A Preliminary Study Of Rater Language Background In The English Writing Assessment

The current study investigated the differences between Cantonese, Putonghua and English speaking teachers in their assessment of English language writing in the Hong Kong context. The rating behaviors of the three groups (19 Cantonese-speaking ESL teachers, 14 Putonghua-speaking EFL teachers and 13 Native English-speaking (NES) teachers) were discussed with regard to their holistic scores, evaluative standards, and their scoring to writers carrying different identities. The One-way ANOVA results indicated no significant difference between the three language groups in their holistic scores. Also, writers'linguistic identities exerted insignificant effects on their scoring. However, the evaluative criteria for written compositions differed markedly among Cantonese, Putonghua and English speaking teachers. A qualitative analysis demon-strated that the rating criteria shared by Cantonese and Putonghua speaking teachers were similar, but diverged from those of their NES colleagues. Overall, Cantonese and Putonghua speaking teachers put more emphasis on linguistic and grammatical accuracy and textual organization, but failed to produce detailed feedback on the na-ture of grammatical errors made as compared to NESs who often provided detailed comments on individual errors. Moreover, English-speaking teachers focused more on intersentential discourse such as content, organization and academic appropriateness than accurate choice of words. Based on the research, implications and further re-search suggestions are provided.