Number of Citations: 0

https://doi.org/10.29060/TAPS.2025-10-4/TT004

Lambert Schuwirth

Professor of Medical Education
NewMedSchool, Australia

We are currently living in a time in which different assessment paradigms co-exist. There are contexts in which assessment is purely seen as a measurement, for instance national licensing. There are contexts in which assessment is predominantly judgement, such as in VIVAs and there are contexts in which these are combined to form an integral programmatic assessment for learning program.

Although a programmatic assessment for learning aims to combine both measurements and judgements, the manner in which all assessment data is synthesised and fed back to the learner is always through judgement and narratives. That is inevitable because numerical outcomes without added narratives are as meaningless as a scientific paper without an introduction, methods and discussion section and only the tables of the results. Numerical outcomes can only drive learning purely by punishment and reward and cannot provide the learner agency or support with meaningful self-regulation of their learning. Modern professionals, however, need capabilities to self-regulate their learning.

Assessment for learning therefore always needs to be programmatic – as opposed to formative assessment which can simply be a test that does not count.

But are narratives defensible? This is a common concern as we tend to trust quantitative outcomes more than qualitative. I find this odd. We trust healthcare as a system, even though the history, physical examination results, pathology and imaging reports and even contextualised lab values are all narratives, as are the diagnosis and management plan.

In an educational context, I would therefore plea that assessment should be more like a diagnostic and therapeutic relationship with the medical student. Assessment should guide them to become the best doctor they can be, and with our intake of the brightest and most hard-working young people the vast majority are. But what about the minority? Yes, they need to be identified as well but focusing an entire system on an issue with a low prevalence (the irremediable student) creates a huge NNT problem and is a waste of money and resources.

But assessment needs to be credible and fair. That is where programmatic assessment differs from a testing approach. Testing defines fairness as equality – everybody receives the same standardised test. Programmatic assessment defines fairness as equity – everybody receives the same quality of assessment. Just like we don’t push all patients through the same template in healthcare, but offer bespoke and high-quality diagnostic and therapeutic care, we should do the same with assessment.