Towards an Error Taxonomy for Student Writing

Zeitschrift für Interkulturellen Fremdsprachenunterricht 19(2), pp. 111–118.

8 Pages Posted: 7 Jan 2015

See all articles by Nikola Dobric

Nikola Dobric

Alpen-Adria-University Klagenfurt - Institut für Anglistik und Amerikanistik

Guenther Sigott

Alpen-Adria-University Klagenfurt - Institut für Anglistik und Amerikanistik

Date Written: October 1, 2014

Abstract

Current practice in writing assessment has tended to divert attention from the notion of error in favour of more global and intuitive descriptions of learner performance. As a result, experts tend to disagree in the way errors in student writing are described. This brings about complications for student feedback as well as for studying the construct validity (cf. Messick 1989) of rater-mediated assessments of student writing by means of computer-corpus-based methodology, which requires reliable anno-tation. In order to alleviate these complications, an error taxonomy is proposed which could serve as a basis for student feedback on the one hand, and as a basis for corpus-based studies into the construct validity of large-scale assessments of writing, on the other.

Keywords: error analysis, error taxonomy, learner corpora, scope, substance, annotation, inter-rater agreement, inter-annotator agreement, rating, validation, feedback

Suggested Citation

Dobric, Nikola and Sigott, Guenther, Towards an Error Taxonomy for Student Writing (October 1, 2014). Zeitschrift für Interkulturellen Fremdsprachenunterricht 19(2), pp. 111–118., Available at SSRN: https://ssrn.com/abstract=2545303

Nikola Dobric (Contact Author)

Alpen-Adria-University Klagenfurt - Institut für Anglistik und Amerikanistik ( email )

Universitätsstrasse 65-67
Klagenfurt, Corinthia 9020
Austria

HOME PAGE: http://www.uni-klu.ac.at/iaa/inhalt/2512.htm

Guenther Sigott

Alpen-Adria-University Klagenfurt - Institut für Anglistik und Amerikanistik ( email )

Universitätsstrasse 65-67
Klagenfurt, Corinthia A-9020
Austria

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
105
Abstract Views
455
Rank
463,225
PlumX Metrics