How Family First made Labor last

Demographic modeling shows Family First voters were composed of two distinct groups.

The first group was what you would have expected from a party founded by religious activists: middle income, professional, evangelical - and Liberal.

But the second group, equal in size, was rusted on Labor voters - agnostic, blue collar, lower income, single parents.

When you rank the correlations of the Family First vote, to determine exactly who lived in Family First neighborhoods, the evangelical religious variables light up like a Christmas tree.

Top religious correlations for Family First were: Pentecostals, Lutherans, Uniting Church (all with high correlations of 0.60 and above), Other Christian, Jehovah’s Witnesses, Church of Christ, Baptists, Salvation Army, Brethren and Mormons.

Of the religions most hostile to Family First, Catholics were the strongest, with correlations of minus 0.36. In addition, Jews and Muslims seemed to agree, along with Hindus, Buddhists and Orthodox Christians, they didn’t like Family First at all.

This may have had something to do with pre-election media publicity for the Victorian Family First leaflet about Satan’s strongholds including: “Brothels, gambling places, bottle shops, mosques, temples - Freemasons, Buddhist, Hindu etc, witchcraft.”

If you insult someone’s religion, or, even worse, threaten to take money out of their pockets, you tend to see the impact highlighted by inferential statistics, which look at what determines the outcome in the House of Representatives, by objectively measuring exactly which demographic groups live in each electorate, and how the same electorates voted.

But, it’s when we get away from religious variables, that the profile of Family First becomes a lot more interesting.

After the very strong positive correlations for the evangelical religions, the large group of agnostics and atheists (no religion or not stated) appears with a significant positive correlation of plus 0.35.

The computer analysis of 500 plus demographic variables, found there were no significant correlations with Family First voters and conventional family couples of any description.

In relation to the actual family status and age, of your typical Family First voter, we saw a single mum or dad, in their late forties or early fifties, with one teenage child.