Who responds? An examination of response rates to a national postal survey of Aboriginal and Torres Strait Islander adults, 2018-2019

Abstract

Abstract Background Evidence on the effectiveness of postal recruitment methods for Indigenous peoples is lacking. Mayi Kuwayu, the National Study of Aboriginal and Torres Strait Islander Wellbeing, uses multi-staged sampling. We aimed to test postal surveys as a primary recruitment method, analysing preliminary response rate data to inform the Study’s ongoing sampling approach. Methods Twenty thousand adults aged ≥16 years were sampled from Aboriginal and Torres Strait Islander people enrolled in the Medicare Australia Enrolment Database. We calculated response rates at 4 and 15 weeks, overall and by age group, gender, state/territory and remoteness. Results The overall response rate was 2.3% (n = 456/20000). Highest response rates were observed among males and females ≥50 years from major cities (6.0, 95%CI 4.4–7.9 and 5.5%, 4.1–7.2, respectively) and regional areas (6.0%, 4.6–7.6 and 6.2%, 4.9–7.7, respectively). Younger age groups and remote areas had lower response rates; all remote age groups < 50 years had a response rate ≤ 0.6%. While most participants responded on the paper surveys, online responses were more common among younger age groups and, respondents with higher education levels and whose first language was not English. Conclusion Using a postal survey, we observed response rates of ≥5.5% among older Aboriginal and Torres Strait Islander adults in major cities and regional areas; response rates were lower in other groups. A two-stage postal distribution approach provided an opportunity to adapt sampling approaches to different demographic groups. Based on initial response rates, the sampling strategy was revised to send postal surveys to groups with higher response rates groups and focus field recruitment strategies on low response groups.