The widespread use of computer conferencing for instructional purposes, both as an adjunct to and a replacement for the traditional classroom, has encouraged teachers and students alike to approach teaching and learning in ways that incorporate collaborative learning and the social construction of knowledge. Discussion and dialog between instructor and students and among students is a key feature of computer conferencing and the foundation of constructivist learning techniques. Computer conferencing can be used both asynchronously, which allows time for reflection between interactions, and synchronously, allowing real-time, interactive chats or open sessions among as many participants as are online simultaneously.

 This study used content analysis to first identify the communication conventions and protocols that real-time, interactive electronic chat users developed in instructional settings. The study also determined that the students recognized a need to use their communication conventions and protocols to communicate clearly and minimize misunderstandings in their online transactions with others. The more obvious conventions included using keywords and names of individuals, shorthand techniques, non-verbal cues in text, and asking questions and seeking clarification.
 
 
 
 

Contents

Introduction
Theoretical Framework
Course Structure and Descriptions
Method
Results
Conclusions
Further Research
References

Introduction

The widespread use of computer conferencing for instructional purposes, both as an adjunct to and a replacement for the traditional classroom, has encouraged teachers and students alike to approach teaching and learning in ways that incorporate collaborative learning and the social construction of knowledge. Discussion and dialog between instructor and students and among students is a key feature of computer conferencing and the foundation of constructivist learning techniques.

Dialog has been recognized by Moore (Moore & Kearsley, 1996) as a determining factor in the amount of transactional distance that exists in most, if not all, instructional events: those taking place in a traditional classroom and those taking place at a distance where instructor or student may never see one another. Transactional distance describes not only a dimension of a physical separation but also a communication gap that must be bridged by dialog in some structured fashion so that shared meaning can be constructed, teaching and learning can occur, and the potential for misunderstanding and miscommunication is significantly diminished. When teaching at a distance some form of communication technology must be used to carry this critically important dialog between teacher and student and among students. In electronic venues, this usually takes one of several forms of computer conferencing.

Computer conferencing can be used both asynchronously, which allows time for reflection between postings, and synchronously, allowing real-time, interactive chats or open sessions among as many participants as are online simultaneously. There are a number of different forms of synchronous online communication, or electronic chats. An electronic chat is a series of real-time, short (usually 1 to 3 lines) text phrases and sentences exchanged with the other chat users who are logged onto the same computer system and facility. The interactions appear as individual lines of text, prefixed by the nickname of the contributor, that scroll up the screen as they are entered.

Internet Relay Chat (IRC) is an example of a synchronous communication program that is international in scope and is available to anyone with access to a client program and who can log on to any one of the IRC servers located across the Internet. IRC can involve large, international groups of people, many of whom may be strangers. Real-time electronic chat is commonly used for recreation and social interaction. On IRC, both topic and tone of discussion is policed only by the participants, with the result that a shared culture has gradually developed that defines communication norms and conventions somewhat differently than they are in face-to-face conversations.

Synchronous on-line communication can also occur in "members only" situations like the "chat rooms" on commercial services like CompuServe or America Online, where the topics and language of appropriate discourse are defined by the online service provider. Synchronous on-line communication is also a feature of closed groupware systems, like LotusNotes, FirstClass and other systems used for computer supported collaborative work in business and education, where the norms of business and academic discourse usually prevail. Users of UNIX and VAX computer systems can communicate privately, one-to-one, using the system facilities called "talk" and "phone" where users type at each other in real time, each in their own half of a horizontally split screen.

Since the late 1980s, computer-supported collaborative work using various groupware programs has supported chats for information exchange, brainstorming, discussion and problem-solving in business settings (Lloyd, 1994). There is increasing interest in the use of chats for instructional purposes through virtual classes such as those at Athena University and Diversity University, using a highly structured MUD or MOO format. Instructional electronic chats (IEC) can also supplement otherwise traditional or distance coursework.

IECs are intriguing to educators as they appear to allow a sense of communicative immediacy and presence that is often lacking in asynchronous computer-mediated communication. Synchronous dialog, if it could be appropriately structured, could go a long way to reducing transactional distance. Questions and concerns could be quickly raised and addressed and misunderstandings sorted out. IECs pose challenges for educators because the verbal and non-verbal communication protocols used in face-to-face or in video-based distance education settings may not be sufficient for quality educational exchange with and among participants. The design of most forms of IRC software is such that multiple, disjointed conversational threads can quickly develop as various members of the group form smaller conversational groups, each focused on their own topic and ignoring, or only intermittently, joining in others. This may result in conversational chaos. Even affective protocols such as emoticons (smilies) used to enhance email communication are not necessarily transferable to an IEC. Day and Batson (1995) note "although at first it may seem difficult to follow the separate strands or topics, classes typically get used to the nonlinear 'flow' of the conversation rather quickly" (p. 29). They do not say how this "getting used to" occurs, nor how students develop and establish the communication conventions necessary for meaning-making in the turbulent flow of on-screen conversation. This is the problem that has intrigued us.

The current study asked the following questions:

1. What communication conventions and protocols do IEC users develop?

2. Do IEC users recognize a need to use these communication conventions and protocols to communicate clearly and minimize misunderstandings in their online transactions with their instructor and their peers?
 

Theoretical Framework

Communication and Discussion in Instructional Settings

 Adult educators frequently use discussion in their teaching to reach one of their primary goals, which is "encouraging adults to undertake intellectually challenging and personal precarious ventures in a non-threatening setting" (Brookfield, 1986, p. 135). Interacting openly with their peers in discussion groups, either face-to-face or in some technologically-mediated fashion, adult learners can practice using the new conceptual tools they are acquiring. Supportive feedback can be immediately forthcoming as they tentatively articulate their changing ideas and opinions, as can rousing arguments as opposing perspectives clash. Learners can be exposed to widely divergent points of view, lifestyles and belief systems and receive encouragement as they enlarge and revise their own world-views.

When using computer-mediated communication, discussion occurs among teachers and learners using several different kinds of text-based electronic communications modalities. Asynchronous electronic communication in its most familiar form takes place with electronic mail, newsgroups, or computer conferences. The message is composed and then sent, to be read by the recipient(s) at a later time. The parties to the transaction do not need to be on-line synchronously (i.e., at the same time). Synchronous communication does require that all the parties to the discussion be on-line and participating at the same time. The familiar name for this activity is "chat," although the discussion appears as lines of text that scroll up the screen, each line preceded by the nickname of the discussant.

The FirstClass^TM computer conferencing program used for the courses described in this paper has facilities for both asynchronous (electronic mail and postings to class conferences or discussion groups) and synchronous (chat) communication. Students can access the FirstClass server via the Internet at times convenient to themselves and work independently on their projects and assignments, read and respond to their electronic mail, and read and post comments to the various open and closed class conferences, as McIsaac and Ralston (1996) describe. For a scheduled chat session, the participants must all be logged into the FirstClass server at the same time, reading and contributing to the ongoing discussion.

Synchronous Computer-mediated Communication

 The use of synchronous electronic communications programs in instruction is relatively new. While it has been used in college writing classes (Day & Batson, 1995) and to teach literature (Harris, 1995), the more common use of synchronous chat programs among undergraduate students is to communicate with one another (Archee, 1993; Newby, 1993) for social and recreational purposes (Aoki, 1995) using IRC and similar public synchronous communications programs.

Public IRC is a text-based, international, message-handling program resident on many Internet servers. Multiple communication channels (similar to citizen's band radio channels) can be created and named (sometimes fancifully) by their creators, thus introducing the meta-message: "Let's make-believe and suspend disbelief" (Ruedenberg, Danet, & Rosenbaum-Tamari, 1995). Generically these channels are variously designated as chatlines or chat rooms and encompass discussion on every conceivable topic. Access via a client program allows users to join and listen in on (read) conversations on multiple channels on multiple servers. With experience, four of five different channels can be attended to at one time (GiZelle, 1996, personal correspondence). Users join the channel(s) of their choice and type in their conversational contributions, one line at a time, and the conversation is distributed, via the servers, to all those who are logged on and listening to (reading) that particular channel.

Synchronous communication program users identify others, often strangers, with similar interests and engage in conversations with them. Users of public synchronous chat programs are customarily identified by a descriptive nickname that is sometimes chosen to "promote a certain image or invite a particular response" (Newby, 1993, p. 35). A nickname can serve as a mask not only to hide identity, but to call attention to the person through the expressive power and imaginativeness of the mask (Ruedenberg et al., 1995). Nicknames and other personal information can be changed at will, so that anonymity can be maintained within IRC programs until users choose to reveal their true identities to each other (Reid, 1991), which may never actually happen (Phillips, 1995).
 
 

Advantages and Limitations of Synchronous On-line Communication

Advantages

Aoki (1995) notes that brainstorming and other activities requiring spontaneity can be handled effectively in a synchronous chat, as can decision-making, that requires a quick turn-around time rather than extended discussion (Siemieniuch & Sinclair, 1994).

Synchronous communication can simulate an instructional environment that is familiar to students, faculty, administration, and funding sources (Fanderclai, 1995). Formal behavior patterns may carry over from the face-to-face classroom, and students may find it easier to orient themselves when surrounded by familiar, albeit virtual, structures like classrooms, libraries, cafes and faculty offices.

Synchronous communication adds the excitement of interacting with others in real time and builds a sense of social presence (Aoki, 1995) - there really are people on the other side of the computer screen - and a heightened sense of involvement in the ongoing communication events (Ruedenberg et al., 1995). This can lead to what Csikszentmihalyi (1977) refers to as a "flow experience" in which action and awareness are fused, the passing of time is unremarked, and the activity itself becomes intrinsically rewarding and deeply engaging.. This sense of involvement and engagement can be critical in building a sense of community among the participants (Reid, 1991).
 
 

Limitations

IRC software provides each participant with a small window (approximately two lines deep across the screen) in which to type their contributions, which is input in its entirety into the conversation when the 'return' or 'enter' key is pressed. Only then can the new input be seen by all participants. Turn-taking in synchronous communication is problematical as there are no observable kinesthetic or para-verbal cues to indicate when someone wants to enter the conversation or to change the subject. Siemieniuch and Sinclair (1994) suggest that verbal and non-verbal protocols currently used in face-to-face meetings may be inadequate for synchronous computer-mediated communication, and perhaps some controls must be built into the software to rationalize the passing of control from one participant to another to permit efficient interaction.

Synchronous communication requires substantial typing skills to communicate effectively, and the conversation may move too fast for non-native speakers of English, who have no time to reflect, frame questions and compose responses as the text incessantly scrolls up the screen (Aoki, 1995).

Frequent users may have skills that novices have not had time to develop: ". . . users are adept at following multiple chains of discussion at once, reading other's responses while typing their own . . the every-changing flow of messages results in considerable cacophony in the number of simultaneous thought-trains expressed on a channel. The closest analogy in the non-networked world would be a pub or a party" (Newby, 1993, p. 35).

The novelty of being in a virtual classroom will not last through an online lecture that appears as line after line of "teacher-talk", input at the teacher's typing speed or ability to cut and paste from a prepared text. Synchronous communications environments are designed for interaction, and students will soon want to DO something (Fanderclai, 1995).

Hiding behind a nickname, habitual "chat" users can develop communication habits that might be disruptive to an instructional setting:

"Protected by the anonymity of the computer medium, and with few social context cues to indicate 'proper' ways to behave, users are able to express and experiment with aspects of their personality that social inhibition would generally encourage them to suppress" (Reid, 1991).

Users of synchronous communication may choose to change the way they express their personalities; the anonymity and consequent sense of being alone can allow them to switch genders (Reid, 1991) or change their age; a quiet person may become expressive, abusive, or explosive. Participants can enjoy a sense of reduced accountability; actions and utterances are often in a playful mode or can erupt into sudden bursts of conflict (Ruedenberg, Danet & Rosenbaum-Tamari, 1995) and the vitriolic, verbal abuse known as "flaming" (Aycock, 1995).
 
 

Communications Conventions in Synchronous Computer-mediated Communication

 Curtis (1992) has commented that the technical aspect of text-based synchronous communication has significant effects on social phenomena, leading to new modes of interaction and new cultural phenomena. The lack of actual physical presence, indeed the often great physical distances between individual participants, demands that a new set of behavior codes be invented if the participants are to make sense of and understand each other (Reid, 1994). In the case of synchronous chat, these are behaviors expressed in text that are designed to present a recognizable self, set a context for the interactions, share affect and meaning, and minimize misunderstanding.

Because synchronous communication users frequently spend many recreational hours chatting with (typing to) each other, communications conventions arise in several forms, including shorthand for common phrases (e.g., btw = by the way; brb = be right back; rotfl = rolling on the floor, laughing).

Ruedenberg et al. (1995) call communication on IRC a veritable forest of symbols as participants use typographic symbols comprised of such humble materials as commas, colons and backslashes to convey affect, as in the use of the ubiquitous emoticons known as the "smiley" :-) and the "frownie":-( among many forms. Aycock (1995) noted that these "orthographic pictures" appear to indicate an "impersonal, but friendly" interest.

Conventional text emphasis cues are used to signal conversational tone and nonverbal communication information through underlining, capitalization for emphasis (read as SHOUTING), misspellings for slang purposes, and exclamation points !!! (Kuehn, 1993).

Aycock (1995), in his discussion of Foucault's notion of fashioning oneself (souci de soi) notes that on-line discussion participants use at least three textual devices in their presentation of self (Goffman, 1959):

• Posting of facts or explanations of techniques, tips, or tricks which displayed a shared interest in a topic. This was understood as an indication of commitment to the ongoing conversational activities.
• Emoting in text - using a word or phrase enclosed in angle-brackets to express emotion, e.g., <sigh>, <grin>, <madememad!!!>, verbalized pauses for thought <hhhhhmmmmmmm>, <er...> or movement and gesture <tiptoeing in quietly>, < ??? - scratching my head!>
• "Flaming" - vulgarity or verbal threats or insults to express a kind of ritualized confrontation that expresses formulaic anger as much as televised sports may express formulaic masculinity.

In addition, key words and personal names may be used to preface comments. In the ever-scrolling lines of text it is difficult to follow conversational threads, so key words are used as indicators of content, or the individual to whom the comments are directed.

Playfulness is indicated by multiple repetitions of phrases or invitations to join a particular channel (Ruedenberg et al., 1995) and textual manipulations signifying jokes, sarcasm, irony, and witticism (Kuehn, 1993).

Disclosure of personal information in an otherwise anonymous textual environment allows others to form an accurate perception of personal presence. The more one discloses personal information, the more others are likely to reciprocate, and the more individuals know about each other, the more likely they are to establish trust, seek support, and thus find satisfaction and to create a sense of being in a safe learning/discussion environment. Without disclosure and interaction nothing happens. Disclosure creates a kind of currency that is often spent to keep interaction moving (Cutler, 1995).

Social presence in cyberspace takes on more of a complexion of reciprocal awareness by others of an individual and the individual's awareness of others. . . to create a mutual sense of interaction that is essential to the feeling that others are there (Cutler, 1995, p. 18). This can be indicated by referring to others by name, and to past, present or future activities that have occurred in the shared communication space or in real life.

Status in face-to-face interactions is often signified, as it is in face-to-face conversation, by who can interrupt whom and who can speak to whom, who takes conversational leadership and who listens, who changes topics or otherwise directs the group's activities.

Transactional Distance in Distance Education

 Moore defines a distance education transaction as "an interplay between people who are teachers and learners, in environments that have the special characteristic of being separate from one another, and a consequent set of special teaching and learning behaviors" (Moore & Kearsley, 1996, p. 200). He goes on to say that distance teaching behaviors can be described in terms of two clusters of variables: dialog and structure. "Dialog" is defined as a positive interplay of words, actions and ideas (or a series of such interactions) between teacher and learner(s) that is purposeful, constructive, and valued by each party to the transaction and directed to increased understanding by the student (p. 201). "Structure" is defined as the extent to which an educational program can accommodate or be responsive to each learner's individual needs. Structure is expressed in the rigidity or flexibility of the course's educational objectives, teaching strategies, and evaluation methods (p. 203). Dialog and structure set the parameters of "transactional distance."

Transactional distance is a continuous, relative variable and describes the distance in technologically-mediated educational transactions. This distance is not only a physical measure but "a distance of understandings and perceptions caused by the geographic distance, that have to be overcome by teachers, learners, and educational organizations if effective, deliberate, planned learning is to occur" (Moore & Kearsley, 1996, p. 200). Transactional distance is a "psychological and communications space to be crossed, a space of potential misunderstanding between the inputs of instructor and those of the learner" (Moore, 1993, p. 22) and among learners as peers. This transactional distance can be as prevalent and disruptive of learning in a large face-to-face classroom (especially when conducted in a large lecture hall) as it can when instructional delivery occurs on-line. Transactional distance is likely to be at its highest in a recorded television program where there is no dialog between teacher and student and "where virtually every activity of the instructor and every minute of time provided for and every piece of content [is] pre-determined" (Moore & Kearsley, p. 203). In a course delivered using interactive technologies where student and teacher are in frequent communication, transactional distance and the misunderstandings and miscommunications it entails is likely to be low. Interactive dialog can permit continuous, structural modifications of course content, pace and activities to suit the learners' individual needs, and students' questions and concerns can be responded to immediately thus significantly reducing transactional distance. By manipulating the media it is possible to increase this dialog among learners and their teachers, and thus reduce transactional distance (Moore, 1993) by reducing the potential for misunderstanding.
 

Course Structure and Descriptions

The courses from which the IEC transcripts for this paper were taken are oriented by constructivist learning principles, so communication may be largely among students as peers who learn and teach one another on-line as they strive to increase their understanding of one another. To facilitate this understanding, the students developed communications conventions to structure their synchronous conversation, reduce the cognitive load, and minimize misunderstandings.

This study involves graduate students studying educational technology and distance education in five courses in the College of Education at Texas A&M University over three semesters during 1996 and 1997. One course was a survey course in educational technology (ET), and the other four were courses in distance education and telecommunications (DL, AT, MT, and TP). The students were masters and doctoral students primarily from education, and their numbers in each course ranged from 9 to 13. All five courses were taught by the same instructor.

Each course was delivered by a combination of two-way interactive (compressed) video conferencing (VTel) and computer conferencing via FirstClass software. The students used this computer conferencing system as both an adjunct and an alternative to class sessions held by video conference. Since 1993, he instructor has used video conferencing and various forms of computer-mediated communication (i.e., email, email distribution lists, VAX-Notes, and computer conferencing via Forum software) to reach her graduate students. These courses are described elsewhere (Murphy, Cathcart, & Kodali, 1997; Murphy et al., 1996; Murphy, Drabier, & Epps, 1997; Yakimovicz & Murphy, 1995).

The instructor was relatively unfamiliar with FirstClass, as the Summer 1996 semester was only the second time that she had used this particular software program. The software had been purchased during the previous semester, and the pioneer users were faculty and students involved in educational technology courses. Technical support for FirstClass was almost non-existent at the beginning.

Students who were enrolled in the five courses included in this study were required to use FirstClass for two purposes: (a) asynchronous communication through computer conferences, email, and document sharing using attached files, and (b) synchronous chats. Asynchronous and synchronous computer conferencing was used for a number of "virtual" class meetings in each of the courses (see Table 1 for a brief description of each course).

Table 1: Descriptions of five courses over three semesters
 

Course Title	Code	Semester / Duration	No. of "virtual" meetings	No. of students
EDTC 689: Theory and Practice of Distance Learning	TP	Summer 1996/  3 weeks	1 out of 9	12
EDTC 602: Educational Technology: Field, Theory, Profession	ET	Fall 1996/  15 weeks	1 out of 15	13
EDTC 664: Management of Instructional Telecommunications Systems	MT	Fall 1996/  15 weeks	3 out of 15	9
EDTC 608: Foundations of Distance Learning	DL	Spring 1997/  15 weeks	7 out of 15	10
EDTC 618: Applications of Telecommunications in Education	AT	Spring 1997/  15 weeks	12 out of 15	10

The nature of IECs differed from course to course and from one chat session to another. During the initial use period, the instructor and students alike were unfamiliar with the process of real-time chats and invented communication protocols during the IECs themselves. At that time, the instructor announced the topic at the beginning of the chat period instead of providing advance notice about the tasks to be accomplished during the chat. Later, as she developed familiarity with the software, the instructor assigned specific topics and tasks to small groups in advance and provided a specific time for the small groups to have private chats before calling the large group together for a chat. Subsequently, students who were required to moderate parts of class sessions delivered by computer-mediated communication emulated the instructor by assigning specific topics and tasks to small groups prior to the large group chat.

During her very first online chat session, the instructor encouraged the students to contribute to the topic without waiting for each other (i.e. taking turns). Without any protocols other than the "netiquette" explained by Mason (1991), the students made off-task remarks, used acronyms and other text-based shortcuts, occasionally reprimanded each other, and turned what was intended as an instructional event into conversational chaos. Over the course of two semesters, the instructor experimented with various degrees and forms of structure in the chat sessions, each time requesting reflection afterwards from the students. The students' reflections, posted in the asynchronous conference. included the need to structure chat time "to make sure we don't 'trip' over each other and that things move swiftly along" and "when there is not enough structure there is little accomplished; when there is too much structure the objective could just as well be accomplished asynchronously."

The third semester AT course was the first course that the instructor ever taught online, in which 12 out of the 15 class meetings were held by a combination of FirstClass asynchronous conferences and synchronous chats, a Listserv discussion with another university, a Web page (http://disted.tamu.edu/~kmurphy/618index.htm) maintained by one of the students, and a Web board (http://disted.tamu.edu/~kmurphy/wwwboard/wwwboard.html) enabling both asynchronous threaded discussions and a real-time chat function. The IECs typically lasted between 60 and 90 minutes, were moderated by the instructor, and were often open-ended, which allowed the students to raise issues that they considered important. During the second IEC (the chat chosen for analysis in this paper), the instructor and students established the agenda and then brainstormed decisions about their responsibilities for moderating the next Listserv discussion topic, how to use the Web board, and plans for a lab session on constructing individual Web pages.
 

Method

Researchers

The research team consisted of the instructor and a graduate student from another university. The instructor had been involved in distance education for over a decade and was experienced in using a variety of forms of computer-mediated communication in both teaching and conducting research. The graduate student was very experienced with moderating Listservs and using chat functions to conduct research, and she participated via FirstClass in two of the initial large group chats and used the chat function for continuing communication with the instructor. The two researchers collaborated throughout the process of data collection, analysis, writing, and rewriting. We downloaded and printed out the relevant electronic file of messages from the asynchronous conferences and the logs of all chats that were available in each of the five courses.

Subjects

The subjects in the IEC under analysis were ten students in AT, one of the distance education courses. This elective course consisted of students who were experienced and interested in telecommunications, some of them being telecommunications professionals working in the field. Adding to the complexity of communication were three students whose native language was not English. Five students had taken a previous class involving FirstClass with the instructor and therefore were more experienced with the software than the others.

An examination of the pre-course surveys revealed that nine of the students had taken at least one course via distance learning prior to the beginning of this semester, and one of the students had taken more than five courses via distance learning. The majority of the students indicated that they preferred to take this course via distance technologies and felt that it would hold their attention well. The students anticipated that they would take more responsibility for their own learning than in more traditional courses and that they would still achieve as much. They expected that active communication and interaction with the instructor and their classmates would be as good as it would be in a traditional class and expected that the course would help them communicate easily with students in other locations.

Data Collection and Analysis

This research is a case study of one chat in the course "Applications of Telecommunications in Education". The data sources included: (a) the transcript of an electronic chat, which lasted approximately 90 minutes, (b) transcripts of an asynchronous discussion in a computer conference about IECs, and (c) pre-course surveys.

We based our research design on those of researchers in computer-mediated communication: Henri (1992), Kuehn (1993), and Levin, Kim, and Riel (1990). We used content analysis to examine the data (Krippendorff, 1980). The codes used in our content analysis were derived from the work of researchers in recreational uses of synchronous chat (IRC, MUDs, MOOs) (e.g., Curtis, 1992; Reid, 1991; Reid, 1994; Ruedenberg et al., 1995). Individually, working from the list of communication conventions that we identified from the literature, we read through the transcript of the IEC, seeking to identify the conventions that the students had used. Simultaneously, we sought to identify any other conventions or protocols that these students may have developed. The next step was to compare our coding and reconcile our discrepancies. For example, while reading the transcripts, we highlighted and assigned code words to individual inputs to identify the types of communication that we thought were taking place. Table 2 lists the codes that we used to identify the types of communication in each unit of analysis, which consisted of a single posting or partial posting in a chat.

Table 2: Codes for Types of Communication in IECs
 

Types of Communication	Code
Continuation of an idea from one input line to another, using ellipses ( . . .)	C
Emoticons (smilies)	E
Flaming	F
Shorthand for common phrases	H
Key words and names of people, like a subject line	K
Facts or techniques for sharing meaning or indicating interest in a topic	M
Playfulness, humor, and similar "asides"	P
Questioning and clarification seeking	Q
Social presence, such as referring to others by name, and to shared activities	S
High status assumed by directing group's activities	T
Conventional text emphasis cues (underlining, punctuation marks, and capital letters to indicate SHOUTING) to signal non-verbal communication cues	V

In addition, we established the following codes to identify quoted messages, which we include in the text precisely as the authors wrote them in the chats, including typing errors. We identify the author of the posting (I = Instructor, and S = Student) and the communication code, which is derived from the list in Table 2.
 

Results

Communication Conventions and Protocols

Our first research question asked, "What communication conventions and protocols do IEC users develop?" Table 3 summarizes the number and percentages of postings to the IEC discussed in this paper. In the table heading, ESL indicates that the student is a non-native speaker of English; FC indicates that the student had previously taken a course with this instructor and had used FirstClass; and CMC indicates a student who had taken a previous course with this instructor - a course that included the use of computer-mediated communication, but not the FirstClass conferencing system.
 
 

Table 3: Number and percent of inputs (postings) by participant in the selected IEC
 

Participants	Participant Code	Gender	ESL	FC	CMC	# of inputs n=380	% of inputs
Instructor	I	F				102	28.8
Student 1	S1	M		*		34	8.9
Student 2	S2	F				17	4.3
Students 3 & 4 1	S3/4	F/F		*		53	13.9
Student 5	S5	F		*		30	7.8
Student 6	S6	M			*	52	13.7
Student 7	S7	M	*	*		14	3.7
Student 8	S8	M	*			17	4.5
Student 9	S9	F				23	6.0
Student 10	S10	F	*	*		38	10.0

¹ Students 3 and 4 had to share a keyboard for this chat

The instructor was responsible for 28.8% of the total inputs (n=380), and the students posted 71.2% of the inputs. Of the student inputs: 42% were made by males, 58% by females (the gender split in the course was 40% male, 60% female, so the contributions were proportional); 25% were made by ESL speakers, 75% by native English speakers (there were 3 ESL students, so their contribution was slightly less than proportional); 61% were made by those who had taken a course using FirstClass previously (5 students had previously taken a class involving FirstClass and one student was taking two courses concurrently, so the representation was proportionate to course membership).

The students implemented conventions and protocols that were useful to them as they struggled to make meaning out of this new form of communication. Examples of these communication conventions and protocols are provided with the one- to two-line entries extracted from the IEC, identified by the individual who posted the input. The content analysis of the IEC indicates that the participants used the following conventions, in order of approximate frequency in the transcript. They are included below, beginning with the most commonly used conventions and ending with those least commonly used.

1. M = (shared) meaning

The students shared facts, helpful hints or techniques to indicate a shared meaning or an interest in the discussion topic:

• When you weave you vallidate the contributions others have made which motivates further contributions (S1)

• UNM brought up a number of tech problems...which for some of us might need defining, searches for solutions, consulting each other, etc. (S5)

2. K = keywords

Many students adopted the convention of typing a keyword or personal name descriptor at the beginning of a line to indicate the subject of the entry:

• UNM: whatever topic we decide, we need to weave it and keep responses (S4)
• Student 10, re terms ...Platforms, different conferencing software(FC and e-mail such as Pine and Elm)...there are many others. (S5)

3. H = shorthand

The students used shorthand as a substitute for common phrases:

• Student 4 UNM is behind TAMU, their capabilities are limited (S8)
• Q&A OK Who weaves? All cannot weave at once. (S1)
• TeleC - Dr. M can we talk to UNM over FC? (S6)

4. S = social presence

The students created a sense of social presence, by referring to each other by name, and by sharing activities, both on- and off-line:

• Student 2 I was there a while ago. Did your life pass before your eyes and a bright light show up a long way away? (S6)
• I feel like we need more interaction with them to make it a truly worthwhile learning experience, not just superficial communication (S4)

5. P = playfulness and humor

The students exhibited their playfulness and humor in attempts to have fun with each other in socially acceptable ways:

• TeleC - Weaving - Quiet Student 1 might do a bang up job. (S6)
• Shhhhh (S1)

6. V = non-verbal cues in text

The participants adopted the use of conventional text emphasis cues (e.g., underlining, punctuation marks, and capital letters that often indicate SHOUTING) to express non-verbal communication:

• Man I got lost! (S2)
• "Cognitive lrning strategies" in my mind refer to learning to learn strategies/methods of learning... therefore, such terms as scaffolding, linking, working with zones of proximal developm (S5)

7. Q = questions

Students felt comfortable enough to pose questions and seek clarification of meaning, often from each other:

• Student 6 do you mind sending me something related to what you will do? It will be a nice way for me to learn (S10)

• What are you talking about (S10)

8. T = status and directions to others

Evidence of assumption of high status is seen in these postings in which students sought to direct the group activities:

• Hey, let's not drift apart in our conversations . . . (S7)
• TeleC - who wil start - Should we not begin with an invitation for questions? We could put forward and initial question, but open the conversation for others at the same time. (S6)

9. C = continuation through ellipses

The participants used ellipses (. . .) to indicate continuation of a thought from one input line to another:

• Telecomp: Nobody knows how to use it yet, most likely, except for a few of you. I posted instructions at the top of the board. People would need to be advised to read them first... (I)
• ... but you'd need to set the example of posting and replying, to demonstrate the threading capabilities. (I)

10. E = emoting and smilies

The participants occasionally expressed emotions by using emoticons such as smilies :

• Student 1, Student 6, do you mean there are NO quiet students in this class ;-) (I)

There were, however, no inputs that could be interpreted as flaming in the transcript of this IEC. A comparison of demographic data from the pre-course surveys with the IEC transcript reveals a positive link between the previous level of use of email and both (a) the frequency of interaction in the IEC, and (b) the number of playful and humorous postings in the IEC. In all, the students developed a repertoire of communication conventions and protocols that they used in this and subsequent IEC.

Need for Using Communication Conventions and Protocols

 The second research question asked, "Do IEC users recognize a need to use these communication conventions and protocols to communicate clearly and minimize misunderstandings in their online transactions with their instructor and their peers?" The following remarks, which are extracted from asynchronous conferences in the AT course, reflect the students' recognition of a need to use a variety of these conventions and protocols to reduce transactional distance.

The students discussed clarity of communication through using keyword descriptors. One student suggested this chat protocol: "Start every line with a referent: if there are two topics underway, use a title to refer to which topic you are chatting about" (S4, 2-10-97). A second student replied to this posting with the following: "The FC protocol was helpful. I tend to get excited about replying and forget to type in the person I am responding to. Thanks I will try to remember this on the next FC Chat" (S3, 2-11-97). Yet another remarked, "the first chat was not well organized but the second one was improved because of the method of mentioning the name of the students who surpose to anwser or respond to a particular statement" (S8, 2-22-97).

Students used metaphors to create shared meanings with their classmates, even explaining new words to reduce possible misunderstandings. "Looking back at the chat, my comments seem like you may have created a monster. The proverbial vaccination with the victrola needle. Student 10, are you familiar with the term Victrola? It is a brand of record player. Being vaccinated with a Victrola needle means a person talks a lot" (S6, 2-20-97). Another student's metaphor related to literature: "To me the synchronous chats are much like Faulkner's use of stream of consciousness in some of his works. It is analogous to the way one thinks…a flow of information and thoughts. Sometimes mine become waterfalls, but most of the time they move at a slow trickle!…like finding water during a Texas drought" (S5, 2-24-97).

Students, particularly those from other countries, recognized the challenges of the constant and rapid flow of text up their screen during chats. For example, one international student remarked, "Well, chat is never my friend. I cannot keep on with the speed of information coming in. Instead of pitching in at wrong time, I prefer to read and contribute when necessary" (S8, 2-22-97). Another student, without the additional burden of English as a second language, still faced problems with his slow typing speed. He commented, "I had developed some strategies to get around my slow typing speed. I also had an idea of how the process worked so I could tolerate more 'noise' on the chat" (S6, 2-20-97).

Another technique that students recognized as important in reducing transactional distance was self-disclosure, perhaps from their desire to present themselves on an equal footing to those without excellent telecommunication skills, as the following excerpt suggests.

"I would be very interested in learning if others find live chats similar to what I ffound or if they find them particularly enjoyable . . . I do not like to read large amounts of text. I think therefore, that some of my unease in the first chat was due to personality and preference" (S6, 2-20-97).

Although we did not include self-disclosure as one of the codes for types of communication in our analysis of the IEC, it is clear from the preceding excerpt that students sought to create a sense of personal presence and establish trust in a safe learning environment.

As the semester progressed, the students continued to use various communication conventions and protocols in their efforts to communicate clearly with their peers and the instructor. As they practiced with these protocols, the students became more proficient in using IECs for brainstorming and group decision-making-two activities that are handled well in synchronous communication.

Conclusions

Much research on the development and use of communication conventions in computer-mediated communication has been conducted on transcripts of asynchronous conferencing in instructional environments (Harasim, 1990; Henri, 1992; Hiltz, 1994). Research on communication conventions in synchronous computer-mediated communication has occurred in settings such as MOOs and IRCs, which are used primarily for social and recreational activities. Not surprisingly, in this synchronous format most of the research has concentrated on the social aspects of communication, such as play and role-taking (Kuehn, 1993; Ruedenberg et al., 1995). The literature on real-time, computer-based communication in instructional settings is sparse, concentrating primarily on MOOs in which scholarly collaboration occurs (Day, Crump, & Rickly, 1996; Fanderclai, 1995), college writing classes (Day & Batson, 1995) and literature classes (Harris, 1995), and forms of exploratory MUDs for primary and secondary students (Woodruff, 1995).

This study first explored the existing literature to identify the communication conventions and protocols that occurred in synchronous chats, matched them against those developed by the students, then ranked the conventions in terms of their frequency of use in one IEC. While the findings are similar to the literature about communication conventions in non-instructional computer-based chat settings, the purpose of IECs in this study was different: the students were engaged in problem-solving and decision-making, which was usually accomplished through discussion and brainstorming. The students demonstrated their developing competencies in communicating with each other and with the instructor during the second large group chat of the semester. It should also be noted that the students in this course typically checked to see if any of their classmates, or their instructor, were logged on when they were, with the result that they engaged in numerous spontaneous small group chats. Such chats helped the students to improve their skills in a non-threatening setting as well as enhance their communication with the individuals involved in the chats. These private chats were not typically saved unless the individuals involved chose to log them.

The study also indicated that the students recognized a need to use their communication conventions and protocols to communicate clearly and minimize misunderstandings in their online transactions with others. Because the class met only three times face-to-face, the regular chat sessions supplemented the ongoing asynchronous communication by providing a simulated face-to-face experience during what would have been regular class meeting times. Because the course required the students to work collaboratively in small and large group endeavors throughout the semester, the chats in FirstClass provided regular opportunities for the students to ask questions, establish work schedules, make decisions, and stay in touch with their classmates and with the instructor. The students' metacognitive comments in the asynchronous conferences reflected their confusion when trying to follow multiple chains of discussion at once. It was particularly trying for the slow typists and non-native speakers of English. At the same time, these metacognitive comments reflect the students' growing skill in using keywords and names of individuals, shorthand techniques, asking questions and seeking clarification, and non-verbal cues in text.

Students were provided with little training or direction in the use of communication conventions and protocols prior to using chats in FirstClass, so the protocols that they developed were based on their prior experience in similar communicative settings (e-mail, classrooms, and face-to-face communication). Consistent with findings of other research (Rohfeld & Hiemstra, 1995), the students acknowledged that the IECs would have been less productive and more difficult without having first established rapport with the other group members in a face-to-face setting. Additionally, as Hillman, Willis, & Gunawardena (1994) suggest, it is imperative to determine how best to help students in advance to be competent with the technology, in this case the chat process, as they typically didn't had the experience of working in a synchronous computer communication setting. Pairing students with keyboard partners of unequal ability in the FirstClass training session appeared to decrease the cognitive load created by having to pay simultaneous attention to content and a new technological process. What structures, like shared keyboards and group work, lead to the social construction of knowledge via IECs?

When asked to debrief IECs online, students in the TP course made such observations as "you can interrupt without interrupting," "I feel I'm in another world," "you can't send too much content in a chat," "a moderator is needed in a chat involving more than 2 people," "good for visual learners and field dependent learners," and "Hey, can we get back on video and hash this out? I find chats very frustrating" (excerpted from the third large group chat in the first semester of using FirstClass, 7-24-96).

Further Research

Further research on IECs needs to include content analyses of chats logged at different times during the semester, chats that focus on specific tasks and topics, and chats in various disciplines, to determine if the conventions and protocols remain constant and useful over varied times and settings. Additionally, the categories for content analysis should be refined to include the category of self-disclosure that we did not include in the current study and other categories as they emerge. Further investigation could determine optimal pedagogical techniques that could be employed effectively in IECs of different kinds. Because of the growing global community fostered in the computer-mediated communication environment, it would also seem important to determine ways to empower non-native speakers of English with the conventions and protocols necessary to communicate easily in IECs.

Finally, the growing acceptance and use of instruction grounded in constructivism implies that instructors and teachers need to know more about how to facilitate IECs that foster communication and learning in such settings. Gunawardena, Anderson, and Lowe (1996) developed a model for constructivist interaction analysis of asynchronous computer conferencing, and this model should be tested in a synchronous chat environment.

References

K. Aoki, 1995. "Synchronous multi-user textual communication in international tele-collaboration," Electronic Journal of Communication, Volume 5, Number 4, at http://www.cios.org/getfile\AOKI_V5N495

R. Archee, 1993. "Using computer mediated communication in an educational context… Educational outcomes and pedagogical lessons of computer conferencing," Electronic Journal of Communication, Volume 3, Number 2, at http://www.cios.org/getfile\ARCHEE_V3N293

A. Aycock, 1995. "Technologies of the self: Foucault and Internet discourse," Journal of Computer-Mediated Communication, Volume 1, Number 2, at http://jcmc.huji.ac.il/vol1/issue2/aycock.html

S. D. Brookfield, 1986. Understanding and facilitating adult learning. San Francisco: Jossey-Bass.

M. Csikszentmihalyi, 1977. Beyond boredom and anxiety. San Francisco: Jossey-Bass.

P. Curtis, 1992. "Mudding: Social phenomena in text-based virtual realities," Intertek, Volume 3, Number 3, at ftp://parcftp.xerox.com/pub/MOO/papers/DIAC92.txt

R. H. Cutler, 1995. "Distributed presence and community in cyberspace," Interpersonal Communication and Technology: A Journal for the 21st Century, Volume 1, Number 2, at http://jan.ucc.nau.edu/~ipct-j/1995/n2/cutler.txt

M. Day and T. Batson, 1995. "The Networked-based writing classroom," In: Z. L. Berge and M. P. Collins (editors), Computer-mediated communication and the online classroom in higher education. Cresskill, N. J.: Hampton Press, pp. 25-46, and at http://129.118.38.138/kairos/1.2/coverweb/cmcmday.html

M. Day, M., E. Crump, E., and R. Rickly, 1996. "Creating a virtual academic community: Scholarship and community in wide-area multiple-user synchronous discussions," In: T. M. Harrison and T. Stephen (editors), Computer networking and scholarly communication in the twenty-first-century university. Albany, N. Y.: State University of New York Press, pp. 291-311, and at http://www.missouri.edu/~wleric/cnsc21/

T. L. Fanderclai, 1995. "MUDs in education: New environments, new pedagogies," SenseMedia Homepage, at http://lucien.sims.berkeley.edu/MOO/muds-in-education.html

E. Goffman, 1959. The Presentation of self in everyday life. New York: Doubleday.

C. N. Gunawardena, T. Anderson, T., and C. A. Lowe, 1996. "Interaction analysis of a global on-line debate and the development of a constructivist interaction analysis model for computer conferencing," paper presented at the Annual Convention of the American Educational Research Association, New York.

L. M. Harasim (editor), 1990. Online education: Perspectives on a new environment. New York: Praeger.

L. D. Harris, 1995. "Dante in MOO space: Using networked virtual reality to teach literature," Electronic Journal of Communication, Volume 5, Number 4, at http://www.cios.org/getfile\HARRIS_V5N495.

F. Henri, 1992. "Computer conferencing and content analysis," In: A. R. Kaye (editor), Collaborative learning through computer conferencing: The Najaden papers. New York: Springer-Verlag, pp. 117-136.

D. C. A. Hillman, D. J. Willis, and C. N. Gunawardena, 1994. "Learner-interface interaction in distance education: An extension of contemporary models and strategies for practitioners," American Journal of Distance Education, Volume 8, Number 2, pp. 30-42.

S. R. Hiltz, 1994. The Virtual classroom: Learning without limits via computer networks. Norwood, N. J.: Ablex.

K. Krippendorff, 1980. Content analysis: An introduction to its methodology. Beverly Hills, Calif.: Sage.

S. A. Kuehn, 1993. "Communication interaction on a BBS: A content analysis," Interpersonal Communication and Technology: A Journal for the 21st Century, Volume 1, Number 2, at http://www2.nau.edu/~ipct-j/1993/n2/kuehn.txt

J. A. Levin, H. Kim, and M. M. Riel, 1990. "Analyzing instructional interactions on electronic message networks," In: L. Harasim (editor), Online education: Perspectives on a new environment. New York: Praeger, pp. 185-213

P. Lloyd (editor), 1994. Groupware in the 21st Century. London: Adamantine Press.

R. Mason, 1991. "Moderating educational computer conferencing," DEOSNEWS, Volume 1, Number 19, Archived as DEOSNEWS 91-00011 on LISTSERV@PSUVM.PSU.EDU

M. S. McIsaac and K. D. Ralston, 1996. "Teaching at a distance using computer conferencing," TechTrends, Volume 41, pp. 48-53.

M. G. Moore, 1993. "Theory of transactional distance," In: D. Keegan (editor), Theoretical principles of distance education. London: Routledge.

M. G. Moore and G. Kearsley, 1996. Distance education: A systems view. New York: Wadsworth.

K. L. Murphy, S. Cathcart, and S. Kodali, 1997. "Integrating distance education technologies in a graduate course," TechTrends, Volume 42, pp. 24-28, and at http://disted.tamu.edu/~kmurphy/techtrd2.htm

K. L. Murphy, L. Cifuentes, A. D. Yakimovicz, R. Segur, S. E. Mahoney, and S. Kodali, 1996. "Students assume the mantle of moderating computer conferences: A case study," American Journal of Distance Education, Volume 10, Number 3, pp. 20-36.

K. L. Murphy, R. Drabier, and M. L. Epps, 1997. "Incorporating computer conferencing into university courses," 1997 Conference Proceedings: Third Annual International Distance Education Conference, pp. 147-155. College Station: Texas A&M University, and at http://disted.tamu.edu/~kmurphy/dec97pap.htm

G. B. Newby, 1993. The Maturation of norms for computer-mediated communication," Internet Research, Volume 3, Number 4, pp. 30-38.

G. M. Phillips and S. B. Barnes, 1995. "Is your epal an ax-murderer?" Interpersonal Communication and Technology: A Journal for the 21st Century, Volume 3, Number 4, pp. 12-41, and at http://jan.ucc.nau.edu/~ipct-j/1995/n4/gmp1.txt

E. M. Reid, 1991. Electropolis: Communication and community on Internet Relay Chat. Honors Thesis, University of Melbourne, Australia, at http://troll.elec.uow.edu.au/~neut/electrop.html

E. M. Reid, 1994. Cultural formations in text-based virtual realities. MA Thesis, University of Melbourne, at http://www.ee.mu.oz.au/papers/emr/cult-form.html

R. W. Rohfeld and R. Hiemstra, 1995. "Moderating discussions in the electronic classroom," In: Z. L. Berge and M. P. Collins, (editors). Computer-mediated communication and the online classroom: Distance education. Cresskill, N. J.: Hampton Press, pp. 91-104.

L. Ruedenberg, B. Danet, and Y. Rosenbaum-Tamari, 1995. "Virtual virtuosos: Play and performance at the computer keyboard," Electronic Journal of Communication, Volume 4, Number 4, at http://www.cios.org/getfile\RUEDEN_V5N495

C. Siemieniuchand M. Sinclair, 1994. "Concurrent engineering: People, organisations and technology or CSCW in manufacturing," In: P. Lloyd (editor), Groupware in the 21st century. London: Adamantine Press.

M. Woodruff, 1995. "Multi-user dungeons enter a new dimension: Applying recreational practices for educational goals," Electronic Journal of Communica