Rating scale impact on EFL essay marking: A mixed-method study

Abstract

Educators often have to choose among different types of rating scales to assess second-language (L2) writing performance. There is little research, however, on how different rating scales affect rater performance. This study employed a mixed-method approach to investigate the effects of two different rating scales on EFL essay scores, rating processes, and raters’ perceptions. Four EFL teachers in Tunisia rated a set of 24 EFL essays silently and two subsets of four essays while thinking aloud using a holistic scale and then a multiple-trait rating scale. The essay scores were analyzed using G-theory while the think-aloud protocols were coded in terms of Cuming, Kantor, and Power's (Cumming, A., Kantor, R., & Powers, D. (2002). Decision making while rating ESL/EFL writing tasks: A descriptive framework. Modern Language Journal, 86 (1), 67–96.) rater decision-making scheme. The holistic scale resulted in higher inter-rater agreement. Raters employed similar processes with both rating scales. Raters were the main source of variability in terms of scores and decision-making behavior. These findings have implications for writing assessment practices and for further research.