Causal Discourse Analyzer: Improving Automated Feedback on Academic ESL Writing

Autor/inn/en	Chukharev-Hudilainen, Evgeny; Saricaoglu, Aysel
Titel	Causal Discourse Analyzer: Improving Automated Feedback on Academic ESL Writing
Quelle	In: Computer Assisted Language Learning, 29 (2016) 3, S.494-516 (23 Seiten) PDF als Volltext Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Zeitschriftenaufsatz
ISSN	0958-8221
DOI	10.1080/09588221.2014.991795
Schlagwörter	Discourse Analysis; Feedback (Response); Undergraduate Students; Accuracy; Recall (Psychology); Attribution Theory; Academic Discourse; Writing (Composition); Writing Evaluation; Computer Assisted Instruction; Essays; Second Language Learning; Teaching Methods; English (Second Language); Second Language Instruction; Error Correction; Writing Instruction; Grammar; Computer Software; Natural Language Processing; Error Analysis (Language); Computational Linguistics; Statistical Analysis + Suchen Sie Ihr Suchwort? Diskursanalyse; Abberufung; Discourse; Diskurs; Schreibübung; Computer based training; Computerunterstützter Unterricht; Essay; Aufsatzunterricht; Zweitsprachenerwerb; Teaching method; Lehrmethode; Unterrichtsmethode; English as second language; English; Second Language; Englisch als Zweitsprache; Fremdsprachenunterricht; Korrektur; Schreibunterricht; Grammatik; Natürliche Sprache; Error analysis; Language; Fehleranalyse; Linguistics; Computerlinguistik; Statistische Analyse
Abstract	Expressing causal relations plays a central role in academic writing. While it is important that writing instructors assess and provide feedback on learners' causal discourse, it could be a very time-consuming task. In this respect, automated writing evaluation (AWE) tools may be helpful. However, to date, there have been no AWE tools capable of evaluating causal discourse. The authors of the present study attempt to fill in this gap by (1) developing an automated causal discourse analyzer and (2) investigating how accurately the analyzer processes learners' causal discourse in academic writing. The accuracy of the analyzer is evaluated on cause-and-effect essays written by 17 non-native undergraduate students. The results indicate precision of 0.93, recall of 0.71, and accuracy of 0.76, which is promising for pedagogical applications of the analyzer, that is, providing learners with automated formative feedback specific to causal discourse. (As Provided).
Anmerkungen	Routledge. Available from: Taylor & Francis, Ltd. 325 Chestnut Street Suite 800, Philadelphia, PA 19106. Tel: 800-354-1420; Fax: 215-625-2940; Web site: http://www.tandf.co.uk/journals
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2020/1/01