Masoomeh Estaji; Reza Morad Sahraee; Monire Shahbaz
Abstract
The present study investigated repair fluency, one of the three main aspects of fluency in Skehan’s model (2003; 2009), across different proficiency levels using the giver test-taker data in a speaking test. There are three main aspects of fluency in the model: Speed fluency, breakdown fluency, ...
Read More
The present study investigated repair fluency, one of the three main aspects of fluency in Skehan’s model (2003; 2009), across different proficiency levels using the giver test-taker data in a speaking test. There are three main aspects of fluency in the model: Speed fluency, breakdown fluency, and repair fluency. To date, there is no research aiming to measure the speech fluency of foreign Persian test takers. Therefore, this study sought to achieve an appropriate rating scale for speech fluency. The study, in particular, has examined some parts of the repair fluency construct in the mentioned model, specifically repetition and false starts, and attempted to determine the extent to which repetitions and false starts varied across different Persian proficiency levels of foreign Persian test-takers. This study transcribed, rated, and labeled repetitions and false starts of 23 foreign Persian language learners taking a speaking test. The spoken data were classified into four proficiency levels according to the Persian Teaching Standard Reference (PTSR), namely pre-intermediate, intermediate, upper-intermediate, and advanced. The duration of recordings was 44 minutes in total (11 minutes for each level). The results were analyzed, using the Kruskal-Wallis test, and revealed that false starts and repetitions did not make a distinction across Persian proficiency levels. The findings imply that raters and rating scale designers do not need to consider repetitions and false starts for identifying or describing different Persian proficiency levels.
