Abstract
Shadowing, a learning method involving the immediate repetition of heard speech, has been reported to improve listening skills, particularly for low-proficiency Second Language (L2) learners. However, the advantage of shadowing for low-proficiency learners might have been an artifact of the statistical methods used, specifically the treatment of pretest scores as a measure of proficiency. To address this issue, the present study first replicated the observed advantage of shadowing for low-proficiency learners by employing listening tests. Additionally, two further analyses were conducted to determine whether the observed advantage was due to the analysis method: one analysis used a different type of test, while the other used vocabulary size as a measure of proficiency instead of pretest scores from a listening test. The results replicated the advantage of shadowing for low-proficiency learners using the original method. However, no such advantage was observed with the different type of test or when vocabulary size was used as the measure of proficiency. Thus, the present study suggests that shadowing may not be exclusively effective for low-proficiency learners but could benefit a broader range of learners with varying proficiency levels.
https://ror.org/01r3kcg18 
