Open Access Research article

A model of the pre-assessment learning effects of assessment is operational in an undergraduate clinical context

Francois J Cilliers1*, Lambert WT Schuwirth2 and Cees PM van der Vleuten3

Author Affiliations

1 Centre for Teaching and Learning, Stellenbosch University, Private Bag X1, Matieland 7602, South Africa

2 Health Professional Education, Flinders Innovations in Clinical Education, School of Medicine, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia

3 Department of Educational Development and Research, University of Maastricht, Universiteitssingel 60, 6229 ER Maastricht, The Netherlands

For all author emails, please log on.

BMC Medical Education 2012, 12:9  doi:10.1186/1472-6920-12-9

Published: 16 March 2012



No validated model exists to explain the learning effects of assessment, a problem when designing and researching assessment for learning. We recently developed a model explaining the pre-assessment learning effects of summative assessment in a theory teaching context. The challenge now is to validate this model. The purpose of this study was to explore whether the model was operational in a clinical context as a first step in this process.


Given the complexity of the model, we adopted a qualitative approach. Data from in-depth interviews with eighteen medical students were subject to content analysis. We utilised a code book developed previously using grounded theory. During analysis, we remained alert to data that might not conform to the coding framework and open to the possibility of deploying inductive coding. Ethical clearance and informed consent were obtained.


The three components of the model i.e., assessment factors, mechanism factors and learning effects were all evident in the clinical context. Associations between these components could all be explained by the model. Interaction with preceptors was identified as a new subcomponent of assessment factors. The model could explain the interrelationships of the three facets of this subcomponent i.e., regular accountability, personal consequences and emotional valence of the learning environment, with previously described components of the model.


The model could be utilized to analyse and explain observations in an assessment context different to that from which it was derived. In the clinical setting, the (negative) influence of preceptors on student learning was particularly prominent. In this setting, learning effects resulted not only from the high-stakes nature of summative assessment but also from personal stakes, e.g. for esteem and agency. The results suggest that to influence student learning, consequences should accrue from assessment that are immediate, concrete and substantial. The model could have utility as a planning or diagnostic tool in practice and research settings.