Citation
Comparing the Language of Intermediate Learners of French in Asynchronous Electronic Communication vs. Face to Face Communication

Material Information

Title:
Comparing the Language of Intermediate Learners of French in Asynchronous Electronic Communication vs. Face to Face Communication
Copyright Date:
2008

Subjects

Subjects / Keywords:
Classrooms ( jstor )
College students ( jstor )
Computer mediated communications ( jstor )
Electronics ( jstor )
Foreign language learning ( jstor )
Language ( jstor )
Nonnative languages ( jstor )
Target audiences ( jstor )
Television commercials ( jstor )
Words ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
4/17/2006

Downloads

This item has the following downloads:


Full Text












COMPARING THE LANGUAGE OF INTERMEDIATE LEARNERS OF FRENCH IN
ASYNCHRONOUS ELECTRONIC COMMUNICATION VS. FACE TO FACE
COMMUNICATION















By

CYRILLE GUILLO


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF ARTS

UNIVERSITY OF FLORIDA


2005















ACKNOWLEDGMENTS

I thank Dr. Antes and Dr. Lord for their constant support and greatest patience.

I thank Dr. Shoaf for suggesting and letting me use the language laboratory.

I thank my parents for always loving me and being supportive of me even though

we were so far apart.

I thank my grand-parents as well for their encouragement and for having been so

good to me all my life.
















TABLE OF CONTENTS

page

ACKNOWLEDGMENTS .................................................. .......................... ii

L IST O F T A B L E S ............................................................................. ................ ......... iv

L IST O F FIG U R E S ............................................................................ ....................... v

A B ST R A C T ..................................................... ...................................................... .. vi

IN T R O D U C T IO N ........................... ...............................................................................

PREVIOUS WORK......................................................................................................

M E T H O D O L O G Y ............................................................................ ...........................12

R research Q uestions............................................ ............................................... 12
P artic ip an ts ................................................................................. ........................... 12
T ask ............... ............ ........... ....................................................... ............. .. 13
A analysis .............. .............. ....................... ........ 15

RESULTS AND CONCLUSIONS ................................................... ......................17

Syntactic Complexity...............................................................17
L exical C om plexity ................................................................... ....................... 29
Im p licatio n s ................................................................................ ...........................3 3

C O N C L U SIO N ......................................................................... ....................................38

PROJECT DESCRIPTION.............................................................41

FORUM SCREENSHOTS ............................................................... .......................45

LIST OF REFERENCES ......................................... ................................................48

BIOGRAPHICAL SKETCH ..................... ....................................50
















LIST OF TABLES

Table page

3.1 : Summary of findings according to the CI................................. ......................19

3.2 : Total num ber of threads............................................................. ........................ 20

3.3 : Sum m ary of the TTR ............................................................ ...................31
















LIST OF FIGURES

Figure page

1: Screenshot of the forum homepage. ........................................ ....................... 45

2: Screenshot of the Project threaded discussions. ...........................................46

3: Screenshot of a thread including (1) emoticons (2) avatars and (3) personalized
signatures.................... ............................................. 47















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Arts

COMPARING THE LANGUAGE OF INTERMEDIATE LEARNERS OF
FRENCH IN ASYNCHRONOUS ELECTRONIC COMMUNICATION VS. FACE TO
FACE COMMUNICATION


By

Cyrille Guillo

December 2005

Chair: Theresa Antes
Cochair: Gillian Lord
Major Department: Romance Languages and Literatures

As technology permeates in the foreign language classroom, teachers have to

determine whether any technology they intend to use can be beneficial to their students.

Electronic forums are a recent technology that teachers may consider using. This

research's purpose is to compare students' language production on an electronic forum to

face-to-face communication. A study group used an electronic forum outside of the

classroom to complete a project based on the cognitive approach; at the same time a

control group performed the same task using in-class face-to-face communication. The

first research question of this study aimed at comparing the grammatical complexity of

the language produced by two groups of students. One group produced their language on

an electronic forum and the other communicated orally. In order to compare these two

groups' grammatical complexity, the Coordination Index (CI), which compares the

number of dependent clauses over the total number of clauses, was compounded. The









results for this variable suggest that the language produced on an electronic forum tends

to be more complex than the language produced orally. The written nature of the

language used on the electronic forum accounts for this result. However, the CI does not

take into account the types of dependent clauses and the data revealed that there was a

similar number of completive clauses in the language produced with both media.

The second research question of this study aimed at comparing the lexical

complexity of the language produced using both media. In order to compare the

language complexity, the Type Token Ratio (TTR), which measures the number of

different words over the total number of words, was used. While it was expected that the

TTR of the language used in the electronic forum would be higher than the one used in

the F2F environment, it was not the case for this research. Indeed, the TTR for both

media were almost identical. Thus, these data suggest that more research be performed in

order to compare both media again with data coming from various tasks.

The third and final question of this research aimed at whether other patterns could

be discovered from this type of data. First of all, it revealed that students display

interesting behaviors on an electronic forum. The first student to write a thread is more

likely to become the leader of the group and the last student to write a thread is more

likely to be the student who will participate the least. Students reported that they did not

enjoy working on the electronic forum claiming that it was not convenient for negotiation

and, indeed, most of them ended up using other media for communication. On the other

hand, most students in the F2F group reported that they enjoyed communicating in the

language laboratory. They enjoyed the new environment and felt that negotiating

meaning among other aspects of the task was beneficial.














CHAPTER 1
INTRODUCTION

Technology in the foreign language classroom is not a new phenomenon. The first

teacher to have used a tape player or a tape recorder brought technology into her/his

classroom. However, when one speaks of technology in the foreign language classroom

nowadays, one speaks of CDs, DVDs, computers and the most recent global phenomenon

associated with computers, that is the World Wide Web. New technology is exciting and

there will always be a teacher or a researcher who will try to incorporate it into his or her

teaching practices. The difficulty becomes to optimize the use of this technology so that

its benefits outweigh or supplement existing practices. Such has been the case for

Computer Mediated Communication (CMC). CMC is the use of a Wide Area Network

(WAN) or a Large Area Network (LAN) to allow communication between two or more

computers. CMC can take the shape of students chatting together from computer to

computer in the same room, such as is used in language laboratories, or it can be students

communicating from remote locations. CMC can be synchronous when students chat

simultaneously with the use of a chatting interface such as instant messengers (AOL

Instant Messenger also called AIMTM, MSN Messenger or ICQTM) or with the use of

voice over IP (Internet Protocol), which allows for voiced conferences over the Internet

thanks to software like Netmeeting TM. Synchronous CMC via chat software displays

similar characteristics to Face-to-Face (F2F) communication. For instance, there is an

important amount of turn-taking and turns are short in general (Warschauer 1996,

Beauvois 1996). CMC can also be asynchronous, meaning that every message will have









a delay and that people will be able to have access to those messages in their own time.

Asynchronous communication is typically associated with written communication as

users can take the time to organize their writing before making it available to its recipient.

Examples of asynchronous CMC include email, email lists (or newsgroup or listservs) or

the use of electronic forums also called Bulleting Boards (BB). Electronic forums differ

from other asynchronous communication as they archive and thread all writing so that

participants can access, select and retrieve any written message at any time, regardless of

the topic or when it was written as long as the discussion is stored on its host. Messages

on electronic forums can be displayed chronologically but, most commonly nowadays,

they are displayed first according to topic and then chronologically (see appendix B).

The present thesis intends to examine the use of an electronic forum to complete a class

project as opposed to a F2F alternative. The goal of this thesis is not to determine

whether one approach is better than the other. Both approaches have advantages and

disadvantages that influence students' performance differently. Therefore, the goal of

this thesis is to compare students' performance using both approaches so as to provide

pedagogical input for teachers. Thus, a project was designed and administered to two

classes of students of intermediate French at the University of Florida. One class was

asked to use a bulletin board software as means of communication while the other

communicated orally in class. The two classes language production were compared

using the Coordination Index (CI) and the Type Token Ratio (TTR) as variables. The

results and a more detailed explanation are presented below.














CHAPTER 2
PREVIOUS WORK

The following articles have been selected from a more complete list concerning

technology in the classroom because of the limited amount of research regarding the use

of bulletin board software as a means of communication. In order to compare F2F and

Electronic discussion, Warschauer (1996) asked the following questions:

1) do second language students participate more equally in small group discussions
held electronically than those held in a traditional F2F manner? 2) if so, who
benefits from this more equal participation? In particular, how are differences in
participation for a F2F mode or an electronic mode related to factors such as
gender, nationality, and age and language proficiency? 3) what are students'
attitudes toward participating in electronic and F2F discussion and how do these
attitudes correlate with changes in amounts of participation? 4) does electronic
discussion include language that is lexically or syntactically more complex than
F2F? 5) what other differences are noted in the language use and interaction style in
the two modes? (p. 10)

Warschauer studied the language production of 16 students of various ages and

nationalities enrolled in an advanced ESL course in a community college in Hawaii. The

F2F data was transcribed and all the transcripts were entered in the Computerized

Analysis Program (CLAN) of the Child Language Data Exchange System (CHILDES);

which was used to count the number of words per speaker and to calculate the Type

Token Ratio (TTR). The transcripts were analyzed to calculate the number of clausal

coordinations and subordinations. He found that there was an increased participation in

the computer mode. The language was also more complex in the electronic mode when

comparing the CI and the TTR. The turn-taking in the conversation mode was more

numerous with short turns and many confirmation checks. The computer exchanges









displayed less direct levels of interaction and students expressed their own ideas as

opposed to directly answer questions. The electronic communication showed more

formal expressions such as transition words. In his conclusion, Warschauer suggested

that further research be performed according to nationality, according to the speaking

fluency of the learners.

This article is relevant for the present research as it presented the two variables that

were used in this study. In other words, it presented the coordination index, which

provides information on the complexity of sentences and the Type Token Ratio, which

provides information on lexical complexity.

Beauvois (1997) examined the affective and social benefits that students can derive

from LAN (Large Area Network) communication. The purpose of her research was to

examine, in controlled conditions, whether a link between written synchronous

communication, via the DaedalusM software, and oral communication could be

established. She wanted to measure the transfer of skills that operated. Her participants

consisted of 83 fourth semester students of French at the University of Tennessee in

Knoxville (UTK) in 1995. The students were randomly assigned with half the class in a

PC laboratory and the other half in a regular classroom. There were 49 females and 34

males. Classes were instructed using a communicative approach. Students were assigned

the completion of tasks based on their readings from the Petit Nicolas a children's book

written by Andr6 Gosciny. At mid-semester and again at the end of the semester, all

students took an oral examination, the grading of which was used to compile a T-test

used to compare the study group and the control group. She found that the LAN group

exceed the control group in grades and that not only was the LAN group better but also









more homogeneous. According to Beauvois, the synchronous communication carried

elements similar to conversation such as a high amount of turn-taking. Furthermore, she

posited that there was a compelling character of the PC message with flashing visual

prompts, which kept students focused. She also noticed that Vygotsky's scaffolding

theory applied to the synchronous communication, that is to say students benefited from

one another's input thus creating a peer teaching environment. Thus, she concluded that

there was a significant amount of transfer of skills from one medium to the other and

suggested that further research be performed at different levels of instruction in addition

to investigating other language skills.

The article is relevant to the present research as it offered several suggestions. First

of all, the variables she had chosen to study were interesting as using grading as a

variable to measure student achievement was considered. One of the problems that

grading pofses is subjectivity. Furthermore, grading cannot provide an accurate

illustration of the type of language that is used in both media. Therefore, the idea of

using grading as a variable was rejected for the present study.

Boyd, Davis, and Ralf Thiede (2000) examined what happens to different features

of discourse when English as Foreign Language (EFL) learners must choose to function

in an ESL situation as shown by the changes that occur in their writing in asynchronous

electronic forums. Thus, Davis and Thiede studied the style shifting of students of ESL

in response to native speaking participants' accommodation to the experience of creating

a learning community online. By pairing native speakers and ESL learners the

researchers tried to emulate mimetism. They found that investigating social practices

such as politeness, authority status or distance is not always simple. Students









participating in asynchronous communication present themselves exclusively in a

positive and polite manner. They can enter the conference at any point, read as little or as

much as they wish, and choose to reply to whomever they want. The conventions on the

forum included exaggerated politeness and signals of approbation with compliments that

showed alignment more than partisanship. In order to measure the replication of

language researchers studied "lexicosyntactic indicators of stylistic emulation." In other

words, they looked at an acquisition scale and at lexical density, which they defined as

the number of lexical words divided by the total number of words.

The two variables that they used were extremely interesting. On the one hand, the

acquisition scale was not applicable as measuring acquisition was not the goal of the

study and native speakers were not introduced as a factor. On the other hand, the

definition for the lexical density was the exact same one as the definition for the Type

Token Ratio that Warschauer (1996) had used. In Davis and Thiede's case, however, it

was used to measure the production of language learners versus that of native speakers on

the same medium, i.e., bulletin board software, which was used for the present study.

In her article, Pellettieri (2000) studied the interaction and the negotiation of

meaning in synchronous CMC. Her research questions included:

1) does the negotiation of meaning occur in task-based synchronous CMC? 2)do
the negotiations facilitate mutual comprehension? 3) do the negotiations push
learners to output modifications that are both meaning and form-focused? 4)do the
negotiated interactions foster the provision of corrective feedback and the
incorporation of target-forms in the subsequent turns? (p. 64)

Her participants were 20 students of Spanish at the University of California at

Davis. They were all native speakers of American English enrolled in intermediate

Spanish. Students' interactions were observed as they functioned in dyads. She found

that negotiation via synchronous CMC facilitated mutual communication and that









negotiations pushed learners to output modification encouraging corrective feedback.

Her conclusion offered pedagogical suggestions on tasks to be designed so that all

participants are required to request and obtained information from one another for

successful task completion so that communication is goal-oriented.

This article offered the perspective of performing a research from a theoretical

point of view. Pellettieri chose to perform her research from the interactionist point of

view. The present research was based on the Cognitive point of view as Skehan (2001)

defines it. The Cognitive approach also calls for goal oriented tasks and includes some

interactionist elements. It is described in greater detail below.

B6hlke (2003) designed his study to verify that participation in a CMC is more

equalizing than F2F participation as was suggested by Kern and Warschauer (2000)

whom he cited. His participants were fourth semester students of German as a foreign

language using a communicative approach. The students participating in chats and F2F

produced discourse from two activities presented on worksheets. Half of the students

used CMC and the other half F2F communication. Contrary to Warschauer who used

broader units of meaning, called T-units, in order to measure students' language

participation, B6hlke used C-units that he considers to be the fundamental elements of

communication, i.e., a C-unit can represent one word only or a whole sentence.

Consequently, the C-unit does not require a verb nor does it require a predicate, it is more

inclusive than the T-unit. He found that group size has to be factored into the equalizing

effect of CMC. Groups of 5 students did not tend to be optimal whereas groups of 4

offered a positive impact. He also measured students' language competencies according

to a scale of stages defined by Tschirner. He found that CMC is indeed more equalizing









at certain stages of language than others. For future research, he suggested that the chat

room should set the ground work for in-class discussion, and that "more research should

be performed on the ideal number of students within a group as well as further

investigated Tschimer's stages.

His discussion on the C-unit was extremely interesting, however, using the C-unit

may be too encompassing as it includes utterances of only one word such as "yes" or

"no" as a unit of meaning. Particularly since in F2F communication turn-taking is much

more important, and using C-units as a construct would create an imbalance with CMC

communication. Therefore, it was necessary to perform the present research on units that

would include a verb and a predicate. Tschimer's stages are also interesting, however,

they only apply to the German language and such a scale was not found for a research

using French. B6hlke claims that chat room should set the ground work for in-class

discussion, however, doing so seems redundant. How useful would it be for students to

carry the same conversation twice? His approach to literature in CMC was also

interesting. Indeed, he was the first to offer opposing views to most of the literature.

For instance, he offered an alternative explanation for the transfer of skills that Beauvois

had suggested, expressing "doubts that the implementation of chat does not imply

transfer of skills since there is no immediate cause and effect relationship with increased

speech proficiency" (p. 70). B6hlke also voiced concerns for using CMC. Such concerns

were expressed by Beauvois (1992) who claimed that students become increasingly

indifferent to the appropriate use of the target language the longer they use the chatroom.

Yet, the present research by Guillo also showed how some students in F2F started

speaking in their native language in the language laboratory. Therefore, it is a concern to









be had in both media. He further quoted Kelm who saw a disadvantage "when students

copy incorrect form from another student's message" (p. 71). It is a right claim although

not only true to CMC since the lack of feedback both from a peer or a native speaker may

lead a learner to acquire an erroneous form of language. Finally, he quoted Bremp

(1990) "it gets frustrating sometimes when a conference gets really busy and you would

have no time to type anything if you worried about reading absolutely everything" (p.

72). Although chatting takes on a form that allows one to realize that we may not be fast

enough or have enough time to read every thing when we are communicating, it is also

true to the oral language. It is not suggested that these concerns should be dismissed for

the present research, on the contrary, they may be used for all communication purpose.

As far as the present research is concerned, an electronic forum was preferred to a chat

software.

Ann, Chenoweth, and N. K. Murday (2003) noticed that students at Camegie-

Mellon Universtity were interested in taking a foreign language course but could not do

so because of scheduling issues. Therefore, they designed an online class to meet those

needs. Thus they wanted to measure whether this online course using CMC would be as

efficient as a regular course. The participants for this research included students enrolled

in French 1. They were all undergraduate students, 12 of them participated in the F2F

evaluation and 8 participated in the online course. On the SAT II French exam, the

online participants averaged 581 points and the F2F participants averaged 556. They all

filled out a General Background Questionnaire (GBQ) and a Technical Background

Questionnaire (TBQ). The students' oral production was measured by interviews

conducted after 5 weeks and at the end of the semester. The researchers gathered data









from a focus group and an interview from one student. They found that in all the testing

done there was no statistically significant difference between the two groups. Thus,

students registered in the online course made similar progress but they offered more

mitigated satisfaction feedback than students in the conventional course.

As with the research done by Beauvois, the manner in which students' performance

could be used as a variable to compare two media was intriguing. To solve the question

of subjectivity, Chenoweth and Murday had several impartial graders. Doing the same

was not feasible within the limits of the present research. Therefore using grades as a

variable even with impartial graders was abandoned.

This literature review offers information on how the use of technology can be

compared to conventional teaching practices. Most of the articles dealt with the use of

synchronous communication as opposed to F2F communication. The researchers'

interests varied from theoretical approaches to systematic statistical research. In the end

their concern is the same, how efficient can CMC technology be in the language

classroom?

Overall the current literature did not present many articles on electronic forums or

other asynchronous communication. No article offered any comparison between

electronic forums and F2F communication. That is because electronic forums are

asynchronous communication and F2F is synchronous. This fundamental difference

makes them difficult to compare. Yet as asynchronous communication can often be

presented as an alternative to in-class F2F communication, it is important to examine how

different those two media are when student language production is concerned.






11


Understanding these differences will allow teachers to make appropriate pedagogical

decisions about which medium to use and how to use them.














CHAPTER 3
METHODOLOGY

Research Questions

We know that CMC is different from F2F communication. Additionally, we know

that asynchronous communication is different from F2F communication. However, we

do not know to what extent these modes of communication are different nor do we know

if there exist similarities between them for foreign language learners. For these reasons,

the following research questions were asked:

1. How does language produced in an electronic forum compare to that of F2F
communication in terms of grammatical complexity?

2. How does language produced in an electronic forum compare to that of F2F
communication in terms of lexical complexity?

3. What patterns, if any, can be noted from both modes of communication?

Participants

The participants in this study consist of two classes of students enrolled in a French

grammar class at the intermediate level at the University of Florida. The University of

Florida code and title for this class is: FRE 2200 Intermediate Grammar. The project

was a class assignment but each student volunteered for the research and signed an

informed consent form approved by the University of Florida Institutional Review Board

02 under the protocol number 2005-U-0295 for use through 06/4/2006. All the data was

kept anonymous. There were 22 students who participated in the protocol, 17 of whom

were women. No other ethnographic data was requested. The textbook used in FRE









2200 was Interactions; chapters 4 and 5 of this textbook provided the language and

cultural impetus for the activity used in this study and detailed in Appendix A.

Task

The activity was created with Skehan's model in mind (Skehan 2001). At the

center of Skehan's Cognitive approach is the action of "Noticing," which first integrates

input into the "working memory" and into the "long term memory." "Noticing" occurs

when input takes on several qualities, namely frequency and salience. The teacher's task

is to create focused input. A particularly interesting aspect of Skehan's model is that it

takes into account learners' internal factors (readiness and individual differences), forcing

a teacher to also take the various learning styles into account.

The application of this model in the classroom requires task-based instruction. In

other words, the CMC activity will follow these guidelines provided by Skehan (2001 pp.

121-152):

* Meaning is primary;
* There is some communication problem to solve;
* There is some sort of relationship to comparable real-world activities;
* Task completion has some priority;
* The assessment of the task is in terms of the outcome.
Furthermore, Skehan (2001 p. 152) specifies that tasks:
* Do not give learners other people's meanings to regurgitate;
* Are not concerned with language display;
* Are not conformity-oriented;
* Are not practice-oriented;
* Do not embed language into materials so that specific structures can be focused
upon.
With all these guidelines in mind, a series of communication-driven tasks for a

CMC activity was created (appendix A). The project is also designed to provide students

with an opportunity to use their language skills through negotiation as well as to develop

an understanding of French culture through the media. The media in question are









television and advertising, particularly how commercials fit into French television

programs.

Commercials provide an idea of the target audience the marketers are trying to

reach. The positioning of the advertising also provides information on this target

audience as it occurs at the particular time when this target audience is most likely

watching television. Some information the advertising provides about the target audience

includes demographics, such as age group, income bracket, geographical location,

education, values, etc. Once the marketer has drawn a sketch of the target audience,

he/she will identify the programs this target audience is the most likely to watch on

television in order to position the advertising so as to maximize the reach, i.e., the

proportion of the target audience that will be sensitive to the advertising message. For

students it translates into an exploration of French television programs and particularly

what French people watch and how this compares to US programming.

Students performed a CMC activity that helped them draw conclusions about

French programs as well as the make-up of the audience for these programs. According

to Skehan, an activity is more efficient if students are provided with an adequate

background and are given a role. For this activity, students were told that they were

members of a marketing company, The Famille Guillo Advertising Agency, whose CEO

was offering them an opportunity to lead a project related to the marketing of a product

and particularly, to the positioning of the advertising of that product. Appendix A

contains a detailed description of the project the students accomplished, for both the

CMC and F2F groups.









Analysis

The present research was performed using two different classes of FRE 2200. The

first class consisted of 15 students distributed within 4 groups, one group of 3 students

and 3 groups of 4 students, who were instructed to use the electronic forum to perform all

their communications. They will be referred to as the Computer Mediated

Communication group or CMC group. The second class consisted of 7 students who met

in the language laboratory twice in the course of the project during regular class sessions.

They will be referred to as the Face to Face group or the F2F group. Data from the CMC

group was retrieved directly from the electronic forum using the copy/paste functions.

They were then compiled in a Microsoft WordTM document in order to perform the

analysis of the data by hand on printed material. The data from the F2F was collected on

the computers of the language laboratory of the University of Florida using the DivaceTM

software. The data from this group consisted of audio files that was transcribed

afterwards on a written Microsoft WordTM document. Thus, this data was subjected to

the researcher's interpretation. Furthermore, on the second session of the project,

eroneous manipulations to save the data were performed by the researcher and the entire

session was lost. Therefore, only half of the data expected to be used was analyzed.

The data was analyzed in relation to two variables. The first variable was used to

identify the syntactic complexity of sentences uttered or written by students according to

a ratio called the Coordination Index (CI). Warschauer (1996) defined this ratio as the

number of dependent clauses divided by the number of total clauses. Further information

is provided in the "Results and Conclusions" section of the present thesis. The second

variable measured the lexical complexity displayed by students in the form of another

ratio called the Type Token Ratio (TTR). Warschauer defined the TTR as the amount of









different words produced divided by the total number of words produced. As for the CI,

further information about the TTR is provided in the "Results and Conclusions" section

of this thesis. Both ratios are similar to averages or probabilities. An average is the part

divided by the whole and a probability is the number of desired outcomes divided by the

total amount of outcomes. Thus both variables will provide us with a general idea of the

patterns, if there are any, present in students' communication as well as allowing us to

make predictions on such behaviors. Another advantage to both variables is that they can

easily be applied to both communication media. As previously stated, one medium is

asynchronous while the other is synchronous, therefore we expect differences. The goal

of this study is to measure these differences and derive pedagogical implications related

to these findings.














CHAPTER 4
RESULTS AND CONCLUSIONS

Syntactic Complexity

In order to determine the coordination index for the two types of samples that were

collected, I first removed all the utterances that were produced in the native languages of

the students. Second, the utterances that pertained to the task demands that requested a

written preparation, particularly the first part of the project, which consisted of providing

a summary and a description of the advertising were eliminated. Since students were

provided with guidelines in the form of questions, they created a written sample and since

the spontaneous language that students created was the object of this research, a written

sample was not desirable. For the CMC groups, it consisted of their first participation to

the thread for the most part. For the F2F groups it consisted of any part that was read

from their course preparation. Any single word feedback such as "oui, non etc.,"

repetitions, etc., were eliminated. Thus, only the most spontaneous language that they

produced while negotiating for the one commercial each group was going to analyze, the

type of program during which they would place their advertising, as well as any off-topic

discussion, was left.

Once, the utterances were narrowed down to usable samples, each clause of the

samples were outlined. The clauses were separated into two categories: Dependent

clauses (D) and Independent clauses (I). A independent clause is a clause that has a

meaning by itself. A dependent clause is a clause that has no meaning by itself. For

example, if we analyze the following sentence "I met her at the restaurant where we had









our first date" there is one independent clause, "I met her at the restaurant," which would

be meaningful by itself, and a dependent clause "where we had our first date," which is

meaningless by itself. In this sentence all the verbs are conjugated and the relative clause

introduced by "where" is easy to determine. However, dependent clauses also include

clauses where the verbs are not conjugated. These clauses include participial clauses and

infinitive clauses. No participial clauses in the samples collected were found, however

there were a few infinitive clauses.

Although attention was paid to the types of dependent clauses present in the data,

this analysis did not reflect these various types. In other words, the coordination index

does not discriminate between relative clauses, completive clauses or other subordinate

clauses. Instead, it offers an insight on the general syntactical complexity of samples.

The coordination index is a ratio of the amount of dependent clauses over the total

amount of clauses. Therefore, the higher the ratio the more complex the sample.

Once, dependent clauses are separated from the independent clauses, each were

counted separately and computed in an ExcelM chart according to the following formula

for the CI:

CI= number of dependent clauses/total number of clauses

Table 3.1 summarizes the results of this data. D represents the dependent clauses, I

represents the independent clauses and CI is the Coordination Index. The data is

presented on a per student basis, a per group basis, and finally the totals were

compounded. Each student was studied in his/her order of appearance for the threaded

participation or the oral exchanges. For every group a total, an average and a standard

deviation was calculated in order to identify individual patterns. Finally, the total,








19



average and standard deviation were calculated for the two different means of


communication so as to identify general patterns. The data highlighted in gray represents


the students whose production was also used for the TTR.


Table 3.1 : Summary of findings according to the CI
l I U1 01 11 Tol:l I -ll II F F DII F 1i Tolal I ,I


Gp 1 7 44 101 0 f6l
S 1
? 11 1 _1 -
Toilal'Gp i 1 0 00

STDCa :: : 0_ :i" i 1
op 2 4 27 63 90 0 300

6 1; 23 16 0 28
__ .i J () 1
Toall'Gp j 1 Jii 0 JO

STOie if'0 1) : 0 i :
,.p ? 16 0 421

11 1 1 1
Tol1d.'Gp -1 -. 1_ J 0

STI e' i0'1 I 0 :' ii 0
Gp i 12 36 61 91 0 371

11 .- '. 1 .
___ 1. 1 il i
T Inal Gp 1 14 ii -

STi I )i 1ii % i _ii 0100


Gp 1 2 1 dl E3 0 226
2 14 4S 62 0 226
3 12 40 c2 0 21
Toial:. p 1 1 i 0

STDes. 1i:: J : 0 00
1p 2 4 12 38 50 0 240
5 156 V 0 .2 6
6 ?2 22 65 0 492
'7 IIU I" I' IZI IIII, II ,)1
TlIOI Ii p II"0 1_ 1 I 0 -

STDeI. 1 : 1I 1. : I. ... ... I1 01 l


TotiI 2;i l2Il 2I:2 0.?3 1 T0o i 2"." ?J29 0., 2 l

STID \e, ..l.. \t.27. ,;i. ).0",? __STDe. ]._i.; l 1..4 ; 1,.i.i2 0.141


The last factor to take into account for this analysis is that the CI does not reflect


language accuracy. In other words, all clauses were counted, regardless of whether they


were grammatically correct. They may not have been introduced by the right pronoun or


the verb may have been incorrectly conjugated. Therefore, when the CI is determined


and compared, the tendencies for students to formulate complex sentences are compared,


without comparing their degree of language accuracy.


Interestingly, as the data shows for every group who used CMC except for group


CMC3, the first person to participate on the forum for each group is also the person who


provided the most data. CMC Student 1 has a total of 101 clauses, CMC student 4 a total









of 90, and student 12 has 97; each one of these students far exceeded the amount of total

clauses of all the other members of their group. This accounts for the very high standard

deviations among the total number of clauses and thus shows a very uneven distribution

of production. The total amount of clauses is roughly correlated to the total number of

threads that each student wrote. Table 3.2 below summerizes the amount of threads that

each students wrote:

Table 3.2 : Total number of threads
# of
threads
Gp 1 1 10
2 4
3 3
Total 17
Gp2 4 24
5 10
6 13
7 17
Total 64
Gp3 8 5
9 6
10 8
11 3
Total 22
Gp 4 12 9
13 4
14 9
15 4
Total 26


Not only did these students participate more and thus provide a larger quantity of

data but they also proved to be the decision-makers and the leaders of each group. They

were the students who prompted their peers to provide feedback. They were also the

motivators to solicit actual work and meet deadlines. Finally, after consulting with their









peers they made the final decisions as to which commercials and programs were going to

be used. Below is an example of student 12's motivational input:

-" Jepense que nous devons choisir ou la pub de Axe ou la pub de Peugot, les deux sontfaciles.
Mais ilfaut que nous choisisions unepub ASAP!!!!

I think that we must either choose the Axe commercial or the Peugeot commercial, both are easy.
But we must choose a commercial ASAP!!!!

The pattern was true to all the groups except for group CMC3. If we examine the

data a little closer we can see that the first student to participate was not the one who

provided either the most data nor the one who wrote the more threads. CMC Student 10

actually did, and CMC student 10 was the leader of her group and made the decisions in

her group. However, this was not always the case within this group. When we look at

the dynamics for this group we observe that CMC student 8, who actually started the

thread, was the initial leader. The leadership was then taken over by CMC student 10.

This data suggests that when a teacher offers a communicative task where students

will be participating on a forum, the first students of each group will be very likely to

become the leaders of their group. Furthermore, the teacher will also be able to expect

that the leader of a group will generally be the one who will participate the most.

What one cannot predict is whether the leader of a group will also be the student

whose language skills are the best. The coordination index is an average that represents

language complexity. In other words, it will show students' tendencies to express

themselves in a more complex form of language. Yet, of all the students that qualified as

group leaders only one, CMC student 1, had a higher CI then the rest of his/her group.

CMC students 4 and 10, with CI's of .300 and .357, respectively were even below their

group averages of .340 and .395, respectively. They were also below the class average

and means of .381 and .362, respectively. What does this suggest? For one, leadership









does not equal language complexity or skills. Second, it may suggest that leaders tend to

offer straight to the point communication.

Inversely, the last student to participate on the forum was also generally the one

who participated the least and thus provided the least data with the exception of CMC

student 7 of group CMC2. This an interesting predictor and, therefore, it is worthwhile

examining what could be the reasons for this pattern. B6hlke (2003) suggested a

correlation between group size and participation. For him, CMC has an equalizing effect

provided that the group size does not exceed five and not be lower than four. He also

suggested that more research be performed on the matter. From the present data one can

conclude that as long as an electronic forum is used as a CMC group size does not appear

to be relevant. Three groups comprised four students whereas one consisted of three

students. Even though group CMC1 consisted of three students, the last one to

participate, CMC student 3 hardly communicated and most of the work was done by

CMC student 1 and 2. In group CMC3 and CMC4 CMC student 11 and CMC student

15 also communicated far less then their peers with 27 and 18 total clauses, respectively.

Their CI was also below average with .333 and .389 respectively albeit not as much

below the average of the groups, which used CMC. Thus group size was not a relevant

predictor nor did it have an equalizing effect in the present data. The slightly below

average CI's of those students who participated the least may suggest that their language

skills may be a corrolary to explain their level of participation. Other factors may include

motivation for performing the task as well as their familiarity and motivation with the

means of communication. However, no questionnaire was offered to inquire about their

familiarity with the technology used, no question was asked about their affective









response to the same technology either. Therefore, no conclusion can be definitely drawn

at this time.

Leadership patterns can not be derived as easily for students who participated in

groups in the F2F tasks. In fact, none can be derived whatsoever. The first students to

participate in both groups did not display more of a leadership ability as the students in

the CMC tasks. Furthermore, it appeared that leadership was more diluted and more

democratic. All students of group 1 offered their opinion agreed together as to what

commercial was going to be used to complete the task. In group F2F2, two students, F2F

students 5 and 6, seemed to argue but the final decision was taken between F2F students

4 and 5. F2F Student 7 participated minimally as is shown in the table: he/she only

provided 7 clauses.

Although CMC students were specifically instructed to conduct all communication

on the forum, the researcher discovered that some of the communication was conducted

via other means such as telephone, email and oral in-class meetings. No data pertaining

to the amount of communication that was performed outside of the forum was collected,

therefore it is impossible to determine what percentage it represented. Testimonies in

students' participation suggested that they had actually communicated by other means.

Below are examples of threads suggesting other means of communication.

"Comme nous avons dit en classes, stade 2 et cd ajourd'hiu sont bons programmes pour passer
nitre pub. "

-As we decided in class, Stade 2 and CD aujourd'hui are good programs to position our commercial



"Je suis confondu de que notre groupefaitpour le rapportfinal. [...] quel est votre e-mail.
Quelqu'un s'il vous plait m'envoie un e-mail a [...] "

-I am confused about what our group should do for the final report. [...] what is your email? Please
someone, send me an email at [...]









This explains the discrepancy among all the groups that participated.
Students' behavior can also be analyzed by looking at other factors, such as the

amount of threads that they wrote combined with their CI. The Bulletin Board software

used provided data that included the amount of threads, the time and the date of each

thread and other statistical data. For instance group CMC1, CMC3 and CMC4 totaled

17, 22 and 26 threads respectively. Their total clauses were 136, 129 and 229

respectively and the average amount of clauses per person was 45, 32 and 57. On the

other hand, group CMC2 had 64 total threads for 288 total clauses and an average of 72

clauses per person. These figures give us some insight on the use of the forum for

communication. Either all groups managed to effectively make decisions, without much

negotiation or they used other means of communication to come to those decisions. The

latter is what happened in reality. As for group CMC2, they used the forum to

communicate often.

One of the consequences of participating on an electronic forum is the community

building effect. Many Internet businesses use that effect to create a community around

their products and also to create brand loyalty. Consequently, community software such

as an electronic forum has developed devices to create loyalty. Such devices are the use

of emoticons (smilies) and avatars personall pictures that appear on their profile every

time they post a thread) along with the necessity to register for a forum before

participating and the use of attractive Graphic User Interfaces (GUI). One of the goals of

the researcher was to create a community around the French class that he taught using the

Forum. Group CMC2 who displayed the necessary characteristics of community

building was a success. Their enthusiasm and their use of the forum was evidence of that

effect. They enjoyed identifying themselves with avatars already provided in the forum,









and they even actively sought additional avatars to include. They also registered on the

forum as indicated using pseudonyms that did not disclose their identity but identified

them through other traits. They also used emoticons extensively to express feelings that

cannot be expressed in the written form. Finally, they continued communicating on the

forum after the project was completed. Other students from the other groups did the

same but not to the extent of the students in group CMC2.

Finally, as we look at the averages for all the groups, we notice that the standard

deviations from the means are pretty high. The dispersion from the amount of total

clauses and the amount of dependent versus independent clauses as indicated by the

standard deviation, indicate a wide variety among those students. On the other hand, this

variety is not reflected in the average CI for all groups. This average is .381 with a mean

at .362 and a standard deviation of .093, which shows a certain homogeneity in the type

of utterances, which is to be expected as all the students were approximately at the same

level in the same class.

The amount of participation for the F2F groups cannot be measured by or with

turn-taking as on the forum. Indeed, many turns only comprise of one word, indicating

acquiescence or feedback, which is one of the fundamental differences between a written

and an oral performance. On the other hand, we can derive participation with the

amount of total clauses.

Group F2F 1 is homogeneous at every level. The CI of each student is very similar

and, as the standard deviation revealed, they all participated equally and produced a

language of an equal level of complexity. Their average CI is .228 and their mean is

exactly that as well, with a standard deviation of .003, which is very narrow. This group









displayed quite an efficient level of communication that would be labeled as "synergy" in

the business world. Synergy is the ability to create seamless communication for the most

efficient business practices. Although such homogeneity is ideal, it is also suspect for it

could be abnormal. Unfortunately, it is almost impossible to know accurately whether or

not it is natural, given the amount of data. Additional data would definitely help in

defining the statistical validity of this pattern. However, when we look at the data, the CI

for F2F students 4 and 5 of group F2F2 with .240 and .246 respectively are not far

removed from the average CI of group F2F 1, which is .228. In fact, they are very close,

which suggests that we could predict a certain homogeneity for the oral medium and that

we need to treat F2F Students 6 and 7 separately.

As the researcher was also the teacher for this class, he felt that there were certain

significant differences that were not measured. For instance it was striking that two of

the students did not participate as much as they did in a regular class setting. These

students may have been influenced by the Novelty Effect (Kern, Warschauer 2000). In

FRE 2200 at the University of Florida, students do not typically use the language

laboratory, thus creating motivation for students who find themselves in a new

environment while using new technology. In this case the language lab offered the

novelty, the technology and the new environment. The novelty effect predicts that

students will react positively to a new environment and that the effect will dwindle as this

environment becomes more and more familiar. In the case of the present research,

however, students did not have time to experience such familiarity, thus the novelty

effect carried over the duration of the task. On average, the researcher felt that all

students in general, and certain students in particular, participated more than they usually









did in class. The Novelty Effect of the language lab was also more equalizing than on the

forum.

F2F student 7 of group F2F2 was the only students who maintained the same level

of participation, which was quasi inexistent in general. This group did not show as much

homogeneity and its dynamics were far different. For instance, although F2F student 6 is

the student who shows the most complex language in the classroom, she was also the

students whose production was eliminated for the most part. Indeed, she produced many

utterances in her native language, i.e., English, and even though the researcher was

monitoring the class, he realized afterwards that she spoke English when he was not

monitoring her group. There were some attempts from her peers to make her speak

French. However, Group Think was not powerful enough to change her behavior. Group

Think is when a deviant member's behavior is reduced or stopped altogether under the

tacit influence of a group majority. The effect was first identified under the Kennedy

administration when dissenting politicians were stopped under penalty of being excluded

from the group. In a language laboratory, it is difficult for an instructor to maintain a

certain level of discipline for the simple reason that every group also requires help and

attention. That is when group dynamics can help. For the present research, groups were

randomly assigned. Perhaps assigning students according to criteria that include

language skills and personality types would have helped in assuring discipline.

F2F student 7 is a very interesting case as she is also the student that present the

highest CI of the oral groups, with .492. This number is extremely high compared to the

other students in the oral groups. Furthermore, when comparing this result with the

results of the CMC group, we notice that it is much higher than the average for all the









groups. It may suggest that this student has a tendency to express herself with more

complex sentences. A good way to measure if that is the case would be to compare her

native speech with the second language speech or/and to extract additional data from her

or his daily language production. Then we could also determine whether her lack of

discipline can be attributed to frustration or lack of confidence in her oral skills.

We expect the language in both media to be different. Whether synchronous or

asynchronous, CMC carries elements of the written language. About synchronous CMC,

Chun (cited in B6hlke 2003) suggests that the CMC environment is less stressful than

oral discussion, because students have more time to think about their utterances and do

not have to worry about their pronunciation. It is all the truer for asynchronous CMC

where the pressure of spontaneity is completely eliminated. Warschauer (1996) showed

that synchronous CMC chat was more complex than face-to-face communication among

his advanced students of English as a second language at the University of Hawaii. The

Hawaii CI for the face-to-face group was .182 and for the CMC group was .475. As a

result we can expect CMC communication to be more syntactically complex than oral

communication. As far as the present research is concerned, the electronic discussions

were completely devoid of negotiation of meaning, which occurred a few times with the

oral groups. As expected, the average CI for the CMC groups was higher than that of the

oral groups, .381 and 278 respectively.

When we look a little closer at the types of dependent clauses produced, we notice

that elements of the oral language permeated almost as much in the forum. One

particular element is the amount of dependent propositions introduced by a variation of

"je pense que" or "je crois que," both translated as "I think." These clauses are









completive clauses and they are part of the oral language because in the written language

students are instructed not to use the first person of the singular "je" or "I." Furthermore,

when one writes an argument one does not use personal opinion or conjecture, instead

one supports one's argument with tangible evidence. There were 62 out of 298 such

completive clauses in the CMC data and 18 out of 98 in the F2F data. Proportionately,

both results are very close with a ratio of .208 for the CMC and a ratio of. 184 for the oral

production, the difference is only .014. This suggests that the proportion of dependent

clause types will be maintained in student language from oral production to CMC

production. In order to determine that this proportion is not just an effect of the CMC

sharing elements of oral production, it would be interesting to compare individual

students' written production to the present data, both for the CMC group and the control

F2F group.

Lexical Complexity

The Type Token Ratio provides allows to determine the semantic diversity students

display. It is determined by identifying all the different words in a given sample and

comparing those words to the total amount of words. Thus we obtain the following

formula:

TTR=Amount of different words/total amount of words

In order to determine the TTR, random samples of 200 word productions from

students in the two media were retrieved. In the interest of time, a limited sample size

allowed for easier analysis. The sample size was determined by three factors. First of

all, the TTR is not a typical ratio like the CI, which resembles a probability or an average

more than the TTR does. Indeed, the TTR requires that each different word be counted,

yet the bigger the sample the less diverse the vocabulary, and an individual's vocabulary









capacity will eventually be reached. Secondly, the data available did not allow for

samples much higher than 200 words (particularly in the oral production groups).

Finally, in the interest of time the samples were reduced to a workable size as the analysis

was going to be performed by hand.

To clarify what is meant by "amount of different words" it is best to use an

example. The sentence "the cat eats the mouse" contains five words total but four

different words as "the" is repeated. Thus, the TTR for this sentence is 4 over 5 or .80.

The difficulty is in determining what constitutes a "different" word. One particular rule

was kept: only words that were semantically different were counted and therefore

morphological or syntactical variations were eliminated. For example, all the variations

of the French definite articles le, la, les, 1' were eliminated after one of them was

encountered once, however, I had to make sure that those words except for 1' were not

variations of the direct or indirect object complements, which would be considered

differently. Other examples included conjugated variations of a verb, for instance, I

eliminated pensions, the first person plural of the verb penser (= to think) after having

seen pense first person of the singular. The importance was to remain consistent when

retrieving all the data. Table 3.3 below summarizes the findings of the TTR:









Table 3.3 : Summary of the TTR
# of # of
different different
CMC word Total TTR F2F word Total TTR
Student 1 92 200 0.460 Student 1 79 200 0.395
Student 4 83 200 0.415 Student 2 61 200 0.305
Student 5 71 200 0.355 Student 3 72 200 0.360
Student 6 77 200 0.385 Student 4 84 200 0.420
Student 8 78 200 0.390 Student 5 82 200 0.410
Student 10 87 200 0.435 Student 6 88 200 0.440
Student 12 59 200 0.295
Total 547 1400 0.391 Total 466 1200 0.388
STDEV 0.054 STDEV 0.049


The same considerations as for the CI applies to the TTR. The TTR does not

reflect language accuracy. In other words, it does not say whether vocabulary is

employed relevantly, nor whether it is spelled or pronounced correctly.

Unlike for the CI, a per group analysis was not performed since not all students

yielded a sufficient amount of workable data. As for the CI, data considered workable

included, any data that was spontaneous, therefore all the written samples that were

requested by the project were excluded. The written samples that were excluded were the

summary and description of the commercial, the rationale on where to position the

commercials, and the final report (see Appendix A). Thus, data from students 1, 4, 5, 6,

8, 10 and 12 of the CMC group was used. Overall, they were all very consistent and

ranged from .295 to .460. The average was .391 with a standard deviation of .054, which

indicates a limited dispersion from the mean. This consistency and limited dispersion

suggest a marked homogeneity among the students' semantic diversity.

Students in the F2F groups showed consistency as well. Their TTR ranged from

.305 to .440, with an average of .388 and a standard deviation of .049, which again

indicated a limited dispersion from the mean. As for the CMC groups this consistency









and limited dispersion suggest a marked homogeneity among the students' semantic

diversity.

Since both media are different and since the CMC carry similarities with written

production, one would expect the TTR to also be different for the two media. The CMC

groups should show a higher TTR, thus a higher semantic diversity than the F2F groups.

Warschauer (1996) also determined the TTR for his advanced students of English as a

foreign language at the University of Hawaii. The F2F group yielded a lower TTR of

.262, compared to the synchronous CMC group's of .301, as expected. Perhaps this

difference can be attributed to the task Warschauer gave. Students in his research were

supposed to answer two "counterbalanced questions" (p. 12). However, one would

reasonably expect to find the same pattern with the present research.

Yet, both media yielded a similar consistency, as we have seen in the group

analyses above. This similarity ended up being much more striking than expected since

the two groups displayed a margin of only .003 between the two TTR of .388 for the F2F

group and .391 for the CMC group. This is reinforced by the two standard deviation of

.049 for the F2F group and .054 for the CMC group. These numbers allow us to

conclude that intermediate students of French at the University of Florida display the

same level of semantic diversity both orally and using asynchronous CMC

communication. It further suggests that students' language on an electronic forum is

similar in semantic diversity to F2F communication contrary to previous affirmations that

asynchronous communication is more diverse than face-to-face communication within

the context of the task provided to the students. However, this may be because students









had used all the vocabulary requested by the task, since the task was specific and

encompassed the vocabulary of two chapters from the students' textbook.

Implications

None of this data suggests that one medium is better than the other. However, there

are some pedagogical benefits to both of them that will be discussed herein. Then, some

suggestions about running the same project or any other that would be similar will be

offered in light of students' behaviors and also in light of their feedback.

The above data suggests that running the project on an electronic forum did offer

some of the benefits that were stated in the introduction. All of the students respected the

rule about posting threads in French. The messages they posted included both

characteristics from the oral language and the written language. Having their messages

posted on the Internet certainly influenced their behavior. None of them required any

particular training as to how to use the forum and most actually used it instinctively. A

few students were even so familiar with the software that they developed their own

profile and adopted interesting behaviors. Some of these behaviors included using a

different colored font for every message that they posted (see appendix B), or the use of

avatars and emoticons. The majority of the students only performed what was asked of

them and, albeit, they did not necessarily find using the forum practical for the duration

of the project. One can deduce that by looking at the number of threads they posted as

well as their messages suggesting other means of communication. Their final feedback on

the project was more obvious. Here are some examples of positive and negative

feedback:

-Pour la plus partjepense queje n'aipas apprecie ceprojet en raison duformat deforum. Mais, le
travail &tait assez facile, seulementje desteste des ordinateurs.









-For the most part I think that I did not appreciate this project because of the format of the forum.
However, tasks were rather easy, I just hate computers.



-Je suis aussipas bon avec les ordinateurs et comprendre le site etait au debut tres dur. Mais c'est
plus facile a ecrire le fiancais pendant qu'a un ordinateur parce que c'est plus rapide pour chercher
des mots que vous ne savezpas.

-I am not very good with computer and understanding the site was very difficult at first but it is
easier to write in French with a computer because it is faster to find words that you do not know.

Overall students reported that they did not particularly enjoy working on the forum.

All the groups agreed that they needed time in class to coordinate better. Yet, members

of group 2 still maintained communication on the forum after the project was over.

As for the F2F group, they enjoyed working in the language laboratory very much.

Here are examples of their feedback:

-Je pense que ce project a etd interessant. J'aimais utiliser les ecouteurs. Je prefere
travailler dans le laboratoire de langues que dans la salle de classes. [...]

-I think that this project was interesting. I liked using the headphones. I prefer
working in the language lab than in the classroom.



-Nous pouvions aussi pratiquer nos competences de grammaire et communication.
Le project etait une bonne idde

-we could also practice our communication and grammatical skills. The project
was a good idea.



-Le project a ete tres bon pour practiquer lefrancais. Nous avons utilise beaucoup
les mots unique de la project. II a ete difficile a comprendre les autres personnel
dans le group quand nous avons decrive les pubs. Mais, c 'etait un bon exercise
pour nous parce que nous avons du comminquer nos idees entire eux. Quand
quelqu 'un ne comprendait pas les choses qu 'un autre disait, cette personnel a df
chercher un autre maniere d'exprimer son idee.

-The project was very good for practicing French. We used many unique words of
the project. It was difficult to understand the other people in the group when we
described the commercials. But, it was a good exercise for us because we had to
communicate our ideas to one another. When someone did not understand the









things that someone else was saying, this person had to look for another way to
express his/her ideas.

The overall feedback on the F2F version of the project was much more positive

than that of the forum. Students did indeed negotiate meaning often, which was a very

beneficial exercise for all of them. The only problem was to accommodate the project

and the time in the language laboratory within a pretty heavy syllabus. One of the

advantages of running the project with a forum was to offer the possibility to do work

outside of the classroom. Thus it was easier to accommodate the project within the

syllabus. On the other hand, having to do extra work outside of class may have been a

reason why students did not enjoy the electronic forum. The final product of the project

was equally satisfactory with both media. However, since so many students ended up

not using the forum for actual negotiation, but only to post their final thoughts, and since

so many groups ended up using other means of communication, it is not clear that they

actually benefited as much from the project as the F2F group did in the laboratory.

Furthermore, although the forum offered controlling possibilities such as time and date

when the thread was posted, it did not allow to control the rest of the students'

negotiations. These negotiations offer the best opportunity to improve students'

language skills therefore it is imperative for the teacher to be present to offer feedback

and control the exchanges. The forum was indeed not as beneficial partly because the

teacher did not add threads to encourage students to participate more. If the teacher had

been more involved and encouraged students to post replies by asking questions for

instance, perhaps students would have used the forum better. The forum could also be

used in combination with a F2F medium. For instance, using the forum for posting work

such as the description of the commercial and for completing the final project would be









beneficial. Students are then at liberty to review what the other groups are doing and can

also check on their progress. This is what a student wrote when she/he was worried that

her team had not yet made a decision as to which commercial to use:

-Vous n'avezpas choisi unepub. Mais c'est 11:52, etpuisje vais choisir unepub pour le group. J'ai
regarded les autres groups et il y a un group qui a choisi "le sculpteur" et il y autre group qui n'apas
choisi unepub mais ils peuvent choisir la pub de Axe maisjepense que il n'ontpas la choisi.
Donc, je vais choisir lapublicite de AXE pour notre group. Si, on peut changer plus tard quand
vous avez repondre, nous changons. Si, non, doncj'ai choisi la pub de AXE pour le group.

-You did not choose a commercial. But it is 11:52pm, and then I am going to
choose a commercial for the group. I looked at the other groups and there is a
group that chose le sculpteur (=the scupltor) and there is another group that did not
choose a commercial but they may choose the Axe commercial but I think they did
not choose it yet. So I am going to choose the commercial AXE for our group. If
we can change later when you answer, we will change. If not, then I chose the
commercial Axe for the group.

Finally, pairing students with native speakers from a French speaking university to

add a cultural exchange component to the project should be considered. Insight from

native speakers would help students understand certain cultural subtleties related to the

media. It is all the more important as we live in an era of global communication.

The CI helped determine that there is a difference in complexity between CMC

and F2F communication, whereby CMC offer more syntactically complex

communication. However, they both proved to be as varied in lexical terms. As one of

the students in the F2F group pointed out, it is possible that negotiation of meaning

pushes students to be more creative in an oral setting. In order to obtain more insight into

both media and the languages produced therein research should be performed using

additional data, such as written samples, to compare both groups so as to find out whether

the type of dependent clause (such as those introduced byje pense que (I think that)) is

maintained proportionately in the written samples, on the forum and in F2F samples.

Thus we would be able to determine whether this proportion is more a factor of the oral









language while using CMC or not. Furthermore, future research on the matter could

include additional variables that pertain to language accuracy. It is important that we

look at variables that can be consistent with both media. In other words, we cannot look

at spelling errors as the oral language in F2F communication does not offer such insight.

Yet we can look at patterns in errors such as preposition mistakes that students would

typically make. We could also look at morphological errors such as verb conjugations,

particularly with the use of the auxiliary Otre ou avoir ("to be" or "to have") in analytical

tenses such as the "pass&-compos6." Finally, we could collect additional data to

determine whether the homogeneity that was found with the TTR variable during the F2F

exchanges was an abnormality or not. This could be done by performing the research

using different tasks that would vary in specificity in order to elicit a larger variety of

vocabulary.














CHAPTER 5
CONCLUSION

The first research question of this study aimed at comparing the grammatical

complexity of the language produced by two groups of students. One group produced

their language on an electronic forum and the other communicated orally. In order to

compare these two groups grammatical complexity, the CI, which compares the number

of dependent clauses over the total number of clauses, was compounded. The results for

this variable suggest that the language produced on an electronic forum tends to be more

complex than the language produced orally. The written nature of the language used on

the electronic forum accounts for this result. However, the CI does not take into account

the types of dependent clauses and the data revealed that there was a similar number of

completive clauses in the language produced with both media. This suggests that more

research be performed in order to compare the quantity of the various types of clauses

between the two media and written and spoken samples of the native language of the

students.

The second research question of this study aimed at comparing the lexical

complexity of the language produced using both media. In order to compare the

language complexity, the TTR, which measures the number of different words over the

total number of words, was used. While it was expected that the TTR of the language

used in the electronic forum would be higher than the one used in the F2F environment, it

was not the case for this research. Indeed, the TTR for both media were almost identical.

One reason for this result may be the fact that at the intermediate level of their language









apprenticeship and for this particular task, students had reached and used the extent of

their vocabulary bank. Thus, this data suggests that more research be performed in order

to compare both media again with data coming from various tasks.

The third and final question of this research aimed at whether other patterns could

be discovered from this type of data. First of all, it revealed that students display

interesting behaviors on an electronic forum. The first student to write a thread is more

likely to become the leader of the group and the last student to write a thread is more

likely to be the student who will participate the least. Students reported that they did not

enjoy working on the electronic forum claiming that it was not convenient for negotiation

and, indeed, most of them ended up using other media for communication. It is possible

that students did not enjoy the extra work to be performed outside of the classroom in

addition to other factors, thus they did not enjoy using the forum as much as if they had

had to perform the task in class. One group of students, however, enjoyed the community

building characteristic of the electronic forum. Most students in the F2F group reported

that they enjoyed communicating in the language laboratory. They enjoyed the new

environment and felt that negotiating meaning among other aspects of the task was

beneficial. In fact, some students even participated much more than in the regular class

setting.

Learning about the differences between the types of language produced using an

electronic forum and F2F communication can help teachers make informed pedagogical

decisions on when alternative to use an alternative and how to use it when they design a

task. This research revealed some of these differences and perhaps similarities.






40


However, additional research is necessary to examine more of these differences for

teachers to be able to make better informed pedagogical decisions.














APPENDIX A
PROJECT DESCRIPTION

FRE 2200: Activite m6diatique
Dear employees,

You are my best team of advertising executives, here at Famille Guillo Advertising
Agency. and I am proud of the work you did on our last campaign. I have been
contacted by several French clients who would like to use your talents for their campaign.
Your job will be fairly easy, as our clients have already shot the commercials. This will
be your first time working in an international environment so be aware of cultural
differences and all your production must be in FRENCH (including the meetings).

Your first task will be for each one of you to choose a commercial and to analyze it
according to our usual guidelines. The most important thing of course will be the
identification of the target audience.

Then, as a team, you will have to decide on a single commercial from the ones that each
one of you will have chosen. Then, individually you will have to research TV
programming and choose two shows that the target audience is most likely to watch so
that we reach as much of that target audience as possible. We want our client's messages
to be as effective as possible.

Once each one of you will have identified the two shows, as a team, you will have to
decide on which two shows will best suit your commercial. Your budget is limited, that
is why I am only allowing two shows.

Finally, you will have to give me a report on all of your progress and all the decisions
you made and why. This report will be no longer than two pages.

Attached, you will find a timeline, a copy of our guidelines, links to the 2 French TV
channels with the largest viewership and the guidelines for the final report.

I almost forgot... Since you are so busy, you will not be able to meet at the same time.
That is why I have requested that a forum be created on our web site. Each one of you is
required to post all of your work on it as well as your team communication. You will
also find the link to the forum as well as guidelines on how to register.

Go ahead and make me and all of the Famille Guillo Advertising Agency proud of you
again.









Respectfully,

Cyrille Guillo
President Directeur General

Timeline for the CMC groups:

All of your communication and production will have to be posted on the forum at the
following website:

http://www.familleguillo.com/forum

You will have to register in order to be able to post. To register click on "Register" at the
top of the page. Then go to "FRE 2200 Projet mediatique" find your group and click
on reply in order to post. Do not start another thread!

1. Pour Vendredi 25 Mars:
Imlividlhlly, choose a commercial:
Go to the following link where you may have to register (it is free):

http://www.pubstv.com


Analyse the commercial according to the advertising guidelines attached.

2. Pour Lundi 28 Mars:
Discuss which commercial your team will choose. You will have to tell your
teammates why you decided on the commercial you chose. As a group you will
choose the commercial that appeals to the team the most

3. Pour Mercredi 30 Mars:
Ihlividhlly, explore these 2 French TV web site to find their programming in order
to identify 2 shows that would best fit your commercial. Your goal is to reach the
largest target audience possible.

http://www.tf.fr

http://www.france2.fr

Write a paragraph for each show explaining your rationale. See the advertising
guidelines below.

4. Pour Vendredi 1 Avril:
As a team, decide on which 2 shows will best reach your audience for your ad
campaign.


5. Pour Lundi 4 Avril:









As a team, work on the final report, the guidelines of which are provided below.

Timeline for the F2F groups:

6. Pour Mercredi 30 Mars:
Iuli'vidldllr, choose a commercial:
Go to the following link where you may have to register (it is free):

http://www.pubstv.com


Analyse the commercial according to the advertising guidelines attached.

7. Mercredi 30 Mars EN CLASSE au Turlington Language Lab:
Discuss which commercial your team will choose. You will have to tell your
teammates why you decided on the commercial you chose. As a group you will
choose the commercial that appeals to the team the most

8. Pour Vendredi ler Avril:
Idulivid dllr, explore these 2 French TV web site to find their programming in order
to identify 2 shows that would best fit your commercial. Your goal is to reach the
largest targeted audience possible.

http://www.tf.fr
http://www.france2.fr
Write a paragraphe for each show explaining your rationale. See the advertising
guidelines below.

9. Vendredi 1 Avril EN CLASSE au Turlington Language Lab:
As a team, decide on which 2 shows will best reach your audience for your ad
campaign.

10. Pour Lundi 4 Avril:
As a team, work on the final report the guidelines of which are provided below.


NO LATE WORK WILL BE ACCEPTED. Failure to provide work will result
in personal grade penalties.

Advertising Guidelines:

In order to analyse a commercial properly you have to first identify the product,
the brand, and the message, then you will provide a short summary describing the
commercial and finally, you will provide a profile of your target audience. For the target
audience profile, use the following major demographic variables and suggested
breakdowns:










Age Sex Revenu Education Occupation Statut Nombre Georgra
(xl000) Marital d'enfants phie
-18 Masulin -15 -Brevet des -Professionel et Marrie(e) 1-2 Rural
18-24 Feminin 15-30 colleges Technique Celibataire 3-4 Urbain
25-34 31-50 5+
35-44 51-70 -Lycee ou -Cadre
45-54 71-90 BEP -Agriculteur
55-60 91-100 -Bac -Profession
60+ 100 -Bac +2 liberal
ou plus -Etudiant
-Sans emploi
-autre

In addition, describe the possible interests and hobbies of the target group. Most of the
time, interests, and hobbies can be derived from the commercial itself. Is it shot in an
artistic manner? Is it showing sports of any kind? If yes which type? Is it showing
wealth? Is it humorous? If yes, what kind of humor? Etc... If you watch the
commercial closely you will learn many things about the people targeted, be attentive and
creative!

Finally, according to your findings, anticipate and tell us what type of shows the target
audience is most likely to watch and when.



Report Guidelines:

The report will include parts.
In the first part you will describe the commercial and why your team decided to
choose it.
In the second part you will provide your rationale for the 2 shows you chose for
your advertising campaign.
In the third part you will discuss the cultural differences or similarities you found
while choosing a commercial and while examining French TV programming. For
instance, would your commercial be efficient in the US? Why or why not? What are the
differences or similarities between French and US shows.


Finally, what did you think of this project? Write one paragraph.





























APPENDIX B

FORUM SCREENSHOTS









-Bak S earh -Favontehs Mneda _- Ilo o Le
Search the Web Psearch Address http:/www, famllegul o.comforum

mac= F E] J Sear-h Highlght M] Options O Pop-ups Allowed Hotmal Messenger [ My M5N


Yphu st ed on 04 Ju205 230 am
The ile 1 Fr 0 Jl 2005 0502 pm
Cyrllle imlllo Forum Index


Cyrille Guillo
Academic Forum

FAQ Search IMemberlst -Uergroups
mProfile You have no new messages Log out [ gulllo ]


Forum Faq's
Moderated forum Contains important nfo about this board Please read beforeresgisterin






Aiatualites
Forum Announcements




ue e pase-t- en Frane et d e monde Qu en penez-vou
ModerCtor Shet -d, The palish Kid
Discussion GFnRrale
-I Forum Puhlque
l.mr. TI, Pd, Kid
Le fronois un Ianogge difficile!
ous aveon de dffleutes sur un point de ramnalre ou de voabulare Ou ave vous toue autre question he a langueo Profte en
Moderatorshds, qera, The Polish Kid
FRE 2240-- Annonce,
Si vou avez des iuestions au sulet du eours en partouller, par e-emple ee que vous devez preparer AU-sl je postere des information urgentes sur ce
iorum

F RE 2240 RARCMC
DlsUten FRANCAIS 1''

FRE 2200 -- Projet Mediatie
.1 Toute discussion dolt se fare en n AI :) Vos group ont 6t asgn- Vou devez vou nscrl re pour p ouvor parlciper o ce forum

Project de Maitrise
Welcome
D i r ecrpt on of this category 004

Directed Reading Summer 2004
cadem dlicuion

n--dt~p... ~ ~l~n


S 05 Sep 2004 1152 pr


2 0 No Post.



s 7 04 No 2004 102- p-


27 Jan 2005 05.03 arm
S ThPoh Kid OD


07 O05 asd23 p
1 10


12 28 o Nov -n02s ;a


26 219 I23 -o 4 :14 D


21 Ap, 2005 12 28 am
7 166 carolk t




1 0o un 00 I 2'56 pm

23e u 2004 02 49 pm

^ .. -- A Q


Figure 1: Screenshot of the forum homepage.






Students need to enter the threaded topic related to their class project in order to




have access to the threaded discussion (1). As the administrator, I am can maintain




students' privacy by denying the access to the forum to any one who is not registered on




the forum and in the class.














45


IV] Go LInk *


ilee wour Posts
View unanrwered Psos
p. t
P.-


IAllll ~8~~ "L]IBa


I-I CC


Irrr.~ Iu


I














1 riic r..,,liu tvio ro0 r mrnt i Tn n Io m' Ao.i ire-rc Ali r ni fc r,,iarcr JI
File Edt View Favorites Tools Help
O Back [N S)5earch MFavorte Medi E 3
Search the Web PSearch Address L http. www famllegullo.comlforuiewforum php f=1& d= 264d87b76ca4DB3d03e7boMb9d4o0e6 Go in "
Ja -I i] Search -.# Hghlght ]pions n Pop- ps Allowed 4E Hotmal Messenoger y M5N


Ps


I RE 2200 -- Projet M4diatique

uses bmwsing mis Emm: quite
Cyrille Guillo Forum Index -> FRE 2200 -- Pro3et M6diatique


Cyrille Guillo
Academic Forum
FAQ GSearoh IMemberhst Usergroups
Profile You have no new messages Log out [ gullo ]


Cyrille Gulllo Forum Index -> FRE 2200 -- Proet diatique
Page 1 of 1




N hew posts [ Locked ] o n posts [ bn ]- En
. hew posts [ Locked ] ho new posts [ Locked ]


-ump to: S le a fr,,m



You can vote In polr In thi forum
You ran rroder~te thli forum


Go to dmnl aon Panelp
powe~d by phpBB 2 0 e @ 2001, 2002 phpBB Gmup


Figure 2: Screenshot of the Project threaded discussions.


After students have clicked on the topic all they have to do is click on the threaded



discussion of their assigned group. A little flag on the left side (1) indicates if there is a



new message they have not read yet and a comment on the right side (2) tells them who


was the last person to post a comment.


SMai alltopics read

Group 2: Laura Carol Kra. 1 21 A 0 128 a2

G oupe 4 : Roemar Ern Lahley Ashley B. 1
Group 3 Andy luda Brittany ine. 2 227
[ Doto page: -- ] 5

S5: egan Jaki A-nnie Carolina.

- je deterte w pubtu.com 0 cdl.zle 17 31 Mar200~o55s am
SLa pub francais 2 Vincel3
Display topi s from previous:


Ilmes aI. GMT


,---. "'J0-b- ,
















File Edt View Favortes Tools Help
Back Search Favo rtes gMeda -
Search the Web I PSearch Address L http./ww fanllegullo.com forumliewtopi phpt=6,postdays=0epostorder=ascitart=45 Go Lins "
S -I ] Search Highlght f Options nPop-ups Allowed Hotmall y Messenger L My M5N

L o ,n. "To the world, you might be only one person, but to one person, you just might be the world

Back to top i S1i [ AIM

Alle.Ycat DOostedl 07 Apr 200s 12 12 -m P-t -ubeEtZ

S 3. Les commercials francals ont sembl6 etre le m6me ue les commercials des Etats Unis. Le commercial de Maurce et Coco est une publicity trees comique pour les
enfants;les deux a France et a des Etats Unis C'est trees attirant a es jeunesses par ce que I ya des animaux et un eune fil dans le publiciteO 3al surprise quea ny a
pas plus des dessins animes sur le weekend S, Sle commercial montrait a des Etats Uns; le commercial arerat a le weekend pendant les dessins animes.
1o ed, 16 Feb 2005
~ots: 13
4. alme le project, mals le a contribute a mon dffration (procrastination) 3'tait amusement avec le project par ce que ajal pu parler avec mes amles dans Iensemble
Jal pense que le rolet atalt beaucoup de travaail mas n'atalt pas difficil. C'Etat tres interessant a voir les publicts francals, et des autres si vous ete moll

A bird may loe a f but where would they live?
Back to top e m nl AIM

5tarlocketl7 DPosted: 07 Apr 2005 12:27 am Pos-t s-ub-e:. no..i. 2ut 1 Bo nourilli

Ce le que je pene sur la tro e me part des questions.n

S2 e dans out est a la revision, ar eemle les ubs et es rorame s ns trouve un ub trs m n urra e u ub n ut
alnes~lle/oalaa ailment le choc ce pub atirera t a quelqu'un. C'est la meme chose pour les programmes francals, is sent tres simlaire come les programmes amercalns II y a
beaucup de dessinantieaour les enants, beaucoup sont americann, et II y a auss des autres rog rammes come les feuilletons, les jeux, ou les vanetes.


Back to top o In ai M 2

starlocketl7 DPo Apr 2005 12:33 am Post subject: Buong orn"l' Zutiil BonjurJl p C E

Qu voudrat post tout ensemble 7 0

J ned: 25 Mar 2005
Po.ts 24

To the word you might be ony one ersn, bu eron, you just might be the world
Back to top ti ALM
D..pl.. P.-.. .fi..r- ... 1. I 'I

Cy rille Guillo Forum Index FRE 2200 -Projet Mediatique Got page Previous 4, 5 Net
Page 4 of 5



Figure 3: Screenshot of a thread including (1) emoticons (2) avatars and (3) personalized

signatures.


Notice the different color fonts.















LIST OF REFERENCES


Beauvois, Margaret (1997), "Write to Speak: the Effects of Electronic Communication on
the Oral Achievement of Fourth Semester French Students," New Ways ofLearning
and Teaching: Focus on Technology and Foreign Language Education, Heinle and
Heinle Publishers, pp 93-115.

Beauvois, Margaret H., and J. Eledge (1996), "Personality Types and Megabytes: Student
Attitudes Towards Computer Mediated Communication (CMC) in the Language
Classroom," Calico Journal, Volume 13 Numbers 2 and 3, pp 27-45.

B6hlke, Olaf (2003), "A Comparison of Student Participation Levels by Group Size and
Language Stages During Chat Room and Face to Face Discussions in German,"
Calico Journal, Volume 21 Number 1, pp 67-87.

Chenoweth, Ann, and N. K. Murday (2003), "Measuring Student Learning in an Online
Course," Calico Journal, Volume 20 Number 2, pp 285-314.

Davis, Boyd, and Ralf Thiede (2000), "Writing into Change: Style Shifting in
Asynchronous Electronic Discourse," Concepts and Practice, Cambridge Applied
Linguistics, pp 87-120.

Kern Richard, and M. Warschauer (2000), "Introduction Theory and Practice of
Network-Based Language Teaching," Concepts and Practice, Cambridge Applied
Linguistics, pp 1-19.

Lantolf, James (Ed) (2000), Sociocultural Theory and Second Language Learning.
Oxford, UK: Oxford University Press.

Pellettieri, Jill (2000), "Negotiation in Cyberspace: The Role of Chatting in the
Development of Grammatical Competence," NBLT: Concepts andPractice,
Cambridge Applied Linguistics, pp 59-86.

Perez, Luisa C. (2003), "Foreign Language Productivity in Synchronous vs.
Asynchronous Computer-Mediated Communication," Calico Journal, Volume 21
Number 1, pp 89-104

Skehan, Peter (2001), A Cognitive Approach to Language Learning, Oxford University
Press.






49


Warschauer, Mark (1996), Comparing Face-to-Face and Electronic Discussion in the
Second Language Classroom," Calico Journal, Volume 13 Numbers 2 and 3, pp 7-
27.

Warschauer, Mark (2000), "Online Learning in Second Language Classrooms," NBLT:
Concepts andPractice, Cambridge Applied Linguistics, pp 41-58.















BIOGRAPHICAL SKETCH

Cyrille Guillo obtained a Master of Arts in English as a Second Language from the

University de Haute Bretagne, Rennes, France. He also obtained a Master of Business

Administration from the University of Missouri, Columbia.




Full Text

PAGE 1

COMPARING THE LANGUAGE OF INTERMEDIATE LEARN ERS OF FRENCH IN ASYNCHRONOUS ELECTRONIC COMMU NICATION VS. FACE TO FACE COMMUNICATION By CYRILLE GUILLO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2005

PAGE 2

ACKNOWLEDGMENTS I thank Dr. Antes and Dr. Lord for their constant support and greatest patience. I thank Dr. Shoaf for suggesting and letting me use the language laboratory. I thank my parents for always loving me and being supportive of me even though we were so far apart. I thank my grand-parents as well for their encouragement and for having been so good to me all my life. ii

PAGE 3

TABLE OF CONTENTS Upage TACKNOWLEDGMENTST..................................................................................................ii TLIST OF TABLEST.............................................................................................................iv TLIST OF FIGUREST.............................................................................................................v TABSTRACTT.......................................................................................................................vi TINTRODUCTIONT...............................................................................................................1 TPREVIOUS WORKT.............................................................................................................3 TMETHODOLOGYT............................................................................................................12 TResearch QuestionsT.....................................................................................................12 TParticipantsT.................................................................................................................12 TTaskT.............................................................................................................................13 TAnalysisT......................................................................................................................15 TRESULTS AND CONCLUSIONST...................................................................................17 TSyntactic ComplexityT.................................................................................................17 TLexical ComplexityT....................................................................................................29 TImplicationsT................................................................................................................33 TCONCLUSIONT..................................................................................................................38 TPROJECT DESCRIPTIONT................................................................................................41 TFORUM SCREENSHOTST................................................................................................45 TLIST OF REFERENCEST...................................................................................................48 TBIOGRAPHICAL SKETCHT.............................................................................................50 iii

PAGE 4

LIST OF TABLES UTable U Upage U T3.1 : Summary of findings according to the CIT..................................................................19 T3.2 : Total number of threadsT.............................................................................................20 T3.3 : Summary of the TTRT.................................................................................................31 iv

PAGE 5

LIST OF FIGURES UFigure U Upage U T1: Screenshot of the forum homepage.T.............................................................................45 T2: Screenshot of the Project threaded discussions.T...........................................................46 T3: Screenshot of a thread including (1) emoticons (2) avatars and (3) personalized signatures.T.................................................................................................................47 v

PAGE 6

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts COMPARING THE LANGUAGE OF INTERMEDIATE LEARNERS OF FRENCH IN ASYNCHRONOUS ELECTRONIC COMMUNICATION VS. FACE TO FACE COMMUNICATION By Cyrille Guillo December 2005 Chair: Theresa Antes Cochair: Gillian Lord Major Department: Romance Languages and Literatures As technology permeates in the foreign language classroom, teachers have to determine whether any technology they intend to use can be beneficial to their students. Electronic forums are a recent technology that teachers may consider using. This researchs purpose is to compare students language production on an electronic forum to face-to-face communication. A study group used an electronic forum outside of the classroom to complete a project based on the cognitive approach; at the same time a control group performed the same task using in-class face-to-face communication. The first research question of this study aimed at comparing the grammatical complexity of the language produced by two groups of students. One group produced their language on an electronic forum and the other communicated orally. In order to compare these two groups grammatical complexity, the Coordination Index (CI), which compares the number of dependent clauses over the total number of clauses, was compounded. The vi

PAGE 7

results for this variable suggest that the language produced on an electronic forum tends to be more complex than the language produced orally. The written nature of the language used on the electronic forum accounts for this result. However, the CI does not take into account the types of dependent clauses and the data revealed that there was a similar number of completive clauses in the language produced with both media. The second research question of this study aimed at comparing the lexical complexity of the language produced using both media. In order to compare the language complexity, the Type Token Ratio (TTR), which measures the number of different words over the total number of words, was used. While it was expected that the TTR of the language used in the electronic forum would be higher than the one used in the F2F environment, it was not the case for this research. Indeed, the TTR for both media were almost identical. Thus, these data suggest that more research be performed in order to compare both media again with data coming from various tasks. The third and final question of this research aimed at whether other patterns could be discovered from this type of data. First of all, it revealed that students display interesting behaviors on an electronic forum. The first student to write a thread is more likely to become the leader of the group and the last student to write a thread is more likely to be the student who will participate the least. Students reported that they did not enjoy working on the electronic forum claiming that it was not convenient for negotiation and, indeed, most of them ended up using other media for communication. On the other hand, most students in the F2F group reported that they enjoyed communicating in the language laboratory. They enjoyed the new environment and felt that negotiating meaning among other aspects of the task was beneficial. vii

PAGE 8

CHAPTER 1 INTRODUCTION Technology in the foreign language classroom is not a new phenomenon. The first teacher to have used a tape player or a tape recorder brought technology into her/his classroom. However, when one speaks of technology in the foreign language classroom nowadays, one speaks of CDs, DVDs, computers and the most recent global phenomenon associated with computers, that is the World Wide Web. New technology is exciting and there will always be a teacher or a researcher who will try to incorporate it into his or her teaching practices. The difficulty becomes to optimize the use of this technology so that its benefits outweigh or supplement existing practices. Such has been the case for Computer Mediated Communication (CMC). CMC is the use of a Wide Area Network (WAN) or a Large Area Network (LAN) to allow communication between two or more computers. CMC can take the shape of students chatting together from computer to computer in the same room, such as is used in language laboratories, or it can be students communicating from remote locations. CMC can be synchronous when students chat simultaneously with the use of a chatting interface such as instant messengers (AOL Instant Messenger also called AIM, MSN Messenger or ICQ) or with the use of voice over IP (Internet Protocol), which allows for voiced conferences over the Internet thanks to software like Netmeeting Synchronous CMC via chat software displays similar characteristics to Face-to-Face (F2F) communication. For instance, there is an important amount of turn-taking and turns are short in general (Warschauer 1996, Beauvois 1996). CMC can also be asynchronous, meaning that every message will have 1

PAGE 9

2 a delay and that people will be able to have access to those messages in their own time. Asynchronous communication is typically associated with written communication as users can take the time to organize their writing before making it available to its recipient. Examples of asynchronous CMC include email, email lists (or newsgroup or listservs) or the use of electronic forums also called Bulleting Boards (BB). Electronic forums differ from other asynchronous communication as they archive and thread all writing so that participants can access, select and retrieve any written message at any time, regardless of the topic or when it was written as long as the discussion is stored on its host. Messages on electronic forums can be displayed chronologically but, most commonly nowadays, they are displayed first according to topic and then chronologically (see appendix B). The present thesis intends to examine the use of an electronic forum to complete a class project as opposed to a F2F alternative. The goal of this thesis is not to determine whether one approach is better than the other. Both approaches have advantages and disadvantages that influence students performance differently. Therefore, the goal of this thesis is to compare students performance using both approaches so as to provide pedagogical input for teachers. Thus, a project was designed and administered to two classes of students of intermediate French at the University of Florida. One class was asked to use a bulletin board software as means of communication while the other communicated orally in class. The two classes language production were compared using the Coordination Index (CI) and the Type Token Ratio (TTR) as variables. The results and a more detailed explanation are presented below.

PAGE 10

CHAPTER 2 PREVIOUS WORK The following articles have been selected from a more complete list concerning technology in the classroom because of the limited amount of research regarding the use of bulletin board software as a means of communication. In order to compare F2F and Electronic discussion, Warschauer (1996) asked the following questions: 1) do second language students participate more equally in small group discussions held electronically than those held in a traditional F2F manner? 2) if so, who benefits from this more equal participation? In particular, how are differences in participation for a F2F mode or an electronic mode related to factors such as gender, nationality, and age and language proficiency? 3) what are students attitudes toward participating in electronic and F2F discussion and how do these attitudes correlate with changes in amounts of participation? 4) does electronic discussion include language that is lexically or syntactically more complex than F2F? 5) what other differences are noted in the language use and interaction style in the two modes? (p. 10) Warschauer studied the language production of 16 students of various ages and nationalities enrolled in an advanced ESL course in a community college in Hawaii. The F2F data was transcribed and all the transcripts were entered in the Computerized Analysis Program (CLAN) of the Child Language Data Exchange System (CHILDES); which was used to count the number of words per speaker and to calculate the Type Token Ratio (TTR). The transcripts were analyzed to calculate the number of clausal coordinations and subordinations. He found that there was an increased participation in the computer mode. The language was also more complex in the electronic mode when comparing the CI and the TTR. The turn-taking in the conversation mode was more numerous with short turns and many confirmation checks. The computer exchanges 3

PAGE 11

4 displayed less direct levels of interaction and students expressed their own ideas as opposed to directly answer questions. The electronic communication showed more formal expressions such as transition words. In his conclusion, Warschauer suggested that further research be performed according to nationality, according to the speaking fluency of the learners. This article is relevant for the present research as it presented the two variables that were used in this study. In other words, it presented the coordination index, which provides information on the complexity of sentences and the Type Token Ratio, which provides information on lexical complexity. Beauvois (1997) examined the affective and social benefits that students can derive from LAN (Large Area Network) communication. The purpose of her research was to examine, in controlled conditions, whether a link between written synchronous communication, via the Daedalus software, and oral communication could be established. She wanted to measure the transfer of skills that operated. Her participants consisted of 83 fourth semester students of French at the University of Tennessee in Knoxville (UTK) in 1995. The students were randomly assigned with half the class in a PC laboratory and the other half in a regular classroom. There were 49 females and 34 males. Classes were instructed using a communicative approach. Students were assigned the completion of tasks based on their readings from the Petit Nicolas a childrens book written by Andr Gosciny. At mid-semester and again at the end of the semester, all students took an oral examination, the grading of which was used to compile a T-test used to compare the study group and the control group. She found that the LAN group exceed the control group in grades and that not only was the LAN group better but also

PAGE 12

5 more homogeneous. According to Beauvois, the synchronous communication carried elements similar to conversation such as a high amount of turn-taking. Furthermore, she posited that there was a compelling character of the PC message with flashing visual prompts, which kept students focused. She also noticed that Vygotskys scaffolding theory applied to the synchronous communication, that is to say students benefited from one anothers input thus creating a peer teaching environment. Thus, she concluded that there was a significant amount of transfer of skills from one medium to the other and suggested that further research be performed at different levels of instruction in addition to investigating other language skills. The article is relevant to the present research as it offered several suggestions. First of all, the variables she had chosen to study were interesting as using grading as a variable to measure student achievement was considered. One of the problems that grading pofses is subjectivity. Furthermore, grading cannot provide an accurate illustration of the type of language that is used in both media. Therefore, the idea of using grading as a variable was rejected for the present study. Boyd, Davis, and Ralf Thiede (2000) examined what happens to different features of discourse when English as Foreign Language (EFL) learners must choose to function in an ESL situation as shown by the changes that occur in their writing in asynchronous electronic forums. Thus, Davis and Thiede studied the style shifting of students of ESL in response to native speaking participants accommodation to the experience of creating a learning community online. By pairing native speakers and ESL learners the researchers tried to emulate mimetism. They found that investigating social practices such as politeness, authority status or distance is not always simple. Students

PAGE 13

6 participating in asynchronous communication present themselves exclusively in a positive and polite manner. They can enter the conference at any point, read as little or as much as they wish, and choose to reply to whomever they want. The conventions on the forum included exaggerated politeness and signals of approbation with compliments that showed alignment more than partisanship. In order to measure the replication of language researchers studied lexicosyntactic indicators of stylistic emulation. In other words, they looked at an acquisition scale and at lexical density, which they defined as the number of lexical words divided by the total number of words. The two variables that they used were extremely interesting. On the one hand, the acquisition scale was not applicable as measuring acquisition was not the goal of the study and native speakers were not introduced as a factor. On the other hand, the definition for the lexical density was the exact same one as the definition for the Type Token Ratio that Warschauer (1996) had used. In Davis and Thiedes case, however, it was used to measure the production of language learners versus that of native speakers on the same medium, i.e., bulletin board software, which was used for the present study. In her article, Pellettieri (2000) studied the interaction and the negotiation of meaning in synchronous CMC. Her research questions included: 1) does the negotiation of meaning occur in task-based synchronous CMC? 2)do the negotiations facilitate mutual comprehension? 3) do the negotiations push learners to output modifications that are both meaning and form-focused? 4)do the negotiated interactions foster the provision of corrective feedback and the incorporation of target-forms in the subsequent turns? (p. 64) Her participants were 20 students of Spanish at the University of California at Davis. They were all native speakers of American English enrolled in intermediate Spanish. Students' interactions were observed as they functioned in dyads. She found that negotiation via synchronous CMC facilitated mutual communication and that

PAGE 14

7 negotiations pushed learners to output modification encouraging corrective feedback. Her conclusion offered pedagogical suggestions on tasks to be designed so that all participants are required to request and obtained information from one another for successful task completion so that communication is goal-oriented. This article offered the perspective of performing a research from a theoretical point of view. Pellettieri chose to perform her research from the interactionist point of view. The present research was based on the Cognitive point of view as Skehan (2001) defines it. The Cognitive approach also calls for goal oriented tasks and includes some interactionist elements. It is described in greater detail below. Bhlke (2003) designed his study to verify that participation in a CMC is more equalizing than F2F participation as was suggested by Kern and Warschauer (2000) whom he cited. His participants were fourth semester students of German as a foreign language using a communicative approach. The students participating in chats and F2F produced discourse from two activities presented on worksheets. Half of the students used CMC and the other half F2F communication. Contrary to Warschauer who used broader units of meaning, called T-units, in order to measure students language participation, Bhlke used C-units that he considers to be the fundamental elements of communication, i.e., a C-unit can represent one word only or a whole sentence. Consequently, the C-unit does not require a verb nor does it require a predicate, it is more inclusive than the T-unit. He found that group size has to be factored into the equalizing effect of CMC. Groups of 5 students did not tend to be optimal whereas groups of 4 offered a positive impact. He also measured students language competencies according to a scale of stages defined by Tschirner. He found that CMC is indeed more equalizing

PAGE 15

8 at certain stages of language than others. For future research, he suggested that the chat room should set the ground work for in-class discussion, and that more research should be performed on the ideal number of students within a group as well as further investigated Tschirners stages. His discussion on the C-unit was extremely interesting, however, using the C-unit may be too encompassing as it includes utterances of only one word such as yes or no as a unit of meaning. Particularly since in F2F communication turn-taking is much more important, and using C-units as a construct would create an imbalance with CMC communication. Therefore, it was necessary to perform the present research on units that would include a verb and a predicate. Tschirners stages are also interesting, however, they only apply to the German language and such a scale was not found for a research using French. Bhlke claims that chat room should set the ground work for in-class discussion, however, doing so seems redundant. How useful would it be for students to carry the same conversation twice? His approach to literature in CMC was also interesting. Indeed, he was the first to offer opposing views to most of the litterature. For instance, he offered an alternative explanation for the transfer of skills that Beauvois had suggested, expressing doubts that the implementation of chat does not imply transfer of skills since there is no immediate cause and effect relationship with increased speech proficiency (p. 70). Bhlke also voiced concerns for using CMC. Such concerns were expressed by Beauvois (1992) who claimed that students become increasingly indifferent to the appropriate use of the target language the longer they use the chatroom. Yet, the present research by Guillo also showed how some students in F2F started speaking in their native language in the language laboratory. Therefore, it is a concern to

PAGE 16

9 be had in both media. He further quoted Kelm who saw a disadvantage when students copy incorrect form from another students message (p. 71). It is a right claim although not only true to CMC since the lack of feedback both from a peer or a native speaker may lead a learner to acquire an erroneous form of language. Finally, he quoted Bremp (1990) it gets frustrating sometimes when a conference gets really busy and you would have no time to type anything if you worried about reading absolutely everything (p. 72). Although chatting takes on a form that allows one to realize that we may not be fast enough or have enough time to read every thing when we are communicating, it is also true to the oral language. It is not suggested that these concerns should be dismissed for the present research, on the contrary, they may be used for all communication purpose. As far as the present research is concerned, an electronic forum was prefered to a chat software. Ann, Chenoweth, and N. K. Murday (2003) noticed that students at Carnegie-Mellon Universtity were interested in taking a foreign language course but could not do so because of scheduling issues. Therefore, they designed an online class to meet those needs. Thus they wanted to measure whether this online course using CMC would be as efficient as a regular course. The participants for this research included students enrolled in French 1. They were all undergraduate students, 12 of them participated in the F2F evaluation and 8 participated in the online course. On the SAT II French exam, the online participants averaged 581 points and the F2F participants averaged 556. They all filled out a General Background Questionnaire (GBQ) and a Technical Background Questionnaire (TBQ). The students oral production was measured by interviews conducted after 5 weeks and at the end of the semester. The researchers gathered data

PAGE 17

10 from a focus group and an interview from one student. They found that in all the testing done there was no statistically significant difference between the two groups. Thus, students registered in the online course made similar progress but they offered more mitigated satisfaction feedback than students in the conventional course. As with the research done by Beauvois, the manner in which students performance could be used as a variable to compare two media was intriguing. To solve the question of subjectivity, Chenoweth and Murday had several impartial graders. Doing the same was not feasible within the limits of the present research. Therefore using grades as a variable even with impartial graders was abandonned. This literature review offers information on how the use of technology can be compared to conventional teaching practices. Most of the articles dealt with the use of synchronous communication as opposed to F2F communication. The researchers interests varied from theoretical approaches to systematic statistical research. In the end their concern is the same, how efficient can CMC technology be in the language classroom? Overall the current literature did not present many articles on electronic forums or other asynchronous communication. No article offered any comparison between electronic forums and F2F communication. That is because electronic forums are asynchronous communication and F2F is synchronous. This fundamental difference makes them difficult to compare. Yet as asynchronous communication can often be presented as an alternative to in-class F2F communication, it is important to examine how different those two media are when student language production is concerned.

PAGE 18

11 Understanding these differences will allow teachers to make appropriate pedagogical decisions about which medium to use and how to use them.

PAGE 19

CHAPTER 3 METHODOLOGY Research Questions We know that CMC is different from F2F communication. Additionally, we know that asynchronous communication is different from F2F communication. However, we do not know to what extent these modes of communication are different nor do we know if there exist similarities between them for foreign language learners. For these reasons, the following research questions were asked: 1. How does language produced in an electronic forum compare to that of F2F communication in terms of grammatical complexity? 2. How does language produced in an electronic forum compare to that of F2F communication in terms of lexical complexity? 3. What patterns, if any, can be noted from both modes of communication? Participants The participants in this study consist of two classes of students enrolled in a French grammar class at the intermediate level at the University of Florida. The University of Florida code and title for this class is: FRE 2200 Intermediate Grammar. The project was a class assignment but each student volunteered for the research and signed an informed consent form approved by the University of Florida Institutional Review Board 02 under the protocol number 2005-U-0295 for use through 06/4/2006. All the data was kept anonymous. There were 22 students who participated in the protocol, 17 of whom were women. No other ethnographic data was requested. The textbook used in FRE 12

PAGE 20

13 2200 was UInteractions U; chapters 4 and 5 of this textbook provided the language and cultural impetus for the activity used in this study and detailed in Appendix A. Task The activity was created with Skehans model in mind (Skehan 2001). At the center of Skehans Cognitive approach is the action of Noticing, which first integrates input into the working memory and into the long term memory. Noticing occurs when input takes on several qualities, namely frequency and salience. The teachers task is to create focused input. A particularly interesting aspect of Skehans model is that it takes into account learners internal factors (readiness and individual differences), forcing a teacher to also take the various learning styles into account. The application of this model in the classroom requires task-based instruction. In other words, the CMC activity will follow these guidelines provided by Skehan (2001 pp. 121-152): Meaning is primary; There is some communication problem to solve; There is some sort of relationship to comparable real-world activities; Task completion has some priority; The assessment of the task is in terms of the outcome. Furthermore, Skehan (2001 p. 152) specifies that tasks: Do not give learners other peoples meanings to regurgitate; Are not concerned with language display; Are not conformity-oriented; Are not practice-oriented; Do not embed language into materials so that specific structures can be focused upon. With all these guidelines in mind, a series of communication-driven tasks for a CMC activity was created (appendix A). The project is also designed to provide students with an opportunity to use their language skills through negotiation as well as to develop an understanding of French culture through the media. The media in question are

PAGE 21

14 television and advertising, particularly how commercials fit into French television programs. Commercials provide an idea of the target audience the marketers are trying to reach. The positioning of the advertising also provides information on this target audience as it occurs at the particular time when this target audience is most likely watching television. Some information the advertising provides about the target audience includes demographics, such as age group, income bracket, geographical location, education, values, etc. Once the marketer has drawn a sketch of the target audience, he/she will identify the programs this target audience is the most likely to watch on television in order to position the advertising so as to maximize the reach, i.e., the proportion of the target audience that will be sensitive to the advertising message. For students it translates into an exploration of French television programs and particularly what French people watch and how this compares to US programming. Students performed a CMC activity that helped them draw conclusions about French programs as well as the make-up of the audience for these programs. According to Skehan, an activity is more efficient if students are provided with an adequate background and are given a role. For this activity, students were told that they were members of a marketing company, The Famille Guillo Advertising Agency, whose CEO was offering them an opportunity to lead a project related to the marketing of a product and particularly, to the positioning of the advertising of that product. Appendix A contains a detailed description of the project the students accomplished, for both the CMC and F2F groups.

PAGE 22

15 Analysis The present research was performed using two different classes of FRE 2200. The first class consisted of 15 students distributed within 4 groups, one group of 3 students and 3 groups of 4 students, who were instructed to use the electronic forum to perform all their communications. They will be referred to as the Computer Mediated Communication group or CMC group. The second class consisted of 7 students who met in the language laboratory twice in the course of the project during regular class sessions. They will be referred to as the Face to Face group or the F2F group. Data from the CMC group was retrieved directly from the electronic forum using the copy/paste functions. They were then compiled in a Microsoft WordPTMP document in order to perform the analysis of the data by hand on printed material. The data from the F2F was collected on the computers of the language laboratory of the University of Florida using the Divace software. The data from this group consisted of audio files that was transcribed afterwards on a written Microsoft Word document. Thus, this data was subjected to the researchers interpretation. Furthermore, on the second session of the project, eroneous manipulations to save the data were performed by the researcher and the entire session was lost. Therefore, only half of the data expected to be used was analyzed. The data was analyzed in relation to two variables. The first variable was used to identify the syntactic complexity of sentences uttered or written by students according to a ratio called the Coordination Index (CI). Warschauer (1996) defined this ratio as the number of dependent clauses divided by the number of total clauses. Further information is provided in the Results and Conclusions section of the present thesis. The second variable measured the lexical complexity displayed by students in the form of another ratio called the Type Token Ratio (TTR). Warschauer defined the TTR as the amount of

PAGE 23

16 different words produced divided by the total number of words produced. As for the CI, further information about the TTR is provided in the Results and Conclusions section of this thesis. Both ratios are similar to averages or probabilities. An average is the part divided by the whole and a probability is the number of desired outcomes divided by the total amount of outcomes. Thus both variables will provide us with a general idea of the patterns, if there are any, present in students communication as well as allowing us to make predictions on such behaviors. Another advantage to both variables is that they can easily be applied to both communication media. As previously stated, one medium is asynchronous while the other is synchronous, therefore we expect differences. The goal of this study is to measure these differences and derive pedagogical implications related to these findings.

PAGE 24

CHAPTER 4 RESULTS AND CONCLUSIONS Syntactic Complexity In order to determine the coordination index for the two types of samples that were collected, I first removed all the utterances that were produced in the native languages of the students. Second, the utterances that pertained to the task demands that requested a written preparation, particularly the first part of the project, which consisted of providing a summary and a description of the advertising were eliminated. Since students were provided with guidelines in the form of questions, they created a written sample and since the spontaneous language that students created was the object of this research, a written sample was not desirable. For the CMC groups, it consisted of their first participation to the thread for the most part. For the F2F groups it consisted of any part that was read from their course preparation. Any single word feedback such as oui, non etc., repetitions, etc., were eliminated. Thus, only the most spontaneous language that they produced while negotiating for the one commercial each group was going to analyze, the type of program during which they would place their advertising, as well as any off-topic discussion, was left. Once, the utterances were narrowed down to usable samples, each clause of the samples were outlined. The clauses were separated into two categories: Dependent clauses (D) and Independent clauses (I). A independent clause is a clause that has a meaning by itself. A dependent clause is a clause that has no meaning by itself. For example, if we analyze the following sentence I met her at the restaurant where we had 17

PAGE 25

18 our first date there is one independent clause, I met her at the restaurant, which would be meaningful by itself, and a dependent clause where we had our first date, which is meaningless by itself. In this sentence all the verbs are conjugated and the relative clause introduced by where is easy to determine. However, dependent clauses also include clauses where the verbs are not conjugated. These clauses include participial clauses and infinitive clauses. No participial clauses in the samples collected were found, however there were a few infinitive clauses. Although attention was paid to the types of dependent clauses present in the data, this analysis did not reflect these various types. In other words, the coordination index does not discriminate between relative clauses, completive clauses or other subordinate clauses. Instead, it offers an insight on the general syntactical complexity of samples. The coordination index is a ratio of the amount of dependent clauses over the total amount of clauses. Therefore, the higher the ratio the more complex the sample. Once, dependent clauses are separated from the independent clauses, each were counted separately and computed in an Excel chart according to the following formula for the CI: CI= number of dependent clauses/total number of clauses Table 3.1 summarizes the results of this data. D represents the dependent clauses, I represents the independent clauses and CI is the Coordination Index. The data is presented on a per student basis, a per group basis, and finally the totals were compounded. Each student was studied in his/her order of appearance for the threaded participation or the oral exchanges. For every group a total, an average and a standard deviation was calculated in order to identify individual patterns. Finally, the total,

PAGE 26

19 average and standard deviation were calculated for the two different means of communication so as to identify general patterns. The data highlighted in gray represents the students whose production was also used for the TTR. Table 3.1 : Summary of findings according to the CI The last factor to take into account for this analysis is that the CI does not reflect language accuracy. In other words, all clauses were counted, regardless of whether they were grammatically correct. They may not have been introduced by the right pronoun or the verb may have been incorrectly conjugated. Therefore, when the CI is determined and compared, the tendencies for students to formulate complex sentences are compared, without comparing their degree of language accuracy. Interestingly, as the data shows for every group who used CMC except for group CMC3, the first person to participate on the forum for each group is also the person who provided the most data. CMC Student 1 has a total of 101 clauses, CMC student 4 a total

PAGE 27

20 of 90, and student 12 has 97; each one of these students far exceeded the amount of total clauses of all the other members of their group. This accounts for the very high standard deviations among the total number of clauses and thus shows a very uneven distribution of production. The total amount of clauses is roughly correlated to the total number of threads that each student wrote. Table 3.2 below summerizes the amount of threads that each students wrote: Table 3.2 : Total number of threads # o f threads Gp 1 1 10 2 4 3 3 Total 17 Gp 2 4 24 5 10 6 13 7 17 Total 64 Gp 3 8 5 9 6 10 8 11 3 Total 22 Gp 4 12 9 13 4 14 9 15 4 Total 26 Not only did these students participate more and thus provide a larger quantity of data but they also proved to be the decision-makers and the leaders of each group. They were the students who prompted their peers to provide feedback. They were also the motivators to solicit actual work and meet deadlines. Finally, after consulting with their

PAGE 28

21 peers they made the final decisions as to which commercials and programs were going to be used. Below is an example of student 12s motivational input: -T Je pense que nous devons choisir ou la pub de Axe ou la pub de Peugot, les deux sont faciles. Mais il faut que nous choisisions une pub ASAP!!!! I think that we must either choose the Axe commercial or the Peugeot commercial, both are easy. But we must choose a commercial ASAP!!!! The pattern was true to all the groups except for group CMC3. If we examine the data a little closer we can see that the first student to participate was not the one who provided either the most data nor the one who wrote the more threads. CMC Student 10 actually did, and CMC student 10 was the leader of her group and made the decisions in her group. However, this was not always the case within this group. When we look at the dynamics for this group we observe that CMC student 8, who actually started the thread, was the initial leader. The leadership was then taken over by CMC student 10. This data suggests that when a teacher offers a communicative task where students will be participating on a forum, the first students of each group will be very likely to become the leaders of their group. Furthermore, the teacher will also be able to expect that the leader of a group will generally be the one who will participate the most. What one cannot predict is whether the leader of a group will also be the student whose language skills are the best. The coordination index is an average that represents language complexity. In other words, it will show students tendencies to express themselves in a more complex form of language. Yet, of all the students that qualified as group leaders only one, CMC student 1, had a higher CI then the rest of his/her group. CMC students 4 and 10, with CIs of .300 and .357, respectively were even below their group averages of .340 and .395, respectively. They were also below the class average and means of .381 and .362, respectively. What does this suggest? For one, leadership

PAGE 29

22 does not equal language complexity or skills. Second, it may suggest that leaders tend to offer straight to the point communication. Inversely, the last student to participate on the forum was also generally the one who participated the least and thus provided the least data with the exception of CMC student 7 of group CMC2. This an interesting predictor and, therefore, it is worthwhile examining what could be the reasons for this pattern. Bhlke (2003) suggested a correlation between group size and participation. For him, CMC has an equalizing effect provided that the group size does not exceed five and not be lower than four. He also suggested that more research be performed on the matter. From the present data one can conclude that as long as an electronic forum is used as a CMC group size does not appear to be relevant. Three groups comprised four students whereas one consisted of three students. Even though group CMC1 consisted of three students, the last one to participate, CMC student 3 hardly communicated and most of the work was done by CMC student 1 and 2. In group CMC3 and CMC4 CMC student 11 and CMC student 15 also communicated far less then their peers with 27 and 18 total clauses, respectively. Their CI was also below average with .333 and .389 respectively albeit not as much below the average of the groups, which used CMC. Thus group size was not a relevant predictor nor did it have an equalizing effect in the present data. The slightly below average CIs of those students who participated the least may suggest that their language skills may be a corrolary to explain their level of participation. Other factors may include motivation for performing the task as well as their familiarity and motivation with the means of communication. However, no questionnaire was offered to inquire about their familiarity with the technology used, no question was asked about their affective

PAGE 30

23 response to the same technology either. Therefore, no conclusion can be definitely drawn at this time. Leadership patterns can not be derived as easily for students who participated in groups in the F2F tasks. In fact, none can be derived whatsoever. The first students to participate in both groups did not display more of a leadership ability as the students in the CMC tasks. Furthermore, it appeared that leadership was more diluted and more democratic. All students of group 1 offered their opinion agreed together as to what commercial was going to be used to complete the task. In group F2F2, two students, F2F students 5 and 6, seemed to argue but the final decision was taken between F2F students 4 and 5. F2F Student 7 participated minimally as is shown in the table: he/she only provided 7 clauses. Although CMC students were specifically instructed to conduct all communication on the forum, the researcher discovered that some of the communication was conducted via other means such as telephone, email and oral in-class meetings. No data pertaining to the amount of communication that was performed outside of the forum was collected, therefore it is impossible to determine what percentage it represented. Testimonies in students participation suggested that they had actually communicated by other means. Below are examples of threads suggesting other means of communication. -Comme nous avons dit en classe, stade 2 et cd ajourd'hiu sont bons programmes pour passer nitre pub. -As we decided in class, Stade 2 and CD aujourdhui are good programs to position our commercial -Je suis confondu de que notre groupe fait pour le rapport final. [] quel est votre e-mail. Quelqu'un s'il vous plat m'envoie un e-mail [] -I am confused about what our group should do for the final report. [] what is your email? Please someone, send me an email at []

PAGE 31

24 This explains the discrepancy among all the groups that participated. Students behavior can also be analyzed by looking at other factors, such as the amount of threads that they wrote combined with their CI. The Bulletin Board software used provided data that included the amount of threads, the time and the date of each thread and other statistical data. For instance group CMC1, CMC3 and CMC4 totaled 17, 22 and 26 threads respectively. Their total clauses were 136, 129 and 229 respectively and the average amount of clauses per person was 45, 32 and 57. On the other hand, group CMC2 had 64 total threads for 288 total clauses and an average of 72 clauses per person. These figures give us some insight on the use of the forum for communication. Either all groups managed to effectively make decisions, without much negotiation or they used other means of communication to come to those decisions. The latter is what happened in reality. As for group CMC2, they used the forum to communicate often. One of the consequences of participating on an electronic forum is the community building effect. Many Internet businesses use that effect to create a community around their products and also to create brand loyalty. Consequently, community software such as an electronic forum has developed devices to create loyalty. Such devices are the use of emoticons (smilies) and avatars (personnal pictures that appear on their profile every time they post a thread) along with the necessity to register for a forum before participating and the use of attractive Graphic User Interfaces (GUI). One of the goals of the researcher was to create a community around the French class that he taught using the Forum. Group CMC2 who displayed the necessary characteristics of community building was a success. Their enthusiasm and their use of the forum was evidence of that effect. They enjoyed identifying themselves with avatars already provided in the forum,

PAGE 32

25 and they even actively sought additional avatars to include. They also registered on the forum as indicated using pseudonyms that did not disclose their identity but identified them through other traits. They also used emoticons extensively to express feelings that cannot be expressed in the written form. Finally, they continued communicating on the forum after the project was completed. Other students from the other groups did the same but not to the extent of the students in group CMC2. Finally, as we look at the averages for all the groups, we notice that the standard deviations from the means are pretty high. The dispersion from the amount of total clauses and the amount of dependent versus independent clauses as indicated by the standard deviation, indicate a wide variety among those students. On the other hand, this variety is not reflected in the average CI for all groups. This average is .381 with a mean at .362 and a standard deviation of .093, which shows a certain homogeneity in the type of utterances, which is to be expected as all the students were approximately at the same level in the same class. The amount of participation for the F2F groups cannot be measured by or with turn-taking as on the forum. Indeed, many turns only comprise of one word, indicating acquiescence or feedback, which is one of the fundamental differences between a written and an oral performance. On the other hand, we can derive participation with the amount of total clauses. Group F2F1 is homogeneous at every level. The CI of each student is very similar and, as the standard deviation revealed, they all participated equally and produced a language of an equal level of complexity. Their average CI is .228 and their mean is exactly that as well, with a standard deviation of .003, which is very narrow. This group

PAGE 33

26 displayed quite an efficient level of communication that would be labeled as synergy in the business world. Synergy is the ability to create seamless communication for the most efficient business practices. Although such homogeneity is ideal, it is also suspect for it could be abnormal. Unfortunately, it is almost impossible to know accurately whether or not it is natural, given the amount of data. Additional data would definitely help in defining the statistical validity of this pattern. However, when we look at the data, the CI for F2F students 4 and 5 of group F2F2 with .240 and .246 respectively are not far removed from the average CI of group F2F1, which is .228. In fact, they are very close, which suggests that we could predict a certain homogeneity for the oral medium and that we need to treat F2F Students 6 and 7 separately. As the researcher was also the teacher for this class, he felt that there were certain significant differences that were not measured. For instance it was striking that two of the students did not participate as much as they did in a regular class setting. These students may have been influenced by the Novelty Effect (Kern, Warschauer 2000). In FRE 2200 at the University of Florida, students do not typically use the language laboratory, thus creating motivation for students who find themselves in a new environment while using new technology. In this case the language lab offered the novelty, the technology and the new environment. The novelty effect predicts that students will react positively to a new environment and that the effect will dwindle as this environment becomes more and more familiar. In the case of the present research, however, students did not have time to experience such familiarity, thus the novelty effect carried over the duration of the task. On average, the researcher felt that all students in general, and certain students in particular, participated more than they usually

PAGE 34

27 did in class. The Novelty Effect of the language lab was also more equalizing than on the forum. F2F student 7 of group F2F2 was the only students who maintained the same level of participation, which was quasi inexistent in general. This group did not show as much homogeneity and its dynamics were far different. For instance, although F2F student 6 is the student who shows the most complex language in the classroom, she was also the students whose production was eliminated for the most part. Indeed, she produced many utterances in her native language, i.e., English, and even though the researcher was monitoring the class, he realized afterwards that she spoke English when he was not monitoring her group. There were some attempts from her peers to make her speak French. However, Group Think was not powerful enough to change her behavior. Group Think is when a deviant members behavior is reduced or stopped altogether under the tacit influence of a group majority. The effect was first identified under the Kennedy administration when dissenting politicians were stopped under penalty of being excluded from the group. In a language laboratory, it is difficult for an instructor to maintain a certain level of discipline for the simple reason that every group also requires help and attention. That is when group dynamics can help. For the present research, groups were randomly assigned. Perhaps assigning students according to criteria that include language skills and personality types would have helped in assuring discipline. F2F student 7 is a very interesting case as she is also the student that present the highest CI of the oral groups, with .492. This number is extremely high compared to the other students in the oral groups. Furthermore, when comparing this result with the results of the CMC group, we notice that it is much higher than the average for all the

PAGE 35

28 groups. It may suggest that this student has a tendency to express herself with more complex sentences. A good way to measure if that is the case would be to compare her native speech with the second language speech or/and to extract additional data from her or his daily language production. Then we could also determine whether her lack of discipline can be attributed to frustration or lack of confidence in her oral skills. We expect the language in both media to be different. Whether synchronous or asynchronous, CMC carries elements of the written language. About synchronous CMC, Chun (cited in Bhlke 2003) suggests that the CMC environment is less stressful than oral discussion, because students have more time to think about their utterances and do not have to worry about their pronunciation. It is all the truer for asynchronous CMC where the pressure of spontaneity is completely eliminated. Warschauer (1996) showed that synchronous CMC chat was more complex than face-to-face communication among his advanced students of English as a second language at the University of Hawaii. The Hawaii CI for the face-to-face group was .182 and for the CMC group was .475. As a result we can expect CMC communication to be more syntactically complex than oral communication. As far as the present research is concerned, the electronic discussions were completely devoid of negotiation of meaning, which occurred a few times with the oral groups. As expected, the average CI for the CMC groups was higher than that of the oral groups, .381 and 278 respectively. When we look a little closer at the types of dependent clauses produced, we notice that elements of the oral language permeated almost as much in the forum. One particular element is the amount of dependent propositions introduced by a variation of je pense que or je crois que, both translated as I think. These clauses are

PAGE 36

29 completive clauses and they are part of the oral language because in the written language students are instructed not to use the first person of the singular je or I. Furthermore, when one writes an argument one does not use personnal opinion or conjecture, instead one supports ones argument with tangible evidence. There were 62 out of 298 such completive clauses in the CMC data and 18 out of 98 in the F2F data. Proportionately, both results are very close with a ratio of .208 for the CMC and a ratio of .184 for the oral production, the difference is only .014. This suggests that the proportion of dependent clause types will be maintained in student language from oral production to CMC production. In order to determine that this proportion is not just an effect of the CMC sharing elements of oral production, it would be interesting to compare individual students written production to the present data, both for the CMC group and the control F2F group. Lexical Complexity The Type Token Ratio provides allows to determine the semantic diversity students display. It is determined by identifying all the different words in a given sample and comparing those words to the total amount of words. Thus we obtain the following formula: TTR=Amount of different words/total amount of words In order to determine the TTR, random samples of 200 word productions from students in the two media were retrieved. In the interest of time, a limited sample size allowed for easier analysis. The sample size was determined by three factors. First of all, the TTR is not a typical ratio like the CI, which resembles a probability or an average more than the TTR does. Indeed, the TTR requires that each different word be counted, yet the bigger the sample the less diverse the vocabulary, and an individuals vocabulary

PAGE 37

30 capacity will eventually be reached. Secondly, the data available did not allow for samples much higher than 200 words (particularly in the oral production groups). Finally, in the interest of time the samples were reduced to a workable size as the analysis was going to be performed by hand. To clarify what is meant by amount of different words it is best to use an example. The sentence the cat eats the mouse contains five words total but four different words as the is repeated. Thus, the TTR for this sentence is 4 over 5 or .80. The difficulty is in determining what constitutes a different word. One particular rule was kept: only words that were semantically different were counted and therefore morphological or syntactical variations were eliminated. For example, all the variations of the French definite articles le, la, les, l were eliminated after one of them was encountered once, however, I had to make sure that those words except for l were not variations of the direct or indirect object complements, which would be considered differently. Other examples included conjugated variations of a verb, for instance, I eliminated pensons, the first person plural of the verb penser (= to think) after having seen pense first person of the singular. The importance was to remain consistent when retrieving all the data. Table 3.3 below summarizes the findings of the TTR:

PAGE 38

31 Table 3.3 : Summary of the TTR CMC # o f different word Total TTR F2F # o f different word Total TTR Student 1 92 200 0.460 Student 1 79 200 0.395 Student 4 83 200 0.415 Student 2 61 200 0.305 Student 5 71 200 0.355 Student 3 72 200 0.360 Student 6 77 200 0.385 Student 4 84 200 0.420 Student 8 78 200 0.390 Student 5 82 200 0.410 Student 10 87 200 0.435 Student 6 88 200 0.440 Student 12 59 200 0.295 Total 547 1400 0.391 Total 466 1200 0.388 STDEV 0.054 STDEV 0.049 The same considerations as for the CI applies to the TTR. The TTR does not reflect language accuracy. In other words, it does not say whether vocabulary is employed relevantly, nor whether it is spelled or pronounced correctly. Unlike for the CI, a per group analysis was not performed since not all students yielded a sufficient amount of workable data. As for the CI, data considered workable included, any data that was spontaneous, therefore all the written samples that were requested by the project were excluded. The written samples that were excluded were the summary and description of the commercial, the rationale on where to position the commercials, and the final report (see Appendix A). Thus, data from students 1, 4, 5, 6, 8, 10 and 12 of the CMC group was used. Overall, they were all very consistent and ranged from .295 to .460. The average was .391 with a standard deviation of .054, which indicates a limited dispersion from the mean. This consistency and limited dispersion suggest a marked homogeneity among the students semantic diversity. Students in the F2F groups showed consistency as well. Their TTR ranged from .305 to .440, with an average of .388 and a standard deviation of .049, which again indicated a limited dispersion from the mean. As for the CMC groups this consistency

PAGE 39

32 and limited dispersion suggest a marked homogeneity among the students semantic diversity. Since both media are different and since the CMC carry similarities with written production, one would expect the TTR to also be different for the two media. The CMC groups should show a higher TTR, thus a higher semantic diversity than the F2F groups. Warschauer (1996) also determined the TTR for his advanced students of English as a foreign language at the University of Hawaii. The F2F group yielded a lower TTR of .262, compared to the synchronous CMC groups of .301, as expected. Perhaps this difference can be attributed to the task Warschauer gave. Students in his research were supposed to answer two counterbalanced questions (p. 12). However, one would reasonably expect to find the same pattern with the present research. Yet, both media yielded a similar consistency, as we have seen in the group analyses above. This similarity ended up being much more striking than expected since the two groups displayed a margin of only .003 between the two TTR of .388 for the F2F group and .391 for the CMC group. This is reinforced by the two standard deviation of .049 for the F2F group and .054 for the CMC group. These numbers allow us to conclude that intermediate students of French at the University of Florida display the same level of semantic diversity both orally and using asynchronous CMC communication. It further suggests that students language on an electronic forum is similar in semantic diversity to F2F communication contrary to previous affirmations that asynchronous communication is more diverse than face-to-face communication within the context of the task provided to the students. However, this may be because students

PAGE 40

33 had used all the vocabulary requested by the task, since the task was specific and encompassed the vocabulary of two chapters from the students textbook. Implications None of this data suggests that one medium is better than the other. However, there are some pedagogical benefits to both of them that will be discussed herein. Then, some suggestions about running the same project or any other that would be similar will be offered in light of students behaviors and also in light of their feedback. The above data suggests that running the project on an electronic forum did offer some of the benefits that were stated in the introduction. All of the students respected the rule about posting threads in French. The messages they posted included both characteristics from the oral language and the written language. Having their messages posted on the Internet certainly influenced their behavior. None of them required any particular training as to how to use the forum and most actually used it instinctively. A few students were even so familiar with the software that they developed their own profile and adopted interesting behaviors. Some of these behaviors included using a different colored font for every message that they posted (see appendix B), or the use of avatars and emoticons. The majority of the students only performed what was asked of them and, albeit, they did not necessarily find using the forum practical for the duration of the project. One can deduce that by looking at the number of threads they posted as well as their messages suggesting other means of communication. Their final feedback on the project was more obvious. Here are some examples of positive and negative feedback: -Pour la plus part je pense que je n'ai pas apprci ce projet en raison du format de forum. Mais, le travail tait assez facile, seulement je desteste des ordinateurs.

PAGE 41

34 -For the most part I think that I did not appreciate this project because of the format of the forum. However, tasks were rather easy, I just hate computers. -Je suis aussi pas bon avec les ordinateurs et comprendre le site tait au dbut trs dur. Mais c'est plus facile crire le franais pendant qu' un ordinateur parce que c'est plus rapide pour chercher des mots que vous ne savez pas. -I am not very good with computer and understanding the site was very difficult at first but it is easier to write in French with a computer because it is faster to find words that you do not know. Overall students reported that they did not particularly enjoy working on the forum. All the groups agreed that they needed time in class to coordinate better. Yet, members of group 2 still maintained communication on the forum after the project was over. As for the F2F group, they enjoyed working in the language laboratory very much. Here are examples of their feedback: -Je pense que ce projet a t intressant. Jaimais utiliser les couteurs. Je prfre travailler dans le laboratoire de langues que dans la salle de classe. [] -I think that this project was interesting. I liked using the headphones. I prefer working in the language lab than in the classroom. -Nous pouvions aussi pratiquer nos comptences de grammaire et commuication. Le projet tait une bonne ide -we could also practice our communication and grammatical skills. The project was a good idea. -Le projet a t trs bon pour practiquer le francais. Nous avons utilis beaucoup les mots uniques de la projet. Il a t difficile comprendre les autres personnes dans le group quand nous avons dcriv les pubs. Mais, ctait un bon exercice pour nous parce que nous avons du comminquer nos ides entre eux. Quand quelquun ne comprendait pas les choses quun autre disait, cette personne a d chercher un autre manire dexprimer son ide. -The project was very good for practicing French. We used many unique words of the project. It was difficult to understand the other people in the group when we described the commercials. But, it was a good exercise for us because we had to communicate our ideas to one another. When someone did not understand the

PAGE 42

35 things that someone else was saying, this person had to look for another way to express his/her ideas. The overall feedback on the F2F version of the project was much more positive than that of the forum. Students did indeed negotiate meaning often, which was a very beneficial exercise for all of them. The only problem was to accommodate the project and the time in the language laboratory within a pretty heavy syllabus. One of the advantages of running the project with a forum was to offer the possibility to do work outside of the classroom. Thus it was easier to accommodate the project within the syllabus. On the other hand, having to do extra work outside of class may have been a reason why students did not enjoy the electronic forum. The final product of the project was equally satisfactory with both media. However, since so many students ended up not using the forum for actual negotiation, but only to post their final thoughts, and since so many groups ended up using other means of communication, it is not clear that they actually benefited as much from the project as the F2F group did in the laboratory. Furthermore, although the forum offered controlling possibilities such as time and date when the thread was posted, it did not allow to control the rest of the students negotiations. These negotiations offer the best opportunity to improve students language skills therefore it is imperative for the teacher to be present to offer feedback and control the exchanges. The forum was indeed not as beneficial partly because the teacher did not add threads to encourage students to participate more. If the teacher had been more involved and encouraged students to post replies by asking questions for instance, perhaps students would have used the forum better. The forum could also be used in combination with a F2F medium. For instance, using the forum for posting work such as the description of the commercial and for completing the final project would be

PAGE 43

36 beneficial. Students are then at liberty to review what the other groups are doing and can also check on their progress. This is what a student wrote when she/he was worried that her team had not yet made a decision as to which commercial to use: -Vous n'avez pas choisi une pub. Mais c'est 11:52, et puis je vais choisir une pub pour le group. J'ai regarde les autres groups et il y a un group qui a choisi "le sculpteur" et il y autre group qui n'a pas choisi une pub mais ils peuvent choisir la pub de Axe mais je pense que il n'ont pas la choisi. Donc, je vais choisir la publicite de AXE pour notre group. Si, on peut changer plus tard quand vous avez repondre, nous changons. Si, non, donc j'ai choisi la pub de AXE pour le group. -You did not choose a commercial. But it is 11:52pm, and then I am going to choose a commercial for the group. I looked at the other groups and there is a group that chose le sculpteur (=the scupltor) and there is another group that did not choose a commercial but they may choose the Axe commercial but I think they did not choose it yet. So I am going to choose the commercial AXE for our group. If we can change later when you answer, we will change. If not, then I chose the commercial Axe for the group. Finally, pairing students with native speakers from a French speaking university to add a cultural exchange component to the project should be considered. Insight from native speakers would help students understand certain cultural subtleties related to the media. It is all the more important as we live in an era of global communication. The CI helped determine that there is a difference in complexity between CMC and F2F communication, whereby CMC offer more syntactically complex communication. However, they both proved to be as varied in lexical terms. As one of the students in the F2F group pointed out, it is possible that negotiation of meaning pushes students to be more creative in an oral setting. In order to obtain more insight into both media and the languages produced therein research should be performed using additional data, such as written samples, to compare both groups so as to find out whether the type of dependent clause (such as those introduced by je pense que (I think that)) is maintained proportionately in the written samples, on the forum and in F2F samples. Thus we would be able to determine whether this proportion is more a factor of the oral

PAGE 44

37 language while using CMC or not. Furthermore, future research on the matter could include additional variables that pertain to language accuracy. It is important that we look at variables that can be consistent with both media. In other words, we cannot look at spelling errors as the oral language in F2F communication does not offer such insight. Yet we can look at patterns in errors such as preposition mistakes that students would typically make. We could also look at morphological errors such as verb conjugations, particularly with the use of the auxiliary tre ou avoir (to be or to have) in analytical tenses such as the pass-compos. Finally, we could collect additional data to determine whether the homogeneity that was found with the TTR variable during the F2F exchanges was an abnormality or not. This could be done by performing the research using different tasks that would vary in specificity in order to elicit a larger variety of vocabulary.

PAGE 45

CHAPTER 5 CONCLUSION The first research question of this study aimed at comparing the grammatical complexity of the language produced by two groups of students. One group produced their language on an electronic forum and the other communicated orally. In order to compare these two groups grammatical complexity, the CI, which compares the number of dependent clauses over the total number of clauses, was compounded. The results for this variable suggest that the language produced on an electronic forum tends to be more complex than the language produced orally. The written nature of the language used on the electronic forum accounts for this result. However, the CI does not take into account the types of dependent clauses and the data revealed that there was a similar number of completive clauses in the language produced with both media. This suggests that more research be performed in order to compare the quantity of the various types of clauses between the two media and written and spoken samples of the native language of the students. The second research question of this study aimed at comparing the lexical complexity of the language produced using both media. In order to compare the language complexity, the TTR, which measures the number of different words over the total number of words, was used. While it was expected that the TTR of the language used in the electronic forum would be higher than the one used in the F2F environment, it was not the case for this research. Indeed, the TTR for both media were almost identical. One reason for this result may be the fact that at the intermediate level of their language 38

PAGE 46

39 apprenticeship and for this particular task, students had reached and used the extent of their vocabulary bank. Thus, this data suggests that more research be performed in order to compare both media again with data coming from various tasks. The third and final question of this research aimed at whether other patterns could be discovered from this type of data. First of all, it revealed that students display interesting behaviors on an electronic forum. The first student to write a thread is more likely to become the leader of the group and the last student to write a thread is more likely to be the student who will participate the least. Students reported that they did not enjoy working on the electronic forum claiming that it was not convenient for negotiation and, indeed, most of them ended up using other media for communication. It is possible that students did not enjoy the extra work to be performed outside of the classroom in addition to other factors, thus they did not enjoy using the forum as much as if they had had to perform the task in class. One group of students, however, enjoyed the community building characteristic of the electronic forum. Most students in the F2F group reported that they enjoyed communicating in the language laboratory. They enjoyed the new environment and felt that negotiating meaning among other aspects of the task was beneficial. In fact, some students even participated much more than in the regular class setting. Learning about the differences between the types of language produced using an electronic forum and F2F communication can help teachers make informed pedagogical decisions on when alternative to use an alternative and how to use it when they design a task. This research revealed some of these differences and perhaps similarities.

PAGE 47

40 However, additional research is necessary to examine more of these differences for teachers to be able to make better informed pedagogical decisions.

PAGE 48

APPENDIX A PROJECT DESCRIPTION FRE 2200: Activit mdiatique Dear employees, You are my best team of advertising executives, here at Famille Guillo Advertising Agency. and I am proud of the work you did on our last campaign. I have been contacted by several French clients who would like to use your talents for their campaign. Your job will be fairly easy, as our clients have already shot the commercials. This will be your first time working in an international environment so be aware of cultural differences and Uall your production must be in FRENCH (including the meetings U). Your first task will be for each one of you to choose a commercial and to analyze it according to our usual guidelines. The most important thing of course will be the identification of the target audience. Then, as a team, you will have to decide on a single commercial from the ones that each one of you will have chosen. Then, individually you will have to research TV programming and choose two shows that the target audience is most likely to watch so that we reach as much of that target audience as possible. We want our clients messages to be as effective as possible. Once each one of you will have identified the two shows, as a team, you will have to decide on which two shows will best suit your commercial. Your budget is limited, that is why I am only allowing two shows. Finally, you will have to give me a report on all of your progress and all the decisions you made and why. This report will be no longer than two pages. Attached, you will find a timeline, a copy of our guidelines, links to the 2 French TV channels with the largest viewership and the guidelines for the final report. I almost forgot Since you are so busy, you will not be able to meet at the same time. That is why I have requested that a forum be created on our web site. Each one of you is required to post all of your work on it as well as your team communication. You will also find the link to the forum as well as guidelines on how to register. Go ahead and make me and all of the Famille Guillo Advertising Agency proud of you again. 41

PAGE 49

42 Respectfully, Cyrille Guillo President Directeur General UTimeline for the CMC groups: All of your communication and production will have to be posted on the forum at the following website: HThttp://www.familleguillo.com/forumTH You will have to register in order to be able to post. To register click on Register at the top of the page. Then go to FRE 2200 Projet mediatique find your group and click on reply in order to post. Do not start another thread! 1. Pour UVendredi 25 Mars U: Individually, choose a commercial: Go to the following link where you may have to register (it is free): HThttp://www.pubstv.comTH Analyse the commercial according to the advertising guidelines attached. 2. Pour ULundi 28 Mars U: Discuss which commercial your team will choose. You will have to tell your teammates why you decided on the commercial you chose. As a group you will choose the commercial that appeals to the team the most 3. Pour UMercredi 30 Mars U: Individually, explore these 2 French TV web site to find their programming in order to identify 2 shows that would best fit your commercial. Your goal is to reach the largest target audience possible. HThttp://www.tf.frTH HThttp://www.france2.frTH Write a paragraph for each show explaining your rationale. See the advertising guidelines below. 4. Pour UVendredi 1 Avril U: As a team, decide on which 2 shows will best reach your audience for your ad campaign. 5. Pour ULundi 4 Avril U:

PAGE 50

43 As a team, work on the final report, the guidelines of which are provided below. UTimeline for the F2F groups: 6. Pour UMercredi 30 Mars U: Individually, choose a commercial: Go to the following link where you may have to register (it is free): HThttp://www.pubstv.comTH Analyse the commercial according to the advertising guidelines attached. 7. UMercredi 30 Mars U EN CLASSE au Turlington Language Lab: Discuss which commercial your team will choose. You will have to tell your teammates why you decided on the commercial you chose. As a group you will choose the commercial that appeals to the team the most 8. Pour UVendredi 1er Avril U: Individually, explore these 2 French TV web site to find their programming in order to identify 2 shows that would best fit your commercial. Your goal is to reach the largest targetted audience possible. HThttp://www.tf.frTH HThttp://www.france2.frTH Write a paragraphe for each show explaining your rationale. See the advertising guidelines below. 9. UVendredi 1 Avril U EN CLASSE au Turlington Language Lab: As a team, decide on which 2 shows will best reach your audience for your ad campaign. 10. Pour ULundi 4 Avril U: As a team, work on the final report the guidelines of which are provided below. NO LATE WORK WILL BE ACCEPTED. Failure to provide work will result in personal grade penalties. UAdvertising Guidelines: In order to analyse a commercial properly you have to first identify the product, the brand, and the message, then you will provide a short summary describing the commercial and finally, you will provide a profile of your target audience. For the target audience profile, use the following major demographic variables and suggested breakdowns:

PAGE 51

44 Age Sex Revenu (x1000) Education Occupation Statut Marital Nombre denfants Georgraphie -18 18-24 25-34 35-44 45-54 55-60 60+ Masulin Feminin -15 15-30 31-50 51-70 71-90 91-100 100+ -Brevet des colleges -Lycee ou BEP -Bac -Bac +2 ou plus -Professionel et Technique -Cadre -Agriculteur -Profession liberale -Etudiant -Sans emploi -autre Marrie(e) Celibataire 1-2 3-4 5+ Rural Urbain In addition, describe the possible interests and hobbies of the target group. Most of the time, interests, and hobbies can be derived from the commercial itself. Is it shot in an artistic manner? Is it showing sports of any kind? If yes which type? Is it showing wealth? Is it humorous? If yes, what kind of humor? Etc If you watch the commercial closely you will learn many things about the people targeted, be attentive and creative! Finally, according to your findings, anticipate and tell us what type of shows the target audience is most likely to watch and when. UReport Guidelines: The report will include 4 parts. In the first part you will describe the commercial and why your team decided to choose it. In the second part you will provide your rationale for the 2 shows you chose for your advertising campaign. In the third part you will discuss the cultural differences or similarities you found while choosing a commercial and while examining French TV programming. For instance, would your commercial be efficient in the US? Why or why not? What are the differences or similarities between French and US shows. Finally, what did you think of this project? Write one paragraph.

PAGE 52

APPENDIX B FORUM SCREENSHOTS 1 Figure 1: Screenshot of the forum homepage. Students need to enter the threaded topic related to their class project in order to have access to the threaded discussion (1). As the administrator, I am can maintain students privacy by denying the access to the forum to any one who is not registered on the forum and in the class. 45

PAGE 53

46 2 1 Figure 2: Screenshot of the Project threaded discussions. After students have clicked on the topic all they have to do is click on the threaded discussion of their assigned group. A little flag on the left side (1) indicates if there is a new message they have not read yet and a comment on the right side (2) tells them who was the last person to post a comment.

PAGE 54

47 3 2 1 Figure 3: Screenshot of a thread including (1) emoticons (2) avatars and (3) personalized signatures. Notice the different color fonts.

PAGE 55

LIST OF REFERENCES Beauvois, Margaret (1997), Write to Speak: the Effects of Electronic Communication on the Oral Achievement of Fourth Semester French Students, New Ways of Learning and Teaching: Focus on Technology and Foreign Language Education, Heinle and Heinle Publishers, pp 93-115. Beauvois, Margaret H., and J. Eledge (1996), Personality Types and Megabytes: Student Attitudes Towards Computer Mediated Communication (CMC) in the Language Classroom," Calico Journal, Volume 13 Numbers 2 and 3, pp 27-45. Bhlke, Olaf (2003), A Comparison of Student Participation Levels by Group Size and Language Stages During Chat Room and Face to Face Discussions in German, Calico Journal, Volume 21 Number 1, pp 67-87. Chenoweth, Ann, and N. K. Murday (2003), Measuring Student Learning in an Online Course, Calico Journal, Volume 20 Number 2, pp 285-314. Davis, Boyd, and Ralf Thiede (2000), Writing into Change: Style Shifting in Asynchronous Electronic Discourse, Concepts and Practice, Cambridge Applied Linguistics, pp 87-120. Kern Richard, and M. Warschauer (2000), Introduction Theory and Practice of Network-Based Language Teaching, Concepts and Practice, Cambridge Applied Linguistics, pp 1-19. Lantolf, James (Ed) (2000), Sociocultural Theory and Second Language Learning. Oxford, UK: Oxford University Press. Pellettieri, Jill (2000), "Negotiation in Cyberspace: The Role of Chatting in the Development of Grammatical Competence," NBLT: Concepts and Practice, Cambridge Applied Linguistics, pp 59-86. Prez, Luisa C. (2003), Foreign Language Productivity in Synchronous vs. Asynchronous Computer-Mediated Communication, Calico Journal, Volume 21 Number 1, pp 89-104 Skehan, Peter (2001), A Cognitive Approach to Language Learning, Oxford University Press. 48

PAGE 56

49 Warschauer, Mark (1996), Comparing Face-to-Face and Electronic Discussion in the Second Language Classroom," Calico Journal, Volume 13 Numbers 2 and 3, pp 7-27. Warschauer, Mark (2000), "Online Learning in Second Language Classrooms," NBLT: Concepts and Practice, Cambridge Applied Linguistics, pp 41-58.

PAGE 57

BIOGRAPHICAL SKETCH Cyrille Guillo obtained a Master of Arts in English as a Second Language from the Universit de Haute Bretagne, Rennes, France. He also obtained a Master of Business Administration from the University of Missouri, Columbia. 50