In 2003, the AEC published the Informal Vote Survey for the House of Representatives 2001 Election. The survey provided an overview of informal voting at the 2001 federal election and focused on variations in informal voting. Factors contributing to voting informality in Australia include differences in State and Federal electoral systems, and sociological factors. The paper detailed the number of informal votes by category. AEC Research Report Number 8 (Appendix of Divisional Summaries/Informality 2004) includes a breakdown of informality by division for 2004.
The key issues from the analysis of the 2001 election were as follows.
The 2003 paper tested the hypothesis that sociological and institutional factors influence informal voting and found that the 'Not fluent in English' variable is both the major predictor of informal voting and statistically highly significant. The 'Persons with low education attainment' variable also correlated with higher levels of informality.
Differences between voting procedures at the federal and state levels, and between the voting systems for the Senate and the House of Representatives, may also contribute to voting informality. In the 'full preferential' system used in elections for the House of Representatives, the elector must number every square on the ballot paper to cast a formal vote, but not all elections in Australia use a full preferential system. New South Wales and Queensland both practise 'optional preferential' voting at the state level. Furthermore, state elections have differing criteria for formality.
Several studies indicate that compulsory voting is may be one of many factors contributing to informality, and that blank ballots may in some cases be the 'functional equivalent of abstention'. These studies show that while the link between compulsory voting and informal voting is difficult to prove, some informal votes could be protest votes from voters who would have otherwise abstained under a non-compulsory system. Informal ballots with either protest slogans or marks making the numbering illegible made up about 6.6% of informal votes in 2001.