Accuracy of peer ratings on the quality of spoken-language interpreting.

The accuracy of peer ratings on students' performance has attracted much attention from higher education researchers. In this study, we attempted to explore the accuracy of peer ratings on the quality of spoken-language interpreting in the context of tertiary-level interpreter training. We sought to understand how different types of peer raters (experienced versus novice) would affect the accuracy of their ratings on three quality dimensions (fidelity, fluency and expression) of bi-directional interpretation (English-to-Chinese and Chinese-to-English). Our data analysis, based on a rater accuracy model, shows that: (a) overall, it was easier for the peer raters to score the English-to-Chinese interpretation accurately than the opposite direction and to assess fidelity accurately than fluency and expression; (b) the experienced peer raters scored the interpretations more accurately than their novice counterparts; and (c) the higher-quality interpretations were easier to be scored accurately than those of lower quality. However, the directionality of interpreting may moderate these relationships. These findings are expected to enrich and deepen our understanding of rater accuracy as a function of distal and proximal factors in the Brunswik's lens model, and to help us better organize peer assessment in the higher educational context. [ABSTRACT FROM AUTHOR]