Education Development Center, Inc.
Center for Children and Technology

Technology's Role in Restructuring for Collaborative Learning

CTE Technical Report Issue No. 8
November 1990

Prepared by:
Denis Newman
Bolt beranek and Newman Inc.

The problem of supporting collaborative learning is placed in the framework of the organizational restructuring of schools. This paper contrasts the organizational impact of two technology systems in terms of: (a) the physical location in the school, (b) the curriculum, and (c) how time is scheduled. Considered first is a class of computer systems for which the function is individualization. In contrast, an environment called Earth Lab is described. Examples of its use in one school illustrate the way that collaborative learning is supported in the three categories: location, curriculum, and time. In conclusion, the complex relationship between school restructuring and the implementation of technology for schools is addressed.

Collaborative learning involves approaches to instruction that are not typical of American schools. Fundamentally, if students are going to collaborate, there must be a task for which
work groups in schools occurs at the level of the schoolwide organization. The technologies of central interest are local area networks (LANs) within the school. The group members do not normally communicate with each other via this network. Instead, they use the LAN to access technology resources, to communicate with the teacher and other groups, and to store and access group products. The classroom work group is thus very different from a group that conducts its work the teacher or workbook or computer is not providing step-by-step instructions. The group must have some autonomy, otherwise there will be no occasion for discussion, group problem solving, question asking, or other processes that account for the benefits of collaborative work in schools. Such open-ended tasks may require changes in the way instruction in a school is structured. The duration of the task may extend beyond the single classroom lesson, since once students begin working with some autonomy, the project may involve new goals that are discovered in the process. Teacher relationships, including distribution of expertise and collaboration among the teaching staff, may change as student projects begin to cross over the compartmentalized curriculum structure. Evaluation of students may also have to move from the typical short-answer tests of individuals to assessments of the group performance of the project itself.

This paper addresses how to get collaborative learning to occur in schools and how computers might support itor at least not inhibit it. The central theme in the paper is that computer support for such primarily via computer with each member in a different physical location.

Software design issues are often considered at the level of the student-computer interaction without concern for issues such as the social organization of the classroom or school. When considering the use of local area networks, we must also consider these organizational issues in technology design. I will begin this paper by placing the problem of collaborative learning in the framework of the organizational restructuring of schools. I then turn to a description of a class of computer systems which are currently popular in the U.S. and which are not designed for collaborative learning, in fact quite the oppositetheir function is individualization. I will then describe, in contrast, some work my colleagues and I have been doing on an environment called Earth Lab. Based on observations of Earth Lab, I will summarize the organizational impact of the system in relation to (a) the physical location in the school, (b) the curriculum, and (c) how time is scheduled. In conclusion, I address the complex relationship between school restructuring and the implementation of technology for schools.

Why Consider School Restructuring

This paper takes a step back from the small group of children working around a table or a computer on some shared learning task in order to enlarge the usual time and space framework that is used in studies of collaboration. Experiments on cooperative learning groups, for example, typically examine learning processes once the group has gotten to work or learning outcomes as soon as the group finishes its work. When we turn to how collaborative learning might be supported by computers, a timeframe much longer than the group's task must be used in order to consider the teacher's work in setting up the task and appropriating the results into future lessons. Teachers are critically important for the support and guidance they provide and for creating the context in which groups of students can work on their own. The support provided by computer technology can be considered in the complex web that includes the classroom interactions as well as interactions around the school.

School restructuring is currently a focus of many school reform efforts in the U.S. that recognize that improving education requires changes in governance, modes of teacher-student interaction, incentives, and methods of evaluation (17, 2). Technology's role in restructuring is very much a matter of debate. While it has been common for technologists to paint scenarios in which schools are transformed by technology, other analysts of school change note that technology has not penetrated beyond the margins of the school system (3, 5). A common explanation for marginal impact is the incompatibility of the teaching styles that are called for by the new technologies and the styles that are at the core of current teaching practices. Collins (4) argues, on the other hand, that the rapidly growing role of technology in the workplace will drive the adoption of technology in schools and spur collateral changes, including a shift from lecture to coaching and from a competitive to a cooperative social structure. These discussions of technology and restructuring often assume an incompatibility of technology and current practices which, as argued in this paper, is not warranted for all computer technologies. But in relation to reform attempts, the debate is raising the critical issues concerning the impediments to change and the possible mechanisms by which technology may have an impact on school structure.

The support of collaboration must be considered both at the level of the collaborative group and the level of the school organization. A combined interest in cognitive change in students and in school restructuring suggests that we look both at the way teachers might collaborate with support from the school administration and at the way students might collaborate with support from their teachers. At both levels the support system is central: Students will not just start collaborating without appropriate tasks, accountability assumptions, and timeframes organized by the teacher. Teachers cannot provide collaborative tasks within the structure of the typical school, which demands that subjects be covered in a particular place and time. I will be focusing on the schoolwide level of the organization in this paper. But it should be clear that the schoolwide structure has direct effects on the students' learning.

At the heart of this argument is the notion that a decompartmentalization of the school's curriculum can be beneficial for student learning. Students will benefit from seeing the connections among topics, such as between math and science or science and writing or writing and geography, and so on. Projects that groups of students undertake can be made more authentic and perhaps more motivating if related to real-world concerns where the disciplinary boundaries do not necessarily hold. Students can also become more motivated if their school work is to a greater extent under their own control rather than tightly controlled by the school schedule. Computer technology can be arranged in the school to help make the cross-disciplinary connections and to help give students flexible access to these connections.

A theory of cognitive change that takes into account the social construction of knowledge may be a useful guide in exploring how learning takes place in the complex organization of a school. Social context can be understood as a tool that can be put in place by the teacher to assist learning, rather than as a background to individual learning processes (12, 9). The critical learning events, in this view, may be the points at which the teacher brings the small groups together and appropriates their contributions into a larger classwide project, or events where the teacher supports the students in combining information from a variety of sources. Also, learning may take place over an extended period of time as students slowly acquire skills with various tools, such as computers, and change their way of thinking across several projects. In this view, the processes of cognitive change occur as part of these events that may be organized at the cross-classroom level well beyond even the group of students who form the collaborative work group. In the next two sections, I will contrast two very different technology systems and attempt to trace the ways in which their impact on the school organization directly conditions the social construction of learning events in the school.

Integrated Learning Systems

Computers in schools have as much potential for reducing collaborative learning as for promoting it. I want to start with an example of a category of school computer systems that discourage collaborative work in order to make the point that there is no sense in which computer technology naturally promotes collaboration. Features of these systems, called integrated learning systems (ILS) or more recently, integrated instructional systems (IIS), can be used to structure a contrast with a different kind of approach, one that promotes the practice of collaborative learning.

James A. Mecklenburger, director of the Institute for the Transfer of Technology to Education, set out the goals of integrated instructional systems in his forward to a report by the EPE Institute on these systems (16):

The IIS idea, which first emerged as a major national project known as PLATO in the 1960s, is this: educators should apply large quantities of computer power cost-effectively to effect large-scale (schoolwide, statewide, even nationwide) improvement of teaching and learning. The dream of IIS developers and usersreflected in this report by both the enthusiasm of current users and by their frustration over the shortcomings of today's systemshas been that computers could:
The report by the EPE Institute, an independent consumer information organization, provides a detailed catalogue of the major integrated learning systems currently on the U.S. market. These systems typically consist of networked personal computers with a fileserver that both delivers programs to the individual workstations and keeps records on the progress of individual students. The IIS software (or "courseware") covers all or most of the school curriculumthus the term "integrated." Systems are sold to school districts at costs ranging from $60,000 to $180,000 for a "lab" of 20 to 30 stations. These systems are growing in popularity according to the EPE report, which estimates that sales doubled in 1989. The vendors' sales reps, often recruited from among retired school district superintendents, are expert at selling to the highest level of the district administration (14). The arguments of the sales force usually focus on achievement gains, especially on standardized tests of basic skills for which the superintendents must answer to the school board and parent community, and which find their way into state and national reports of school achievement. These arguments are made in spite of the doubts expressed by reputable researchers about the validity of the IIS claims for achievement gains (7). Often state and federal funds that are targeted for disadvantaged students can be used directly for purchasing these systems because of the strong relation between the students identified for these special programs and low test scores.

The EPE report documents a generally high level of satisfaction with IIS systems among the teacher and student end users. The high level of student satisfaction may have resulted from the interactivity in comparison to workbooks and paper-and-pencil drills. And from the teachers' point of view, the automatic individualization and record keeping of student progress helps them get through those classroom tasks. The report notes that in most of the sites where observations and interviews were conducted, the IIS system was the first exposure teachers and students had had to computers. The EPE researchers also detected a consistent undercurrent of dissatisfaction. They note that "teachers want more software options, and (along with students) they want IIS lessons to be more varied and less repetitious in instructional style (or, as students put it 'less boring')" (p. 295). There was also some concern with the lack of openness of the systems, which generally do not accommodate software from other vendors. Overwhelmingly, teachers say they would like to have the IIS stations in their own classrooms rather than in a lab so the work can be better integrated with their curriculum.

In general, however, the IISs can be considered a successful implementation of computers: one that serves a perceived need and fits in well with the practices of the school. In fact, their fit with the existing school structure provides a framework for looking at the ways the computer support for collaborative learning may require restructuring of the school. We can examine three prominent features of IIS practice: location, curriculum, and the timeframe of the IIS tasks.


The EPE report notes that it is an all but universal practice to put the IIS in a lab. There is nothing inherent in the technology or even the pedagogy that requires this configuration. The local area network cabling can extend throughout the school. But IISs are sold as "labs" in units of about 30, which is sufficient to simultaneously accommodate a whole class brought into the lab. Having the IIS centralized in a lab means that the system can be more easily managed by a school computer lab teacher, or nonteaching paraprofessional, who can retain the school's technical expertise. For a school starting with a low level of expertise, the centralization reduces the training costs that would otherwise be necessary. The lab arrangement also makes it possible for regular classroom teachers to leave the class with the computer teacher for the lab period and take a preparation period or spend the period with half the class not accommodated by a smaller lab. In most cases, however, teachers accompany their students to the lab and attempt to connect their classroom lessons with the material learned from the IIS, according to the EPE report.

The 30 computers in a lab scenario assumes that students will be working at the computers individually since it provides a 1 to 1 ratio. In many IISs, students also wear headsets as part of their computer interactions, so are further cut off from peer interaction. The teacher's role is to help individuals who are having difficulty with some aspect of the computer-presented task. While it is argued that the one-to-one instruction that is afforded by having students occupied with computer tutorials offers teachers an opportunity for individualized tutorial dialogue with students who are often ignored in whole class teaching (15), others have pointed out that teachers can experience this mode of classroom work as deskilling, since the overall direction of instruction and the tasks presented are controlled by the computers (13) .

Individualization of instruction is the fundamental premise of these systems, so there is certainly no attempt to promote collaboration. As the use of these systems evolves over time, schools may begin distributing workstations among classrooms rather than centralizing them in a lab, in which case some of the prominent features of IIS practice may change (7).


Another feature that provides a good fit to the organizational structure of the typical school is the content organization of the courseware. It is straightforward to categorize the courseware into the standard school subjects: math, reading, language arts, or science. This is no different from typical textbooks, which are designed to cover the material in a particular recognized subject area. In fact, IISs are often tied explicitly to the major textbooks. The IISs reviewed by EPE were all rated on "breadth/scope/coverage" in the major curriculum areas. While not all IISs got "A" ratings in all areas, the application of the rating scheme speaks to the intent of these systems. Also, like the typical textbook, the content is, for the most part, presented as facts or procedures to be mastered in sequence. While several IISs provide tools such as word processors and built-in calculators, the presentation of the courseware is predominantly designed to match the sequence of topics in the textbooks.

The content and its sequencing is an integral part of the management and evaluation functions. Discrete tasks that result in a single correct answer can be evaluated by the system itself. More open-ended tasks requiring, for example, formulation of the problem or research into sources outside the computer, or any kind of free-form response, cannot be handled by the system. The system cannot provide the teacher with interactions that are needed to support open-ended tasks, nor can it evaluate the responses.

The management function that tracks students through the sequence of topics depends on individualization. Given the heavy use of automatic tracking of individual student progress, there is a strong disincentive for group work. While a system could be designed to track groups of students (e.g., simply by providing feedback and evaluation to the group based on its responses as though it were an individual), none of the IISs have that capability. Collaborative learning is thus hindered both by the individualized tracking system and by the fact that the tasks represent small steps in a learning hierarchy which do not call for the kinds of open-ended problem solving for which collaborative groups are well suited.

The division of topics into clearly defined subject areas also eliminates the need for teachers handling different subjects to collaborate or for a teacher in a self-contained classroom to consider the integration of learning across subjects. In this respect, too, IISs strongly support the structure of most schools where teachers are not expected to know much about what other teachers are doing. A major advantage of lISs, in comparison to other approaches to school technology, is that teachers and students can get started using the system with little technical or subject matter training and without adopting new styles of teaching or interacting with other teachers.


A common complaint of IIS users, according to the EPE report, is that many systems will not pick up exactly where the student left off in the previous session. Tasks are intended to be completed within the timeframe of a single period in the computer lab (usually about 45 minutes). Single tasks, such as an arithmetic problem, can usually be done in a matter of a minutes and so many can be completed in a period, but nevertheless, these short tasks may still overlap the end of a period and need to be picked up later. IISs are designed on the assumption that each period will be self-contained. The tasks do not require preparation prior to the computer lab period and call for a minimum of technical capability on the part of the student. This design allows it to fit nicely into the usual school structure that is divided into periods devoted to discrete topics and, in the upper grades, taught by different teachers. Teachers can thus schedule the use of the computer lab into the preexisting slots in the day.

The short tasks of the IIS are very similar to the kinds of tasks found on the standardized tests used extensively for evaluation. The short, carefully constrained answer slots are ideal for automatic scoring in both instances. And in both cases, it is not expected that students will collaborate on producing answers. The timeframe assumed by IISs is thus well suited to both the individualized instructional practice and the structure of the school day schedule, which is broken down into discrete periods. Collaborative tasks may create problems both for the usual evaluation procedures, as noted above, and for the fundamental structure of the school day as compartmentalized into separate periods.

The Earth Lab Project

For the last four years, the Earth Lab project has been designing, implementing, and observing the effects of a local area network system intended to facilitate collaborative work in elementary school earth science. Our plan was to create a prototype LAN system and demonstrate it in a New York City public school using an earth science curriculum. The pedagogical rationale was that students should use technology the way real scientists do: to communicate and share data; that is, to collaborate. The school is a public elementary school (grades 3 to 6) located in central Harlem, New York City. The school population of approximately 700 students is predominantly black with a minority of Hispanic and other groups. The school's achievement scores are about average for New York City but lower than the national averages. With a few exceptions, the staff took a traditional approach to teaching through whole-class lessons, textbook reading, and worksheet drill. Under normal circumstances, the school would have been a likely customer for an integrated instructional system. However, the school's computer teacher, who had a different vision, was able to play a leadership role and make use of the technology provided by the project.

A year-long formative experiment began in the fall of 1986. In the initial setup, a LAN connected the 25 Apple lIe computers in the school to a hard drive, which allowed for central storage of data, text, and programs. The Bank Street Writer word processing program was enhanced with an electronic mail system (8). The Bank Street Filer was another basic tool which made it possible for students to create databases that could be accessed from any computer in the school. Along with the technology, we introduced a year-long earth science curriculum designed in collaboration with the teachers (1). The formative experiment took as its goal an increase in the frequency of collaborative work among students. At least for the one year in which systematic research was funded, we were prepared to modify the design of the technology, introduce new software, develop curriculum materials, and conduct staff development workshops as needed (10). After the first year, the school obtained an additional 20 Apple IIGS computers through an award from Apple Computer Inc., and over the last few years has added several other computers, including five Macintosh computers. Several other application programs are in use on the network, including "hypermedia" systems, LogoWriter, telecommunication programs, and Macintosh programs, including desktop publishing tools.

Databases are used extensively both within and outside the the earth science curriculum. During the lunch hour, students can be found inventing databases of their favorite action figures. In social studies, students research almanacs and other sources to fill in databases about countries of the world and figures from black history. In earth science, they examine databases of dinosaur fossils and earthquakes and create databases of the weather readings and indicators of seasonal change that small groups of students collect over a period of several months.

The primary means for supporting collaborative groups is the Earth Lab's network interface, which makes it easy for individuals or groups to store and retrieve data pertaining to their projects. The work of the projectin the form of text, database, graphics, and code filesis stored in workspaces, which are folders or directories on the network fileserver. These workspaces, available to any computer on the school LAN, give groups a location for their work together. Students and teachers can be assigned to any number of workspaces. For example, workspaces are set up for pairs of students to work on writing assignments together. Other workspaces serve schoolwide clubs or other projects. Each individual also has a personal workspace. In the first year of the project, the science teacher, who had the students for two periods a week, had the class form into groups of three or four for the purpose of conducting investigations in the science lab. The science groups gave themselves names that were used for group workspaces on the network. Students shared different data with different students or groups in the school; for instance, a science group, a noon hour club, and the whole class. The current Earth Lab network system is designed to present the same information when students are on either Macintosh or Apple II computers.

When our project began, our explicit goal was to create a classroom environment in which students used technology the way scientists didfor collaborative work. Our analysis of what actually happened led us to a broader conception of how the local area network technology can function. While direct support of collaborative work groups is still important, we have increasingly become interested in the decompartmentalization of the school that can result from this kind of use of a local area network. Teachers are better able to collaborate, students are better able to carry their work from one context to another, and the computer lab is increasingly used in a heterogeneous manner with several projects or groups from different classes working simultaneously. This restructuring supports both individual and group work and contributes to a sense of community in the school. The following examples taken from our observations at the school illustrate these changes within the same categories used to describe the integrated instructional systems.


From an initial 25 computers, the school's network has grown to about 50 in two separate labs: a satellite lab in a small room off one of the classrooms and network connections in several other classrooms into which computers can be moved as needed. When teachers bring their class to the lab, they stay with the class rather than hand it over to the computer teacher. The computer teacher works with the teacher to develop activities that can be continued back in class. The way the project workspaces were set up for groups and individuals helped to develop a sense of continuity not possible with IISs. The following stories illustrate some of the ways this worked.

We expected that projects would be started while the class was in the computer lab and would be continued in the classroom. We found that students were taking this flexibility another step. For example, we observed two girls working on a book report at a computer in a small room off their classroom. They had not finished when the teacher announced that it was time for the class to go upstairs to the computer lab. Instead of dropping their work, they brought their notes with them and asked if they could continue their work at a lab computer while the rest of the class worked on other assignments. The students logged on and called up the file on which they had been working. The network makes the boundaries between classrooms and class periods more permeable. This permeability was used by the students to pursue their tasks on their own initiative.

Several students from different classes and different grades were editors for the school newspaper. The newspaper had a workspace on the network that students used for storing articles and other material for the newspaper. Beyond the editorial group, many students around the school contributed articles to the newspaper by sending them as messages through the electronic mail system to the editors. The common workspace made it easy for the editorial group to work at different times and places on the newspaper. The ease with which any student could contribute to the newspaper and the identity of the group task that was supported by the workspace widened participation. Students became familiar with the network's function as a data organizer so that when other school projects, such as editing a video newscast, were started, students thought it quite sensible to create a workspace for their scripts, plans, and edit lists.

Although the computer labs, housed in adjacent rooms, have enough computers to accommodate a class with a 1 to 1 ratio, we seldom see the computers used that way. Usually students work in pairs or small groups. Often a small group will use more than one computer simultaneously. Since there are spare computers, students from other classrooms that do not have their own computers and teachers on their prep periods also come to the lab to work on various projects. As a result, activities in the labs are very heterogeneous.

Students frequently work in groups and often work with more than one computer. For example, a group of students was using the Bank Street Writer to compose a letter to students in Australia with whom they had been telecommunicating. One of their members suggested that they include some of the data from their math project in the report. A second student turned to an unused computer at the next desk. She called up the Bank Street Filer, compiled the report and "printed" it to the group workspace on the network, converting it into a word processing file where it was easily merged into the letter they were preparing. The completed letter was mailed on the LAN to the person responsible for portaging it to Australia. lt was possible for the students to create a "multitasking system" out of the two computers because they knew, in terms of the workspace, where one computer had to save the data in order for the other computer to find it. For the students who were accustomed to sharing data on the network, it was quite obvious how to do it.

The fact that the "tool" applications (as opposed to games or drills, i.e., content specific programs) were used heavily made group and individual projects the appropriate mode of computer use. The Earth Lab interface, which displays for students and teachers lists of their project workspaces, enables them to work on any of their projects whenever they have the opportunity and inclination at any of the networked machines available to them. With this greater continuity over space and time, students can take greater initiative in following through with work on a project.


The earth science curriculum developed for the initial field test, and the curriculum materials that the teachers have continued to develop over subsequent years, have been interdisciplinary. As they worked on weather and seasonal change, students made connections to physics, math, writing, and social studies. The network system made classroom projects easier to manage and promoted collaboration among the teachers.

Over the initial year of observations, there was substantial movement from whole-class teaching toward more collaborative work in small groups. We found that the science groups, which had been formed to work together in the science lab, were being used by the classroom teachers for a variety of social studies research activities, some of which were unrelated to the earth science curriculum. The network system produced these changes in an unexpected way: It made it possible for all teachers to assign classroom work to the groups created by the science teacher. The science group workspaces were a convenient means for organizing small-group projects in other curriculum areas. The science groups became a resource across the school. There had never been a mechanism by which a social organizational structure created by one teacher in this school could be used by other teachers as a resource for managing instruction (11).

At the beginning of the field-test year, some teachers in this essentially traditional school had doubts about the students' capabilities for handling the autonomy involved in small-group work. Having the small-group workspaces on the network helped communicate to the teacher community that students were expected to do collaborative work. Where interdisciplinary projects become a more common feature of the curriculum, the workspaces can give the students a clearer group identity or sense of project continuity and thus help in the classroom management. Instead of greater centralized control of individualized instruction, as is common in integrated instructional systems, control can be distributed to the students. We suspect that the solution to the teachers' difficulties in managing instruction involving collaboration among the students is to provide students with tools with which they can assume some of the burden, rather than to provide teachers with tools with which to gain greater control.

But students are not simply going off on their own projects unconnected to any curriculum goals of the teachers. The network also makes it easier for the teachers to appropriate the output of the small groups into whole-class activities so as to go back and forth between individual or group work and integrative projects that combine the work of the small groups. The LAN technology seemed naturally to invite coordination in which students contributed to some larger quest for knowledge, since it was easier to give common access to the same shared database than to maintain separate copies for each individual or group (11). An example is provided by our weather data collection project. The data had been collected by a rotating group of students throughout the school year from a small rooftop weather station and entered into a database in a shared weather folder. The database was later used to discover correlations between such variables as pressure and cloudiness. The whole dataset became the object of group discussion of relationships that cannot be found in the individual contributions. This coordination around a shared database was a new activity structure that emerged with the LAN technology.

The Earth Lab network made no attempt to provide a technological solution to the problem of assessing student progress or grading student projects, which is the central function of integrated instructional systems afforded by the hierarchical nature of the courseware. We have, however, begun exploring the use of group and individual workspaces as portfolios of student work. The notion of a portfolio is receiving growing attention among U.S. educators as an alternative means of assessment in which the stages of work on a project collected in a portfolio can provide insight to both the teacher and the student about the state of their work and, in retrospect, about the process of learning (6). The workspaces currently serve as archives of the group or individual project work and so can serve the function of a portfolio in this sense.


The project workspaces provide continuity over time as well as location. Projects involving collecting weather data and data on seasonal change extended over many months. In some cases, projects may extend over years as new cohorts of students move through the school curriculum. The continuity over time that is developing in the school may have an important impact on what students are able to do as they gain technical skills with the computer tools available for their project work. The following examples suggest the nature of that impact.

One science group was analyzing the weather data using the Bank Street Filer database manager. They were trying to support a theory suggested by their impression that the last winter had been much milder than the previous one. They were comparing their data with those collected by the previous year's sixth grade class. When their theory was not supported by averages in the report generated with the filer program, one of the students checked the data entered by their classmates. Several temperatures seemed unrealistic; for example, several January days with highs of zero degrees. Suspecting an error attributable to missing values, they sent electronic messages to representatives of several other classes who had kept similar records and obtained actual data for the days in question.

Many students used the system extensively for their own work in addition to the work assigned as part of instruction. Much of the student-initiated work occurred during lunch and after school when the computer lab remained open. Students were able to pursue their own writing, data-collection, or programming projects. While some educational games were also available during these extra periods, many students chose to pursue projects rather than to play games. Students developed a sense of ownership of their workspaces to a greater extent than we anticipated. Some students accumulated hundreds of files in their personal workspaces over the course of a year. Earth Lab was also used for student-initiated collaborative work; for example, on science fair projects, which were a source of great pride for many groups of students. Their appropriation of the technology also was evident in end-of-the-year interviews. For example, one student suggested enabling students to determine who could have access to a workspace because she did not have easy access to work she was doing with another student. Her suggestion was based on a need that arose in her own attempts at collaboration. Taking her suggestion one step further, our more recent implementation has made it easy for students to create their own workspaces as well as to determine who would be able to use the files in the workspace.

The use of tool software requires a greater initial investment in order to bring students up to speed with the technology than is required for integrated instructional systems, which present small tasks and simple interactions with the technology. However, the availability of the Earth Lab system to students over a period of years and the consistency of the available tools has made it increasingly easy for teachers to introduce long-term projects as part of their curriculum. In the first year of operation, the sixth grade class spent several months on fairly simple introductory projects designed to familiarize them with the word processing, database, and communication tools. Several years later, teachers were able to start immediately with substantial projects. Although students enter sixth grade with widely varying levels of expertise, due to the uneven use of the technology among the fifth grade teachers, enough students have the necessary expertise to support the start up of projects. In one case, the class began early in the year to collect data on the length of their shadows. Students who were familiar with the Filer entered all the data into a database, which was then available for all the students to explore. So students who were unfamiliar with the tool were introduced to it in the context of substantial data that they had helped to collect.

The school in which Earth Lab has been operating for four years is now engaged in a major restructuring involving the creation of a school-within-the-school that will focus on an experiment in teacher collaboration. Approximately one fifth of the students have voluntarily signed up for a "Computer Mini-School," which will involve six classrooms spanning grades 3 through 6. The classrooms, four of which bridge between grades, will be heterogeneous with respect to achievement levels. Each year, new students will be recruited for the third grade slots, and the program is expected to slowly expand horizontally to additional classes per grade. There is no shortage of volunteers spanning the range of achievement levels found in the school. A portion of the school's computer technology will serve this vertical slice. The Mini-School will continue to take advantage of the technology in the ways that have been illustrated. The choice of forming a vertical slice is important for developing continuity over a long period and building the skills and expectations among both students and teachers. It is expected that the sixth grade teachers will be continually challenged by the growing skills of their incoming classes.


Common to these observations was the perception that students had a place in the computer systemnamely, their individual and group workspaces containing their project datathat was not dependent on having an individual computer or being in a particular classroom. Both students and teachers made use of the workspaces to bridge between school contexts. The extent to which both teachers and students appropriated the system into their own work gives us reason to believe that systems like Earth Lab are sustainable in the educational environment. The system that we installed in the school was to some extent modelled on the use of technology in research labs, but the system that emerged had characteristics quite specific to schools; for example, the coordination of small groups, the teacher collaboration, and the use of workspaces rather than personal workstations. Our formative experiment succeeded in using technology to increase the likelihood of collaborative work groups, but we also discovered that a critical function of LAN technology in schools is to make a more seamless connection between school contexts. In this respect, we found ourselves focusing more on the level of the school organization and the collaboration among teachers than on the level of students' working collaboratively around a computer.


This paper has contrasted two very different uses of local area network technology in schools attempting to trace their relationship to the way the school organizes instruction, the way teachers work together, and the opportunities students have for engaging in long-term, open-ended projects that are appropriate for collaborative learning. The relationship between technology and the organization of instruction is complex. An integrated instructional system certainly does not cause schools to have a compartmentalized curriculum sequence and division of labor among teachers. It may, however, provide strong support for those tendencies and place barriers to teachers who may wish to change their mode of instruction. At the same time, the use of these systems can be modified quite radically, for example, by distributing the computers among the classrooms where they might become more integrated with the ongoing work of the classroom. In the same way, a system like Earth Lab does not cause collaborative, interdisciplinary project work to happen, but for schools that are moving in that direction, it can provide some useful tools and mechanisms.

In planning a replication of the Earth Lab environment in other schools, we face the complex relationship between the technology system and the existing structure of instruction in the new schools. Typically, when schools acquire technology, the approach is to purchase a new lab, even when the technology is to be devoted to word processing or programming or earth science projects. The lab becomes the convenient unit for administering the new computers and for conceptualizing the need when arguing for it to the administrators who will make the decision. In developing the materials for the replication of Earth Lab, we were encouraged to think in terms of a technology/curriculum package that could be implemented in a lab that was devoted to science instruction at a particular grade level. Distribution of the computers around the school could not be part of the package, and it was unreasonable to think in terms of developing an integrated curriculum. From a purely administrative point of view, the people in charge of technology acquisition have no authority over school structure, so they necessarily appropriate the technology to the existing school structure. The restructuring that the Computer Mini-School is engaged in cannot be provided to the school in the usual technology/curriculum package. The flexibility of location and time, the collaboratively constructed interdisciplinary curriculum, and the provision for student access to the tools over an extended period are critical components of the environment in which the collaborative projects can emerge. There remains the possibility, however, that a local area network system like Earth Lab can help in slowly subverting a rigid structure by supporting teachers and students in developing and propagating collaborative projects.

Author's Note

For their comments on earlier drafts, I am grateful to Paul Reese and Jan Ellis. Preparation of this paper was supported by the Center for Technology in Education under Grant # 1-135562167-Al from the Office of Educational Research and Improvement, U.S. Department of Education, to Bank Street College of Education. The Earth Lab project was supported by the National Science Foundation under Grant # MDR 8550449 and by Apple Computer Inc. External Research.


1. Brienne, D., & Goldman, S.V. (1989, April). Networking: How it has enhanced science classes in New York schools, and how it can enhance classes in your school too. Classroom Computer Learning, 44-53.

2. Carnegie Forum on Education and the Economy. (1986). A nation prepared: Teachers for the 21st Century. New York: Carnegie Corporation of New York.

3. Cohen, D. K. (1988). Educational technology and school organization. In R. S. Nickerson & P. Zodhiates (Eds.), Technology and education: Looking toward 2020. Hillsdale, NJ: Erlbaum.

4. Collins, A. (in press). The role of computer technology in restructuring schools. In K. Sheingold & M. Tucker (Eds.), Restructuring for learning with technology.

5. Cuban, L. (1986). Teachers and machines. New York: Teachers College Press.

6. Gardner, H. (in press). Assessment in context: The alternative to standardized testing. In B. R. Gifford & M.C. O'Connor (Eds.), Future assessments: Changing views of aptitude, achievement and instruction. Boston: Kluwer.

7. Mageau, T. (1990). ILS: Its new role in schools. Electronic Learning, 10(1), 22-32.

8. Newman, D. (1990). Cognitive and technical issues in the design of educational computer networking. In L. Harasim (Ed.), Online education: Perspectives on a new medium. New York: Praeger.

9. Newman, D. (1990). Using social context for science teaching. In M. Gardner, J. Greeno, F. Reif, & A. Schoenfeld (Eds.), Toward a scientific practice of science education. Hillsdale, NJ: Erlbaum.

10. Newman, D. (1990). Opportunities for research on the organizational impact of school computers. Educational Researcher, 19(3), 8-13 .

11. Newman, D., Goldman, S. V., Brienne, D., Jackson, I., & Magzamen, S. (1989). Peer collaboration in computer-mediated science investigations. Journal of Educational Computing Research, 5(2),151-166

12. Newman, D., Griffin, P., & Cole, M. (1990). The construction zone: Working for cognitive change in school. Cambridge: Cambridge University Press.

13. Pearlman, R.(1990, August 10). American Federation of Teachers. Personal communication.

14. Resta, P. (1990). Factors to consider in selecting and purchasing integrated instructional systems. In M. Sherry (Ed.), The integrated instructional systems report. Water Mill, NY: EPE Institute.

15. Schofield, J., Evans-Rhodes, D., & Huber, B. R. (1990). Artificial intelligence in the classroom: The impact of a computer-based tutor on teachers and students. Social Science Computer Review, 8(1).

16. Sherry, M. (Ed.). (1990). The integrated instructional systems report. Water Mill, NY: EPE Institute.

17. Sizer,T. R (1990). Horace's compromise. Boston: Houghton-Mifflin.

Paper presented at the NATO Advanced Research Workshop on Computer Supported Collaborative Learning, Maratea, Italy, September 1989. To appear in C. O'Malley (Ed.), Computer supported collaborative learning, Springer-Verlag. Preparation of this paper was supported by the Center for Technology in Education under Grant # 1-135562167-AI from the Office of Educational Research and Improvement, U.S. Department of Education, to Bank Street College of Education.

[ Home | About CCT | Projects | Newsletters | Reports | Staff | Links | EDC Home ]

Last Update: 11/18/96
Comments on the CCT Web site: Webspinner.

©1996 Education Development Center, Inc. All Rights Reserved.