|
|
|
|
Investigations
of collaborative design and the dynamics of learning communities Alex
Cuthbert Cognition
& Development University
of California at Berkeley <http://www.kie.berkeley.edu/people/alex>
My
research seeks to (a) understand how communities of students mutually revise
their ideas and (b) develop strategies for encouraging this process of
refinement and consensus-building. In
recent years, researchers have made advances in explaining the connection
between learning and social interaction (Vygotsky, 1934; Piaget, 1970; Saxe,
Gearhart, & Paduano, 1993; Petty & Cacioppo, 1986; Scardemalia &
Bereiter, 1991). However, it is still unclear how these and other theories of
learning inform practice (Landauer, 1991). My research investigates students’
learning processes and knowledge attainment in online learning communities,
while making contributions to the social and technical organization of classroom
practice. I have developed a template (figure 1) for “design projects” which require consensus-building and refinement of solutions. Design projects require elaboration of the initial problem definition and frequently involve collaboration between individuals with diverse backgrounds and perspectives. Figure
1. Activity map for design projects This
template provides an overall structure for students without requiring a rigid,
step-by-step approach. Students access resources and make contributions at
different levels of abstraction throughout a project, participating in online
discussions in sections three and six and adding annotated Internet sites to a
resource library in section four. My research provides evidence that this
structure encourages students to refine their ideas collectively and
independently over the course of the project. I briefly summarize the two goals
and associated hypotheses that guided the development of my approach to
supporting collaborative design projects: 1.
Encourage mutual refinement of ideas. My hypothesis is that students who
revise their ideas will have a more linked and connected understanding of the
various scientific concepts associated with a project. Eliciting
student ideas is an important initial step in most constructive learning
activities. In some situations, such as brainstorming, the goal is to generate
as many ideas as possible before evaluating them. Having a broader problem
definition should help students differentiate between alternatives, thereby
increasing the potential for comparison and critique. Whether comparison and
critique activities help students develop linked and connected explanations to
support their designs is an open question. To begin, we can see if having access
to a range of alternatives encourages students to compare and critique ideas
which are distinct from their own. Measuring appropriation of ideas is difficult
and requires that students have the opportunity to apply those ideas in new
situations, such as peer review. 1a.
Represent ideas in an accessible and equitable manner. To encourage principled
consideration of alternatives, ideas should be represented in an accessible and
equitable manner. My hypothesis is that representing ideas equitably will help
students evaluate ideas based on their relationship with other ideas rather than
on social cues. Assessing how students interpret and use representations is
difficult. However, we can begin with crude measures from computer-mediated
communication that assess interest in comments based on response rates. In
addition, we would expect that ideas would be evaluated based on a combination
of social cues and content depending on the amount of social information
provided. Hence, the importance of creating detailed case studies of how ideas
are transformed as they circulate through a community of learners. 1b.
Differentiate between perspectives. One of the most important dimensions of
collaborative learning is the potential for students to recognize views that are
distinct from their own. Helping students differentiate between perspectives is
challenging since students tend to view other ideas in relation to their own
perspective. Identifying and differentiating between different models is an
important part of this process. As researchers, we can provide situations where
students reflect on the alternatives they are considering and the principles
they are using to select between them. Our assessments need to examine both the
range of alternatives considered and the methods students use for selecting
between them. 2.
Develop a common set of criteria for making decisions. The variable
quality of information on the Internet requires teachers to help students
develop critique skills for evaluating evidence (Cuthbert, 1996). Similarly,
mutual refinement requires a common set of criteria for comparing and
contrasting alternatives. Shared criteria refers to a set of requirements or
guidelines for interpreting ideas and making decisions (though “common” is
perhaps a more accurate term.) Criteria operate at different levels, from
determining the adequacy of a piece of evidence to the coherence of a design.
From a knowledge integration perspective, criteria should help students make
informed decisions using scientific principles. We can measure the extent to
which students have internalized criteria and correlate this measurement with
their success at developing coherent designs and arguments. My
hypothesis is that having a common set of criteria would help students provide
constructive criticism for their peers as well as help them select evidence for
supporting their own design decisions. By encouraging application of criteria in
new situations, we can assess the degree to which students internalize the
criteria. In turn, this analysis will help us understand the developmental
processes of collaborative knowledge building, providing insights into
strategies for refining the computer-based representations and supports for
these learning processes. * * * * * older version of Cuthbert's position paper: Assessing Knowledge Building Environments
Alex Cuthbert Assessment is almost always tied to a set of criteria for evaluation. So when we talk of "assessment", we need to begin by specifying the goals we are trying to achieve. I have identified seven goals for knowledge building environments (KBEs) which I describe below. For each goal, I provide a short description of the theoretical grounding for the goal and a method for assessing that goal. As a researcher, I have been involved in the design and construction of the WISE environment from which many of these positions evolved. My personal view on assessment of these environments is that we need to develop a common set of criteria for assessing KBEs, linked to the following goals: Goal one: Elicit students' ideasIn some situations, such as brainstorming, the goal is to generate as many ideas as possible before evaluating them. The hypothesis is that having a broader problem definition will help students differentiate between alternatives, thereby increasing the potential for comparison and critique. To assess whether we have encouraged students' to generate their own ideas, we can count the number of ideas generated within groups of students and compare that with individual and smaller groups of students. More importantly, we can see if having access to a range of alternatives encourages students to compare and critique ideas which are distinct from their own. Measuring appropriation of ideas is difficult and requires comparison across a longer time period, such as that afforded by extended projects and design tasks. Goal two: Reduce planning problemsPlanning is one of the perennial problems students have as evidenced by the number of hands which are routinely raised to ask, "What should we do next?" Reducing planning problems should help students focus on the content and rationale of the decisions they make. Achieving alignment between students problem-solving paths and the structures to support those activities (e.g., checklists, learning cycles, etc) is a challenging task. The hypothesis is that by providing students with an organizational framework rather than a prescription of the steps they need to take that KBEs can accommodate a variety of trajectories through the activity. To assess whether an approach has worked, we can analyze whether the students' approaches were supported by the framework (e.g., were the questions they needed answered addressed in a timely manner.) Goal three: Represent ideas equitablyGoal three is the representational counterpart of eliciting student opinions (the first goal.) The hypothesis is that representing ideas equitably will help students evaluate ideas based on their relationship with other ideas rather than on social cues. Assessing how students interpret and use representations is difficult. However, we can begin with crude measures from computer-mediated communication that assess interest in comments based on response rates. In addition, we would expect that ideas would be evaluated based on a combination of factors including social cues. Hence, the importance of allowing anonymous contributions to discussions. Goal four: Encourage sharing of resourcesCollaborative and cooperative learning environments permit the sharing resources and the building of a community knowledge base. The challenge is to provide ways for students to locate and use resources effectively. Differences in how students approach problems complicate this goal. One solution is to have multiple ways of accessing the same information (e.g., through contributor, area of specialization, or by searching comments for specific phrases.) Annotating resources such as Internet sites can make those resources easier for students to interpret. Perhaps more importantly, this process encourages them to reflect upon the utility, understandability, and validity of the material they are submitting. Assessment of the extent to which students share resources needs to include a metric for the way in which the information is appropriated (e.g., copy-replace, aggregate, elaborate.) Goal five: Develop shared criteriaDeveloping shared criteria can help students evaluate evidence. Shared criteria refers to a set of requirements or guidelines for interpreting ideas and making decisions. Criteria operate at different levels, from determining the adequacy of a piece of evidence to the coherence of a design. From a knowledge integration perspective, criteria should help students make informed decisions using scientific principles. We can measure the extent to which students have internalized criteria and correlate this measurement with their success at developing coherent designs and arguments. Goal six: Differentiate between perspectivesOne of the most important dimensions of cooperative learning is the potential for students to recognize views that are distinct from their own. From a knowledge integration perspective, identifying and differentiating between different models is a central part of learning. As researchers, we can provide situations where students reflect on the alternatives they are considering and the principles they are using to select between them. Assessments can examine both the range of alternatives and the methods students use for selecting between them. Goal seven: Manage complexityMost approaches to the problem of managing complexity involve either (a) linking representations at different levels of detail or (b) iteratively unfolding functionality as learners become more experienced. For example, the gIBIS (graphical issue-based information system) depicted in figure 1 provides a node-based view, a news-group view, and the text of the currently selected comment. Selecting a comment in the news-group window or the graphical, node window updates the selection in the other windows.
Figure 1. gIBIS approach to managing the complexity of discussions The ability to filter conversations using criteria such as "all open questions" or "contributions made by X" provides other useful navigational tools which help select specific sets of comments for viewing. Other strategies for handling the complexity of large discussions include the ability to (a) collapse or expand groups of related comments, (b) gray out or hide older comments, and (c) designate discussion threads as "asides" or "off-topic" comments.
|
