Impact of Redistributions in South Australia and Victoria

I’ve spent a bit of time calculating the impact of the redistributions in districts throughout South Australia and Victoria (which took place after the 2010 election) on each district. The table below shows the impact of the redistributions for Victoria state for Labor, the Liberal group, the Greens and Family First (click on the table for an easier read):

Victoria Redistribution Table

The biggest changes took place in the McEwen district, which saw a significant shift to Labor. The Greens will lose about 1.2 percentage points in the Melbourne district, the only seat in Parliment they currently hold. Unless there is a major reduction in Green support however, they should hold this seat in the upcoming election. The Casey district tightened up in favor of Labor but the Liberals are likely to hold this seat.

The next table shows the effect of the redistribution in South Australia state:

South Australia Redistribution

The redistribution’s effect was much less in South Australia than in Victoria with only the Hindmarsh district appearing to experience any significant potential impact.

The next stage in my work is to input the effect of these changes into my model and then calculate how the race in districts with no incumbent should change compared with 2010. With these figures in hand, I will then calibrate the 2010 polls with my model and with that calibration in hand, I will be able to make my first forecast for the upcoming election.

Building an Election Forecasting Model

I wanted to discuss in this post the process of creating an election forecasting model for the upcoming Australian election. I have covered some of the steps in the process in earlier posting but I will briefly go over these points here for the sake of completeness.

The first step is understanding the Australian election system, most importantly its preferential voting system. This system produces viable minor parties with the potential of holding the balance of power in individual district elections. Four minor parties presently have this potential. It is possible from election data to precisely measure the effect each of these viable minor parties have on individual elections. This work has been completed.

The second step is to understand demographic and other drivers which explain the share of votes received by the various parties. I have made  statistical models using district level demographic data from the 2011 Census for the two major party groups, the Greens and for the most important of the minor parties to accomplish this. These models use the standard ordinary least squares method. I used the Durbin spatial model to test for spatial autocorrelation across districts but found that the evidence for such autocorrelation is weak.

The third step is to account for changes in district composition since the last election. Two states, South Australia and Victoria, have had nearly all of their districts redistributed since the 2010 election. The changes were somewhat minor in the 11 South Australian districts but were quite significant in many of Victoria’s 37 districts. I have used local polling place results for the 2010 election and redistribution information from the Australian Electoral Commission to estimate the impact of these district changes on the various districts. This work is in progress.

The fourth step is to use polling data to estimate the status of the election at the national and state levels at any given moment in time. I use a Baysian approach to update and adjust existing polling information. The Newspoll group provides regular polling at the national level and periodic state polling. Newspoll’s final polling in the 2007 and 2010 elections was quite accurate and it seems reasonable to make use of this group’s polls in the analysis. The polling estimation is then inputted into the model produced in the second step above to make district level estimations.

These district level estimations are then the fifth step in the process and are estimations of the first preference percentages for the major party groups and for the Greens and Family First parties. Since the full candidate slates will not be available until about 30 days before the election it will be difficult to estimate percentages for the Christian Democrats and Liberal Democrats until we know whether and where they are running candidates. It is also possible that one or two new political parties will emerge in the process and we can estimate the impact of such new parties from the national and state level polling data as well as from media information regarding individual candidates and district races.

Finally, I use the information developed in step one above to estimate the preference distributions for the Greens and other significant minor parties. This produces the final results for each district. I anticipate making a first estimation of the number of seats won by each party shortly.

Model Observations

My modelling of support for the major party’s candidates is taking shape. I have used district level demographic data from the 2010 election to model the support base for the major party groups and the Greens. Four demographic data points have proven to be important in predicting the support base for the major party groups:

1. Percentage of professionals and managers in the district work force. These are people with an elite status in the laborforce.

2. Income. Income is a very tricky measure because low income and high income voters tend to support the same party in Australia. Thus if one models the effect of income as linear such a model would produce an inconclusive result. I have dealt with this problem by taking some point “near” the average median income across districts and calculating the median income distance from this point.

3. Immigrants from non-English countries, non-western. A significant percentage of immigrants to Australia come from English speaking Commonwealth countries most importantly New Zealand, the UK and South Africa. To a lesser extent there are immigrants to Australia from western native speaking English countries such as Ireland, Canada and the US. My assumption is that immigrants from such countries are likely to have general political views in line with Australians and, thus, such immigrants can be treated the same as native Australians. I have used detailed district level demographic data to separate out such native, western, English speaking immigrants from those from other countries. The result is a measure of the number of people that can be called “ethnic” living in each district.

4. Never married. These people tend to be young.

The percentage of professionals and managers in the district work force is very strongly tied with support for the two major party groups and for the Greens. Districts with high percentages of professionals and managers are likely to have an increased support for Liberals while the opposite is true for Labor. Greens also draw a significant percentage of their votes from the same group of professionals and managers supporting Liberals. This suggests that Liberals and Greens are, to some degree, drawing their support from the same population. This is very interesting. Labor’s support, on the other hand, tends to decline with an increasing percentage of professionals and managers in a given district.

I noted above that I calculated the distance from a point “near” the average median income across districts. This calculation showed that districts with median incomes “near” the average median income tended to support Labor while incomes that were higher and lower tended to support Liberal group candidates.

Immigrants from non-English speaking, non-Western countries tend to support Labor. It is interesting to note than these immigrants appear to have very little interest in Green candidates.

Never married people tend to support both Labor and the Greens which suggests that, to some extent, these left-of-center parties are in conflict with each other for this group of voters.

In the end, my analysis suggests that Greens are actually more likely to take votes from Liberals rather than from Labor a result that is counterintuitive. On the other hand, my analysis also suggests that Family First voters are more likely to take votes away from Labor rather than Liberals.

I am now getting closer to creating a model of the 2013 Australian election. The only thing now lying between this and such a model is an analysis of redistricting that has taken place in Victoria and South Australia States.