Researchers argue that written feedback (CR) promotes language accuracy and therefore contributes to second language learning and skill development.
In their study carried out to describe the retroactive practices of teachers of French as a language of instruction (L1) and as a second language (L2) in Quebec, Ahlem Ammar, Daniel Daigle and Pascale Lefrançois report that the majority of the CR provided is indirect (encouraging learners to correct themselves) and that direct CR (rephrasing the learner’s error) is sometimes used to correct syntactic errors. It is important to educate teachers about the effects of their feedback practices on L2 learning.
Due to different methodological limitations, previous research does not provide conclusive answers, hence the relevance of this study which aims to assess the effects of three CR techniques, including two indirect – underlining or underlining + metalinguistic index – and one direct – provide the correct form on learning French as a second language. Given the importance of revision in writing, the effects of two revision conditions (individual or collaborative) are studied.
Finally, in light of the research indicating that the effects of CR are moderated by variables specific to the learner and the language, this project studies the moderating effects of the level of the learner and the type of error. .
In order to meet the three research objectives, five secondary French second language classes, enriched context, participated in this quasi-experimental study comprising a pre-test, an experimental intervention, an immediate post-test and a deferred post-test.
Among other things, the results indicate that while the two indirect techniques favor the greatest short-term gains, the underlining + metalinguistic index group is the only one to have obtained a change in language precision over time. Collaborative review allows for a better error detection and repair rate than individual review. The effect of the techniques depends on the level of language skills of the learners.
Ahlem Ammar, Université de Montréal
Deposit of the research report: October 2019