Educational Technology & Society 3(3) 2000
ISSN 1436-4522

Teacher Collaboration in a Networked Community

Daniel R. Dunlap, Dennis C. Neale, John M. Carroll
Center for Human-Computer Interaction
Department of Computer Science, Virginia Tech
660 McBryde Hall, Blacksburg, VA 24060 USA
Tel: +1 540 231 7542


In this paper, we examine the collaboration problems teachers encountered in the course of instructing students using collaborative computer software to connect distributed classrooms. We describe issues surrounding teacher collaborations arising from three types of sources: Organizational chaos of teaching, physical and temporal dispersion of events and causes, and individualism in teaching practices. We illustrate how our situation presents new challenges and opportunities to better facilitate and motivate teacher collaborations, ones that may become typical in the future of education and thus are important to analyze now. Finally we describe some technological solutions to problems arising from traditional and novel teacher collaborations.

Keywords: Teacher collaboration, Teaching practices, Collaborative technology, Distributed instruction


Auburn Middle School, Friday 12:45 pm: An eighth grade science teacher hustles in and out of the Ethernet-wired log cabin she uses for a classroom in rural Southwest Virginia. She is preparing her sixth period physical science students for a round of computer collaborations with a group of students she has never met or seen. “Folks, go to your group and find the people and click on ‘em.”About 20 miles away in small town fed by the local university, a physics teacher sets the day’s agenda for his sixth period class. “Today is research project day. Those of you who have been working on the computers, you should use the same computers you've been using.” In a matter of minutes, students in these two schools will be collaborating real-time with one another via computer-mediated audio, video, and shared text. These technologically mediated interactions have important implications for students, but what do they mean to teachers?

National Science Education Standards emphasize the need for innovative ways of improving collaboration among teachers, suggesting that teachers develop new forms of collaboration that help them reflect on their practices and curricula (NRC, 1996); however, educators have long recognized a number of traditional problems with teacher collaboration. Cultural norms of individualism, structural conditions in schools, teachers’ differing pedagogical orientations, as well as the absence of common beliefs and expectations or “shared professional identity” all stifle teachers’ abilities to collaborate (Mitchell 1997, Rosenholtz 1989). Organizational structures of schooling, “how school space is arranged; how content and students are organized in grade levels; how time is allotted to tasks; and how organizational rules govern the behavior and performance of both adults and students,” have meant that instructional practices over the last century have remained consistently teacher-centered despite various reform efforts (Cuban 1993, 17). Open classroom reforms in the 1970’s did little to alter the seclusion and independence of classroom instruction. As such, teachers have traditionally had limited impetus to share and collaborate over instruction. Distributed teachers have had even less opportunity to collaborate.

The Learning in Networked Communities (LiNC) project encourages distributed teacher collaboration by developing and deploying innovative collaborative technology in science classrooms. A major outcome of this project is the collaborative software called The Virtual School. While the LiNC project began as a study of collaborative software use by students, it generated a great deal of understanding of and attention to new problems associated with teacher collaborations. The student collaborations facilitated by LiNC obligated teachers to collaborate; however, teacher collaboration has never been a well-established or well-supported practice. In designing and implementing networked-based student collaborative activities, LiNC teachers themselves faced new kinds of collaborative challenges and opportunities. In these novel collaborative arrangements, teachers confronted unique problems relating to awareness and assessment of student activities, communication among teachers, scheduling and planning for different subjects and schools, as well as pedagogical and stylistic differences.

In this paper, we describe some of the novel opportunities and challenges to teacher collaboration that the Virtual School brought to teachers. We illustrate how network technologies can and will open up classrooms and teachers in ways previously unavailable, but in doing so we note how these new prospects involve novel challenges to teacher collaborations likely to become typical in the future. We identify ways that researchers and developers of collaborative classroom software can help teachers address these new challenges in the course of improving their own research methods. We conclude by suggesting ways that researchers and developers of collaborative technologies can recognize and address the new problems and opportunities provided by these new classroom tools and thresholds.


The LiNC Project

The Virtual School

A significant portion of the LiNC project was devoted to the development and evaluation of the Virtual School. This software is a continually evolving Java-based network application designed to allow students to collaborate across classroom boundaries (Koenemann et al. 1999). Students used a variety of communication and collaborative tools integrated into the Virtual School to work and learn in cooperative groups across grade level, subject, and school.

The client side of the Virtual School application consists of a main control window that integrates email, chat, video-conferencing, application sharing, whiteboard, and collaborative writing tools. Figure 1 represents a computer screen shot of a typical collaboration using the Virtual School.


Figure 1. A typical collaboration with the Virtual School involved chat (lower left), video-conferencing (upper right), and the collaborative notebook (lower right). Group members, communication features, collaborative notebooks, and notices are displayed and accessed using the main panel (upper left)


Students logged in to the application with individual user names, passwords, and group identity. Three students to one computer was a typical ratio. Group login allowed all group members who were jointly using a machine to be noted while restricting group access to documents assigned to individual members. Synchronous and asynchronous modalities allowed distributed students to coordinate collaborative group projects.



Four classrooms from Montgomery County Public Schools in southwest Virginia were involved in the project. Two high school physics teachers and two middle school physical science teachers participated in the initial activities. Student activities ranged from student-to-student mentoring that occurred over several days, to long-term group experiments spanning several grading periods. The collaborative exercises were designed to teach science principles and concepts and facilitate the use of computer technology in collaboration.

 Student groups in the four different classrooms who were concurrently logged-in to the application could see a listing of all present group members. By clicking on a student’s name and then a button in the main window, students could initiate synchronous group communication via chat and videoconferencing. Students and teachers also used the integrated email client, but the heart of the Virtual School rested in the collaborative notebook that allowed students to jointly author and edit electronic media across classroom groups.


Researchers and Data

Investigators came from diverse backgrounds, including computer science, cognitive science, and human factors engineering. The first author has a background in educational research and science and technology studies and actually taught prior to the study for several years in one of the very same classrooms and courses being studied. This mix of background and direct experience in the school system provided researchers insight into particular dimensions of the teachers’ work.

We developed and applied our own mixed methods research approach (see Tashakkori & Teddlie 1998) to capture and document the distributed group activity involving the Virtual School. Students in each classroom were videotaped as the collaborative groups worked in front of the computer with their teacher. Each side of the video for a group was transcribed with time-stamp and integrated into a script with the corresponding computer server logs (see Neale & Carroll 1999 for comprehensive coverage of the methodology). Thus, a comprehensive text file documenting the dialog and machine activity of the groups was generated for each period of collaboration. This process provided a detailed view of the activities of the students, but with it, we also captured a great deal of teachers’ activity. In addition, students and teachers were surveyed prior to and following each set of collaborative activities. Students and teachers were also given semi-structured interviews throughout the study.

Investigators and teachers identified critical incidents in the progress and development of the group activity. These incidents were noted and discussed using the Collaborative Critical Incident Tool (CCIT). This tool generated a rich source of data about the teachers’ attitudes, beliefs, problems, and practices (Neale, Dunlap, Isenhour, & Carroll 2000; Carroll, Neale, & Isenhour submitted). Discussion threads were initiated with descriptions of critical incidents drawn from the collaborative activities.

Much of the resulting information from interviews with teachers and continuing dialog from the CCIT was fed back into observations through ongoing contextual inquiry that was conducted during the course of the field observations. In the process of coding and analyzing these observations of student collaborations, understandings about teacher collaborations emerged as a principal concern. As a result, we began to focus attention on how teacher collaboration practices shaped student activities.


Teacher Collaboration Challenges

Teachers located in different classrooms in different schools experience very different environments seen in students, local culture, school administration, and setting. Computer technology allows new forms of collaboration over time and place, bridging differences and breaking down spatial and temporal barriers. Electronic collaboration tools “allow learners control of time and place when and where learning transpires” (Gamas & Nordquist 1997, 17). This control by students forced teachers to reconsider how they monitor, interact, control, and assess student work.

Use of the Virtual School forced LiNC teachers to collaborate with each other and to do so in new ways. In order to coordinate the use of the rich modes of collaboration offered by the Virtual School, teachers created parallel curriculum across school, age, and ability with integrated tasks, schedules, grading, deadlines, and benchmarks. Teachers coordinated synchronous and asynchronous interactions of distributed groups. The interactions typically consisted of student group activities that involved videoconference meetings supplemented with chat and email while editing collaborative notebook sections.

We identified a number of novel opportunities and impediments arising from three sources that teachers negotiated in their dealings with the Virtual School activities and planning: (1) Organizational issues involving scheduling and administration of classes, (2) awareness of student practices and related difficulties arising from physical and temporal dispersion of events and causes, and (3) reliance on individualistic norms and practices including those involving grading, planning, and coordinating collaborative work.


Organizational Chaos

Teachers will likely agree that teaching in public school has always been somewhat chaotic. Slight adjustments on one level can cause major disruptions on another level. Coordinating schedules across schools compounds many of these problems. As a result, teachers encounter significantly new and greater challenges when collaborating with one another. Computer-mediated activities bring new abilities and problems to teachers’ routine work; and therefore, tools like the Virtual School that connect distant classrooms can demand radical shifts in instructional practices. Collaboration required a great deal of face-to-face work, much more than teachers expected. Despite significant planning, teachers in LiNC confronted several organizational challenges resulting from their need to coordinate Virtual School activities. These included conflicts and shifts in school scheduling, administration, network and computer resources, planning, and assessment.

Even schools in the same system can operate on different schedules, and for small school systems, coordinating science class schedules across schools requires a great deal of additional effort and planning. Despite cooperative administration and extensive pre-planning, teachers may be unable to precisely align schedules. In order to take advantage of the synchronous capabilities of the technology, teachers needed to synchronize their work. Major difficulties arose when a teacher was unable to be assigned the class schedule she needed to conduct synchronous collaborative activities.

School system administration is also complex and distributed. There are few central information systems that give personnel the ability to realize and control the implications and effects of administrative decisions. Commitment from school administrators early in the project assured researchers that teachers and classrooms participating in the collaborations would have a common time slot. Over a five-year period, this goal was never fully achieved. Teachers had little control over or knowledge of scheduling prior to it being announced as a final mandate. Administrations routinely kept schedules “secret” prior to their announcement to meet conflicting demands. As a result, teachers were forced to use planning periods, pull students from other classes, and complete collaborative project work after school. In many cases, this was testimony to the resourcefulness of teachers, that is, their ability to negotiate and improvise in the face of less than ideal circumstances.

Network technology adds new opportunities, but it also adds a new dimension to this organizational chaos. For example, a seemingly benign decision was made by school administrators to route the school system’s Ethernet connections through a firewall for practical reasons other than security and filtering. This caused problems for a variety of LiNC and non-LiNC teacher and student activities. For instance, the firewall eliminated the ability to videoconference outside of the school systems network. Unfortunately, this came at a time when the teachers were gearing up for a second year of a large project involving community mentors that would benefit greatly from the ability to videoconference. As a result, much of the mentoring was eliminated. One teacher explains, “I finally had to drop the requirement that projects have an outside mentor because of the problems maintaining a relationship with them.” This significantly stifled the quality and extent of involvement of the mentors with the Virtual School technology, and it also increased chaos in the teachers’ planning as some were compelled to mediate mentor involvement in ways other than video-conferencing.

Teachers were strongly encouraged into collaborating and at the same time confined to their closed classrooms and routines. Although they were each allotted a planning period, their schedules allowed no mutual time to plan and coordinate activities with teachers at other schools. Teachers met after school on their own time. Ironically, teachers felt the need to meet face-to-face to coordinate efforts effectively rather than meet remotely using the Virtual School. In order to meet face-to-face, LiNC teachers had to drive across the county. While they met more frequently as the project progressed, it was still not enough to allow sufficient communication. Circumstances were extremely dynamic. Without the ability to talk daily, informally in hallways and immediately following class, major breakdowns in communication were inevitable.

One striking example of organizational chaos ensuing from teachers’ lack of ability to communicate frequently about the project activities centered on the use of deadlines or due dates for student groups by teachers. A shift in emphasis from short-term to long-term projects resulted in a greater need for common deadlines mainly because of the degree to which work became distributed among students and sub-groups. Subsequent shifts in deadlines became a major source of problems. One teacher explained the situation forthrightly.

This does seem to be the nature of the beast when we teachers are trying to decide when to make things due. I'll admit too much tinkering with dates because "stuff happens." Here's what I mean in a fictitious example. Let's say [another teacher] and I agree on a Monday deadline for the paper. Because of another nightmare in my room, such as a lab we are working on or a test, I "secretly" tell my kids it is ok if they can't get it until Tuesday. For me, this has been the source of vague and frustrating deadlines.

Though the specifics of the example provided may be “fictitious,” the general problem was not. Clearly, deadlines were and are problematic for students, collaboration or not. The word, “secretly” in the above passage suggests that teachers do not have the ability to make each seemingly minor adjustment known to their colleagues. Unfortunately, minor adjustments can lead to major disruptions in tightly coupled collaborations.

The Virtual School afforded teachers the ability to easily connect synchronous and asynchronous work of remote groups. Remote groups in many cases produced a group-authored product in the end, but in doing so, students and teachers needed to coordinate deadlines in order to avoid missing or unfinished pieces of the groups’ work. Teachers wrestled with how to establish deadlines in Virtual School projects given conflicting local versus distributed constraints.

Maybe we need to set a target date which has some built in latitude-perhaps a week in which to turn it in. I did this for my final projects. The kids could turn them in any time during a particular week. Each classroom has its own quirky situations. Add to that building level considerations, and you can see why the four teachers don't have time to consult with each other every time we feel the need to adjust a deadline. Obviously it makes for complications when there are two groups working remotely in collaboration, but I honestly don't see how to get around it.

When synchronous and asynchronous group work is linked as with much Virtual School work, small changes in deadlines can cause serious planning difficulties. Encouraging inter-reliance through the collaborative features of the technology increased the chance of deeper collaborations, but it also increased the chaos of organizing the group work by extending and distributing the network of inter-relations among classrooms.


Physical and Temporal Dispersion

A number of issues related to teachers' awareness of student activities across remote collaborative groups arise as a direct result of the physical and temporal dispersion of events and causes triggered by Virtual School activities. Tracking student groups locally in the classroom is difficult; spreading out groups among different classrooms and periods via computer-mediated communications, and directing student attention toward computer-mediated group work further detaches teachers from usual channels of awareness of group activity.

When student interaction is directed at a computer screen rather than the teacher or co-present peer, teachers have limited view of the interaction. As a result, remote teachers may have very different perceptions of the same computer-mediated interaction. For example in one long-term project, students were having difficulty collaborating. The evaluation data clearly showed the groups were struggling with their interactions. The frequency of chat and email decreased significantly as the activity progressed to the point where little or no attempt to communicate was initiated by either side. Some communications were blatantly hostile, and planning was intentionally non-collaborative in that they made decisions to work on independent products unlike other groups. Work in the collaborative Notebook done by one side was ignored and simply replaced by the other side. Conversations between one side and their teacher revealed the students had very negative experiences with the interaction. However, the other teacher involved had very different information about the group’s work.

While the first side kept the teacher appraised of the interaction, the other side worked much more independently and kept to themselves. They reported their progress and stayed engaged and busy in their work in class. Thus, while the first teacher saw this group as a poor example of collaboration, the second teacher, having little access to many of the interactions between the groups, considered their work exemplary since they required little guidance and reported little difficulty while documenting considerable progress. In fact, by the standards manifest in their classroom, they were learning and getting their work done. This difference in perception was partly due to differences in teaching style discussed in the next section, but there clearly were significant differences in what each teacher knew about the interactions of the remote group members.

Teachers, like most of us, have not had much practice at developing assessment strategies for distributed student work. Traditionally, teachers have little experience assessing qualities of interaction when it is directly visible and audible, and while teachers in LiNC have considerable experience and interest in cooperative and group instruction, computer mediation makes much of the interaction invisible to the teacher. As one teacher put it,

It also took me [longer] to "tune in" to what was being done in Virtual School than in group work because I couldn't listen in selectively before arriving for a visual check. In many cases it became a juggling act to be sure to visit each work group in a class period. Without that attention, some drifted off into fun and games and others explored blind alleys.

Teachers were forced to more actively monitor group work and to provide more individual attention to an overwhelming classroom, but unless they asked the right people the right questions at the right time, misleading perceptions were inevitable. Lack of experience and context limited teachers’ ability to know the right kinds of questions to ask. Students experienced awareness problems themselves because they lacked an understanding of the remote teachers’ expectations and requirements for the group. In facing these awareness problems, teachers were often forced to reflect on their own habits, practices, and “technical culture” (Rosenholtz 1989).



Traditionally, teachers have been accustomed to and reliant on individualistic norms and practices (Huberman, 1993; Kain, 1996; Mitchell, 1997). This suggests that greater autonomy in teaching means less opportunity for collaboration; however, autonomy in teaching practices does not necessarily oppose collaborative goals. In fact, some argue that people can become more autonomous through working effectively and cooperatively with others. This is Abraham Maslow's (1973) notion of self-actualization, but it also resonates with some of Robert Bellah’s (Bellah, et al 1986) views.

The use of collaborative computing confronted teachers in LiNC with a number of particular problems concerning grading, planning, and coordinating collaborative activities of students. Teachers were coupled with other teachers by the collaborations of their students, and their normal individualistic teaching practices often conflicted due to the nature and level of the teacher collaboration demanded of them. In traditional collaborative arrangements, teachers are often co-located in the same classroom or school. As such, they have direct access to a wide variety of common cues and conditions that inform their practices.

Computers allow teachers in diverse settings to collaborate, but compared to face-to-face interaction, they also obscure and disconnect teachers from one another. As mentioned above, LiNC teachers changed deadlines, requirements, and procedures without notifying each other. Teachers prepared students differently. They disagreed on the degree of student autonomy in planning, and they varied in the degree of intervention and types of help provided to the collaborative groups. Despite agreement on a common grading rubric, teachers reverted to individual ways of assessing group projects. While computers joined their work, teachers did not have the luxury of interacting in person on a daily basis as with same school collaborations to which they have been accustomed.

“Isolated settings compel teachers in opposite direction—toward norms of self-reliance. Under these circumstances, requests and offers of assistance seem far less apt to occur” (Rosenholtz 1989, 44). LiNC teachers recognized their isolation as teachers and, to some degree, hoped that the technology could alleviate some of it. As one teacher explained, “I don't get to see [the other Physics teacher] but maybe two workdays out of the year. So normally we would be geographically isolated. [The technology] provides at least an enhanced level of contact between what's going on in the two rooms.”

However, with the enhanced levels of contact and interaction at a distance provided by the collaborative technology came greater opportunities to clash over individuality and idiosyncrasies. Two key examples of such clashes in the LiNC project arose as a result of the tightly knit but distant collaborations engendered by the use of the technology. Because of the physical disconnection of teachers’ work and the demands of joining students with computers, differences in teachers’ styles and assessment goals became extremely important.

Although LiNC teachers shared a great deal of general pedagogical beliefs and attitudes, they varied significantly in how they enacted the curriculum in their particular classrooms. All LiNC teachers valued hands-on and student-guided projects, and they all valued the curricular and “real-world” technical skills encouraged by the projects. Nevertheless, teachers were distinctly aware of key differences in their teaching styles, even among teachers that taught the same subject. Despite their pedagogical similarities, teachers considered these stylistic differences among the greatest obstacles to successful computer-mediated collaboration. Teachers emphasized and valued different types of results and content and attempted to serve different student populations. Teachers related these differences directly to problems with student collaborations.

[Another teacher] teaches a much harder physics class, and I teach a much easier physics class. But those two different approaches can cause difficulty when we try to get our kids to work together on things that are content. And we had that specific problem two years ago. Right about this time two years ago we did something on friction. It was a terrible problem because [the other teacher’s] kids had a different goal than my kids. Not to say that [the other teacher] is wrong and I’m right, but we had different goals for our students; and so we’re trying to get them to get different things out of the experience. And none of us are willing to sacrifice our kids to it all and just say let’s do what they’re doing, and we just won’t worry about it.

This passage illustrates how strongly teachers feel about the pedagogical decisions they make in response to their individual style and circumstance. Two key examples of differences stemming from variations in teaching style involved 1) interaction among students and teachers and 2) collaborative assessment of student groups.

There was significant variation in the manner in which teachers interacted with computer mediated collaborative groups. One teacher was able and determined to closely monitor, guide, and “stay on top of” the students. This teacher normally had a small number of collaborative groups to manage and few other students in the classroom to monitor. In addition, this was a middle school teacher with “an elementary school background.” This meant that her training prepared her to assume a much more active role in monitoring and facilitating student learning than that expected of more mature students. This teacher expressed very strong motivation to work closely with students to facilitate interaction over computers.

It is very important to interact with the students during the time on Virtual School or while working in groups on their project. Students must sense the importance we as teachers feel, so they will in turn feel it is significant enough to work on. They are not only learning the physics involved in the projects, but a sense of responsibility to other people and a responsibility to get a job done. If there is a commitment to a group, then I believe we as teachers should intervene and remind them of their responsibilities.

The remarks of the middle school teacher above about intervention are in clear contrast to those of the high school teacher who said “the kids are basically in control themselves and come to me with questions when they have them. Then again, for me, they're high school students ... seniors even!” One big reason given for this vast difference in teacher practice is the age level of the students involved. In the next passage, the teacher further explains why persistent intervention in computer collaborations can be a bad idea.

As for remote collaboration ... I have done little to intervene. My reason is that I think it ought NOT be necessary. Remote collaboration is enough of a nightmare to coordinate and manage from a planning end. Adding more elements only makes it that much more unlikely to succeed or be implemented.

These views represent opposite poles with respect to views on intervention, and other LiNC teachers fell somewhere in between these poles. Since teachers have little if any spare time, circumstances that require significant time often conflict with their desire to know and control as much as possible about what the students are doing, but style and beliefs contribute significantly to how and when teachers interact with their students. In addition to differences in teacher intervention, assessment was also a point of contention.

A number of educators, including LiNC teachers, appreciated collaborations on grading for the benefits they offer teachers. Kain (1996) describes its glories in one case. “Through their collaboration, staff members turned grading, a process that is usually completely private and mysterious, into a problem solving opportunity; opened up their own practices to analysis, offering the possibility of growth; and moved toward a willingness to reexamine some of the standards of traditional American education in a spirit that offered the promise of productive change.” Similarly, the National Science Education Standards (1996, 67) state “teachers also need to have the opportunities to collaborate with other teachers to evaluate student work—developing, refining, and applying criteria for evaluation.” LiNC teachers also expressed positive sentiment about the value of collaboration over grading. One teacher explains, “I think that grading or an assessment component is part of the planning that has to happen if you're going to have a successful collaboration.”

The Virtual School technology and the LiNC project allowed and even encouraged collaborations across age level and subject. The collaborative products introduced by the technology as well as the project goals supported joint assessment strategies; however, despite their willingness, teachers struggled with collaborative assessment. As one teacher explained, “[The issue] was what is important to each of us.”

Teachers noted that talking about and coming to consensus over grading was a “new experience” for them, but it is not clear that coming to consensus actually helped teachers grade better. While awareness of differences allowed teachers to anticipate and address the deeper issues, negotiating differences was frustrating to teachers. In the end, some contention about the commonality of the results of the collaborative grading remained. Teachers said that there was incongruity in the actual grading used. “[One teacher] actually in the end ended up using a different system. I don’t know why, but he didn’t use the rubric that we specified here.” Another teacher said, “Grading the results was a difficulty because we each granted merit to differing qualities. It was necessary to sit down side by side to evaluate the results. Even then, reaching agreement required give and take on what to count. Unfortunately this consensus did not last as reportedly some of the grades were later altered.” This further illustrates that despite their involvement in thoroughly collaborative activities, teachers maintained their stylistic assessment choices. Thus, it is not clear what advantages closely coupled collaborative grading brought to this group of teachers. This prompts the question, “to what extent can computer supported collaborative learning allow for flexibility and relaxed constraints in collaborative assessment strategies?”


Lessons and Solutions for Teacher Collaboration

Although collaborative computing offers new opportunities, “teachers cannot be expected to reorganize classrooms to provide this type of experience for students while they themselves remain in the isolation of the traditional classroom” (Reil, 1996). Two key features of computer-supported collaboration exacerbated the already chaotic organizational problems of teachers. 1) Collaborations among distant teachers promoted contention related to clashes of individuality and style, and 2) on-line technologies disperse events and their causes in space and time. In this section, we describe a number of steps we took to address these two critical themes. We focus on describing two interrelated areas of our work regarding 1) notification and feedback for participants, and 2) design and planning of learning activities. We show how tools designed to facilitate teacher collaboration can derive from the same technology and even be integrated directly into software designed for student group work (Isenhour, Rosson, & Carroll, submitted); however, as problems encountered in our study illustrate, teachers need to be more connected with the technology. Teachers need to derive information from, monitor the use of, and utilize the special resources that computer network technology makes available. Moreover, planning for remote collaborative work needs to be more flexible and relaxed to allow for instructional differences.

To combat the additional organizational chaos and conflict engendered by collaborative technology, teachers need ways to stabilize and filter information about computer-mediated collaborations. They need quick and easy access to the information that is most salient to their planning. They need notification of salient information and feedback about student progress and problems as well as teaching changes, and they need ways of managing computer group work that allow for a better understanding of the actions of computer-mediated groups and their instructors and a better ability for teachers to make adjustments.


Notification and Feedback

The collaborative Notebook at the heart of the Virtual School grants students novel abilities to work cooperatively on a joint product. The ability to edit work asynchronously across remote sites caused a number of the problems addressed above. The difficulties pointed to a need for users to notify, be notified, and obtain feedback about the work of group members. A number of features of the Virtual School were designed to notify teachers of student activities as well notify students of other group members’ activities; however teachers are not habituated to electronic interactions. Rather, teachers normally monitor, grade, and give feedback on physically tangible objects like paper and face-to-face discussion. As such, teachers required students to turn in printouts of materials to be graded, and teachers did not get into the habit of using the Virtual School to check up on project progress.

Since the collaborative Notebooks utilized HTML and were maintained centrally on a server, it was easy to allow the notebook sections to be viewed on-line with a standard web browser. Teachers often grade at home, and this was an obvious way teachers could follow student progress without requiring the Virtual School software to be loaded. Teachers could “look in” to periodically monitor progress. Similarly, the Virtual School chat feature maintained previous chat dialog, so users could refer back over days or weeks to prior chat conversations. Of course, the utility of such feedback mechanisms is contingent upon an effort to either look up notebook sections on-line or to peruse past strings of chat dialog. Unfortunately, teachers never made use of it.

The Notice Board represents a much more passive feature that was developed to supply feedback for students, but it could also provide teachers with a great deal of information about student work. The Notice Board, shown in Figure 1 above, appears at the bottom of the main window and consists of a list of recent Virtual School actions performed by students or teachers including any editing and annotation of collaborative notebook sections and email sent. While the tool was initially thought to be visible and useful, indications of its actual use by students and teachers were disappointing to investigators. This was puzzling because of the clear signs that both students and teachers needed this information. We were left struggling with ways of increasing the tools utility but at the same time realizing that a larger problem might concern changing the behavior of participants with respect to the ways they accessed and responded to notification.

One possible explanation for the lack of Notice Board use may have been because it did not provide salient information for users. While teachers might be interested in notification upon the completion of various tasks, the Virtual School had no automatic way of distinguishing tasks for users. The way it was implemented, notification was intended to supply groups with information about other group members’ recent activity. Absent and remote members could catch up; however, participants also needed to know which actions required their responses. For example, in a number of cases, remote mentors failed to realize that work had been added and awaited comment. Had they checked the Notice Board, they might have noted that sections had been edited, but unless notified by email, they would not have known that they needed to respond. Issues like these led directly to the development of the planning tool discussed below. In addition, several other features were developed to allow teachers and students to get better feedback from Virtual School activity.

Since the Collaborative Notebooks consisted of communal spaces, contribution to it by group members sometimes became a point of contention. Initially, teachers had no way of knowing which group or group member entered a particular passage in the notebook; however, since the Virtual School server kept track of users, it readily allowed authorship to be tracked. By providing access to this information by means of an option that changed the color of the text according to origin of the author, users were better able to follow the development of collaborative work across remote sites.

Information collected at the server was also useful in another way. Computer events, such as chat, email, and notebook edits, logged at the server gave investigators the ability to reconstruct a script of the distributed group activities. These descriptions allowed investigators to identify critical incidents that were fed into an on-line forum developed specifically for discussion of Virtual School activity. The Collaborative Critical Incident Tool used the same logging procedure and Java technology to provide a discussion forum that was capable of tracking users. Teachers were automatically informed of new postings, and discussion threads were personalized for each user by noting new, unread, and additions to the teacher discussions. Thus, the information from server logs was fed back to teachers.


Design and planning

In addition to notification and feedback, there were two major steps taken in the areas of design and planning to address organizational problems of teaching and the physical and temporal dispersion associated with the Virtual School. The first was the adaptation of prepared materials with structured activities, and the second was the development of a collaborative planning tool.

Participants recognized the great need for tools and activities that facilitate and develop project planning skills of students in their group work. Poor planning and division of labor was a symptom of both proximal and remote group work. Many of these problems related to interdependence in distributed groups. Reliance on remote members encourages deeper collaboration, but it can also lead to the collapse of group work when one side fails to deliver on time. This situation reflects the need to build flexibility and autonomy in the activities of remote collaborations so that one side can continue and progress despite failures of remote members. One way that we addressed this was through the introduction of pre-made learning materials that provided structured steps for learners but also allowed and facilitated collaborative activities.

The most successful example of this involved LEGO Dactaä (Cyber Toys 1997) materials and activities that were adapted to use with the Virtual School. By providing specific content, manipulative materials, procedures, and impetus for collaboration, the collaborative activities became grounded in material interactions that sustained the learning. Students using these activities were more engaged with the group learning during the class period. Students spent a greater amount of class time communicating and collaborating with remote members, and students spent more time manipulating materials. Students also expressed greater feelings of control over the pace and direction of their learning despite the greater degree of external structure to their work. In addition, the pre-developed learning activities provided periodic questions that were answered as a group in the collaborative notebook. These served as periodic sources of feedback for remote members and teachers about the groups’ learning processes, and as such, they provided more directly accessible outcomes for feedback and assessment.

Although the prescribed structure limited the creative flexibility and self-direction afforded to students, students were allowed and encouraged to experiment and develop materials that went beyond those given. On the other hand, the use of pre-developed materials had definite limitations. First, it required considerable preparation work. The materials needed to be obtained, adapted, and loaded in the Virtual School for use, and that required significant amount of scanning images and text for collaborative group use. Second, the greater structure entailed by the LEGO activities meant less open-ended student control of the learning processes and less room for student creativity and direction in the learning activities. Where the technology-mediated group learning freed the teacher from direct instruction, it also required that the student take a more active role in the process. This required external structure, but such structure came at the cost of student initiation and control that was also a highly valued outcome.

In another direct attempt to address student group planning, a collaborative planning tool was developed for the Virtual School. The planning tool supports collaborative scheduling and task decomposition by providing two views of the project task layout (Figure 2). The timeline allows users to view sub-tasks in relation to the primary task, and the outline view allows users to organize the tasks hierarchically. The tool allows teachers to initiate project planning by creating and scheduling primary or superordinate tasks and goals so that students can work with teachers to organize, identify, and delegate subtasks among group members. Since the development of the tool emerged out of real problems that students encountered, it was more recently developed, and as such, it needs more trial. However, several important issues arise from its development and preliminary use.


Figure 2. The collaborative planning tool provides a hierarchical outline view and a Gantt chart view of the organization of the project tasks


Prior to the implementation of the planning tool, students and teachers used makeshift, confusing, and non-collaborative methods of planning their project. The Virtual School notebook provided a way for the group to collaboratively edit notes on planning and scheduling, but the collaborative notebook proved to be a cumbersome and cryptic manner of updating, scheduling, and providing notice of progress for the various parts of the group project. Other Virtual School features like the notice board and annotation also provided poor tactics for allowing groups to keeping track of group planning and progress. Moreover, students often seriously lack the planning and time-management skills required of long-term collaborations. Even with good tools, students need to learn how to develop and execute planning. Good collaborative planning can require extensive iteration and negotiation. Consultation with expert or instructor is also important to good planning processes. As such, a planning tool for students needs to facilitate teacher interaction and input at all phases of the process.

The integration of the planning tool directly into the Virtual School provides a central structure needed to help them identify points of notification and response. It also allows students and teachers the ability to keep track of student progress and activity according to structured plans that could be continually negotiated. In addition, the tool allows for alternatives to traditional lockstep planning that introduced problems with LiNC collaborations. Rather, the tool provides teachers ways of designing phases with checkpoints into distributed group projects. In this way, it relaxes the constraints of distributed group instruction while providing mechanisms for assessing and monitoring progress of project work.

However, long-term planning requires skills and practices that often need to be learned by students and thus modeled by instructors. While the planning tool offers a way to model the planning process, execution of long-term collaborative plans can require considerable development of group skills. Students cannot always be expected to possess habits conducive to using such a collaborative planning tool. In order to take advantage of the features of the planning tool, students need to consistently check, update, and revise the plan. If students fail to consult the tool regularly, the information soon becomes obsolete and even harder to update; however, when students use the tools, teachers also need to use them. Teachers need to develop those skills associated with monitoring and interacting in computer-mediated student group processes. Moreover, teachers may need to change their practices to better utilize such tools for their own purposes. Like other features of the technology, the planning tool draws on a complex set of social skills and practices for its success, and this applies similarly to both students and teachers.



Although collaborative technologies ironically can introduce new barriers to teacher collaboration, we believe that on balance they will help us to address the long-standing problems of teacher collaboration. The great hope for computer mediated teacher collaborations is that they can break down the isolation and lack of coordination in teachers’ work. Clearly new computer tools provide novel functions that allow teachers to work together in entirely new ways; however, while they connect teachers, they do so in ways very different than traditional face-to-face teacher collaborations. These new kinds of remote connections among teachers are likely to become more and more common as computer-networking technology takes on a greater role in schools. As such, the problems experienced by LiNC teachers that we have described will likely become more prevalent and pervasive in teaching if steps are not taken to address them.

For a number of reasons, like the severe time constraints of teaching in public schools, teachers in the LiNC project failed or neglected to use the technology for their own instructional and assessment needs. They made little or no attempt to annotate student work. They did not use video-conferencing or chat except when overseeing student activities. Until much later in the project, they did not use collaborative notebook to provide instructions, and they did not access features like authorship or pay attention to the general development of electronic versions of student work. Learning to use the software and using it regularly takes extra time, a luxury that teachers rarely enjoy.

In order for teachers to better utilize computer collaboration, they need to have available a variety of new mechanisms for notifying, planning, and getting feedback; however, this is not enough to assure that they can anticipate and respond to the contingencies of their new roles. Teachers need ways of developing their practices and curriculum that allow them take advantage of collaborative computing so that the curriculum and technology co-evolve in a productive manner. In addition, developers need to recognize the importance of integrating student and teacher functionality in software tools. Most collaborative learning technology focuses on the learner, but teacher practices must also be considered in the design of technology for schools. The LiNC project and the Virtual School technology reveal ways that the development of collaborative computing can help teachers not only prepare for the networked classrooms of tomorrow but also address collaboration problems of the past.

New computer technologies, new teaching standards, and theories compel teachers to collaborate in entirely new ways. On-line collaborative tools offer innovative ways of facilitating teacher collaborations, but many teaching conditions and practices discourage and complicate these new kinds of collaborative arrangements. Studies such as ours help researchers and developers design collaborative technology for classroom use, but while this technology generates new opportunities and problems, research and development must be situated in the context of traditional and cultural practices that characterize teachers’ work. Collaborative technologies need to be supplemented with efficient and accessible modes of communication, and they need to remain flexible in their design in order to accommodate differences between teachers and classrooms. In this way, networked computer technologies can provide new opportunities for teacher collaborations by addressing new as well as traditional structural and practical obstacles to teacher collaboration.

The LiNC project is partially supported by the National Science Foundation, the Hitachi Foundation, and the Office of Naval Research. Corporate sponsors include Apple Computer, IBM, and Sun Microsystems.



  • Bellah, R., Madsen, R., Sullivan, W., Swindler, A. & Tipton, S. (1986). Habits of the heart: Individualism and commitment in American life, Berkeley, CA: University of California Press.
  • Carroll, J. M., Neale, D. C. & Isenhour, P. L. (submitted). The collaborative critical incident tool: Supporting reflection and evaluation in a web community.
  • Cuban, L. (1993). How Teachers Taught: Constancy and Change in American Classrooms 1890-1980, New York: Longman.
  • Cyber Toys Inc. (1997). LEGOâ CAD: Simple Machines, Autodesk Inc,
  • Dewey, J. (1938). Experience and Education, New York: Collier Books.
  • Gamas, W. & Nordquist, N. (1997). Expanding Learning Opportunities Through On-Line Technology. NASSP Bulletin, 81 (592), 16-22.
  • Huberman, M. (1993). The Model of Independent Artisan in Teachers' Professional Relations. In Little, J. W. & McLaughlin, M. W. (Eds.) Teachers' work: individuals, colleagues, and contexts, New York: Teachers College Press, 11-50.
  • Isenhour, P. L., Rosson, M. B. & Carroll, J. M. (submitted). Supporting asynchronous collaboration and late joining in Java groupware. Interacting with Computers.
  • Kain, D. L. (1996). Looking beneath the surface: Teacher collaboration through the lens of grading practices. Teachers College Record, 97 (Summer), 569-587.
  • Koenemann, J., Carroll, J. M., Shaffer, C. A., Rosson, M. B. & Abrams, M. (1999). Designing collaborative applications for classroom use: The LiNC Project. In Druin, A. (Ed.) The design of children’s technology, San Francisco: Morgan-Kaufmann, 99-123.
  • Laughton, S. (1996). The design and use of Internet-mediated communication applications in education: An ethnographic study, Ph.D. Dissertation, Department of Computer Science, Virginia Tech, Blacksburg, VA.
  • Maslow, A. H. (1998). Toward a psychology of being, Third edition, New York: John Wiley.
  • Mitchell, A. (1997). Teacher Identity: A key to Increased Collaboration. Action in teacher education, 19 (3), 1-14.
  • NRC, National Research Council (1996). National Science Education Standards, Washington, DC: National Academy Press.
  • Neale, D. C. & Carroll, J. M. (1999). Multi-faceted Evaluation for Complex, Distributed Activities. In Roschelle J. (Ed.) CSCL 99 Proceedings, Mahwah, N.J.: Lawrence Erlbaum Associates, 425-433.
  • Neale, D. C., Dunlap, D. R., Isenhour, P. L. and Carroll, J. M. (2000). Collaborative critical incident development. Proceedings of the 40th annual meeting of the Human Factors and Ergonomics Society, Santa Monica, CA: Human Factors and Ergonomics Society.
  • Reil, M. (1996). Cross classroom collaborations: Communication and education. In Koschmann, T. (Ed.) CSCL: Theory and practice, Mahwah, N.J.: Lawrence Erlbaum Associates, 187-207.
  • Rosenholtz, S. J. (1989). Teachers' workplace: the social organization of schools, New York: Longman.
  • Tashakkori, A. & Teddlie, C. (1998). Mixed methodology: Combining qualitative and quantitative approaches, Thousand Oaks, CA: SAGE.