38 3.1 Optimising Model Design: Addressing Limitations in Models A report by the Athena Forum22 on the results of a survey of 1087 UK university STEM departments23 about the practice and use of Work Allocation Models (WAMs) summarises the negative characteristics of models as: • too rigid and crude • not flexible enough to accommodate changes • promoting a ‘bean counting’ approach • encouraging laziness in staff • overly dependent on who implements the model • unable to take account of differential working speeds • disruption of the model due to student factors • university not responsive to repeated overload. The HEPI report builds upon these indicators of poor practice, noting that a primary cause of poor mental health in higher education institutions is: ‘Excessive workloads and workload models which frequently under-count time necessary for fulfilling tasks, and many tasks prove invisible to the workload assessors’. 22 The Athena Forum was established in 2007/8 as an independent committee to provide an expert voice on issues of women in STEMM (Science, Technology, Engineering, Mathematics, and Medicine). Further, it reports that stress emerges through workloadmodels [which] fail to build in adequate ‘headroom’ for emergency cover, let alone time for activities that emerge during the year. For a model to be “fit for purpose” and optimised for an individual it should ensure that it can flex to accommodate changes, whether arising from disruption (such as that caused by the Covid-19 pandemic) or to changes in targets (such as higher than expected achievement in student enrolments). This therefore requires a model which can pull on accurate authoritative data and create scenarios in real-time, to see where overload can be mitigated. One emerging thread – which the Athena Forum report mentions as poor practice - is the over-dependence on who exactly implements the model. All the models reviewed include a lesser or greater degree of judgment and discretion on final allocations by the academic lead (a department head, for example). Some of this decision-making may inevitably be based on confidential management information and is, rightly, not within the purview of the department. This requires management decisions therefore to be consistent, fair, and where individual decisions on allocations are made, for decision-making to be evidence-based and open to appropriate challenge . 23 The survey resulted in 265 responses giving a total response rate of 24%.