The use of clinical simulations in assessment
Article first published online: 27 OCT 2003
Volume 37, Issue Supplement s1, pages 65–71, November 2003
How to Cite
Schuwirth, L. W. T. and Van Der Vleuten, C. P. M. (2003), The use of clinical simulations in assessment. Medical Education, 37: 65–71. doi: 10.1046/j.1365-2923.37.s1.8.x
- Issue published online: 27 OCT 2003
- Article first published online: 27 OCT 2003
- Received 31 March 2003; editorial comments to authors 13 June 2003; accepted for publication 18 July 2003
- education, medical, undergraduate/*methods;
- *patient simulation;
- clinical competence/*standards/economics;
- educational measurement
Context Simulation-based testing methods have been developed to meet the need for assessment procedures that are both authentic and well-structured. It is widely acknowledged that, although the authenticity of a procedure may be a contributing factor to its validity, authenticity alone never is a sufficient factor.
Aim In this paper we describe the mainstream development of various simulation-based approaches, with their strengths and weaknesses. The purpose is not to provide a review based on an extensive meta-analysis but to present crucial factors in the development of these methods and their implications for current and future developments.
Method The description of these simulation-based instruments uses a subdivision according to the layers of Miller's pyramid. Written and computer-based simulations are aimed at measuring the ‘knows how’ layer, observation-based techniques such as standardised patient-based examinations and objective structured clinical examinations target the ‘shows how’ layer and performance practice measures assess performance at the ‘does’ layer.
Conclusion In all simulations, case specificity was found to pose the most prominent threat to reliability, while too much structure threatened to trivialise the assessment. The conclusion is that authentic and reliable assessment is predicated on a wise balance between efficiency and adequate content sampling.