Nationally representative estimates of food insecurity during COVID-19: An investigation of the Food Insecurity Experience Scale in Australia

The purpose of this paper was to measure the prevalence of food insecurity (FI) in Australia during the COVID-19 pandemic using the 8-item Food Insecurity Experience Scale (FIES). Employing the Household, Income and Labour Dynamics in Australia (HILDA) survey, Rasch models alongside a discrete method were used to investigate the severity of FI and robustness of the 8-item FIES during 2020-2021. Our results indicate that during 2020-2021, 6.47% of Australians experienced moderate or severe FI and 2.93% experienced severe FI. Validation tests following United Nations guidelines show similar weighted and unweighted results and consistency with the discrete method. Infit and outfit metrics, alongside measures of reliability and residual correlations, were all within the expected range. We conclude that the FIES administered as part of Wave 20 of the HILDA survey was a suitable instrument for measuring the prevalence of FI at a representative population level. Although Australia has been measuring FI for 40 years, it has been characterised by measurement inconsistency, inadequacy and infrequency, resulting in insufficient data for policymaking. Reliable representative population estimates of the extent of FI in Australia provides critical intelligence for an effective policy response.