Federal Election 2016
Finity has cracked the political code when it comes to using ‘big data’ to map how the major parties’ policies will impact Australian households’ voting preferences ahead of the Federal Election on 2 July.
Defin’d, which launched in April and is being piloted by companies in the retail, insurance and banking sectors, has the ability to profile Australian voters by electorate, to identify their key concerns and indicate their voting preferences, “down to neighbourhoods of around 150 households#”, according to Colin Brigstock, one of the architects of Defin’d – Finity’s big data analytics consulting service.
“Powered by over 40,000 lines of code and incorporating state-of-the-art modelling and simulation techniques, this platform can segment the Australian population down to small tightly defined segments, not previously possible – and all in a few clicks and in a matter of seconds,” said Colin.
Using a cutting-edge, cloud-based platform developed over the past year, Defin’d accurately segments Australia by demography, behaviours and life stage – providing a rich level of information about the profile of the households and individuals in each and every neighbourhood.
By combining this with data from previous Federal elections published by the Australian Electoral Commission, Defin’d provides a clear view of how voting preferences vary neighbourhood by neighbourhood.
Select an electorate on the Defin’d election insights webpage and see it depict Australia’s voting preferences.
In the ‘marginal seats’ (Eden-Monaro, Robertson, Banks, Deakin, Petrie and Hindmarsh), Defin’d identifies that these electorates have a lower proportion of neighbourhoods with either strong Labor or strong Coalition voting preferences.
When various data sources are combined with the demographic profiling ability of Defin’d, we see which households would be most heavily influenced to political party messages about key issues such as proposed changes to superannuation, child care rebates or tax cuts. Defin’d delivers powerful insights about the potential for a shift in voting preferences for an electorate overall and in each local neighbourhood.
“This type of analysis is particularly useful in the marginal electorates where only modest swings in voting patterns might be enough for the seat to change hands,” said Colin.
# Neighbourhood is defined by the ABS SA1 statistical area classification which classifies Australia into over 54,000 clusters each which generally comprise around 150 households.