Methodology
The Centre for Population Health is guided by principles that place real-world relevance, diversity of experience, and methodological integrity at its core. Here you can read more about the values and approaches that shape the way we think and work.
Methodology at the Centre for Population Health
The Centre for Population Health conducts research that aims to be relevant – not only to science, but to the people it concerns. This shapes how we work, from how we formulate our research questions to how we structure collaborations and interpret findings.
A core element of our working method is a broad understanding of what kinds of experience are relevant to research. We draw on Killackey's 4L framework, published in The Lancet Psychiatry in 2023, which distinguishes between four types of experience: lived (personal experience of mental illness), loved (the experience of carers and family members), laboured (professional experience of working with people with mental illness), and learned (scientific and academic knowledge gained through education and research). The framework was developed to counter tokenistic forms of involvement, and rests on the idea that different forms of experience complement one another and together strengthen the quality and relevance of research. For us, this functions as the DNA of our working method: we actively seek to incorporate all four experiential dimensions in the design and conduct of our projects, treating these perspectives as equally valid contributions rather than inputs ranked in a hierarchy.
Another hallmark of our work is that we do not shy away from scientific disagreement - we use it constructively. We engage in adversarial collaboration, a research model in which researchers who disagree on an empirical question join forces to design and conduct a study capable of resolving, or at least clarifying, the nature of that disagreement. The concept is particularly associated with Daniel Kahneman, who argued that this represents a more productive way of handling scientific controversy than the conventional cycle of critique and rebuttal in journals - where parties rarely change their positions. The core of the model is that the parties agree in advance on what is to be tested, which methods and analyses will be used, and what outcomes would support each position. A neutral third party is often involved as mediator or arbiter. Results are published jointly, typically with separate commentaries in which each party interprets the findings. The aim is to discipline disagreement methodologically: once both parties have approved the design before data are collected, neither can subsequently dismiss the results as methodologically flawed or biased. This reduces the scope for motivated reasoning and what Kahneman called angry science. The model has gained renewed relevance in the context of the replication crisis and the rise of preregistration, which share much of the same underlying logic. Experience suggests that adversarial collaborations rarely end with anyone fully conceding defeat, but they consistently produce more precise delineations of what the disagreement actually consists of - and data that both sides must contend with.
Taken together, these ways of working reflect an ambition for research that is open, courageous, and grounded in real lives.
Further reading on the methodology can be found in the references below:
Kahneman, D. (2003). Experiences of collaborative research. American Psychologist, 58(9), 723–730.