Distinguishing features in scoring L2 Chinese speaking performance: How do they work?

Abstract

For Chinese as a second language (L2 Chinese), there has been little research into distinguishing features (Fulcher, 1996; Iwashita et al., 2008) used in scoring L2 Chinese speaking performance. The study reported here investigates the relationship between the distinguishing features of L2 Chinese spoken performances and the scores awarded by raters to the performances using holistic scoring. Seven distinguishing features – representing four major categories of Pronunciation, Fluency, Vocabulary and Grammar in the L2 Chinese speaking construct – were employed. An L2 Chinese speaking test was developed to assess the overall communicative ability in L2 spoken Chinese within an academic context. Speech samples of 66 candidates on the speaking test (i.e. 9 minutes’ speech length for each candidate) were analyzed in terms of the seven distinguishing features, with correlations and standard multiple regression being employed. Results showed that, first, each of the seven distinguishing features was significantly correlated to the scores, producing large or medium effect sizes; second, 79% and 77% of the variance in the scores could be explained by the distinguishing features (incorporating word tokens and word types respectively) in two regression analyses. The current study has established a link between distinguishing features and scores, contributing empirical evidence of candidate performance to the validation of assessing speaking proficiency in the L2 Chinese context.