Comparison of HILDA Survey Estimates With the 2006–2021 Censuses

Abstract

Issue

The Household, Income and Labour Dynamics in Australia (HILDA) Survey has provided researchers and policymakers with invaluable insights into Australian life for more than two decades. Nevertheless, nonresponse, attrition and shifts in population coverage can affect how representative the sample is over time.

Methods

We evaluate the HILDA Survey's representativeness by comparing its estimates against those from the Census across four time periods and four age groups.

Findings

We find strong alignment across many of the variables examined. For the most part, observed differences can be explained by differences in the questions asked, recall periods or different collection methodologies. Recent immigrants are a concern as they can only join the sample by living with a HILDA household or be included via a top-up sample. Sizeable differences are identified in the proportion of immigrants aged 15–24 and 25–44 in particular, which were corrected in 2011 with the addition of a general top-up sample.

Implications

We conclude that the HILDA Survey is largely representative of the Australian population and note that the coverage of recent immigrants is in the process of being addressed. Additional immigrant-specific samples are being added in 2024 and 2025 and regular immigrant sample top-ups are needed thereafter.