Abstract

This study examined the impact of EFL learners’ proficiency differences on their dyadic interaction on the internet. The participants included 94 university students, who were divided into four pairs: the higher proficiency sender (HS)-lower proficiency receiver (LR) pair, the lower proficiency sender (LS)-higher proficiency receiver (HR) pair, the HS-HR pair, and the LS-LR pair. They were asked to perform an asymmetrical convergent task. In the first part of the data analysis, the differences in the number of words produced and in the ratio of resolved referential problems were examined between the HS-LR pair and the LS-HR pair. To provide a more detailed comparison, the differences between the HS and the LS and between the HR and the LR were also calculated. In the second part, comparisons were made between the HS in the HS-LR pair and the HS in the HS-HR pair, between the LS in the LS-HR pair and the LS in the LS-LR pair, between the HR in the LS-HR pair and the HR in the HS-HR pair, and between the LR in the HS-LR pair and the LR in the LS-LR pair. It was found that the differences between the HR in the LS-HR pair and the LR in the HS-LR pair were significant, while those between other groups were not. Finally, the necessity of questioning two common views regarding pair work was indicated, and the potential for a more synthetic model of dyadic interaction was discussed.

Suggested citation