The workshop, "Collaborating on the Design and Assessment of Knowledge-Building Environments in the 2000's", at CSCL '99 was attended by about 60 participants from all over the world. Most of the workshop took place in 5 subgroups that carried on lively discussions. The subgroups reported back to the whole group. The following outlines are based on those reports.
These outlines are based on my notes from the slides and reports; please send me additions and improvements to these outlines at Gerry.Stahl@Colorado.edu .
This group dealt with the question of how we can capture learning trajectories, both of individuals and of groups. The two big questions are "what" to capture and "how" to capture it.
What to capture of learning trajectories:
 | "with"
Quality of Content
| ||||||||||
| "with"
Types of Process
| ||||||||||
| "of"
Outcomes (residue)
|
How to capture
| social network analysis |
| discourse analysis |
| content analysis |
Paralleling individual and group trajectories
Collaboration
| Sasha Barab will collect references from workshop participants and send them to Gerry, who will post them to the website. |
| Zahra Punja will contribute pointers to some international collaborations. |
This group focused on two transitions:
Curriculum should make room for new curricular goals to emerge. This may conflict with the original goal of curriculum as the organization of facts to be transmitted. Assessment can address some of the "process" goals (often left out of the explicit curriculum). Curriculum becomes two-part, monitoring both the group process and the individual effort / self-directed learning.
Technology should be designed to support portfolio assessment, curriculum design ad the use of artifacts
Some issues, problems and questions:
| The shelf-life of knowledge is rapidly shrinking. In engineering fields, the validity of much knowledge now lasts 1.5 years compared to 5 years in the past. |
| Absorption time is shorter: experts have very little time to spend for just-in-time learning. |
| The classroom model is not effective in industry; work is knowledge-enabled rather than knowledge-driven. |
| Learning must be just-in-time and just-enough to meet the needs of work tasks. |
| Search tools are not adequate to meet the needs. |
| Different people have different learning styles and different companies have different reward systems. |
| How can learning be learner-centered, yet benefit from the community? |
| How can academic and industrial research be reconciled? |
Some areas for possible answers:
|
Software tools like DocReview can facilitate on-going discussion |
|
Open architectures can facilitate combining tools. |
|
Problems have to be defined clearly. |
|
Meta-tagging of knowledge objects may facilitate searching. |
Targeted outcomes of knowledge-building communities:
| collaborative skills |
| group organizational awareness |
| comfort with distributed knowledge and how to assess it |
| learner agency |
| trust |
| engagement in process/intrinsic motivation |
| "usefulness" of process and products for organization |
Specifications for systems design:
| marks community norms |
| structure and annotate statements and claims |
| discuss wide array of artifacts |
| qualitative assessment of artifacts correlated to cognitive processes (e.g., usage data) |
| evoking and marking the context of artifacts |
Question:
| how
to define collaborative learning in terms of knowledge-building?
|
Adoption can involve a variety of populations, approaches, goals and practices;
"successful adoption" varies and depends on the goals and the evaluation methods.
Some common barriers to adoption are:
| motivation (incentive, requirements, priorities/time) |
| accessibility, usability |
| communication mechanisms (e.g., synchronous, audio support in KBEs) |
| privacy and ownership concerns (private vs. public space) |
| lack of a common language or goals |
| data mining (search, recommender systems) |
| variety of represented document types (simulations, …) |
| need for integration |
Although no one chose to be in this group, an interesting discussion took place during the reports. Marlene Scardamalia argued that a group interaction must produce something identifiable and new that has been learned in order to count as a knowledge-building event. Dan Suthers pointed out that his position paper defined knowledge-building environments as combining discourse + artifacts + representations. So the result of knowledge-building discourse should be an artifact that may be expressed using a specific representation. Marlene responded that it is important that the artifact produced be something new, not an existing artifact that was used in the discourse (e.g., a textbook). We might conclude then that collaborative learning is a goal-directed group process through which new knowledge is formed and is encapsulated in a new or modified artifact, such as a document. Perhaps these minutes can serve as such an artifact to preserve something of the collaborative learning that took place at the workshop.
Go to top of this page
Return to Gerry Stahl's Home Page
Send email to Gerry.Stahl@drexel.edu
This page last modified on January 05, 2004
