146 / 2018-07-11 11:35:52

Predictive Validity of Retelling across Languages in Aptitude Testing for Interpreting

Predictive validity; retelling across languages,construct, aptitude testing

全体主题 > 与语言测试与评价相关的其他议题

As a form of aptitude testing widely adopted by the major international interpreter training institutions, retelling across languages has been long considered as a valid test, capable of testing candidates on their linguistic, cognitive and communicative abilities. However, most of the existing research on this topic is only derived from the consensus among or the professional experience of the researchers. There is still a paucity of grounded research on the constructs and the validity of aptitude testing for interpreting, retelling across languages in particular. Drawing on the relevant theories and knowledge in the field of cognitive psychology as well as utilizing an expert interview, this paper first investigates the constructs of retelling across languages as well as designs an analytical rubric and band descriptors for grading. Through a longitudinal experimental study across four interpreter training schools in China, it then tries to explore the predictive validity of the retelling test as a whole as well as to what extent each of the four major parameters (fidelity, language, coherence and delivery) predicts the trainees’ end of semester interpreting performance. Using multiple linear regression analysis, the research finds that retelling as a whole is a highly predictive test, with language proficiency as a more powerful predictor (β=.44) than trainees’ performance on fidelity(β=.30) in Chinese to English retelling and fidelity as a more powerful predictor(β=.43) than trainees’ performance on language(β=.30) in English to Chinese retelling. It is hoped that the findings of this paper will shed light on the understanding of the intrinsic characteristics of aptitude testing and serve as a valuable reference for the design, implementation and assessment of aptitude testing and interpreter training at large.