Unraveling idea development in discourse trajectories
Bibliographic record
Abstract
With the present paper we want to shed light onto an issue that is central within the knowledge building theory but only little studied -the development of ideas in collaborative learning discourse.Starting from the construction of a network of explicit and implicit relations between ideas, we apply a scientometric method to tackle the temporality of collaborative processes based on the structure of successive ideas.The resulting discourse trajectories are shown to give a holistic and also a detailed view on how knowledge advances when their interpretation is combined with a qualitative analysis of the content of the ideas and their relations.The weighted relevance of relations between ideas enables the identification of sub-topics in the discourse, important ideas, and influence or uptake events. IntroductionHow is knowledge about the world created and advanced?Philosophers have spent enormous efforts to answer this question (e.g.Popper, 1972).Knowledge building is an approach from the learning sciences that attempts to build on contemporary answers from philosophical inquiries and research on expertise (Bereiter, 2002;Bereiter & Scardamalia, 1993;Scardamalia & Bereiter, 2006) to engage students in the kinds of knowledge work that are widely assumed important in the 21st century, including ability to collaborate, deal with novelty, and solve ill-structured problems.At the heart of knowledge building is a computermediated collaborative discourse that is oriented toward idea improvement.Following Popper's (1972) theory of objective knowledge, knowledge-building theory considers ideas as "real" objects that can be critiqued, tested and modified, much like how real objects like bicycles undergo these processes (Bereiter & Scardamalia, 2003).Ideas do not reside in the minds of participants but take on lives on their own in this discourse.Hong, Chen, Chang, Liao and Chan (2009) emphasized that the "idea-centered" educational design is enabled through the Knowledge Forum software by allowing interaction around ideas regardless of the discussion threading.Hence, the process of idea development is fundamental for understanding knowledge building.Moreover, focus on processes is widespread in the field of computer-supported collaborative learning (CSCL), reflecting the dynamic nature of discourse as an object of study.However, despite this acknowledged role, there is a dearth of analytical approaches for investigating the dynamic development of ideas in knowledge-building discourse.Therefore, the main goal of this paper is to provide an example of studying what we call a discourse trajectory, i.e. the genuine process characteristics of a discourse based on idea development over time.In order to do so, we first outline briefly previous research and then present a new methodological technique and its application to knowledge building discourse. Related ResearchA starting point to studying a discourse process is the evaluation of surface indicators of participation and communication like number, length of contributions, etc. (Strijbos, Kirschner and Martens, 2004).Such results can automatically be evaluated from the log file data of the software but they are regarded as only very basic descriptors of a collaborative process.Most studies of knowledge building have followed a content analysis approach (Chi, 1997;Gunawardena, Lowe, & Anderson, 1997;Henri, 1992), where qualitative data is segmented into idea units and these are coded for their cognitive, metacognitive, social, motivational and other aspects.The frequency of the assigned codes is then statistically
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.053 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.007 |
| Science and technology studies | 0.002 | 0.007 |
| Scholarly communication | 0.010 | 0.025 |
| Open science | 0.002 | 0.008 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".