Just when you thought it was not possible to gain attention with a teaching & learning paper that doesn't pivot around AI, fresh insights have arrived from left field, reminding us that public faith in universities depends (in part) on fundamental improvements in assessment measures.
Monash University’s Tim Fawns and colleagues Dave Boud and Phill Dawson from Deakin’s CRADLE team have written a cracker of a paper proposing a framework to better demonstrate and evaluate what a student has learned, in terms that even Future Campus can understand.
This is genuinely interesting stuff. The authors note that ‘doubt persists about whether students who pass these assessments possess the capabilities specified in the learning outcomes.”
They note that the credibility of degrees and the university system relies on effective and evidence-based assessment, with the implication being that improving assessment rigour is critical to building confidence in universities and degree qualifications.
“Interpretation of assessment evidence can misrepresent students’ achievements,” they note, after providing examples of scenarios where students could receive marks that are not a fair representation of achieving learning outcomes.
“Over-extrapolation from limited evidence, under-recognition of certain outcomes or reliance on inappropriate proxies all compromise defensible claims about student achievement,” they say.
The authors therefore propose a framework for improving the use of indirect evidence in four categories: product, process, performance and practice, identifying opportunities for implementation and also limitations.
“Standards-based assessment heightens the need for better evidence of the claims we make about student learning,” the authors write.
“Assessing learning outcomes is not simply a matter of crafting careful tasks or applying suitable evaluation methods. It requires deeper consideration of how learning is represented, evidenced and interpreted within an assessment system. The 4Ps framework (product, process, practice, performance) draws attention to the different kinds of proxies through which student achievement of learning is evidenced. In doing so, it shifts the conversation beyond superficial alignment to how assessment systems can generate credible inferences about what students know and can do.”