An Investigation of Rater's Assessment of Test Takers' Speaking Skill: The Effect of Gender and Delivery Platform

Abstract The present study was conducted in order to examine the effect of gender on the rating regarding oral proficiency and also the difference between video and audio performances of the EFL test takers. Three main instruments were used to achieve the mentioned goal. 60 EFL learners studying in Kish Institute in Zanjan were randomly selected to take a sample of TOEFL test including only listening and speaking skills. Oxford Placement Test was also given as a standard placement test in order to make sure of the participants homogeneity in terms of English proficiency. The participants were assigned into three group of levels based on their scores including elementary, intermediate and advanced. The speaking section of the test was rated by the researcher of the study along with one other colleague. The learners' performance was recorded in audio and video recording format. The speaking test included two tasks including a description and a narrative one. The raters were female and male as the factor of gender was considered as one of the main variables of the study. The data were analyzed through different statistics including two-way ANOVA and independent sample t-test and also intra class correlation was done to be sure of the consistency of the rating process. The study revealed that gender of the rater can have a significant effect on the scores they award to their learners while rating an oral proficiency on a speaking test. Also, it shed light on a significant difference between the audio and video performances on a speaking test as the delivery platform. The findings of the study are hoped to be beneficial in the field of language assessments in the context of Iran and the researcher hoped to broaden the related literature particularly regarding Iranian EFL learners and raters.