«Beyond Threaded Discourse JAMES HEWITT Ontario Institute for Studies in Education University of Toronto 252 Bloor Street West Toronto, Ontario, ...»
International Jl. of Educational Telecommunications (2001) 7(3), 207-221
Beyond Threaded Discourse
Ontario Institute for Studies in Education
University of Toronto
252 Bloor Street West
Toronto, Ontario, Canada M5S 1V6
The educational potential of asynchronous, computer-medi-
ated conferencing is well documented. Opportunities for in-
creased group interaction, more equitable communication patterns, higher degrees of reflection, and time-and-place-in- dependent discussions are some of the benefits cited by re- searchers. This article focuses on one of the apparent limita- tions of the medium: the lack of support for convergent pro- cesses. Threaded online environments support electronic conversations that expand and branch, but provide few facil- ities for drawing together discourse in meaningful ways. The implications of this restriction are explored in two studies.
The first study analyzes the degree to which students and in- structors write convergent notes (e.g., notes that synthesize or summarize ideas) in three graduate-level computer confer- encing courses. The second study explores student percep- tions relating to their own synthesizing and summarizing practices. The results suggest that online participants rarely engage in convergent processes in spite of widespread agree- ment that such efforts confer educational benefits. Possible explanations for this phenomenon are discussed.
Recent advances in computer and telecommunications technologies have raised new possibilities for distance education. Increasingly, the Inter- net is the medium of choice for delivering course materials and supporting interaction between teachers and learners. One of the more common distance education technologies is asynchronous computer-mediated conferencing 208 Hewitt (CMC). In a typical CMC course, individuals dial-in to a central database from anywhere in the world and view the writings of their teacher and classmates. Responses to these writings can then be crafted and stored to the course database for others to read. In this fashion, whole-class discus- sions can take place without having to coordinate a common meeting time or meeting place. Instead, people can participate from home and organize class time around their individual schedules (Kaye, 1989; Harasim, 1987, 1989).
Studies of CMC as an alternative to traditional instruction suggest that online environments affect more than just the “where” and the “when” of course-taking. They also change the nature of classroom discourse. In a regular classroom, discussion is sequential and transient; only one person (generally) speaks at a time and the content of the conversation is not preserved.
CMC, on the other hand, allows everyone to “talk” (write) at once because there is no need for turn taking (Mason & Kaye, 1990). This allows shy, and less vocal students to participate without risk of interruption (Davie, 1988), and reduces the possibility of a dynamic individual dominating the conversation (Eastmond, 1994; Tuckey, 1993). Furthermore, since all discourse is preserved electronically, participants can easily revisit old ideas and reflect longer on new ideas before committing them to public scrutiny (Levinson, 1990; Mason & Kaye, 1990). Thus, asynchronous CMC has the potential to be a highly social, egalitarian, and deliberative medium (Harasim, 1989). It would be a mistake to interpret this form of distance education as simply conventional instruction delivered remotely. CMC provides affordances for a fundamentally different form of learning, one that engages students as active, reflective participants in an electronically-linked community.
Despite the promising nature of CMC-based instruction, a number of problems are often associated with online courses. Frequently cited concerns include Information Overload (Paulo, 1999), the challenge of conveying emotion in a text-only environment (Hiltz, 1986), and the lack of support for convergent (e.g., synthesizing and summarizing) processes (Eastmond, 1994). The latter issue is the subject of the following investigation.
Researchers and online instructors have known for a long time that it is difficult to bring ideas together in a CMC environment (e.g., Beckwith, 1987;
Harasim, 1990; Eastmond, 1994). Yet there have been no studies that have analyzed the problem in depth. This article begins an exploration of this issue by examining the incidence of convergent processes in three coursebased computer conferences at the University of Toronto. This analysis is accompanied by a study of student perceptions relating to their own efforts to synthesize and summarize ideas. Subsequent sections discuss the implications of these findings and possible avenues of research for designers of next-generation discourse environments.
209 Beyond Threaded Discourse
The data for this study were drawn from three graduate-level online courses at the University of Toronto. Class sizes ranged from 15 students to 24 students (21 students on average). A different instructor taught each course, but the instructional formats were similar: each engaged students in a series of issue-based, exploratory discussions on matters relating to education.
Each of the three classes used a conventional threaded, web-based computer conferencing environment for their online activities. Threading allows an author to designate a newly written note as a response to an earlier note. This makes it easier for readers of the conference to find and follow conversational chains. For example, in Figure 1, a list of indented note titles illustrates the threaded relationship between eight notes in a web-based conference. Notes 1 and 2 begin two separate threads. Notes 3 and 5 were saved as responses to note 1, notes 7 and 8 were saved as responses to note 3, and so forth. The hierarchical structure allows the reader to trace a conversational branch (e.g., note 1, note 3, note 7) without being distracted by unrelated notes from other threads (e.g., note 4). It also allows the class to simultaneously pursue multiple avenues of inquiry without confusion.
Note 2. The learning paradox by Jim Dunlop Note 4.
What knowledge are we born with? by Linda Morrison Figure 1. A typical threaded organization Students and instructors in each class had full access to the contents of their class conference. All participants could initiate new threads. Interaction was asynchronous; people could login at times of their own choosing to contribute to the course discussions. To respond to a particular note, participants would display the note on the screen and then click on a “Reply” button. They would then compose their response and save it to the conference.
In this fashion, members of a class could engage in extended online discussions with one another. On average, each student contributed between two and three notes a week.
210 Hewitt The organization of threaded discourse (Figure 1) might lead a casual observer to conclude that online discussions develop along branching paths that never come together again. However, it is important to distinguish between the hierarchical structure imposed by the system and the linkages that are implicit in the text of the notes. Online discussions may be much more intertwined and interrelated than the threaded representation indicates. For example, a learner may choose to write a note that summarizes a thread, or reconciles two, previously distinct, lines of inquiry. Such notes are convergent because they cut across the hierarchical structure of a conference by pulling together ideas from multiple sources. To identify convergent notes, and to determine their frequency in computer conferences, the following
qualitative rating scheme was developed:
1. Standalone: A standalone note introduces new ideas to the conference and does not build on the ideas of other notes. Typically, a standalone note is one that begins a new thread.
2. Add-on: As the name suggests, an add-on note builds on the ideas of one other note in the conference. Typically, these are notes in which one person responds to an idea that someone else has introduced. Such notes are not considered to be attempts at convergence because convergence (by definition) involves bringing together ideas from at least two different sources.
3. Multiple reference without convergence: These notes make reference to two or more previous notes, but not in a way that could be considered an attempt at convergence. For example, the phrase, “I agree with Eli’s response to Zoe’s note,” makes mention of two notes, but does not discuss the content of notes or how they interrelate.
4. Convergent: A convergent note is one that discusses (if only briefly) some of the ideas expressed in two or more other notes in the conference.
Two raters analyzed a total of 830 notes from the three distance education courses. Notes were extracted from main discussion areas only; special areas reserved for practice or announcements were not included in the study. The inter-rater reliability was.90 and disagreements were settled by discussion among the raters. The results are displayed in Table 1. Ninetyfour percent of the students’ notes were assigned to one of the first two categories. These notes either made no reference to other messages in the conference, 211 Beyond Threaded Discourse or they made a reference to a single previous submission. Four percent of the submissions contained multiple references and only 2% were rated as attempts at convergence.
The performance of the course instructors varied. One instructor chose not to participate in the mainline conferences at all. Another instructor participated regularly, but even this individual only produced a handful of notes that could be categorized as convergent.
The results suggest that none of the three computer conferences contained many convergent notes, even though the course instructors had different instructional styles. While there was a considerable amount of interaction between participants, virtually all it could be characterized as add-on style responses. Few people tried to produce a more sophisticated kind of note that tied together ideas from different sources.
In an attempt to further understand these results, a follow-up questionnaire was administered to 32 students. Each person was asked a variety of questions pertaining to his or her online habits. The key findings were as
All students said that they would personally benefit from higher levels !
of synthesis and summarization in their class computer conference;
81% of the students said that they personally never make an effort to !
synthesize or summarize ideas from different notes during their computer conferencing sessions;
75% of the students said that the notion of synthesizing or summarizing !
ideas from different notes rarely or never occurs to them; and 50% of the students said that they rarely or never check to see how their !
These results suggest an interesting discrepancy between student desires and student actions. All of the students reported that more synthesizing and summarizing would be beneficial. At the same time, most of the learners acknowledged that they rarely wrote synthesizing notes themselves. In fact, three-quarters of the students indicated that the notion of writing a synthesizing or summarizing note rarely or never occurred to them.
There may be a number of reasons why students rarely work at consolidating ideas in a computer conference. One possibility is that students see synthesizing and summarizing as the teacher’s responsibility. Another possibility is that convergent operations are more intellectual taxing than simply extending an existing thread. Consequently, people may avoid the complex business of drawing together ideas, since they can make other kinds of contributions with less effort. However, the “conservation of effort” hypothesis does not fully explain why so many students rarely (or never) consider writing a summary. This article hypothesizes that there may be other factors at work. In particular, it is proposed that the lack of convergence is partially tied to the “reply” protocol in most computer conferencing programs. The reply feature of CMC environments—indeed, the very word “reply”—suggests a practice of writing responses to individual notes. Consequently, people are less likely to take the multiple note perspective required for convergent operations. The problem is not simply that CMC systems make it inconvenient to write notes that bring together ideas from different sources. Rather the problem is that such operations are less likely to come to mind when “reply to this note” is the only support for interaction offered by the software. Consequently, as the questionnaire data indicates, online participants may start to develop a rather limited idea of what online interaction entails. They think primarily in terms of targeted responses to particular notes, and rarely consider summarizing and synthesizing (which require a multiple note focus) as alternatives.
Interestingly, even when attempts are made to summarize findings, such efforts may be hampered by their low visibility. For example, the indented, hierarchical organization of Figure 1 precludes the possibility of a new note simultaneously extending the, “Where is knowledge?” thread and, “The learning paradox” thread since each new note can only be electronically linked to one of its predecessors. Consequently, if a summary note is linked to the “Where is knowledge?” thread, people reading other threads 213 Beyond Threaded Discourse are less likely to be aware of the summary’s existence, or its relevance to those discussions. This may limit the summary’s impact and effectiveness.