The Community Hopes, Fears and Actions Survey: Survey method, sample representativeness and data quality

Abstract

This paper describes and discusses the method used to collect data on the hopes, fears and actions of Australians in relation to the tax system in 2000. Data were collected from June through to October 2000 by means of a national survey of Australians who were randomly selected from the publicly available electoral rolls. A response rate of 29% yielded 2040 questionnaires that could be used for further analysis. Diagnostic statistics comparing the sample with Australian Bureau of Statistics population estimates on age, sex, occupation and education suggested that the sample was broadly representative of the population, but with some bias in favour of those involved in occupations in which reading and writing skills are integral. The survey also underrepresented younger age groups, a bias that is shared with many other social surveys of this kind. A number of regression models were run to find out if responses were affected by anonymity, time taken to respond to the survey and the introduction of the goods and services tax (GST) during the survey period. No evidence was forthcoming to suggest a direct relationship between any of these variables and tax-related attitudes and behaviours. We conclude that these data provide a satisfactory base for examining the relationships outlined and discussed in the Centre for Tax System Integrity Working Papers No. 2 and No. 3 (Braithwaite, 2001; Braithwaite, Reinhart, Mearns & Graham, 2001).