This is a guest post from Dr Sylvia Anne Mackie, Lecturer in Research Communication at Swinburne University of Technology, and is based on her published work reviewing the available research into the mental health of PhD candidates.
There have been a number of recent research articles, and even more journalistic opinions shared about the experiences of PhD students in university environments, that have indicated that poor mental health is widespread, and that a high proportion of PhD students are at risk of developing common mental health problems.
My colleague Prof. Glen Bates and I conducted a scoping review partly to get ideas from the literature about improving supervision procedures in ways that could bolster PhD candidates’ mental health. As it turned out, though, we didn’t find much that was of immediate practical use—a lot of the research was about ‘risk factors’ or ‘potential stressors’, but not much was about what could be done to ameliorate them. This makes sense with hindsight because researchers are still working out the mental health risk factors in this arena, how they might relate to each other, how they should be classified, and so on.
In public health research, they say that you need to align the theory of the change with the theory of the problem (Stokols, 1996; Wong et al, 2013). And one of the things our review shows is that the theory of this problem is underdeveloped.
The supervisory relationship obviously represents a key difference between the staff-student relationships of undergraduate or coursework-based learning, and higher research degrees, and some studies have looked at ways in which it could potentially cause or exacerbate stress. There have been suggestions that problems may relate to supervisors’ lack of experience, to their inaccessibility or inattention to the candidate or to other aspects of their leadership capacity. Ascribing direct causality is problematic, though, for various reasons including the possibility that the supervisory relationship may just be providing a setting for other stress-causing aspects of the doctorate to become evident, rather than giving rise to the stress in the first place. So, it’s complicated.
To get more traction on the supervision issue, researchers have looked at research on organisational behaviour, in particular how leaders can reduce ambiguity and uncertainty in complex organisational settings. Studies using workplace criteria have focused more on PhD candidates as research workers within a wider system and their health-related rights as members of such systems. Workplace factors like ‘job demand’, (or workload) ‘financial duress’, ‘work-family conflict’ and ‘job control’ have indeed been found to be stressors in the doctoral context; in fact often they’re more salient than supervision (Levecque, et al, 2017).
There is widespread agreement that the supervisory relationship could present opportunities for awareness raising, destigmatising and promotion of mental health interventions. However, supervisors are not trained counsellors and the extent to which they should be held accountable for the mental health outcomes of their students is unclear. It’s also important to note that supervisors themselves are in a high-risk group for mental health problems (see this related blog post), as are academics in general, and much is unknown about how mental health problems play out in research ‘ecologies’ (Guthrie et al, 2017). So there’s a need to look at the ecosystem as a whole.
The literature also points to problems with ‘transparency of university processes’, such as ‘unwritten rules’, ‘unclear expectations’, ‘closed decision-making’, ‘poor procedural communication’ and ‘problems in departmental culture’ that affect PhD candidates’ experiences (Appel and Dahlgren, 2003). These suggest a need for changes in the university context that would lie beyond—and yet probably intersect with—the supervisory relationship. Indeed, the overlapping nature of these areas shows that the complexity of doctoral training systems could be muddying the water. We need more investigation into how these systems could be redesigned for healthier levels of clarity and openness.
Our final point is that trialled interventions have not been well aligned with what is known about the range of potential stressors in the doctoral environment. This is the knottiest part of the discussion but the underlying point is that programs and interventions need to tackle the things that can make a difference. If financial duress, role conflict, high workload and low role control are big risk factors for common psychiatric disorders among PhD candidates, shouldn’t universities be focusing more on ameliorating those? And if so how?