In times of sector and curricular reform, university assessment policies assume heightened importance as key drivers of educational transformation (Medland 2016). While in recent years the emergence of generative artificial intelligence has driven a sharp focus on how to reform assessment practices to assure educational integrity (Lodge et al. 2023), there are few studies that explore how assessment more broadly can be a vital part of institutional transformation – and for improving student outcomes.
This presentation outlines a case study of how one Australian university engaged in substantial reform to their assessment policy and practice with a view to driving authentic, manageable and scaffolded assessment across the institution, and improved student outcomes. This assessment reform was a key facet of a larger curriculum change at the university, where the traditional delivery model of 13-week trimesters, lectures and examinations was replaced with an immersive block model that enables students to focus on just one or two units – and assessments – at a time over 6-week terms (Roche et al. 2024). The existing assessment parameters were no longer fit for purpose for this new model of learning, necessitating a complete rethink of how (and also why) we assessed our students. A number of long-held assumptions and practices around student assessment were upended, including assessment number, volume, timing, integrity, marking turnaround, feedback, and a reliance – arguably an over-reliance – on face-to-face invigilated examinations.
The presentation describes six assessment principles that were implemented in policy to form the foundation and framework for assessing learning in the immersive block model, and the guidance that was embedded in policy to delineate how each principle should be approached. Assessment procedures and guidelines were also instituted to drive, at a practical level, how assessment should take effect in the classroom.
Data evidencing how these changes have affected student outcomes are also presented. Comparing student grades in the traditional trimester model (n = 47,120) with those in the immersive block model (n = 32,062), a 10.8% increase in pass rates was observed. We also found statistically significant improvements over time in student satisfaction (n = 9,752) with assessment tasks, objectives and performance standards, workload, the development of valued skills and attributes, and units overall. A key implication is that deep assessment reform aimed towards increasing the authenticity, manageability and scaffolding of assessment can make a significant positive difference to student achievement and satisfaction at a whole-of-institution scale.