In 2001 the IELTS interview format and criteria were revised. A major change was the shift from a single global scale to a set of four analytic scales focusing on different aspects of oral proficiency. This study is concerned with the validity of the analytic rating scales. Through a combination of stimulated verbal report data and questionnaire data, this study seeks to analyse how IELTS examiners interpret the scales and how they apply them to samples of candidate performance.
This study addresses the following questions:
- How do examiners interpret the scales and what performance features are salient to their judgements?
- How easy is it for examiners to differentiate levels of performance in relation to each of the scales?
- What problems do examiners identify when attempting to make rating decisions?
Experienced IELTS examiners were asked to provide verbal reports after listening to, and rating a set of the interviews. Each examiner also completed a detailed questionnaire about their reactions to the approach to assessment. The data were transcribed, coded and analysed according to the research questions guiding the study.
Findings showed that, in contrast with their use of the earlier holistic scale (Brown, 2000), the examiners adhered closely to the descriptors when rating. In general, the examiners found the scales easy to interpret and apply. Problems that they identified related to overlap between the scales, a lack of clear distinction between levels, and the inference-based nature of some criteria. Examiners reported the most difficulty with the Fluency and Coherence scale, and there were concerns that the Pronunciation scale did not adequately differentiate levels of proficiency.