Abstract
Aims: the assessment of individual competence in medical education is about finding a balance between having sufficient resources to make valid and reliable decisions and not using more resources than necessary. Sequential assessment, in which more resources are used for borderline performing candidates than for poorly or clearly satisfactorily performing candidates, can be used to achieve that balance. Although sequential assessment is commonly associated with larger groups of candidates to be assessed, in many practical settings numbers of candidates may be small.Objective: this article presents a single case design with a statistical model for the assessment of individual competence that can be used regardless of the number of candidates.Method: a worked example of a solution that can be used for an individual candidate, using simulated data in the zero-cost Open Source statistical program R version 4.0.5., is provided.Results: the aforementioned solution provides statistics that can be used to make pass/fail decisions at the level of the individual candidate as well as to make decisions regarding the length and timing of an exam (or parts thereof) for the individual candidate.Conclusion: the solution provided can help to reduce resources needed for assessment to a considerable extent while maximizing resources for borderline candidates. This facilitates both decision making and cost reduction in assessment.
Highlights
Apart from informing subsequent learning and practice, one of the functions of the assessment of competence in medicine and other high-stakes settings is to make valid and reliable decisions regarding progression, and longer exams are commonly associated with higher validity and reliability [1]
The standard in assessment in medical education is usually based on the concept of borderline candidate, that is: a candidate passing by the skin of their teeth, demonstrating a performance almost at the expected level with some minor issues that would not result in concerns about patient safety
The answer is the same to all three questions: these too are options in sequential assessment, and the setup of any assessment – sequential or not – should always be decided on in light of the context in which the assessment is to take place
Summary
Apart from informing subsequent learning and practice, one of the functions of the assessment of competence in medicine and other high-stakes settings is to make valid and reliable decisions regarding progression, and longer exams are commonly associated with higher validity and reliability [1]. Given tremendous pressures on our healthcare systems, not using more resources for assessment than necessary is imperative. Validity and reliability are not constant across the performance range [2]. Sequential assessment, referred to as sequential testing, can be used to address these issues and works as follows. Performance in an assessment can range from excellent to very poor and candidates demonstrating satisfactory or better overall performance (i.e., sufficiently competent) should pass while candidates demonstrating below-standard overall performance should not pass the assessment. The standard in assessment in medical education is usually based on the concept of borderline candidate, that is: a candidate passing by the skin of their teeth, demonstrating a performance almost at the expected level with some minor issues that would not result in concerns about patient safety. The further away a candidate’s performance from that standard, the fewer resources are needed to decide whether the performance meets the standard
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
