This article is from the source 'guardian' and was first published or seen on . It last changed over 40 days ago and won't be checked again for changes.
You can find the current article at its original source at http://www.theguardian.com/news/datablog/2013/sep/09/who-voted-for-palmer-united
The article has changed 3 times. There is an RSS feed of changes available.
| Version 0 | Version 1 |
|---|---|
| Palmer United party: who voted for them? | Palmer United party: who voted for them? |
| (35 minutes later) | |
| The Australian election resulted in a swing away from the incumbent Labor party, but it wasn't always to the Coalition. | The Australian election resulted in a swing away from the incumbent Labor party, but it wasn't always to the Coalition. |
| With 76% of the vote counted, the Palmer United party (PUP) got 5.68% of the national vote. The bulk of this appears to have come from the 4.14% swing against the ALP, as the Coalition had only a marginal primary vote increase of 1.74%. Thus, the election of Tony Abbott seems not to be due to a growth in popular support for the Coalition, but more to a rejection of Labor. | With 76% of the vote counted, the Palmer United party (PUP) got 5.68% of the national vote. The bulk of this appears to have come from the 4.14% swing against the ALP, as the Coalition had only a marginal primary vote increase of 1.74%. Thus, the election of Tony Abbott seems not to be due to a growth in popular support for the Coalition, but more to a rejection of Labor. |
| Given that, what do we know about the voters who swung to PUP? | Given that, what do we know about the voters who swung to PUP? |
| Using census data to create profiles of people in different electorates, it's possible to see if there's a correlation between votes for certain parties and certain categories of people. (For more information on the exact methods I used, check out the method section at the end of this blog.) | Using census data to create profiles of people in different electorates, it's possible to see if there's a correlation between votes for certain parties and certain categories of people. (For more information on the exact methods I used, check out the method section at the end of this blog.) |
| The first general trend was that the higher the PUP vote, the lower the education level of the electorate. I should stress that I'm measuring the correlation between two factors, and this does not necessarily equate to causation. | The first general trend was that the higher the PUP vote, the lower the education level of the electorate. I should stress that I'm measuring the correlation between two factors, and this does not necessarily equate to causation. |
| The education scale is made up of the proportion of people with bachelor degree, graduate diploma or postgraduate degree. | The education scale is made up of the proportion of people with bachelor degree, graduate diploma or postgraduate degree. |
| The second trend was a bell curve on the rural-urban scale: the PUP vote was highest in suburban areas. Suburban electorates in the middle of the scale had the highest PUP voting electorates clustered. There were a few high results in rural Australia too. Rural is on the right and urban on the left of the graph. | |
| There was a significant correlation between the Palmer vote and socioeconomic status. The higher the PUP vote, the lower the socioeconomic status of the electorate. | There was a significant correlation between the Palmer vote and socioeconomic status. The higher the PUP vote, the lower the socioeconomic status of the electorate. |
| The socioeconomic status scale is made up of the proportion of people who work in finance, the proportion in a professional occupation, the proportion of people with weekly incomes over $1,500, and the proportion of people with monthly mortgage repayments over $3,000. | The socioeconomic status scale is made up of the proportion of people who work in finance, the proportion in a professional occupation, the proportion of people with weekly incomes over $1,500, and the proportion of people with monthly mortgage repayments over $3,000. |
| The PUP vote was also highest in economically declining electorates, with a cluster of high votes between 0 and 60 on the scale. | The PUP vote was also highest in economically declining electorates, with a cluster of high votes between 0 and 60 on the scale. |
| This suggests economic voting may have been a factor this election. The economic scale is made up of the proportion of people who are unemployed and looking for full-time work, the proportion of people who work part-time, and the proportion of people who earn more than $1,250 a week. There is a negative relationship between unemployment and high wages. | This suggests economic voting may have been a factor this election. The economic scale is made up of the proportion of people who are unemployed and looking for full-time work, the proportion of people who work part-time, and the proportion of people who earn more than $1,250 a week. There is a negative relationship between unemployment and high wages. |
| However, it must be remembered that it is not the fact that the economy is declining that matters to the voter, but the perception of the economy's trajectory. | However, it must be remembered that it is not the fact that the economy is declining that matters to the voter, but the perception of the economy's trajectory. |
| So what does this tell us overall about PUP voters? | So what does this tell us overall about PUP voters? |
| They’re in neighbourhoods with low levels of education. They tend to be in rural and suburban seats, in neighbourhoods with low levels of people working in professional occupations, where there are low rates of people on weekly wages over $1,250, high rates of unemployment and high rates of people working part-time. | They’re in neighbourhoods with low levels of education. They tend to be in rural and suburban seats, in neighbourhoods with low levels of people working in professional occupations, where there are low rates of people on weekly wages over $1,250, high rates of unemployment and high rates of people working part-time. |
| This suggests that the vote for the PUP is a vote of frustration from those perceiving economic decline, those who live in a local area in which few people access high-paying jobs, those who experience unemployment with little employment prospects. The high part-time work and unemployment rates reflect a lack of demand for labour where the PUP vote is high. | This suggests that the vote for the PUP is a vote of frustration from those perceiving economic decline, those who live in a local area in which few people access high-paying jobs, those who experience unemployment with little employment prospects. The high part-time work and unemployment rates reflect a lack of demand for labour where the PUP vote is high. |
| A question Labor may want to ask itself, when assessing its governance and electoral strategy, is: how has it improved economic prospects in those specific electorates where the Palmer vote spiked? | A question Labor may want to ask itself, when assessing its governance and electoral strategy, is: how has it improved economic prospects in those specific electorates where the Palmer vote spiked? |
| Method | Method |
| To determine what sort of people are in each electorate I factor analysed ABS 2011 census data to find a correlation between different variables that indicate a profile. A scale from 0 to 100 was formed to rank the electorates by a profile. For example, to make a rural-urban scale, variables such as proportion of people who work in mining, proportion of people working in fisheries and agriculture, and proportion of households with more than three vehicles are combined. | To determine what sort of people are in each electorate I factor analysed ABS 2011 census data to find a correlation between different variables that indicate a profile. A scale from 0 to 100 was formed to rank the electorates by a profile. For example, to make a rural-urban scale, variables such as proportion of people who work in mining, proportion of people working in fisheries and agriculture, and proportion of households with more than three vehicles are combined. |
| These factor scores are then correlated to see if there is a statistically significant linear relationship between a vote and the demographic profile, and then a scatter plot is produced. | These factor scores are then correlated to see if there is a statistically significant linear relationship between a vote and the demographic profile, and then a scatter plot is produced. |
| A far more accurate understanding of who swung should be available in a few months after the Australian Electoral Study for 2013, run by the ANU, is completed. | A far more accurate understanding of who swung should be available in a few months after the Australian Electoral Study for 2013, run by the ANU, is completed. |