A comparison between holistic and analytic assessment of speaking

Abstract

—As speaking is an intricate productive skill, its evaluation raises many interesting issues. In this study, the main purpose is to find the most suitable procedure that is useful to assess speaking, by making a comparison between the holistic and the analytic scales. Thus, twenty oral performances were recorded and given to five raters who assessed them holistically and analytically, according to four distinct tasks at the Higher Institute of Languages in Gabès. The raters’ gradings were compared according to four basic parameters, namely the mean, the range, inter-rater and intra-rater reliability. Despite the fact that both methods yielded low reliability rates, it was clear that there were many differences between the two scales, since they reflected a discrepancy at the level of the four parameters as well as at the level of tasks. Apart from the differences, it was noted that the holistic scale is more useful, reliable and consistent as far as speaking assessment is concerned.