Lehrstuhl für Empirische Pädagogik und Pädagogische Psychologie (EN)
print

Language Selection

Breadcrumb Navigation


Content

Problem-oriented learning in higher education

Collaboration Scripts for Online Discussions

Using Natural Language Processing (NLP) to Analyse and Facilitate Online Discussions

Training of complex skills in medical education

Collaboration Scripts for Online Discussions

An essential problem of computer-supported collaborative learning is the often low argumentative quality of online discussions. The research project “Collaboration Scripts for Online Discussions” investigates how collaboration scripts can facilitate argumentative knowledge construction in online discussions. Argumentative knowledge construction means the construction of domain-specific and domain-general knowledge through collaborative argumentation. Collaboration scripts can support learners in argumentative knowledge construction by specifying, sequencing and distributing roles and activities for learning. In our studies, we investigate both processes and outcomes of argumentative knowledge construction based on a multi-dimensional framework. Within this framework, argumentative knowledge construction is being analyzed with respect to epistemic activities, the formal quality of argumentation, and social modes of co-construction including transactivity aspects, i.e. learners’ mutual reference in online discussions. In collaboration with the Carnegie-Mellon-University, we aim to automatize this time-consuming method to analyze natural discourse corpora. Several collaboration scripts have been developed and investigated that specifically facilitate single dimensions of argumentative knowledge construction in online discussions. On the one hand, the findings show that the investigated scripts do have the desired main effects. For instance, scripts facilitating transactivity actually help learners to better refer to each others’ contributions and also to acquire substantially more knowledge than learners without this script. Some scripts, on the other hand, can have unwanted side effects. An epistemic script, for instance, facilitated learners in solving the learning task, but had detrimental effects on knowledge acquisition. Furthermore, we investigated fading of scripts as well as the application of scripts in different domains, such as computer science and medicine at the University of Tübingen.

Collaborators:

Carolyn Rosé, Carnegie-Mellon-University, Pittsburgh, USA, Automatic Analysis of Natural Discourse Corporatop

Using Natural Language Processing (NLP) to Analyse and Facilitate Online Discussions

Analyzing the variety of different facets of learners’ interaction that are important for their learning is a time consuming and effortful process. Improving automated analyses of such highly valued processes of collaborative learning by adapting and applying recent text classification technologies would make it a less arduous task to obtain insights from corpus data. It also holds the potential for enabling substantially improved on-line instruction both by providing teachers and facilitators with reports about the groups they are moderating and by scaffolding technology as in the emerging area of context sensitive collaborative learning support triggered dynamically on an as-needed basis. To improve the quality of automatic classification technologies, we collaborate with the Carnegie-Mellon-University, the Technische Universität Darmstadt, and the University of Stuttgart.

Collaborators:

  • Iryna Gurevych, Technische Universität Darmstadt, Educational Natural Language Processing (E-NLP) &Ubiquitous knowledge processing
  • Carolyn Rosé, Carnegie-Mellon-University, Pittsburgh, USA, Automatic Analysis of Natural Discourse Corpora
  • Hinrich Schütze, Institute for Natural Language Processing, University of Stuttgart, Statistical Natural Language Processing
  • Max Mühlhäuser, Technische Universität Darmstadt, eLearning, Multimodal Interactiontop

Training of complex skills in medical education

Project members of the chair:

Collborators:

  • Prof. Dr. Mathias Siebeck
  • Florian Pilz, M.A. Paed