   |
 |
"Collaborative Inquiry in Networked Communities: Lessons From the Alice Testbed" Table of Contents Background Network Science Model Scalability The Software Innovation Testbed Projects: Developing a Model What Have We Learned About Network Science? What Have We Learned About Scalability? Summary Appendix: Description of Testbed Projects Bibliography |
In Bruce and Rubinıs study of QUILL, a computer program designed to help upper elementary student to develop as writers, they reflect,
It is no simple matter to look beyond preconceptions to see what
happens when computers are used in education. ... The process
of adopting a computer-based innovation is often a gradual one,
so that characterizing the use of innovation requires long-term
study. The relevant variables are often unknown at the beginning.
And much of what students learn, especially with significant
innovations, is not captured by standard assessment techniques,
so that new methods of evaluation must be developed.
(Bruce and Rubin, 1993, p. 216)
Newman (1990) makes a similar point.
... the unit of analysis is the classroom or school, hence it is necessary
to experiment in real settings over a period of time sufficient for the
environment to appropriate the technology
In short, innovation in general, and educational technology innovation in particular, is typically a gradual process, one that must be studied over time with great attention to the variety of contexts that shape the way that the innovation develops.
Needless to say, the initial months that the Alice Testbed has been operational have not given sufficient time to make general assessments of the viability and effectiveness of Network Science as a model of science learning. Rather, we are in a position to discuss issues and obstacles that have emerged across the initial testbed projects. Our ability to understanding and address these issues will significantly shape the testbed in the future.
Participation and completion rates. While participation rates are consistently high (90% of better) among teachers who have signed up for a project, we wondered about the few teachers who did not show up at all on-line. For example, in the EPA/Nebraska Wetlands project, 29 teachers signed up initially and 26 participated. This raised an important question about obstacles to participation. We attempted to surveyed those few who did not participate. Among the Wetlands teachers, one declined to participate because his classıs schedule did not allow participation in the live video uplinks that this project was using, and he did not want to participate partially. Another was not able to participate because the phone line provided did not permit toll calls. Of the 26 Wetlands teachers who participated in any way, 21 submitted data to the database (81%). We have not yet examined completion rates for any of the projects.
How do these rates compare with similar projects for whom telecommunications is limited to messages? Riel (1994) reports that, in her ³task phase analysis² of 110 classes participating in Learning Circles on AT&Tıs Learning Network, most responded to a ³roll call² at the beginning of the session, 86% of the classes sent one or more messages during the initial two weeks, and 75% completed the project.
Scheduling. Anyone who has taught knows how difficult scheduling issues can be at any one school. Students leave early for athletic games, or are gone for an entire day for a field trip in another class. This winter, especially, classes were canceled for weather reasonsin some cases, up to a week at a time. When considering how complex the scheduling issues are for any one school, we were not surprised to observe the dissatisfaction expressed by teachers when faced with rigid schedules for the testbed projects. In addition to teacher dissatisfaction, failure of teachers to meet schedules appears as a common problem for projects that try to tie teachers to a very tight schedule.
Projects are beginning to experiment with less restrictive schedules. In the EPA/Nebraska Amphibians study, which began in March, a schedule has being developed that will allow classes to submit their data about amphibians populations over the course of a month. This schedule allows for the very different climate variations from north to south, which results in the emergence of the amphibians at different times.
Perhaps the most interesting example of a teacher unable to meet the schedule for an introductory message came from a Global Lab teacher in Moscow in September '93. This teacher wrote an apology to her teammates, explaining that her initial submission would be late because her school was near Russian White House, and because of the fighting, school would not be in session for a few more days.
The issue of scheduling of curriculum has another dimension as well. When the curriculum is very full, classes are kept busy with their activities and little time is available for classes to discuss their work with each other. This over-scheduling was reported by teachers in both the Wetlands and Global Lab projects. The Global Lab project has since been substantially pared down, with the goal of giving teachers and students more time to discuss ideas with other classes.
Curriculum. The writing of curriculum for Network Science has presented some difficulties in the initial projects. Some projects have asked people to write curricula who may be experienced curriculum writers, but who are not familiar with telecommunications nor with the specific software being used. There is also need for experience in writing data templates that lend themselves to appropriate graph and map displays.
In response, the Alice project has begun to request review of each group's curriculum, with the goal of encouraging additional consideration about appropriate use of telecommunications. The project has also outlined, and is in the process of writing, a Handbook for Curriculum Writers and Administrators that documents strategies for telecommunications and reflects what has been learned in these initial field trials. Examples:
The importance of classes having an introductory unit, which
has the dual purpose of beginning to build a sense of community
and also serves to introduce them to the technology.
Clarify the different types of telecommunication activities: e-mail
sent to one class, to a cluster, or to the entire set of participating
classrooms; data tables sent to one class, to a cluster, or to the entire
set of participating classrooms; submission of data to a server; and
retrieval of data from the server.
Avoid close dependency on schedule. A schedule must be flexible
and allow for the unpredictable events that are certain to occur.
Schedules may be helpful in encouraging participation, but classes
must be able to submit late and still continue to participate.
Design for data submission. Our initial experience shows that the present design for submitting data is not sufficiently flexible. For many purposes, it is desirable to create a table with one row per site containing multiple fields. The user fills out one set of fields (i.e., part of one row), submits the data for the other classes to analyze, and then submits a second time when all the fields are completely filled. Under the present design, the network manager had to manually delete rows from the database to avoid duplicate information. We are considering giving each submission an identifier so that, when users make multiple submissions of this type, each submission will automatically update the previous entry.
Tools for project management. Network managers report that it is very difficult to know what is happening in multiple classrooms across the project. In response to their needs, we developed a database reporting system which sends a message each morning (or at any requested interval) about all submissions to a database during the previous 24 hours. With this information, we have been able to track participation more readily. We are currently working on understanding other tasks of the network manager, and adapting our database design to automate them when possible.
Multiple technologies. We are looking at the effect of using multiple technologies within a testbed project. One notable success has been the Total Column Ozonometer, a simple tool designed and distributed by Global Lab. On the other hand, the EPA/Nebraska Wetlands project used live video uplinks twice a week with an audio bridge. Some of the teachers reported that this technology was very inflexible in its scheduling demands; some reported that they used videotapes of the broadcasts with their classes, and others did not make use of the broadcasts at all. The second project organized by EPA/Nebraska is not using live video links; they may distribute information by videotape. Further work in the testbed projects will continue to look at what additional technologies enhance Network Science.
Electronic Forums for Teachers. All of the testbed projects have set up "electronic forums" to encourage discussion among teachers. These typically consist of a "mailing group" or other straightforward means for teachers to send e-mail messages to the full group of all the other participating teachers. Projects use their electronic forums in different ways. One early project used the forum to pass administrative messages about the project. However, there was enough uncertainty about who was on-line that the network manager backed up e-mail with faxes and telephone contact. Reflecting its three years of experience, Global Lab has trained some seasoned participants to function as moderators; they answer questions, share experiences, and encourage reflection about good teaching.
We raise a concern about whether the participants on projects that are limited to two or three months are a sufficient or effective community for electronic communication. Teachers need to have the time in contact to establish common purposes and develop confidence in their ability to participate in discussions. If they do establish these prerequisites, the project is ending just as they are up and running.
As an alternative, we suggest that teachers in Network Science projects be included as part of an on-going electronic community focused on Network Science. While for some purposes they will clearly want to talk only with other members of the same project, for many other purposes they will want to talk to a larger group that will have a continuing existence. An electronic forum modeled on LabNet (which focuses on supporting teacher in the practice of project-enhanced science learning) may serve teachers better.
Teacher change. Through electronic forums, we have seen examples of the impact of one excellent teacher on a group of teachers and classes. In this case, the teacher had participated in previous TERC telecommunications projects and was asked by the network manager to act informally as a "moderator" on the forum, asking and answering questions and encouraging discussion among the teachers. One example of his work in this role shows him encouraging thoughtful reflection about students' work:
Our presentations were completed today and I would like to share
the researchable questions with you. I am very much interested
in the questions your students formed and hope you will post
them to the Forum.
[10 examples of student questions listed] ...
The second aspect also concerns Questions. With extra time it's
possible to have students spend the effort to work a couple questions
at the same time and choose the one that appears to be working a
little better. Itıs then possible to form and reform the questions. It takes
time for one teacher to work with the groups to do this. My students
this year seem to want to get the question and just get it done without
the thought required of more involved question forming. I think if
they knew I was willing to spend the extra time, they might have been
more receptive to spend more time also.
(Note from a field test teacher, 3/8/94, KN "Human Body" curriculum)
In this message, he talked about his class's experiences; in other messages, he recommended articles and books, and asked for and gave advice on how to deal with particulars situation. One measure of his effectiveness is the response of other teachers. His messages prompted many responses and succeeded in generating discussion among the participating teachers. By the end of the project, many teachers sent messages to thank him for his input and ask to continue the "electronic relationship" after the project finished (although the project network would not be available to them).
Beyond networks encouraging and assisting teachers in their professional growth, this example also demonstrates another impact of networked learning. The teacher who wrote this message had already participated in another Network Science project, and was bringing the experience and knowledge to his new group. We will continue to watch for evidence of synergy among projects, as shown in this example.
Better connections among local sites. Finally, in observing projects, we have observed that the current design of Alice makes collaboration easier among distant schools than among schools who are in the same building. In the current design, each class in a building would need its own modem and phone lineeven if a local area network were available. We believe that software for collaborative inquiry needs to work as well within a school (connected on a local area network) as among schools that are in distant parts of the world (connected on a wide area network).