نسیم قنبری

Since 1960s writing assessment has been considered as an important trend of inquiry by many researchers. This study aims at investigating the rating behaviors of 8 raters including 4 experienced and 4 novice raters, using Jacobs’ et al EFL Composition Profile. In fact, the main rationales behind this study are clarifying the times both experienced and novice raters refer to the rating scale, the stages they go through and number and types of the behaviors these raters show, finally ending in a quasi-model (trend) in an EFL context. Think Aloud protocols were used to codify the raters’ comments and statements in the rating process. Results showed that the novices referred many times to the scale after finishing their behaviors of looking for mistakes and before assigning the scores to each component of the scale separately; whereas, the experienced raters referred to the rating only one time at the end of their rating. It was also found that the experienced raters, underwent 7 rating stages, but the novice raters underwent only 5 stages lacking experienced raters’ pre-reading and post-scoring stages. Furthermore, the study indicated that the experienced raters’ number of behaviors was higher than those of the novice raters who adopted macro-strategies attending mostly to the lower-order aspects of essays while the experienced raters adopted micro-strategies attending mostly to the higher-order aspects of essays. By conducting this study we expanded upon previous research in several ways. The findings have implications for writing assessment practices and for further research.