The AEC retains ballot papers, sorted by polling place, from each election. Informal ballots from 2004 were sorted by informality and categorised by informality type. This data is compiled by electoral divisions and published in a separate attachment to this paper (Appendix: Research Report Number 7–Analysis of the Increase in Informality During the House of Representatives 2004 Election–Divisional Summaries).
The results were entered into a database at polling place level and aggregated to divisional and state levels. Changes from 2001 to 2004 within categories of informal votes were compared and tested against other factors, including the number of candidates on the ballot.
Changes in informality were not compared with changes in sociological factors as census data has not changed since 2001. Instead, this paper focuses on other independent variables that may account for the increase in informal voting. The paper also looks at sociological factors that might influence the types of informality, particularly ballots intentionally made informal with marks and slogans.
The paper uses a multiple regression model to correlate the increase in the informal vote against the number of candidates per ballot, and to correlate informality within categories with the socioeconomic index produced by the Australian Bureau of Statistics.