Online Collaboration and the Role of the Instructor in Two University Credit Courses

Xinchun Wang and Lucio Teles

Abstract: This paper examines the patterns of online interactions among learners and the role of the instructors in two university credit courses, one in Arts and the other in Business Administration. These courses were delivered entirely online in Spring 1997 through Virtual-U, a Web-based environment developed at Simon Fraser University, Canada. Computer conference transcripts were analyzed to identify patterns of interactions among the participants. Separate interviews with the course instructors were carried out to investigate their role in course delivery. The results of the analysis showed a high level of online interactions among students, who benefited from building on each otherÕs ideas in the active conference discussions. The results also indicated increased student responsibility in initiating online discussions. The instructors acted as facilitators throughout the courses and the online teaching experience has influenced their attitudes toward teaching in the traditional face-to-face mode.

Keywords: Interaction Patterns, Online Collaboration, Student Responsibility, Instructional Design, Facilitator.

1. Introduction

Among the most cited advantages of the communications networks used in education are the promotion of social interaction among learners, particularly for those students who are often marginalized (Faigley, 1990, cited in Warschauer, Turbee & Roberts, 1994, Wang & Hurst, 1997); increased student responsibility and autonomy; change in the traditional instructorÕs role as authority, and improved learning skills (Warschauer, Turbee & Roberts, 1994; Kern, 1995a). Collaboration among learners has also been considered an important criterion in evaluating the success of an online course (Duin & Hansen, 1994, Berge,1996). When collaborative tasks are well designed and supported, there is increased motivation and learning (Harasim et al., 1995, Teles & Rylands, 1998).

In online environments, instructors tend to act as a facilitators rather than as knowledge providers or lecturers. Students and instructors form "online communities of learners" to build knowledge in online social environment s (Scardamalia & Bereiter, 1994).

There have been evaluations of the advantages and drawbacks of the use of computer-mediated communications in pedagogy based on the social theories of Vygotsky and Bakhtin. "According to Vygotsky, all learning is inherently social in nature. His theory of the zone of proximal development posits that learners benefit most from social interactions concerning tasks that they cannot do alone but can do in collaboration with more knowledgeable or more experienced individuals" (Kern, 1995a, p, 4). Duin & Hansen (1994) use BakhtinÕs social theory to support their model of social interaction and social construction. "A speaker gives voice to a thought, an utterance, this utterance, though representing the ideas of an individual, reflects a social environment that is shared. The listener interprets the utterances in a purposeful, conscious act, in terms of his or her own concept of the social context, in terms of what the words mean to him or her individually. Therefore, speech and writing are dialogical in that the meaning of an utterance is created by both the speaker/writer and listener/reader through social interaction". Furthermore, "all dialogue should have the goal of eliciting further dialogue, not shutting it off with pronouncements of truth or other forms of closure" (Duin & Hansen, 1994). The usefulness of computer networks in education should therefore be assessed in terms of whether they promote or inhibit social interaction and knowledge building.

Although computer-mediated communication has the potential to enable online collaboration in learning, the instructional design of the course and the instructorÕs commitment to the promotion of the collaborative environment are essential for success. "What distinguishes online instruction from entertainment or recreation is the purposefulness of the designers and developers in provoking certain intelligent responses to the learning materials, context, and environment" (Berge, 1996, p.1).

Thus, the core issue for instructors who want to implement quality teaching and learning online must be to address principles of instructional design and moderation. This paper examines the patterns of interactions and online collaboration among students in two university credit courses offered through Virtual-U and how these patterns are affected by instructional design. The changing role of the instructor will also be investigated. Data for analysis include the transcripts of conference messages and interviews with instructors. The specific questions addressed are:

1. How do students interact with each other in conference discussions? How does online collaboration affect student learning?

2. What roles do the instructors play in these two online courses? How do they encourage studentsÕ online collaboration?

3. Does teaching online influence instructorsÕ attitudes towards teaching in the face-to-face classroom? If so, how?

2. Virtual-U and Course Information

Virtual-U is an environment designed to deliver university courses and training programs. It operates via an Internet browser and offers asynchronous communication features that allow participants to contribute at convenient times. Its core function lies in the computer-conferencing system (called V-Groups) where instructors and students participate in course activities and exchange messages. Its other features, which include Course Editor, Grade Book, and System Administration Tools, provide a flexible framework to support active learning, collaboration, and knowledge building. The online courses were offered through the Centre for Distance Education (CDE). Print course materials including, for example, readers and assignments dates were sent out from CDE as they are for all regular distance education courses.

The two courses included in the present study were selected from the undergraduate courses delivered entirely online in the Spring Semester, 1997, from January to April, at Simon Fraser University, Canada. They were Dancing in Cyberspace: Creating with the Virtual Body (hereafter, Dance) and Information Analysis and System Design (hereafter, Business). The former offered by the School of the Contemporary Arts, and the latter, by the Faculty of Business Administration.

The Dance course was designed to introduce the virtual body and its creative potential in cyberspace. Students used LifeForms, a software for 3-D human figure animation, to analyze and create simulations of human movements. Thirty-three students participated in the course (23 credit students and 10 special audits). The credit students were regularly enrolled students at Simon Fraser University, and the non-credit participants were from elsewhere in Canada, the U.S., Spain, and France. In the Business course, students learned the process of analyzing business information technology requirements and designed an information system to meet the needs of these new requirements. The course had 50 students, three of them in Finland. Each course was custom designed and included a number of conferences created by the instructors. Students were required to post messages on given topics and to comment on peersÕ messages within a given period. As the two courses were both offered entirely online, the conferences where students posted their comments on readings and exchanged comments on assignments were the most important component of the courses.

3. Conferences and Quantitative Data

Figure 1. Number of messages by the students and the instructors.

First, quantitative analysis of conference messages was conducted to provide the number of messages produced in the discussions. The Dance course had seven conferences, and the students and instructors posted a total number of 985 messages. These conferences were divided into two major types. One focused on the discussions of topics from the readings. The other included exchanges of ideas on how to use the software (LifeForms) to create animation and comments on peersÕ animation assignments.

The Business course had 24 active conferences that were chronologically sequenced ( Week 1, Week 2, and so on) as the course progressed. A total of 1,063 messages were generated. Both courses also included a ŌHelpÕ conference to provide online support for solving technological problems students encountered in using the system. The total number of messages contributed by both the students and the instructors (including the two Teaching AssistantsÕ messages) for the two courses are given in Figure 1.

4. Patterns of Interactions

To gain a better understanding of the learning process and evaluate the results of the online courses, it is important to go beyond the amount of participation and undertake qualitative analysis. Previous studies indicate that content analysis of computer conferencing messages help us to understand the learning process (Harasim & Winkelmans, 1990, Henri, 1992, Hiltz, 1997, Levin, Kim & Riel, 1990).

Researchers have been working to establish a theoretical framework for qualitative analysis and to improve research methodologies in order to use the rich content area of computer-conferencing messages. Previous studies indicate that the dynamic aspects of group communication and cognitive behavior in the learning process must be taken into account because collaborative learning is one of the most important characteristics of computer conferencing. Learners clearly build upon each other's knowledge in their contributions (Harasim, 1993, Henri, 1992, Bullen, 1997). The focus so far has been on the degree of interactive participation and the patterns of interactions. As online collaborative learning becomes more common, new indicators to assess collaborative group work and class participation will are needed.

The degree of interaction has also been defined differently as a three-step process in which the level of references differ. Bretz (1983) defines interactivity as a three-step process: step 1. communication of information; step 2. a first response to this information; step 3. a second response relating to the first. Henri (1992) had identified two types of participation: independent and interactive depending on whether or not the participants make references to other messages. Independent messages are those that deal with the topic of the discussion but make no reference (either implicit or explicit) to any other messages. Interactive messages are those that refer to other messages by responding to, elaborating, and building on them while dealing with the topics of discussion. Henri further distinguished between explicit interaction, where the content of the message refers specifically to one or more messages, and implicit reference, where the content obviously refers to one or more messages or ideas but does not specifically mention the connection. Bullen (1997) analyzed the mean and the total number of independent and interactive messages to determine the degree to which students are building on each otherÕs contributions in their interactive participation.

The categorizations of messages in terms of interactions obviously vary according to the different emphases of the researchers. Similarly, the unit of analysis used differs. Sometimes, the whole message is used as the unit, and elsewhere it has been subdivided into discrete units of meaning for the purpose of the study.

Here the unit for analysis is the complete message to examine the patterns of interactions in the online conferences. The analysis was carried out with the aim of identifying two levels of interaction. The first level characterizes the interaction of the students and instructors with the content of the courses in the form of their responses to the required readings and to lectures or assignments posted by the instructors. The other level examines the interpersonal interaction, especially student with student collaboration, which is the most important part of the interaction online. The messages were classified into the five categories described below.

Coding Criteria

Conference messages were analyzed manually to identify patterns of interactions among the participants. Both courses lasted 13 weeks and had created a total of 30 conferences. The analysis of all the conference messages is desirable but time constraints made it impossible. Therefore, representative conferences were chosen with the aim of including a variety of them. For the Dance course, two out of six conferences were selected. One focused on the discussion of the reading topics and the other dealt with using the software to create animations. For the Business course, six out of twenty-four conferences were selected and analyzed.

Five categories of messages were defined to capture the patterns of interactions in online discussions. They are 1) Students' responses to course topics, 2) Student to student responses, 3) StudentsÕ questions, 4) Instructors' responses to students, 5) InstructorÕs assignments/comments. The following is a brief description of the criteria for the classification of the messages.

1. Students' responses to course topics

Messages in this category were studentsÕ responses to the topic assignments (representing person-content interaction). They were generally the required postings for the weekly discussion and were mostly aimed at the whole class or group, including the instructors. These messages were not directed to a particular student or participant.

2. Student to student responses

Messages of this category were directed from one student to another (indicating interpersonal interaction). The identification was based on the direct mention of the name(s) of the student(s) in the text or in the heading of the message. The content of the message further helped to confirm the classification. These messages were usually comments on an earlier message by another student. They were the most interactive in nature, and the exchange of messages among peers often went on for a few rounds. They were, however, all coded in this category whether the response to an earlier message was first, second, or even third round.

3. StudentsÕ questions

These messages basically consisted of questions on course topics and technical problems. All the initiating questions (first round) were coded as this category whether they were directed to a particular peer, the instructors, or the whole group. The reply or replies to the questions are coded as category "Student to student responses" or "InstructorsÕ responses to students", depending on the person who replied.

4. InstructorÕs responses to students

Instructors' replies to studentsÕ questions and instructors' comments on studentsÕ earlier messages fall into this category (indicating interpersonal interaction). They include the instructorsÕ comments on a particular studentÕs earlier message (direct mention of the name of the student). Instructors' responses to students' questions are also coded as this category. These messages by the instructors were not initiated by the instructors themselves but by the earlier messages sent by an individual student.

5. Instructors' assignment/comment

These messages refer either to the instructors' introduction to the topics of the weekly discussion or to the summaries of the topic discussion. Sometimes they are assignments or explanation of the assignments.

It is important to remember that the unit for coding is the message. Thus, a message may have multiple functions, but it was counted only once in one of the above five categories. The judgment was based on the main function of the message as suggested by the heading and the context. For example, a message that was a student reply to an earlier student question or comment and was directed to that particular student might also bring in or refer to other students or the instructor. It was still counted only once as the category of "Student to student response" based on the main function of this message.

Results

A total of 631 messages of the selected conferences were analyzed and classified into the 5 categories. The results are summarized in Table 1 and Table 2 respectively.

Table 1. Five categories of messages indicating patterns of interactions in two conferences of the Dance course

A one-way repeated measures ANOVA was carried out to investigate the differences between the interaction patterns (category of messages). The effect of category was significant [F(7,4) = 5.809 p < .005].The post hoc Tukey (HSD) revealed (p < .05) that the number of messages in the category of "Student to student responses" was significantly more than the number in all the other categories except "Students response to the topics." StudentsÕ messages to each other were significantly greater than instructorsÕ responses and comments. Yet, although the ratio of number of "StudentsÕ response to the topics" and "Student to student responses" was 1:2 (20 vs. 42.3), the difference did not reach the level of significance. The level of person-content interaction on the studentsÕ part was also high, especially for the Business course. Students posted a substantial number of messages that were responses to the topics, the first round of messages that provided basis for the follow-up interpersonal interaction when students commented on each otherÕs points.

Table 2. Five categories of messages indicating patterns of interactions in six conferences of the Business course

The majority of the messages were "Student to student responses." Peers responded to each otherÕs messages to make comments that carried on the discussions. Often the discussion went on with several rounds of messages back and forth among the participants. Students found the online environment different from the traditional classrooms from the very beginning of the course and demonstrated their preferences for and interest in online collaboration. One student from the Dance course commented in her message:

I hope the online learning will provide the interaction and support that is missing from traditional lecture style education or distance learning. I believe that we will have an opportunity to interact in the conferences and learn from each other more than we could in a place-based environment. I have already discovered that messages in this conference to be very fascinating and thought-provoking. I think that sharing our ideas and working collaboratively in the technology will prove to be fun and interesting.

As anticipated by this student, the Dance course discussion turned out to be very interactive among the students. The topic conference was in the form of group discussion in which students made comments on one antherÕs messages to build upon each otherÕs ideas. In the animation conference, students worked as a group as well as in dyad mode when partners were assigned to critique each otherÕs animation. The impact of this collaboration by the students on learning is discussed later in the interview with the instructor.

The Business course conferences were structured chronologically as the course progressed. Each weekly conference contained a few topics for discussion. The conference usually began with the instructorÕs posting the weekÕs lecture notes followed by one or two studentsÕ summarized "Top Ten" points for discussion. These points were concerned with the topic and the reading for the week. Students then posted their responses to the "Top Ten." Usually every student posted at least one message to respond to the topic points. The discussion often focused on one or two points that formed the key issues of the week. In the next round of messages, students often expressed differing opinions about the key issues and commented on other studentsÕ postings.

The results of the transcript analysis indicated that both courses were successful in promoting a supportive online learning environment in which students willingly interacted and collaborated with each other in the discussions. If such collaboration promoted and facilitated learning, as is often claimed by educators using social theories, what direct impact did it make on learning in terms of fulfilling the goals of the course requirements? How exactly did the students benefit from such collaboration online in terms of their learning experiences and the outcomes of the courses? The instructors were interviewed to address these questions.

5. The InstructorsÕ Roles and Online Collaboration

The role of the instructor in an online course is often described as a moderator and facilitator rather than as a knowledge provider. To act in these roles, instructors must create the necessary pedagogical and social conditions for successful collaboration among the students in promoting learning and fulfilling the course tasks. Instead of delivering formal lectures, instructors set clear objectives for discussion, encourage participation, keep discussions on track, and weave together different treads and the course components (Berge, 1996). We will examine the role of the instructors of these two courses and discuss how they influenced studentsÕ collaboration.

The transcript analysis showed that the amount of online interpersonal interaction was high for both courses. The motivation for such collaboration online, however, is not yet clear. Furthermore, it is important to know the impact of such collaboration on learning outcomes. In addition, it is necessary to explore how instructors encourage online collaboration. With these questions in mind, we interviewed the two instructors.

Both instructors emphasized that instructional design was important for the success of the course in any mode. However, teaching an online course requires even more careful planning. As the instructor of the Dance course put it, "because it was an online course, it had to be more structured." She attributed the success of her course to "a more structured teaching style". The results seemed to be in agreement with BergeÕs (1996) claim that "what distinguishes online instruction from entertainment or recreation is the purposefulness of the designers and developers in provoking certain intelligent responses to the learning materials, context, and environment."

Students need guidance in the online course to participate in the discussions and to build on each othersÕ contributions. The instructors of both courses emphasized studentsÕ responsibility in the form of participation and collaboration as the crucial part of the online course. Specific requirements were set out before the course started. In both courses, students were required to post their comments on each otherÕs messages on a regular basis, and the importance of participation was reflected in the course grading system.

The Dance course had two main forms of collaboration: dyads to create animation, and group discussion to discuss course readings. In the course outline, the instructors specified the number of messages students were required to post for the discussion within a certain period of time as well as the length of the messages. Discussion groups were formed for the reading topic conference, and the dyad mode was used for the more technical task of creating animation using the LifeForms software. Students created animations and sent them to the partner for comments and made changes on each otherÕs creations. Given the nature of the tasks, the work produced was the joint efforts of the two students. Obviously, collaboration is the crucial requirements for carrying out tasks of this kind. The instructor was amazed by the success of the cooperation and amount of support the students gave each other, especially by the fact that students were very receptive of each otherÕs critiques. She commented: "I found students were very receptive to the criticism they received from their classmates whereas they were not as receptive to the criticism received from the teacher. Without any reminding and without me making any rules, they were supportive."

The Business course had similar discussion groups for the reading tasks. Students were required to moderate the discussion in turns by summarizing the instructorÕs lectures into the Top 10 points that were used as the focus for the weekÕs discussion. It worked so effectively that, in fact, all the discussions were initiated by the students while the instructors performed just like participants. The instructors confirmed that they never passed any value judgment on the studentsÕ work as a course supervisor. As the Dance course instructor commented, "I think itÕs a good idea for the instructor not to set up himself as a model of how to critique. I wouldnÕt have said this before I taught" [this course]". The result seemed to support the claim that long, elaborate, logically coherent sequence of comments [by the instructors] yield silence while open-ended remarks and examples elicit comments and help to encourage more discussion (Berge, 1996).

The success of these two courses further confirmed that instructors do not necessarily have to demonstrate their roles as the authority and knowledge provider in online courses. Furthermore, instead of fearing of the absence of such an authority, the instructors were very positive to the change of the roles and were amazed to observe the increased amount of student responsibility in learning. As the Dance course instructor commented: "It [student collaboration] worked very well and I was surprised at how well it worked. It worked very well that I think that it goes back to the old idea that it helps you if you have to think critically about someone elseÕs work, it helps to improve your own work." Indeed, learners benefited from collaborations concerning tasks that they cannot do alone but can do in collaboration with more knowledgeable or more experienced individuals.

The methodology of encouraging collaboration in teaching was not new to the instructors, who have both taught for many years in the traditional classrooms. In fact, they had both applied this teaching style in teaching face-to-face mode. Yet, both agreed that the online environment created a better and more effective use of this methodology and, therefore, that it is a unique feature of the online environment. The Dance course instructor made this comment:

The methodology that you step back and become a facilitator is a methodology we have all used but not a methodology that we generally use throughout the entire course face-to-face. You use different methodology, you tell them the answers, you lecture, and you then step back as a facilitator for different things. With the online course, you are a facilitator throughout the entire course.

The power of the online collaborative model seemed to have changed the instructorsÕ view toward teaching and to have had a profound impact on their teaching style. For example, the instructor of the business course commented:

Teaching the online course changed the way I taught the place based course. Before I taught the online course, I would walk into the place based course and I would start the session off by trying to tell them something, by trying to bring a new topic out and then I would try and draw them out. After the online course, I found that the students had some really neat examples. After teaching the online course, I would walk into the place based course and tell the students that they were responsible to come up with ideas for the discussion.

When asked to evaluate the learning outcomes of the courses, both instructors agreed that the students achieved the course objectives and that the results were comparable to the face-to-face course. In fact, the Dance course instructor claimed that students "accomplished a lot more and did better than those in a similar course taught face-to-face." For the business course, the students also performed better in their examinations on concept-related questions than those in the same course taught face-to-face. The instructor attributed this achievement to the benefits of the online discussion in which students built upon each otherÕs contributions. However, the students in the online Business course did not do better on the examination questions concerning the technical content, such as drawing flow charts. This finding may indicate that certain aspects of the course content may still need face-to-face sessions that involve more direct hands-on demonstrations in teaching. Overall, the examination scores for the online Business course was comparable to the same course taught face-to-face.

6. Conclusion

Transcript analysis showed a high level of interpersonal interactions among the students in computer conferences of both courses. Students benefited from commenting and building upon each otherÕs ideas in the online discussion. This collaborative learning environment helped to promote learning as students cooperated to fulfil tasks that require team efforts. The results also indicated that students were more responsible and more active in the online course: they initiated topics for discussions and critiqued each otherÕs work on a regular basis. The increased student responsibility and autonomy improved learning skills and reduced the authority of the instructors. According to the instructors, the learning outcomes in both courses were comparable to and, in certain aspects, better than the similar courses taught in the traditional face-to-face mode.

The collaborative approaches used in the two courses seemed to confirm that these approaches can have a significant impact on learning and knowledge building. Instructors were also required to develop and learn new skills to moderate and to facilitate online discussion.

The results of the present study also indicated that the success of the online courses requires careful and systematic instructional design. The instructors should set specific goals for the course and structure the conferences. The results suggested that for a successful online course, the role of the instructor is to act as a facilitator throughout the entire course. Instead of feeling frustrated by the change of the roles, the instructors were extremely positive. The impact of such change was found to go beyond the teaching of the online courses as the instructors proceeded to apply this new teaching style in their place based teaching environment.

References

Berge, Z. L (1996). The role of the online instructor/facilitator. On the Internet @http://www.mindspring.com/~profjer/article/role.htm

Bretz, R. (1983). Media for interactive communication. London: Sage.

Bullen, M. (1997). A case study of participation and critical thinking in a university-level course delivered by computer conferencing. Unpublished doctoral dissertation, University of British Columbia, British Columbia, Canada.

Duin, H & Hansen, C. (1994). Reading and writing on computer networks as social construction and social interaction. In Selfe, C. & Hilligoss, S. (Eds.) Literacy and computers: The complications of teaching and learning with technology, (pp. 89-112). New York: The Modern Language Association.

Harasim, L. (1993). Collaborating in cyberspace: Using computer conferences as a group learning environment. Interactive Learning Environments, 3 (2), 119-130.

Harasim, L, Hiltz, R., Teles, L. & Turoff, M. (1995). Learning networks: A Field Guide to Teaching and Learning Online. Cambridge: MIT Press.

Harasim, L. and Winkelmans, T. (1990). Computer-Mediated Scholarly Communication. Knowledge: Creation, Diffusion, Utilization, V. 11 # 4, June 1990.

Henri, F. (1992). Computer conferencing and content analysis. In Kaye (Ed.), Collaborative learning through computer conferencing (pp. 117-136). London: Springer-Verlag.

Hiltz, S. R. (1997). The virtual classroom. ACM Journal, September 1997.

Kern, R. ( 1995a). Computer-mediated communication: Using e-mail exchanges to explore personal histories in two cultures. Unpublished Manuscript.

Kern, R. (1995b). Restructuring classroom interaction with networked computers: Effects on quantity and characteristics of language production. The Modern Language Journal, 79: 457- 476.

Levin, J., Kim, H., and Riel, M. (1990). Analyzing instructional interactions on electronic message networks. In Harasim L. (Ed.), Online Education, (pp. 185-214). New York: Praeger.

Scardamalia & Bereiter (1994). Computer support for knowledge-building communities. The Journal of the Learning Sciences, 3(3), 265-283

Teles, L. & Rylands, J. (1988). The Infoshare module: using collaborative asynchronous training to improve Web search skills. Journal of Educational Media International, 35(3).

Wang, X., & Hurst, R. (1997). An empirical study of computer-assisted discussion: Effects on social interaction and group dynamics. Paper presented at FLEAT III Conference, Victoria, British Columbia, Canada, August 1997.

Warschauer, M., Turbee, L., & Roberts, B. (1994). Computer learning networks and student empowerment. (Research Note #10) Honolulu, HI: University of Hawai'i, Second Language Teaching & Learning Center.

Acknowledgements

The authors thank Dr. Jane Cowan for her editorial comments.