Je-LKS: Journal of E-Learning and Knowledge Society (Oct 2019)

User ratings as a predictor of linguistic feedback quality in Question and Answer portals

  • Simone Torsani

DOI
https://doi.org/10.20368/1971-8829/1135045
Journal volume & issue
Vol. 15, no. 3

Abstract

Read online

Social Question and Answer portals allow users to post and answer questions on different issues, among which foreign languages. The present paper focuses on feedback requests, i.e. questions in which users of the site ask for linguistic feedback on short sentences or phrases. In particular, it reports on a research on how reliable is the evaluation of answers provided by the portal’s users to identify correct and good linguistic feedback. An observational approach was adopted on about 600 answers in the Italian version of Yahoo! Answers. Each feedback was evaluated by two expert teachers and their rating was then compared with the evaluation provided by the site’s users. Results show that, while the correlation between the votes of the community and the rating of the experts is rather weak, answers with a positive evaluation generally contain a correct feedback. We conclude, therefore, that caution must be exercised when using users’ evaluation as guidance on feedback choice.

Keywords