The effects of recasts and metalinguistic corrective feedback on grammar acquisition of postgraduate ESL learners / Mandana Rohollahzadeh Ebadi

The issue of efficacy of oral corrective feedback on grammar acquisition of second language learners in an English language classroom setting remains controversial in recent years due to the implicit and explicit types of corrective feedback and different positions of interface toward implicit and explicit knowledge. This study investigated the impact of implicit corrective feedback in the form of recasts and explicit corrective feedback in the form of metalinguistic information on grammar acquisition of ESL learners. The quantitative study was conducted at the University of Malaya Centre for Continuing Education in Kuala Lumpur with 136 female and male international postgraduate students at lower-intermediate level of English proficiency involving one control and two experimental groups. The relative efficacy of both types of corrective feedback was assessed by using the Elicited Oral Imitation Test (EOIT) and Timed Grammaticality Judgment Test (TGJT) for measuring implicit knowledge; and Untimed Grammaticality Judgment Test (UGJT) and Metalinguistic Knowledge Test (MKT) for measuring explicit knowledge in pretest and posttest sessions before and after the intervention program. To analyze the data, ANCOVA and Post Hoc analysis (Scheffe‘s test) were carried out. The practical implications of the results suggest that both recasts and metalinguistic corrective feedback have a significant effect on developing grammar acquisition of ESL learners. This significant effect is seen not only in the EOIT and TGJT tests, but also in the UGJT and MKT tests. The effectiveness for different grammatical features varied from the small effect size (i.e., Modals can, have to) to moderate (i.e., Past tense, Present perfect) and approximately large effect size (i.e., Comparatives, Unreal conditionals). Moreover, further analysis shows that overall the students in the metalinguistic group scored significantly higher than the recast group. However, this outperformance for most of the target structures of the study was iv significant (i.e., Regular past tense, Present perfect with since & for, Comparatives, and Unreal conditionals) but for some of them was not (i.e., Modal can & Modal have to) and this demonstrates that explicit corrective feedback may benefit more for some structures than others. Methodologically, the study by conducting separate measurement of implicit and explicit knowledge shed light on the effectiveness of corrective feedback on developing both knowledge about the language (i.e., explicit knowledge) and knowledge of the language (i.e., implicit knowledge). The study also provided fresh empirical evidence to support the weak interface position toward implicit and explicit knowledge of ESL learners and in turn proposed pedagogical practices followed by researchers and practitioners adhering to this position.