Gerry Stahl's professional & personal web network

-- GerryStahl.net --

| Home Page | eLibrary | Publications | CSCL | Teaching | Research | Sculpture | Salt Marsh | Personal |

Volume 10. Essays in Group-Cognitive Science

Essays, case studies and documentation related to group cognition as a scientific enterprise. The essays discuss the structure of discourse in collaborative online mathematics, including longer structures built upon adjacency pairs of conversational utterances or chat response pairs of postings. This structure has been captured in coding schemes used in certain studies within the VMT Project.

download volume here

* Download PDF free for reading online or printing: science.pdf
(This is the best version: up-to-date, complete, full color, hi-res.)

* Download iBook version free for iPad, Mac, etc.: science.epub

* Download MOBI version free for Kindle: science.mobi

* Order paperback from Lulu at printing cost ($14.20): Lulu page for Gerry Stahl

* Order paperback from Amazon at printing cost ($14.20): Amazon page for Gerry Stahl

table of contents

Part A: Toward a Science of Group Cognition
Group Cognition as a Foundation for the New Science of Learning
A View of Computer-Supported Collaborative Learning Research and its Lessons
Part B: Theories of Group Cognition
How I View Learning and Thinking in CSCL Groups
Theories of Group Cognition: Foundations for CSCL and CSCW
Part C: Analyzing Group Cognition
How a Virtual Math Team Structured its Problem Solving
Interaction Analysis of a Biology Chat
Coding Scheme for Sequential Discourse
Methodological issues in developing a multi-dimensional coding procedure for small group chat communication
A Multi-Dimensional Coding Scheme for Mathematical Collaboration

from the introduction

The VMT Project represents a paradigmatic CSCL effort. It developed innovative technologies, collaborative pedagogies, school-oriented curricular materials, analytic approaches and theoretical formulations. It supported group learning in online small groups of students in schools settings. The essays collected below consider methodological approaches, issues, techniques and materials. This is the best source for understanding the VMT data-analysis methodology, as it developed over 12 years. Such an overview was not possible in focused papers or even in thematic books. Yet, it is something that may be of particular interest for CSCL researchers and graduate students. The current volume collects essays and supporting materials from throughout the VMT Project (2002-2015). These were key documents in determining and recording the development of the project’s scientific method, which did not, however, find a place in the volumes mentioned above. 

Perhaps one of the most significant findings of the VMT Project involved an analysis of the structure of problem-solving discourse. This finding was never discussed in any of the four major publications. It is documented extensively here. It builds on the central discovery of Conversation Analysis: that conversation is driven by (and thereby structured by) “adjacency pairs” of utterances, which elicit and respond to each other (like question/answer pairs). In mathematical problem solving, however, it is important to look at larger argumentative structures than such adjacency pairs.

In VMT, we found that the larger structures of online problem-solving discussions were built out of response pairs of chat postings. Furthermore, in their interactions, students tended to structure their discussions implicitly into a hierarchy of structures: events, sessions, themes, discourse moves, response pairs, postings and indexical references. Conversation Analysis had worked out a well-defined understanding of much of this structure. This suggested that a coding scheme could be defined for coding online mathematical problem solving, grounded in such an analysis.

In CSCL, it is common to turn to coding schemes to provide a seemingly “objective,” scientific approach to analysis of data. Various coding schemes may be useful for pursuing specific research questions, such as comparing two experimental conditions quantitatively. However, the coding schemes tend to have various problems: they are not theoretically grounded in the structure of interaction, they are not applicable in a general way, they require indoctrination of coders. Above all, they typically lose the sequentiality of the temporal flow of the discussion, which is perhaps the most important characteristic of collaborative interaction and discourse in general.

In this volume, several coding schemes are defined, based on the response structure of interaction. By coding response pairs rather than single utterances and by situating them in the hierarchical structure of dialog, these approaches focused on the small-group unit of analysis, rather than on individuals. They are illustrated by case studies of VMT data. This is another contribution of this book.These coding schemes are not documented anywhere else.