We ran our 18th annual COMPASS Colloquium on Monday 26 August 2024, hosted by StatsNZ at 8 Willis Street, Wellington. Please see the presentations from the day below.
Presentations
Introduction to COMPASS
Associate Professor Barry Milne, Director, COMPASS Research Centre
Mapping Settlement Trajectories of 25,000 People from Refugee Backgrounds in New Zealand
Professor Jay Marlowe, Centre for Asia Pacific Refugee Studies, University of Auckland
This study presents longitudinal income trajectories and mobility patterns of nearly 25,000 adults from refugee backgrounds, for four distinct subgroups: quota, convention, family reunification, and asylum seeker. Using data from the Integrated Data Infrastructure (IDI), I describe these groups’ demographic and socioeconomic characteristics and show outcome disparities with the general New Zealand population and other immigrants. Several policy implications for supporting positive settlement outcomes are suggested, particularly emphasising the first four to six years of settlement.
Long-Term Health Outcomes for Rugby Union Players: Results From the Kumanu Tāngata Study
Dr Stephanie D’Souza, COMPASS Research Centre
The Kumanu Tāngata retrospective cohort study uses linked historical rugby register and health data within the Integrated Data Infrastructure to investigate the long-term health outcomes associated with participation in high-level rugby. Players (n = 12,861) are compared to males from the general population (n = 2,394,300), matched by age, ethnicity, and birthplace. This presentation will showcase the findings on neurodegenerative disease and mortality (all-cause and cause-specific) outcomes.
Life-Course Impact of Chronic Health Conditions: A Family and Whānau Perspective
Dr Lisa Underwood, COMPASS Research Centre
The Life-Course project aimed to understand challenges facing New Zealand families affected by Long Term Health Conditions (LTHCs). We analysed data from over 3,000,000 people, living in 1,000,000 families. Here we present findings on individuals, families, and children affected by LTHCs, and outcomes for partners of people who experienced a stroke or traumatic brain injury, parents of autistic children, and young carers.
The Better Start Microsimulation Model
Associate Professor Barry Milne, Director, COMPASS Research Centre
An important question to ponder when considering a policy change or intervention is, “Will it work?” It is expensive to try it out in the real world, but simulation can model the impact on a ‘virtual’ world. This presentation describes microsimulation – a specific form of simulation, and The Better Start Model, which models interventions undertaken as part of the National Science Challenge: “A Better Start”.
Addressing Linkage Errors in the IDI
Eileen Li, COMPASS Research Centre
Probabilistic linkage introduces linkage errors due to internal and external factors. In many instances, research and analysis conducted using IDI data assume perfect linkage without considering the potential impact of linkage errors. This talk discusses potential approaches to identifying linkage errors using Census 2013 and Education datasets and illustrated the effect of linkage errors on linked data analysis.
A Legacy of Survey Research
Martin von Randow, COMPASS Research Centre
Alongside our large scale work with administrative data and the IDI, we have a long history of data collection, as well as a focus on the preservation of survey data. From medical care surveys predating COMPASS itself and the inception of the New Zealand Social Science Data Service, we have gone on to run the New Zealand Election Study and the International Social Survey Programme for New Zealand, as well as a number of smaller surveys with community collaborators.
Novel Applications of the IDI: Adding Longitudinal Capability to Te Kupenga
Tori Diamond, iNZight Analytics
This project extends Te Kupenga, a nationally-representative post-censal survey of Māori with culturally informed variables, by linking to administrative data sets at different time periods. Outcomes include Ambulatory Sensitive Hospitalisations and COVID-19 vaccinations, using individual, household and geographic determinants. This creates robust statistical processes to transform the survey into a cohort study. The project uses wellbeing indicators to evaluate potential impacts on Māori health inequities.
Gastric Cancer Incidence Among Level 2 Pacific Ethnicities Using the IDI
Nicole Satherley, iNZight Analytics
This presentation outlines findings from a research project on cancer incidence and mortality reporting for Level 2 Pacific ethnicities using the IDI. Administrative data were used to provide additional information on ethnicity, and potentially relevant variables not measured in the New Zealand Cancer Registry, including country of birth and migration. Strengths and limitations of the methods used, particularly when working with small population sizes, are discussed.
Testing for Treatment Inequities in Multiple Sclerosis Using Integrated Data
Natalia Boven, COMPASS Research Centre
Multiple sclerosis is a chronic inflammatory central nervous system disease. We used the IDI to test for treatment inequities in prescriptions of disease-modifying therapies among those with service use suggestive of multiple sclerosis. Disparities in prescriptions for disease-modifying therapies were evident across ethnic groups and levels of socioeconomic position.