Aptitude in interpreting: predicting competences in a reliable way

摘要

Aptitude testing is a special type of assessment where performance is judged by a predictive, rather than a concurrent criterion. Aptitude tests are critical in admission procedures, where they help make decisions as to the candidate’s ability to acquire a skill within a given timeframe. The present study is part of a larger research project (2007-2010), aiming to investigate predictors of specific skilled linguistic behaviour, namely interpreting. We briefly discuss current aptitude tests (Timarová and Ungoed-Thomas, 2007), which are based on face validity/authenticity, but display little predictive validity (Gringiani, 1990). We then report on our replication of a study by Gerver et al. (1989) which identified several predictors. Five types of language processing ability tests (logical memory, text memory, cloze, error detection, synonym generation) were administered to 250 foreign language and interpreting students at Lessius University College (Antwerp) and Charles Unversity (Prague), and their performance was correlated with end-of-year exam results. Our findings do not confirm Gerver et al.’s conclusions: none of the assumed predictors came out as significant in our study. With more tests and a larger amount of students tested in the follow-up of this project, we want to check if this weak (or rather non-existent) significance remains the same, this is to say if we can increase ecological validity. We discuss the implications of this particular study and the more general specificities of aptitude research, including validity, reliability, score range restrictions and generalisability across training programmes. We finally want to find out if their exist common instruments of measurement that can be used to determine aptitude in both sign and spoken language interpreting (Shaw et al., 2008).