Native- and Nonnative-English-Speaking Raters’ Assessment Behavior in the Evaluation of NEAT Essay Writing Samples

This study examines the assessment behaviors between Native-English-speaking rater (NESR) and Non-native-English-speaking rater (NNESR) in the evaluation of National English Ability Test (NEAT) writing samples. Their assessment behaviors are similar in overall achievement but differ from one domain to another. The participants of this study were 110 freshmen high school students in Gyeonggi, Korea, who learned English essay writing from a NNES teacher for six months. 110 students received an instruction of an English-persuasive-essay-writing, and each participant composed a five-paragraph persuasive essay given by NEAT before and after the instruction. Both NESR and NNESR assigned the scores on each student’s essay and analyzed the data based on the given criteria composed of the following four domains in NEAT writing tasks: (1) task completion, (2) content, (3) organization, and (4) language use. Findings portray that NESR assigned a relatively low score and was stringent about three domains of content, organization, and language use while NNESR was generous in grading all four domains. The results showed that NESR and NNESR demonstrated different assessment behaviors about each domain due to diverse variables. The findings suggest that ways to improve the reliability between NESR and NNESR need to be further studied.