A comparative discourse study of simulated clinical roleplays in two assessment contexts: Validating a specific-purpose language test. (April 2016)

• Record Type: Journal Article
• Authors: Woodward-Kron, Robyn; Elder, Catherine
• Other Names: Elder Catherine guest-editor.
• Is Part Of: Language testing. Volume 33:Number 2(2016:Apr.)
• Page Start: 251
• Page End: 270
• Publication Date: 2016-04
• Subjects: Assessing language for specific purposes -- discourse analysis -- medical discourse -- Occupational English Test -- Objective Structure Clinical Examination (OSCE) -- roleplay interaction -- systemic functional linguistics; Language and languages -- Ability testing -- Periodicals; Language and languages -- Examinations -- Periodicals; 407.6
• DOI: 10.1177/0265532215607399
• Languages: English
• ISSNs: 0265-5322

Abstract

The aim of this paper is to investigate from a discourse analytic perspective task authenticity in the speaking component of the Occupational English Test (OET), an English language screening test for clinicians designed to reflect the language demands of health professional–patient communication. The study compares the OET speaking sub-test roleplay performances of 12 doctors who were successful OET candidates with practice Objective Structured Clinical Examination (OSCE) roleplay performances of 12 international medical graduates (IMGs) preparing for the Australian Medical Council clinical examination. The premise for the comparison is that the OSCE roleplays can represent communication practices that are valued within the medical profession; therefore a finding of similarity in the discourse structure across the OET and the OSCE roleplays could be taken as supporting the validity of the OET as a tool for eliciting relevant communication skills in the medical profession. The study draws on genre theory as developed in Systemic Functional Linguistics (SFL) in order to compare the roleplay discourse structure and the linguistic realizations of the two tasks. In particular, it examines the role relationships of the participants (i.e. the tenor of the discourse), and the ways in which content is represented (i.e. the field of the discourse) by roleplay participants. The findings reveal some key similarities but also important differences. Although both tests inevitably fall short in terms of authentic representation of real world interactions, the findings suggest that the OET task, for a range of reasons including time allowances, training of test interlocutors, and the limits of contextual information provided to candidates, constrains candidate topic exploration and treatment negotiation, compared to the OSCE format. The paper concludes with proposals for mitigating these limitations in the interests of enhancing the OET's capacity to elicit more professionally relevant language and communication skills.