Loughborough University
Browse

Domain-specific formative feedback through domain-independent diagram matching

Download (179.58 kB)
conference contribution
posted on 2009-04-30, 13:41 authored by Christos Tselonis, John Sargeant
As part of our Human-Computer Collaborative (HCC) approach to assessment, we seek representations of answers and marking judgements which can be applied to a wide variety of situations. In this paper we introduce such a representation, which we call a gree, and discuss an initial practical application of grees for formative feedback. An experiment was carried out in which students were asked to construct an answer while receiving interactive feedback and then complete a short survey. The results show that it is possible to give effective domain-specific formative feedback based on a domain-independent internal representation or “metaformat”. This work builds on results we have previously presented on domain-independent diagram matching based on heuristic matching of graphs. Grees provide much greater flexibility, with a wide variety of potential applications. We discuss some problems which need to be overcome before we can realise their full potential.

History

School

  • University Academic and Administrative Support

Department

  • Professional Development

Research Unit

  • CAA Conference

Citation

TSELONIS, C. and SARGEANT, J., 2007. Domain-specific formative feedback through domain-independent diagram matching. IN: Khandia, F. (ed.). 11th CAA International Computer Assisted Assessment Conference : Proceedings of the Conference on 10th and 11th July 2007 at Loughborough University. Loughborough : Lougborough University, pp. 403-420

Publisher

© Loughborough University

Version

  • VoR (Version of Record)

Publication date

2007

Notes

This is a conference paper.

ISBN

0956957268

Language

  • en

Usage metrics

    Loughborough Publications

    Categories

    No categories selected

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC