One of the problems which workload modelling creates is the ubiquitous use of ‘clock-time’ as the sole measure of work activities. Models typically make use of an annual quota of a given number of hours and as such are using only clock time as a way of calculating the amount of work that people do. This is a very abstract conceptualization of time which does not consider how that time is distributed across the year or how much work can be expected with any particular hour or day. Here, I consider three additional characteristics of academic work which will impact on people's perceptions and experiences of the amount of work they're doing. They are rhythm, the contrasting temporal characteristics of different academic activities, and the tendency for ‘shadow work’.
Work within an academic context includes responsibility for a whole series of different activities, each of which has a natural rhythm. For example, if we are teaching part of a module then we will probably teach at the same time each week over a period of time, for example 2 hours every Thursday afternoon over a course of 10 weeks. There is a natural rhythm to the activity being undertaken and may extend to the time taken to plan and organise for the sessions, as well as the time needed for tutorials and assessments. As such, all the activities that relate to that work create a rhythmic character. However, this is only one group of activities out of the large number of activities that an academic will be responsible for. Each activity will have a different rhythm which characterizes the work that needs completing and the various activities will begin and end at different points across the year. Therefore, as well as thinking about the number of total hours we work over an academic year it becomes very important to consider the rhythmic nature of the various activities and hence the distribution of work over an annual cycle. This is especially the case for individuals who are working across more than one programme as it may be the case that the rhythms of the different programmes do not fit together particularly well leaving occasions where there is an unsustainable amount of work to do. For these individuals any given programme leader may not recognise the problem because in their sphere of influence the work for that individual is not terribly great. As a result, some academics can end up with periods during the year with extremely high temporal densities (Wajcman, 2013). In other parts of the year where rhythms do not occur together an individual may have a lower temporal density leading to a more manageable workload, but this will tend to be due to good luck rather than careful planning.
This means that given the rhythmic and complex nature of the activities academics are asked to undertake it is imperative that any approach to workload modelling not only accounts for the correct number of hours, but also considers the distribution of the work over the course of the year. This is extremely important as if it is the case that an individual does not have extended periods of low temporal density they may not be able to engage with research in a meaningful manner because they don't have the longer periods of time that are required to sustain this type of activity.
The last point above highlights another problem in trying to create a workload model. It needs to be able to differentiate between different activities each of which require a different relationship to time. Because modelling relies on clock-time it sees time as being homogenous and infinitely divisible, hence if we are completing an hour-long activity, modelling will simply assume that if we only had 10 minutes we could do a sixth of that amount of work. The type of work is not identified as an important contextual variable. As a result, it could be argued that if we have teaching throughout the year regularly leaving 2 or 3 hours available in the middle of the day then that free time could go towards the allocation an individual has for research. In other words, over the course of one week if there are 2 hours free in the middle of the day then it could be argued that 10 hours of research time is available. But obviously, whilst 10 hours might be a useful period to engage in research work, several 2-hour periods are not.
This is another issue workload modelling needs to overcome in relation to activities and how long would you spend on them; it needs a qualitative element that understands that different activities have a different form and tempo, some of which require extended periods of clock-time if they are to be useful. Half an hour in the middle of the day may be useful for checking and replying to e-mails. However, if I am to engage properly with an area of research then I will probably need at least half a day, and preferably one or more days of concerted focused effort, to ensure that I can begin to make sense of the research that I'm involved in. If these variable temporal characteristics are taken seriously this then begins to create a much more complex process to create some form of model that will be usable and is deemed to be acceptable by academics.
The final issue I identify above is that of what I call ‘shadow-work’. As suggested in a previous post, most workload models categorise academic work into one of three areas: research, teaching and administration. However, for many academics there is another important, and often time consuming, area of activity which tends not to be included within allocation models. It is not uncommon for pastoral care to be subsumed into teaching allocations with one or two hours per module for tutorials. But these are not pastoral occasions, they are opportunities for a student and a tutor to consider and extend work that is in process or has been finished so the student is able to develop their academic expertise within the subject.
Some students may have a range of problems beyond the discipline which require academics to spend extra amounts of time in a more pastoral-orientated role, trying to ensure that the students’ well-being is looked after. Models never capture this portion of work, so if tutors end up with four or five students over the course of the year with whom they spend a considerable period of time offering pastoral support, this is work which is not captured, creating ‘shadow-work’ within the academy.
Pastoral ‘shadow-work’ might be particularly intense for individuals leading modules and programmes. In most models these individuals have a larger allocation of time for the programme leader role, but these extra hours reflect the academic related administration that needs to be completed; it does not take into account the often quite considerable pastoral roles that these leaders fulfil. Consequently, this means that individuals who are running programmes may actually begin to face insurmountable workload problems at certain points during the year.
Workload allocation for pastoral roles is a very difficult area to begin to include within models because models themselves are predicated on trying to create fully efficient allocations of time for individual work. When it comes to pastoral roles there has to be an acceptance that much of the work is contingent on need and the associated activity will always suffer from being unpredictable, in one academic year there might be little additional work that needs to be done because the programme leader is fortunate to have few individuals encountering difficulties, but in another year there may be a great deal of pastoral work needing to be undertaken. If this is to be a truly reflective element of a workload allocation model it requires a level of contingent time to be included within the workload model for all staff and, in particular, for those responsible for a large programme or module.
The above reflections suggest that the nature of work allocation models needs to change, with serious consideration of a number of additional issues if they are to begin to move towards being truly reflective of academic work. Both the rhythm and the intensity of activities needs to be reflected in modelling to make academic timescapes sustainable, and in addition, activities which currently do not figure in many models also need to be identified and included, even where this means that the resultant allocations cannot be fully ‘efficient’.
Wajcman, J (2013) Pressed for Time: The Acceleration of Life in Digital Capitalism. Chicago: University of Chicago Press.