COMPASS seminars & events

COMPASS seminar series

COMPASS's seminar series for 2018 will take place Wednesdays 2-3pm in the Pat Hanan room of the Arts 2 Building (207.501). We will be sharing weekly allocations with the Department of Politics and International Relations, supporting each other's speakers on a roughly alternating basis.

COMPASS events

Distinguished Visitor: Professor Julia Lane of New York University gave a seminar, a public lecture, and a radio interview during her time at UoA in June.

Big Data Day: Dr Nichola Shackleton organised a day of presentations on projects involving Big Data, 28 June.

Epidemic polio mortality in New Zealand, 1916–1949


Dr Heather Battles

18 July 2018 

Poliomyelitis emerged as an epidemic disease in the late 19th and early 20th centuries. The traditional model of its epidemiology is based on the hygiene hypothesis, which links the rise of polio to improved hygiene and sanitation and reduced crowding. However, according to the Intensive Exposure (IE) hypothesis, exposure to the poliovirus within the home leads to increased severity of infection; this makes crowding an important risk factor.

New Zealand polio mortality data for 415 individuals were collected from non-Maori death registrations for the epidemic periods of 1916, 1924–25, 1936–37 and 1947–49. Analysis to date shows mixed support for the IE hypothesis. I compare these results to previous findings from a study of polio mortality in southern Ontario, Canada. I also take a closer look at the 1916 epidemic, which resulted in over 1,000 notified cases and 125 deaths among Pākehā alone, in addition to an unknown number of cases and deaths among Māori. Scholarship on the once-forgotten 1918 influenza pandemic has illuminated how intimately linked that disease was to the War. Here I investigate the evidence for biological and social connections between the War and the 1916 polio epidemic.

Heather Battles completed her BA in Anthropology and History at the University of Victoria in BC, Canada, in 2005, before moving to Ontario for graduate studies. She completed her Masters in the Anthropology of Health and her PhD in Biological Anthropology, both at McMaster University. Her doctoral dissertation used historical records to examine the shifting social, geographical, and demographic patterns of polio mortality in southern Ontario in the early 20th century. She took up her current position in Biological Anthropology at the University of Auckland in 2014, beginning her ongoing research into polio mortality in New Zealand. She takes an inter-/multi-disciplinary approach to the study of epidemics, combining historical demography, infectious disease ecology, medical anthropology, and social history.

Nevertheless, She Disrupted: The Development of New Pasteurs


Professor Julia Lane (with Britta Glennon, Ridhima Sodhi, and Matt Ross)

20 June 2018

Little is known about the factors contributing to successful innovation because there is little data about the individual factors. Hypotheses have included:

  • Supportive networks and collaboration
  • Organisational environment
  • Levels of grant funding
  • Large and diverse teams.

Analyses have also been plagued by selection bias. Research funding offers a useful case study because new rich data including longitudinal information on individuals, projects and organisational structure has recently become available. This paper makes use of an interesting change in innovation patterns by gender to shed light on factors contributing to innovation. While women patent at a rate that is 40–50% of that of men, there have been increasing rates of patenting among young women, and increasing rates of patenting among female grant-funded inventors.

We combine the new data with new measures of disruptive patents to identify "New Pasteurs". We find that female "New Pasteurs" work in larger teams, are more likely to work with other female "New Pasteurs" and are more mobile.

Julia Lane is a Professor at NYU Wagner, the NYU Center for Urban Science and Progress, and an NYU Provostial Fellow for Innovation Analytics. Her research focuses on big data’s role in government and public policy. Previous to this, Dr Lane initiated and led the creation and permanent establishment of the Longitudinal Employer-Household Dynamics Program at the U.S. Census Bureau. This program began as a small two year ASA Census Bureau fellowship and evolved into the first large-scale linked employer-employee dataset in the United States. It is now a permanent Census Bureau program. Dr Lane has published over 70 articles in leading economics journals and authored and edited numerous books. She has been the recipient of more than $75 million in grants from foundations, government agencies, and international organizations. Dr Lane is the recipient of the 2014 Julius Shiskin award and the 2014 Roger Herriot award.

Using the Index of Multiple Deprivation to Understand Drivers of Deprivation: Waikato case study

Dr Daniel Exeter

Associate Professor Dan Exeter and Rachael McMillan

30 May 2018

The Index of Multiple Deprivation (IMD) is a new measure of area deprivation for New Zealand. In this talk, the IMD and its development is briefly outlined and compared to NZDep2013, a census-based index of deprivation, before we explore the drivers of deprivation in the Waikato Region. Our analysis shows that overall, the IMD and NZDep2013 provide similar depictions of the Waikato Region however, a closer investigation using the Domains of the IMD suggests that the Waikato comprises some very disparate communities.

At the sub-regional level, no two communities have the same mix of drivers and some experience significant deprivation. This work suggests that central government, districts, social providers and others will need to consider the different drivers in each locality and how the underlying drivers work together to deepen deprivation in communities. Interventions will therefore need to be targeted to address the unique factors in each community.

Dan Exeter works in Epidemiology and Biostatistics at the School of Population Health, the University of Auckland. He is a quantitative health geographer and has a background in Geographical Information Systems and spatial analysis. Dan's research uses large data sets such as the census or routine health databases to identify the occurrence of, or potential solutions to, inequalities in health. Dan's current research interests focus on the use of data zones and the Index of Multiple Deprivation to explore health and social outcomes. He leads a Marsden-funded project investigating measures of socioeconomic position among older people, and is a co-investigator on the VIEW2020 HRC programme, investigating inequities in CVD treatment, management, and outcomes using big data.

Modelling career trajectories of cricket players using Gaussian processes


Oliver Stevenson

23 May 2018

In the sport of cricket, variations in a player's batting ability can usually be measured on two scales:

  • Short-term changes observed during a single innings, which can span multiple days; and
  • Long-term changes observed over entire playing careers, which can span decades.

To measure short-term, within-innings variation, a Bayesian survival analysis method is derived and used to fit a model that predicts how the batting abilities of professional cricketers change during an innings. A second model is then fitted using a Gaussian process to measure and predict between-innings variations in ability. Given the high dimensionality of the Gaussian process model, and for ease of model comparison, models are fitted using nested sampling.

Generally speaking, the results support an anecdotal description of a typical sporting career. Young players tend to begin their careers with some raw but undeveloped ability, which improves over time as they gain experience and participate in specialised training and coaching regimes. Eventually, players reach the peak of their careers, after which ability tends to decline. However, continual fluctuations in ability are commonly observed for many players, likely due to external factors such as injury and player form, which can lead to multiple peaks during a long career.

The results provide more accurate quantifications of a player's batting ability at any given point of their career, compared with traditional cricketing metrics. This has practical implications in terms of talent identification and team selection policy.

Oliver Stevenson is currently a PhD candidate in the Department of Statistics at the University of Auckland. He completed a Bachelor of Science degree in 2014 at the University of Otago, majoring in statistics and minoring in psychology. In 2015 he returned to his hometown of Auckland, where he has since completed Bachelor of Science (Honours) and Master of Science degrees, at the University of Auckland. His main interests are in statistical applications in sport, particularly in cricket, which is the main focus of his PhD work.

(4.0 MB, PDF)

Ending the cardiovascular disease epidemic in New Zealand


Dr Corina Grey

16 May 2018

Despite steep declines in death rates over the past 40 years, cardiovascular disease remains a leading cause of mortality and an important contributor to differences in life expectancy between Māori, Pacific, and non-Māori non-Pacific New Zealanders.

Recent studies have demonstrated declines in the incidence and mortality of ischaemic heart disease and stroke in all ethnic groups, but there are indications that declines have been slower in Pacific people, and particularly Pacific women. If current trends continue, the already significant relative inequities in cardiovascular outcomes for Pacific people will widen, with adverse societal consequences.

This presentation will discuss recent research demonstrating ethnic inequalities in cardiovascular outcomes and put forward some ideas on how we, as researchers, scientists, and public health advocates, could initiate changes that may help to accelerate declines in cardiovascular disease for Māori and Pacific people.

Corina Grey is a public health physician and Research Fellow at Waitemata and Auckland DHBs and the University of Auckland (Section of Epidemiology and Biostatistics). She is part of the VIEW (Vascular Informatics Using Epidemiology and the Web) research team, and her research is focused on the epidemiology of acute coronary disease.

(1.2 MB, PDF)

Uses and Application of Qualitative Research Methods in Policy Evaluations


Assistant Professor Alasdair Jones

28 March 2018 

I will introduce a programme of work I will be undertaking during my visit to the Public Policy Institute (in conjunction with COMPASS) at the University of Auckland. This work will comprise a review and synthesis of existing approaches to research design for public policy evaluation that integrate a qualitative research component. My interest in this area stems from my involvement in a mixed-methods evaluation of the public health impact of a concessionary bus and tram fare scheme for school children in London (the ‘zip card’). Out of this study, my colleagues and I became interested in how the integration of quasi-experimental and inductive designs in evaluation might be conceptualised. I will make the case for one particular conceptual approach to thinking through the relationship between qualitative analysis and causal inference in evaluation. This argument will be grounded in a discussion of some of the findings of the ‘zip card’ evaluation.

More information on this work can be found at

Alasdair Jones is an Assistant Professor in Qualitative Research Methodology at the London School of Economics and the Programme Director for the Department of Methodology’s MSc Social Research Methods. He is also an Associate at LSE Cities. He is a sociologist by training, and has conducted research in a number of thematic areas including urban public space, the transport-health nexus, experiences of living in planned developments, and social care. These studies have involved a variety of methodological approaches and Alasdair has particular interests in the use of qualitative methods in evaluation studies.

Alasdair is visiting the University of Auckland on a UK National Centre for Research Methods ‘International Visitor Exchange Scheme’ award. Prior to working at the LSE, Alasdair held research positions at the University of Hertfordshire and the London School of Hygiene & Tropical Medicine. He has also previously undertaken a Fulbright Scholarship at the Center for Ethnographic Research, UC Berkeley, and a Visiting Fellowship at the City Futures Research Centre, UNSW.

POPNZ – People’s Online Panel for New Zealand


Dr Barry Milne

28 February 2018

Online panels are samples of individuals with whom different online surveys are conducted over time. Currently all online panels in New Zealand are ‘non-probabilistic’ (opt-in) panels, populated by individuals who volunteer their services. The statistical validity and robustness of findings from non-probabilistic panels have been questioned. ‘Probabilistic’ online panels overcome these statistical shortcomings by randomly sampling individuals from the population of interest.

I will describe a new, national-probability online panel we are in the process of establishing: the People’s Online Panel for New Zealand (POPNZ). POPNZ is a survey panel service that academics, government, and other appropriately qualified researchers conducting public good research can utilise to understand the opinions of a representative panel of New Zealanders. I will describe the characteristics of this panel, how it can be used, and the interest researchers are already showing in POPNZ.

Dr Barry Milne is the Director of COMPASS Research Centre. He has a Masters degree in Psychology from the University of Otago, and a PhD in Psychiatric Epidemiology from Kings College London. His main interests are in life-course research, survey research, and the use of large administrative data sets to answer policy and research questions.


Seminars archive