The Impact of Task Types and Rating Methods on Iranian EFL Learners’ Speaking Scores

Abstract

Speaking assessment is quite challenging as there are many factors in addition to speakers' ability contributing to how well someone can speak a language. In this study, 40 male and female upper intermediate EFL learners, with age range of 15-26, selected through ECCE test. They performed four different types of speaking tasks (explaining, problem-solving, story-telling, and picture-describing). These tasks were rated by two raters using two scoring methods: holistic and analytic. The one-way repeated measures ANOVAs, paired-sample t-tests and Pearson Product Moment Correlations illustrated that the task types and rating methods didn't have any significant effects on learners' speaking scores as far as problem-solving, explaining, and picture-describing tasks were concerned. Yet, the rating methods represented to have some effects on story-telling task. The findings also indicated a significant correlation between holistic and analytic ratings of the problem-solving, picture-describing and story-telling tasks while the reverse was true for the explaining task.