Subsequently,
learners’ general perception of CMC in comparison with face-to-face traditional class
is also introduced in detail.
2.1 Text-based Computer-Mediated Communication
Over the past decade, numerous studies have attempted to quantify the features
of how text-based CMC facilitates foreign language learning and enhances linguistic
competence (Blake, 2000; De la Fuente, 2002; Spitzer, 1989; Warschauer, 1996). In
the subsequent sections, three phases of language learning in text-based CMC
including, vocabulary, reading & writing, and speaking are discussed respectively.
The reasons why one of the receptive skill, listening is absent in the following part is
that little attention has been paid to voice-based CMC, leading to the absence of
listening ability from discussions.
2.1.1 Vocabulary in Text-based CMC
The Interaction Hypothesis proposed by Long and Robinson (1998) states that L2
learners’ meaning negotiation with other speakers crucially enhances second language
acquisition. With the emergence of the hypothesis, there has been many literatures
proving and ensuring the positive effect of text-based CMC on vocabulary learning
(Blake, 2000; De la Fuente, 2002; Ellis & He, 1999; He, 1998). Ellis and He (1999)
found that the ‘superior dialogic interaction’ in a ‘modified output group’ resulted in
the higher levels of receptive and productive acquisition of L2 words, accounting for
that negotiation focusing on lexical aspects of the language may be beneficial for L2
vocabulary acquisition.
By applying Interaction Hypothesis to vocabulary learning, Blake (2000)
subsequently concluded in his paper that incidental negotiation in text-based CMC
commonly occurred, especially dealing with learners’ lexical confusions. In Blake’s
study (2000), 50 intermediate Spanish learners were asked to have networked
discussion in dyads through a synchronous text-based chat program, Remote
Technical Assistance (RTA). The result indicated that up to 75% to 95% of all
negotiations were for lexical; in other words, the major part of negotiations was
triggered by lexical confusion. Therefore, the researcher claims that because CMC
raises L2 learners’ metalinguistic awareness of their L2 vocabulary development and
generates learners’ interaction on lexicons, L2 vocabulary can subsequently be
modified and improved in such a text-based learning environment. Subsequent to
Blake’s experiment (2000), De la Fuente (2003) even centered the sutdy on examining
the different potential effects of negotiation of meaning on L2 vocabulary
development between text-based computer-mediated interaction and face-to-face one.
20 elementary Spanish learners randomly assigned into Oral Interaction Group (OIG)
and Virtual Chat group (VCG) were asks to finish tasks with different information
gaps. Before the treatment, a pretest of vocabulary was administered to select 14
target Spanish words that learners had never learned before. The result showed that
face-to-face as well as CMC text-based interaction was equally effective in
developing written receptive and productive acquisition and retention of L2
vocabulary. However, in the promotion of oral acquisition of L2 words, and above all,
from the productive aspect, the latter group seemed to be less effective when
compared with face-to-face interaction. De la Fuente (2003) concluded that “cognitive
factors such as attention and depth of processing are the key to unveiling what
elements in the negotiation process facilitate L2 vocabulary development through
CMC, synchronous interactive tasks” (p. 74).
2.1.2 Reading and Writing in Text-based CMC
By the same token, in reference to the development of reading and writing ability,
text-based CMC such as email exchanges can positively enhance learners’ reading as
well as writing ability. As Robb (1996) stated, “electronic penpals or ‘keypals’ is a
highly motivating way to get valuable practice in both reading and writing” (p. 8).
Warschauer (1996) claimed that compared with face-to-face discussion, the language
in text-based CMC has been more lexically and syntactically complex due to the
written nature. He even further suggested that text-based CMC seem to be a good
vehicle for raising writing ability to greater levels of complexity and improving
learners’ overall writing (Cooper & Selfe, 1990; Mabrito, 1989; Spitzer, 1989;
Warschauer, 1996). Moreover, in Sullivan and Pratt’s (1996) study, they involved 38
intermediate ESL participants in the research examining learners’ attitudes, writing
apprehension, and writing quality in the two different writing environments,
computer-assisted class (text-based CMC) and oral one (traditional face-to-face class).
The quantitative analyses showed that the two writing environments might have no
strong effect on attitudes toward writing while the writing quality reaching the
significant level (0.08) was indeed improved in the computer-assisted classroom. The
study concluded that more than fifteen-week participation in networked computers
was more beneficial to the development of student writing than that in oral discussion.
In addition, Coniam and Wong’s (2004) focused their research on the correlation
between use of text-based CMC and the development of writing ability. 26 students in
one Hong Kong English-medium-instruction secondary school were involved in the
study and were divided into experimental group with the use of ICQ (15) and control
group (11). The experimental group was asked to accumulate a total of at least 20
hours of CMC with the attention to the target grammar, ‘only one verb with time and
number marking per main clause’. The result indicated that text-based CMC group
had promoted the use of English to exchange complex ideas, accounting for the higher
occurrence rate of complex sentences in learners’ use of English. Moreover, compared
with the control group, the writings of CMC group also showed the decrease of errors
in units with single verb, suggesting a better control of the target grammar. The effect
of text-based CMC on improving writing ability, consequently, is both theoretically
and empirically proved.
2.1.3 Speaking in Text-based CMC
Recently there has been growing interest in CMC because it can not only benefit
the writing but also the oral ability in second or foreign language learning. Text-based
CMC shares abundant of similar characteristics with spoken communication and
produces parallel advantages of oral discussions without temporal and spatial
constraints imposed by the classroom setting (Blake, 2000). Foto’s research (2004)
claimed that even ACMC in the form of email exchanges, despite the decrease of
non-verbal (i.e. facial expressions) and verbal cues (i.e. rising and falling of
intonation), could be as interactive as speech interactions. Furthermore, CMC is a
forum where demands of real time communication can be compensated for through a
set of affordances that the medium offers foreign language learners and teachers
(Meskill & Anthony, 2005). Many researchers and teachers (Abrams, 2003; Bölke,
2003; Chun, 1998; Kern, 1995; Salaberry, 2000) believe that text-based CMC can
effectively enhance learners’ spoken performance in the second language because
students tend to have greater equity in participation and produce more language with
more complex structure in text-based CMC than in face-to-face context.
Kern’s study (1995), for example, illustrated the effect of CMC on quantity and
variety of language oral production with the experiment comparing forty French
learners’ oral production during InterChange session (text-based CMC) and during an
face-to-face oral class discussion. The result indicated that overall students made from
two to three and a half times more turns in the InterChange sessions than the
traditional one. Then, consistent with the higher number of turns, students produced
two to four times more sentences with greater variety of discourse functions during
the CMC sessions than during the oral discussion. In the light of Kern’s claim,
Beauvois (1997) conducted a longitudinal study of text-based SCMC and L2 oral
proficiency with 83 participants, suggesting that learners in text-based CMC
outperformed those in non-CMC groups in oral exams in respect of pronunciation,
grammatical accuracy, lexical choice and accuracy, and content.
Not only had the above-mentioned linguistic but also pragmatic knowledge of
the target language made essential contribution to L2 oral proficiency. Chen (2005) in
her recent paper concerning about the use of text-based CMC to develop EFL
learners’ communicative competence (Hymes, 1972). With more and more attention
paid to communicative competence and approach, communication has been
recognized as an important learning objective in most current approaches to foreign
language teaching. The mastery of linguistic forms as well as the knowledge of social
factors plays an important role in communication. Chen concluded that the integration
of text-based CMC into EFL learning can offer learners more authentic input and
more opportunities to be engaged in the target sociocultural contexts (Light, 1989). In
effect, both linguistic and pragmatic knowledge can be enhanced for learners to
develop higher L2 oral proficiency.
In addition to comparing the effect of text-based CMC and face-to-face
discussion on oral proficiency, within text-based CMC itself, task types also greatly
influence the amount of learners’ negotiation, which has something to do with the
effect of text-based CMC on the development of oral proficiency. In Blake’s study
(2000), participants were asked to carry out three tasks types by means of SCMC,
including jigsaw, information-gap, and decision-making. The researcher reported that
jigsaw tasks accounted for 93% and 78% of the total negotiations for the two learning
sessions respectively. He then further concluded that jigsaw tasks promoted and
elicited more negotiations in accordance with what the previous study had predicted
(Pica, Kanagy & Falodum, 1993). Moreover, Smith (2003a) involved fourteen
nonnative-nonnative dyads to collaboratively complete 4 communicative tasks,
composed by 2 jigsaw and 2 decision-making tasks, in the form of SCMC. Partly
different from Blake’s conclusion, however, the result showed that compared with the
jigsaw tasks, learners negotiated a significantly higher percentage of turns in the
decision-making tasks. Statistically speaking, up to 78% of the target items were
negotiated during the decision-making tasks while only 22% were negotiated during
the jigsaw tasks. The researcher further concluded that task-type definitely influences
the extent to which learners engaged in negotiation, but the claim is not necessarily
true in the mode of face-to-face discussions. Furthermore, Peterson (2006) also
carried out a study aiming at shedding more light on the relationship between
task-type and level and numbers of negotiations in SCMC. Twenty-four intermediate
level EFL participants were invited to undertake a variety of language tasks in SCMC.
The data highlighted and confirmed the influence of task types on the quantity of
negotiations once again. There were more negotiation occurred in the
decision-making tasks than in the jigsaw as well as opinion-exchange tasks, which
was consistent with Smith’s (2003a) study. However, in line with what the previous
literature suggests (Smith, 2003a; Pica, 1993), in face-to-face contexts, jigsaw tasks
elicit a higher degree of negotiation than other task types.
Regardless of what task types was involved in those studies, it has been widely
and empirically recognized in literature that one blessing virtue of CMC, more
specifically, lies in the development of second language oral proficiency indirectly
mainly through the text-based CMC with written channel. Nevertheless, the ideal
parallel transfer from networked written ability into oral proficiency remains in doubt.
As the result of indirect transfer between written and oral, language abilities originally
belong to different paradigms in language learning; some unique features or even
genres have drawn many researchers’ attention. Therefore, the underlying
communicative features in text-based CMC accompanied with the influence of CMC
language on communication process and empirical comparison between CMC
language and spoken/written language will be discussed in detail.
The linguistic characteristics in text-based CMC were classified by Murray
(2000) into four major points: ‘similarity to spoken or written language’, ‘use of
simplified registers’, ‘structure of organization’ and ‘mechanisms for topic cohesion’
(p.400). In the field of text-based ACMC, email and electronic text-based bulletin
board are pervasively used nowadays. Based on Yate’s corpus of email (1996) and the
researcher’s personal e-messages, Crystal (2001) suggested that in spite of some
variations of terminology, individual emails consisted of a series of functional
elements, such as headers, greetings, and farewells, which were principally similar to
those found in traditional letters and memos (Flynn and Flynn, 1998; Gain, 1998; Li,
2000). Yet, the language style in email was uniquely different from both spoken and
written languages. Some emails were highly telegrammatic in style and analogies
between email and other forms of communication were drawn by writers to orientate
email in part of communication. Email was situated right between a conversation and
a letter (Hale& Scanlon, 1999; Naughton, 1999), and in the final analysis, was not
either spoken or written style. Crystal (2001) in his book even concludes that “The
consensus seems to be that it is, formally and functionally unique” (p.125). By the
same token, synchronous text-based electronic discourse is claimed to incorporate
features both from spoken and written languages (Crystal, 2001; Weininger & Shield,
2001) and to be a different, ‘medium-dependent register’ (Cherny, 1999; Weininger &
Shield, 2003, p. 330). Moreover, in M.J. Weininger & L. Shield’s (2003) research,
they further compared non-native-speaker (NNS) discourse with native-speaker (NS)
MOO-discourse (synchronous text-based CMC) and found that both native and
non-native speakers shared the same patterns in discourse and that even perceived
different formality of context and situation, NNS MOO-discourse still varied in a
similar way as native speakers did, which was in line with those results shown in
other literature (Cherny, 1999; Schwienhorst, 2004). The conformation of NS’ and
NNS’ discourse clearly proved that the uniqueness of linguistic characteristics in
text-based CMC was universal and cross-cultural. Concerning about ACMC as well as
SCMC, some studies indicate that text-based CMC should be recognized as spoken
genre because compared with face-to-face conversations, text-based CMC contains
greater complexity in lexical and syntax level of language (Coniam & Wong, 2004;
Warschauer, 1996). Notwithstanding being rejected in the spoken genre, interestingly,
research has been suggested that language in both text-based SCMC and ACMC tend
to be more ‘speech-like’ than ‘written’ (Ferrera et al., 1991; Yates, 1996).
Besides the language style of text-based CMC, one of the main differences
between speech and text-based CMC lies in the use of pronouns (Fowler and Gunther,
1979; Chafe and Danielewicz (1987). In Yates’ corpus-based study (1993), the result
showed that despite its similarity to writing in terms of overall frequency of pronoun
use, text-based CMC was quite different from writing in how pronouns of each type
(1st, 2nd, and 3rd person pronouns) are distributed. Then, from the aspect of relative
proportional usage of each type of pronoun, text-based CMC was much closer to
spoken language. Nevertheless, text-based CMC made greater proportional use of first
and second person pronouns (64%) than speech (58%) and writing (27%) did. In
addition to pronoun use, the study also suggested that the usage of modals in
text-based CMC was significantly higher than that of either speech or writing. On the
other hand, similar to the case with pronoun use, the overall relative frequencies of
modal usage across different semantic groups in text-based CMC were closest to
speech, which supported Ferrera’s study (1991) as well.
After the brief introduction to the linguistic features in text-based CMC, some
underlying communicative features appear to influence the communication process.
Text-based SCMC, for example, occurs without the significant delay of time but on
written channel; hence, participants do not have to take temporally sequential turns
since oral interruption is impossible in the written mode (Beauvois, 1992, 1998; Kelm,
1992; Kitade, 2000). Moreover, due to the lack of the aural (i.e. intonation) and visual
(i.e. a smile or a frown) paralinguistic cues in the face-to-face conversation, both
listener and speaker’s orientation to the discourse such as agreement or annoyance
cannot be precisely signified (Kieler et al., 1984). Finally, knowledge of social factors
that is crucial in communicative competence is hardly presented in the context of
text-based CMC without visual and aural paralinguistic cues for the distinguishability
of gender, ethnic background, and other social factors, especially when participants
use pseudo names (Hayne, Rice & Licker, 1994; Marjanovic, 1999; Selfe & Meyer,
1991; Sullivan, 1998).
Most literature consents to the positive effect of text-based CMC on
foreign/second language learning up to present although the difference between
text-based CMC and oral communication is surely existed, verifying that the practice
in the form of text-based CMC might not necessarily be properly transferred to the
development of oral proficiency which requires learners to directly ‘talk’ to with
interlocutors on spoken channel rather than to type the language. Based on the
previous studies, it appears that a number of studies have addressed the question of
how well the oral abilities can be indirectly improved in the text-based CMC context
by means of measuring syntactic complexity, lexical density, and diversity that
learners display. However, perhaps restricted by the limited bandwidth that barely
support the fluent online voice chat in the past, little research (Chang & Yang, 2006)
in the field of language learning to date have focused on this issue directly and
explicitly through voice-based SCMC as well as ACMC.
2.2 Synchronous vs. Asynchronous text-based CMC
Since the positive effect of CMC on language learning in comparison with the
face-to-face communication has been recognized according to the literature for the
decades, some researchers (Abrams, 2003; Sotillo, 2000) begin to turn their attention
to the profounder understanding on the combination as well as comparison between
the effect of ACMC and SCMC on language learning.
According to the numerous discussions either about the effect of ACMC or
SCMC on language learning, it could be reasonably inferred that the combination of
the two in language learning might be effective as well. Then, Harris & Wambeam’s
(1996) experimental study provide a solid evidence for the inference. The authors
tried to address the issue of how a combination of both text-based ACMC (e-mail) and
text-based SCMC (MOO) activities enhanced participants’ opportunities to improve
their writing skills. Aside from participants’ writing skills, learners’ attitudes toward
text-based CMC learning context were carefully observed through the three
Likert-styled questionnaires. The authors divided the participants who enrolled in the
university first-year writing classes into three groups; two experimental groups were
assigned to have synchronous (on Diversity University MOO) and asynchronous
(Internet journals) communication over the Internet. As for the control groups, 17
university students were involved and follow the same syllabi as the experimental
classes without engaging in the any CMC and participating in MOO sessions. In line
with other literature reporting the effect of text-based SCMC and ACMC on language
learning respectively, the result suggested that the experimental groups were more
effective than the traditional experience of the control group. The frequency of
participation was much higher for the experimental group (78%) than for the control
group (18%); that is, the combination of the two modes of text-based CMC also led to
more contributions toward the online journals and brought on learners’ enjoyment of
writing.
Concerning with the comparative effect between text-based SCMC and
text-based ACMC on language learning, Sotillo (2000) conducted an experiment with
the two modes of text-based CMC. In Sotillo’s exploratory study, two instructors and
25 students from advanced ESL writing classes were involved and completed four
synchronous discussions and posted 105 messages to an asynchronous threaded
discussion about reading assignments. Through the collected data of learners’ output,
the author quantitatively and qualitatively analyzed discourse functions and syntactic
complexity between the two modes. In terms of the discourse functions, the results
showed that more greetings, imperatives, requests for clarification and information,
and adversarial moves were included in SCMC data included, while primarily topic
initiation moves, questions, student responses to questions and comments about
postings were contained in ACMC data. Then, in respect of syntactic complexity,
t-units, error-free t-units, total number of words, subordinate clauses, total embedded
subordinate clauses and t-unit length were adopted as measurements. For synchronous
discussions, on one hand, Sotillo concluded that teacher’s domination of discussion
was significantly decreased in SCMC, and that the specific topics of interest
embedded in the multiple threads of SCMC mostly engaged learners’ attention. Also,
more attention was paid to fluency than to grammatical accuracy and the
conversations tended to become more informal. On the other hand, during the
asynchronous discussions, student output was lengthier and more syntactically
complex, which seemed to facilitate the structure as well as meaning of the target
language. The researcher concluded that both modes of text-based CMC could be
exploited for different pedagogical purposes, enhancing the language acquisition
process accordingly.
Subsequent to Sotillo’s exploratory research, Abrams (2003) also investigated
different effect of text-based SCMC and ACMC on improving oral performance in
German, adding the face-to-face communication as a control group. 96 students in a
Germany university were divided into three groups: the first group (control group)
had regular class face-to-face activities with no CMC treatment; the second group
engaged in text-based SCMC whereas the third participated in text-based ACMC.
Slightly different from Sotillo’s qualitative as well as quantitative study, Abrams
conducted a quantitative research, making use of various linguistic features such as
lexical richness, lexical density, syntactic complexity, and amount of language output,
which can reflect the increasing of communicative sophistication to evaluate
participants’ development of oral proficiency. The result showed that text-based
SCMC group outperformed those in the control group while text-based ACMC group
indeed produced significantly less output than their peers. The surprising outcome
indicated that participants in text-based ACMC were less motivated to participate in
the discussion perhaps due to the delayed nature of the interaction.
While the comparison between the effect of text-based SCMC and ACMC start
to unfold, the comparative effect of voiced-based SCMC and ACMC on language
acquisition, which may especially contribute to the development of oral proficiency
and possibly benefit pedagogical process , is still far from being unveiled.
2.3 Learners’ Perception of CMC
An early development in the history of student empowerment in language
learning was the claim of L.S. Vygotsky (1978) that social discourse between people
causes the growth of “inner voice” that offers people sense of mastery and that plays
an extremely important role in language learning. Besides, it is also generally believed
that second language is best learned through active negotiation of meaning (Ellis,
1996). Therefore, collaborative learning has been constantly associated with a process
of empowerment by educators and psychologists (Belmont, 1989; Bruner, 1987;
Cooper and Selfe, 1990; Kollock & Smith, 1996). Many researchers agree that along
with the mechanics in computer-mediated communication, learners will have much
better opportunities for initiating and controlling their own language learning
(Harasim, 1986; McComb, 1993; Santoro and Kuehn, 1988). The computer learning
network facilitates language learners to collaborate with each other in dealing with
different learning projects in single classroom or in various setting around the world.
Speaking of the blessing part of CMC, accordingly, many recent studies have
suggested the following virtues for CMC: (a) increasing amount and equality of
participation (Chapelle, 1994); (b) increasing the quantity of learner output (Kitade,
2000); (c) increasing quality of learner output (Beauvois 1998; Kitade, 2000).
Warschauer’s (1996) simply classified advantages that CMC brings in as “student
autonomy, equality, and learning skills” (pp. 3). These blessing strengths are closely
related to learners’ positive perception of the new language learning environment,
CMC.
In terms of student autonomy, abundant studies have suggested that participants
in CMC feel more involved in the development of ideas, in determining the way to
discuss topics, and in selection of the topics they are interested in (Baynton, 1987;
Garrison and Kelm, 1992; Kern, 1995; Lawrence, 2002; Ortega, 1997). In other words,
CMC has been positively recognized as having the ability to improve foreign
language learning, presumably by creating a more harmonious and collaborative
learning context. Lawrence (2002) observed a Canadian secondary French core
classroom where e-mail exchange with the other two French-speaking partnered
schools through student-designed online newsletter was used on second language
learning. Lawrence concluded that because students were allowed to post topics in a
great variety and had to cooperate with multi e-mail partners, they showed greater
interest in networked based discussion. Apart from Lawrence’s study, Yeung (2002)
collected learners’ opinion on asynchronous use of CMC, indicating that most learners
tended to believe that CMC components offer flexible learning context and claimed
that the flexibility indeed improved the level and quality of their learning. Learners
who engaged in asynchronous CMC such as e-mail or electronic bulletin board,
without the traditional limitations in class time or office hours, consequently, can play
the role as initiators to open up the discussions in anytime and any place with their
teachers or with their peers, which resulted in greatly increased both
student-to-teacher and student-to-student interaction (Harasim, 1986; Hartman et al.,
1991; McComb, 1993) and leaded to further opportunities for learners’ self
expressions (Kinkead, 1987).
Moreover, learner-to-learner exchanges in CMC appear to be more interactive
(Blake, 2000; Darhower, 2000; Kern, 1995; Warschauer, 1996) than learner-to-teacher
exchanges. Kern (1995) also indicated that direct student-to-student interaction could
greatly stimulate students’ interests in discussion and decrease students’ reliance on
teachers. He further showed that the most important of all, learner-to-learner
interaction can reduce communication anxiety because of the similar language
proficiency and nearly equal social status. In Kern’s open-ended questionnaires, quite
a few participants reported that they feel freer in the informal setting with peers as
interlocutors. In comparison with learner-to-teacher exchanges, Chun (1994) found
that second language students were in favor of the direct interaction with peers instead
of the teacher or the moderator in the networked text-based forum. In addition, the
results suggested that learners appear to display more interactive competence in the
context of CMC than in traditional face-to-face discussions. Although
learner-to-learner exchange is considered promising and brings about greater
interaction between learners, some classroom practitioners and researchers worry that
learner-to-learner discussion, especially participants belonging to the same level of
language proficiency, will reinforce non-target-like language, or “the blind lead the
blind” (Blake, 2000, p133). That is, learners may encounter the “blind” danger,
suffering from the poor quality of language without the grammatical correction from
teachers who are supposed to have superior knowledge of the target language (Kern,
1995). However, Porter (1986) pointed out that error incorporation in the data of his
study was quite rare. Furthermore, although Blake (2000) admitted that the ‘blind’
danger warrant close attention, he also reported that in the examination of all 50
participants’ transcripts on text-based CMC, no ill-formed language was acquired by
learners in their incidental negotiations. Even in Swain’s experiment (1998), she
stated that less than ten percent of the total negotiations recorded in his study
contained the transmission of incorrect solutions.
In subsequent years numerous studies were carried out on the effect of
synchronous CMC, reported that using synchronous conferencing leads to a shift in
authority from instructors to learners, which resembles the advantages of ACMC in
development of learners’ autonomy. In Pratt and Sullivan’s (1994) comparative study
between traditional oral discussion and electronic one done in EFL classroom at the
University of Puerto Rico, they discovered that approximately 85% of the
conversational turns were taken by learners in the CMC group, while learners only
took in charge of 35% of conversational turns in the traditional oral discussions.
Similar reports were made by Keln’s (1992) that up to 92% of comments in computer
conferencing in a Portuguese class were made by students themselves. Following
Keln’s research, Chun (1994) conducted a large-scale experiment on 14 computer
conferencing sessions in university German classes. The analysis showed that
approximately 88% of students’ comments and questions were directed to their peers
and only the remaining 12% were directed to the instructor, manifesting the fact the
student autonomy increases the interaction between peers in synchronous CMC as
well as in asynchronous CMC. Different from the traditional language learning
settings, the expansion of students’ autonomy in CMC context provides learners with
more independent opportunity to unfold the discussion on topics corresponding to
their interest. Furthermore, with the network communication, learners have the access
to carry out long distance communication, using the foreign or second language for
authentic communication with other language learners or native speakers throughout
the world, resulting the greater student enthusiasm, initiative, and personal
commitment (Paramaskas, 1993).
Besides students’ autonomy, several studies (Finholt, Kiesler and Sproull, 1986;
Selfe and Meyer, 1991; Sproull & Kiesler, 1991) have found that because learners
favored the atmosphere created in CMC, where everyone is treated fairly, more
balanced and equal participation of language learning among students in the context
of CMC. The computer as a new medium for conversation reduces the bias from
“static social contexts” (Warschauer, 1996, p5) such as gender, race, handicap and
different social status that might negatively influence the equal participation in
interaction during face-to-face conversation in the traditional classroom setting.
Flores’s (1990) and Selfe’s (1990) studies, for example, suggested that in the context
of CMC women’s participation in classroom discussion be equalized compared that of
men who generally are supposed to speak and interrupt more (Spender, 1980).
Moreover, Korenman & Wyatt’s (1996) questionnaire study also showed that CMC
occurred across social and hierarchical institutional boundaries; tenured and
untenured faculty, students, administrators all made great contributions to the
discussion forum because of the unique and nearly anxiety-free environment.
In regard to “dynamic social context” (Warschauer, 1996, p5) in the traditional
classroom, facial expressions like frowning and inactive attitude toward the
conversation such as hesitation may hinder introverted learners from expressing
themselves freely when they sense that their comments are being evaluated (Finholt,
Kiesler and Sproull, 1986). Contrarily, CMC allows extroverted as well as introverted
learners enough time and space to make personal contributions to the discussion in
language courses, waving the enormous pressure in the traditional classroom where
students who answer questions quickly and correctly tend to be favored (Tella, 1992).
Consequently, CMC is believed to have the ability of neutralizing the effect from the
dynamic social context and classroom (Warschauer, 1996). Several studies have
reached the agreement on the effect of CMC on developing learners’ equality in
language learning. In Pratt and Sullivan’s (1994) study, overwhelmingly 100% of
students participated in the CMC groups whereas only half of students got involved in
face-to-face discussion groups. Kern (1993) likewise found that compared with all
students participated in the CMC groups, five students dominated the face-to-face
discussion and the other four students did nothing during the period of discussion.
Moreover, Kroonenberg (1995) also pointed out that shy students were likely to refuse
speak up in the traditional class but tended to have ‘fingers flying across the
keyboard’, becoming more willing to participate in the subsequent oral discussions on
the same topics.
So far, learners’ general perceptions of CMC, either in ACMC or SCMC and the
comparison between learners’ perceptions of face-to-face traditional class and those of
CMC have been widely examined. Nevertheless, few studies focus on clarifying
learners’ possibly different perceptions of voice-bvased ACMC and SCMC. That is,
according to the literature, in the comparison between the two modes of CMC, the
development of learners’ language proficiency in stead of learners’ attitudes toward
SCMC and ACMC tends to be one and the only main target to be explored. Therefore,
the present study not only examines the effect of SCMC and ACMC on the
enhancement of learners’ linguistic competence but also investigates learners’
perception of the two modes of CMC.
2.4 Research Questions
On the basis of the literature just reviewed, the effect of CMC on enhancing
language learning, especially in respect of oral proficiency, has been positively
recognized when compared with the face-to-face discussion. In addition, some studies
further indicate that in terms of oral proficiency, learners in synchronous text-based
CMC seemingly tend to outperform those in asynchronous text-based CMC (Abrams,
2003). Consequently, it is hypothesized in the present study that in line with
text-based CMC, even in the voice-based CMC, learners in the synchronous context
would outperform those in the asynchronous context. Moreover, in terms of learners’
affective domain, while there is an assumption that learners express positive attitudes
towards the use of text-based CMC (D., Coniam & R., Wong, 2004; Hsu, 2005), little
empirical evidence has been found to focus on users’ attitudes toward voice-based
CMC. Therefore, the effect and learners’ attitude about voice-based CMC will be
taken into consideration within the present study. The following are the three research
questions.
1. Is EFL learners’ general oral proficiency significantly enhanced after the treatment
of voice-based CMC?
a. Is EFL learners’ oral proficiency in terms of pronunciation and fluency
significantly improved after the treatment of voice-based CMC?
b. Is EFL learners’ oral proficiency in terms of the accuracy of grammar and
content significantly improved after the treatment of voice-based CMC?
2. Is there a significant difference of the effects that voice-based SCMC and ACMC
had on EFL learners’ general oral proficiency?
a. In terms of pronunciation and fluency, is there a significant difference
between voice-based ACMC group and SCMC group?
b. In terms of the accuracy of grammar and content, is there a significant
difference between voice-based ACMC group and SCMC group?
3. What is EFL learners’ perception of voice-based CMC?
