Sign In

Communications of the ACM

Supporting community and building social capital

What Makes Learning Networks Effective?

Learning networks are groups of people who use the Internet and Web to communicate and collaborate in order to build and share knowledge [3]. The most frequent application is online college courses, but they could also occur at the grade school level (see Amy Bruckman's article in this section), as learning communities within professional societies or other organizations, or to support public participation in issue-oriented discussions.

Here, we summarize what is known about making learning networks effective as a means of teaching and learning at the post-secondary level, and then look at three issues in more detail. Whether or not an effective learning community will emerge from the potential created by course-based online connectivity depends upon the extent to which various aspects of interactivity are developed:

  • Interactivity between student and instructor,
  • Among the members of the class as they engage in discussions and collaborative work, and
  • Between the learners and the software.

One aspect of developing effective student-instructor interaction is whether or not the instructor is able to establish swift trust during the first week of the course. Collaborative learning strategies are crucial for online learning to be effective. Appropriate software might include online quizzes with diagnostic feedback, or special structures that can support interaction among large learning communities, such as hundreds of participants involved in developing and discussing issues in a social decision-support system.

Back to Top

Effectiveness in (Asynchronous) Learning Networks

One term increasingly applied to online education that uses computer-mediated communication (CMC) to connect the learners and the instructor(s) via the Internet is asynchronous learning networks, or ALNs (see Email and computer conferencing (also called "threaded" or "structured" group discussion) are examples of asynchronous systems in which sending and receiving are usually separated in time. The "learning network" refers to the community of learners as well as to the communications networks and the Internet that links them. ALNs may employ some synchronous interaction, but most of the interaction is asynchronous.

For two decades, a research team at NJIT has been involved in constructing a specific version of an ALN called the "Virtual Classroom," and studying its use in a wide variety of courses. In addition, a compendium and analysis of the results of all empirical studies of the effectiveness of ALNs is constructed as part of a "WebCenter for ALN Effectiveness Research" (see www.alnresearch. org). Nineteen studies have been identified that compare ALNs to traditional face-to-face courses on the same campus. These studies employ objective measures of student learning (for example, grades) about as frequently as subjective measures (survey responses by students). The evidence is overwhelming that ALNs tend to be as effective or more effective than traditional modes of course delivery at the university level [5]. However, there is considerable variation in the degrees of effectiveness and consistency on all measures. Here, we recommend the three most important factors, all designed to maximize interactivity.

Recommendation 1. Promote instructor-student interaction by establishing swift trust. In our NJIT studies, one of the consistently strongest considerations of students' overall evaluation of an ALN course is whether or not they felt the use of the medium had resulted in better communication with their professor than in traditional courses that meet only once or twice a week. In order for this to occur, faculty must be online frequently, and they must learn how to build a learning community in this new environment. The instructor/facilitator must reconceptualize his or her role as a teacher. The instructor must create a set of situations and reward structures that encourage students to look upon their interactions with their peers as valuable resources for learning rather than to focus on memorizing lecture-type material presented by an instructor.

We have conducted and analyzed 20 interviews with experienced ALN faculty and have identified some of the things that occur in the process of becoming a virtual professor [1]. Our interviews suggest the roles enacted by instructors in traditional settings are also enacted in ALN environments, though each of these roles is transformed. The specific changes in faculty roles related to cognitive, affective, and managerial activities.

The cognitive role, which relates to mental processes of learning, information storage, and thinking, shifts to one of deeper cognitive complexity for virtual professors. As one humanities instructor put it, "I read the question and this allows me to really sit back and think about giving a reasonable response rather than something that's right there." A management professor explained a similar experience with students: "They have to think about the materials, digest it, and internalize it."

The affective role, which relates to influencing the relationships between students and the instructor and the classroom atmosphere, requires them to find new tools to express emotion since nonverbal cues such as smiles and gestures and gaze are missing. Still, they find the relationship with students more intimate. "I have a sense of some of the issues and problems that students face in their work lives. They tell me about it in the Virtual Classroom." The managerial role, which deals with class and course management, requires greater attention to detail, more structure, and additional student monitoring. Faculty also spoke of building community, relationships, and trust. Especially important is the establishment of swift trust during the first week or two of the course.

Swift trust is a concept developed by Meyerson et al. [6] for temporary teams whose existence is formed around a clear purpose and common task with a finite life span. Its elements include a willingness to suspend doubt about whether others who are strangers can be counted on in order to get to work on the group's task, and a positive expectation that the group activity will be beneficial. It is built and maintained by a high level of activity and responsiveness [1]. In order to build swift trust at the beginning of a course, the instructor needs to structure clear contributions for each student to make, help them cope with any technical or task uncertainties, and model and encourage response to each others' contributions. Early encouragement of social communications (and explicit statements of commitment, excitement, and optimism) also strengthen trust. Finally, the members' initial actions as well as their responses to one another are critical to trust development. It is crucial that the faculty member guides this process during the students' first interactions with the class online.

Recommendation 2. Develop collaborative learning activities. Collaborative learning is a process that emphasizes group or cooperative efforts among faculty and students. ALNs can best enrich educational options when they serve as a means to create the feeling of a true class or group of people who are learning together, and to structure and support a carefully planned series of collaborative learning activities that constitute the assignments for the course [5]. At SUNY, for instance, data on thousands of online students show that students who report higher levels of interaction with classmates in online classes have higher perceived learning [2].

An example of collaborative learning is the online seminar where the students become the teachers. Individuals or small groups of students are responsible for selecting topics; reading material not assigned to the rest of the class; preparing written summaries of the most important ideas in the material, including links to resources on the Web; and leading a discussion on the topic.

Another example is the collaborative exam where students are required to identify key concepts or skills in each module of the course, to construct exam questions, and to answer each other's questions. Students are thus made partners in deciding what course material is important to know. Other examples of collaborative learning strategies suited to the ALN environment include debates, group projects, case study discussions, simulation and role-playing exercises, sharing of solutions to different homework problems, and collaborative composition of essays, stories, or research plans.

In organizing collaborative activities, attention must be paid to the size of the interacting group. Some assignments, such as seminar presentations, collaborative building of exam questions, and debates, may involve an entire class, at least up to 60 students or so. For others, the class must be broken down into smaller groups. Group sizes of about three to six seem to work best for group projects.

Recommendation 3. Generate active participation with appropriate software. Among the types of software beneficial for creating active participation and supporting interactivity are quiz routines designed for self-testing and mastery learning. These routines would include diagnostic feedback and multimedia simulations, such as an online laboratory for organic chemistry or an electronics circuits course. Such software supports interaction between the student and a program, rather than among students and faculty. For example, students at Michigan State [7] who took a class using ALN in combination with a computer-assisted quiz program (with immediate diagnostic feedback) received midterm grades 20% higher compared to students in traditional classes. In addition, streaming video or audio plus PowerPoint presentations or hypertext materials posted on Web pages are often used to deliver lecture-type materials. These can be useful additions to the core activities in effective online courses that focus on supporting communication.

Appropriate software is also needed to support the communication within the learning community. At a minimum, one needs a computer conferencing system that provides a self-organizing structure via separate discussions for each topic or activity in a course and within each discussion organizes the dialogue into threads with roots and replies. There are many other desirable features, and many of them are not included in any systems commercially available[8]. In particular, current CMC systems are inadequate to support months-long communication in medium- to large-sized groups, such as might occur in a course for 300 students, or in a public debate about policy alternatives. What is needed for such large-scale social decision-support processes [9] is a set of templates that systematically solicit options and alternatives, build relationship diagrams, and solicit quantitative (voting) types of feedback as well as qualitative (text) reactions.

Back to Top


There are a million or more students today enrolled in courses involving online learning. Faculties are learning how to be effective virtual professors in this environment, students are learning how to work with their peers to build and share knowledge, and the software available to support such processes is improving. Challenges remain, however, particularly in terms of software to adequately support relatively large learning communities.

Back to Top


1. Coppola, N., Hiltz, S.R., and Rotter, N. Building trust in virtual teams. In Proceedings of IEEE Professional Communication Society 2001 International Professional Communication Conference. (Santa Fe, NM, Oct. 2001). IEEE Computer Society, Washington, D.C., 353366.

2. Fredericksen, E., Pickett, A., Shea, P. and Pelz, W. Student satisfaction and perceived learning with on-line courses: Principles and examples from the SUNY Learning Network. JALN, 4, 2 (Sept. 2000);

3. Harasim, L., Hiltz, S.R., Teles, L., and Turoff, M. Learning Networks: A Field Guide to Teaching and Learning Online. MIT Press, Cambridge, MA, 1995.

4. Hiltz, S.R., Coppola, N., Rotter, N., Turoff, M., and Fich, R.B. Measuring the importance of collaborative learning for the effectiveness of ALN: A multi-measure, multi-method approach. JALN 4, 2 (Sept. 2000);

5. Hiltz, S.R., Zhang, Y., and Turoff, M. Studies of effectiveness of learning networks. Presented at the Sloan ALN Workshop (Lake George, NY, Sept. 2001); to appear in JALN.

6. Meyerson, D., Weick, K.E., and Kramer, R.M. Swift trust and temporary systems. R.M. Kramer and T.R. Tyler, Eds. Trust in Organizations. Sage, Thousand Oaks, CA, Sage, 1996, 166-195.

7. Thoennessen, M., Kashy, E., Tsai, Y. and Davis, N.E. Impacts of asynchronous learning networks in large lecture classes. Group Decision and Negotiation 8 (1999), 371384.

8. Turoff, M. and Hiltz, S.R. Software design and the future of the virtual classroom. J. Information Technology for Teacher Education 4, 2 (1995), 197-215.

9. Turoff, M. Hiltz, S.R., Cho, H.K., Li, Z., and Wang, Y. Social decision support systems (SDSS). In Proceedings of the 35th Hawaii International Conference on Systems Sciences. IEEE Computer Society, Washingto, D.C. (CD ROM).

Back to Top


Starr Roxanne Hiltz ( is Distinguished Professor of Information Systems at New Jersey Institute of Technology, Newark, NJ.

Murray Turoff ( is Distinguished Professor of Information Systems at New Jersey Institute of Technology, Newark, NJ.

Back to Top


Sponsors include the Annenberg/CPB project, the Alfred P. Sloan Foundation, the National Science Foundation (NSF-IRI-9015236), the Center for Multi-Media Research at NJIT, the Center for Pervasive Information Technology, IBM, and UPS. The opinions and findings reported here are those of the authors and not necessarily of the sponsors.

2002 ACM0002-0782/04/0100$5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright2002 ACM, Inc.


No entries found