Professor Luke Prendergast is Deputy Dean of the School of Computing, Engineering and Mathematical Sciences and a Professor in the Department of Mathematical and Physical Sciences.
He is a statistician who specialises in robust statistics, meta-analysis and dimension reduction. Professor Prendergast collaborates with researchers across a range of disciplines, providing biostatistics expertise to health and wellbeing projects.
We asked him about his career and his contribution to FitSkills, a National Health and Medical Research Council (NHMRC)-funded partnership project that examines the impact of peer-mentored exercise programs for young people with disability. FitSkills is also part of the Australian Centre for Health, Independence, Economic Participation and Value Enhanced Care for adolescents and young adults with cerebral palsy, an NHMRC-funded Centre for Research Excellence.
Focus on application
“Being a statistician who is involved in scientific research is one of the most interesting and dynamic areas to contribute to.”
“While statisticians can work on research that develops new methodologies, we can also contribute our expertise to a huge range of other disciplines. As a statistician in the Department of Mathematical and Physical Sciences, I have been able to seek out collaborations where I can have an impact in areas that are interesting and meaningful to me. I have worked on several projects that improve human health and wellbeing and, in recent years, on projects that can positively impact the lives of children and young people with disability.”
“It’s been a privilege to be associated with FitSkills, a research project led by Chief Investigator and Professor of Physiotherapy, Nora Shields.”
“The aim of this project is to determine whether supervised exercise programs in local community gyms can improve the quality of life and exercise participation of young people with disability. It brings together researchers, disability partner organisations, local councils and community gyms. My role is to identify how much data we need to collect in clinical trials, and analysing that data when it comes in. I make the most of the datasets that are collected.”
Finding statistical strength
“A challenge for anyone seeking funding for clinical trials like FitSkills is to ensure that the sample size is suitable to test the hypothesis.”
“The larger the study, the more expensive it is. Studies that are too small do not provide enough evidence to substantiate the hypothesis and often lack diversity. The FitSkills project had promising data from a small sample version of the trial. Plans were then made for a new trial to study the effect of exercise across eight community gyms, using a staggered approach, which was an efficient way to undertake the research and account for things like seasonal effects. These variables are complex, and standard tools to determine an adequate sample size were not available at the time. I was able to use the pilot data to propose a suitable model and program a large simulation that, when run thousands of times, was able to help us to determine the sample size needed.”
Research with impact
“Multidisciplinary research teams can provide diverse outputs and insights.”
“One of our PhD students working on the project did a Master of Data Science at La Trobe. She has created an interactive web application consisting of a map of greater Melbourne, community gym locations, and locations of young people with disability who are registered with our partner organisations. The application can be used to determine the feasibility of a program like FitSkills, providing detailed analyses of time and distance to travel to a nearby community gym.
If a statistician came into our department in the junior stages of their career, I would say ‘go out and work with people.’ There is a new cohort of statisticians coming through who are interested in getting their hands dirty with real data. That’s exciting.”
Find out more about the Department of Mathematical and Physical Sciences.