Category Archives: Uncategorized

I’ve moved

In the off chance you know me, have stumbled across my blog, and we haven’t spoken lately, I’ve moved. I am now a lecturer in Political science at the United States Studies Centre, at the University of Sydney.

My details are here (also FYI – they jumped the gun on the Dr, officially it’s almost Dr*).



*My thesis passed with very minor edits. Graduating later in the year.

Papers presented at the 2016 APSA conference

I’m just finishing up at the 2016 Australian Political Studies Association annual conference in Sydney, and thought I’d post the slides of the papers I presented with colleagues Andrea Carson, Marija Taflaga and Yannick Dufresne.

The first paper looked at the demographic and environmental predictors of concern about immigration in Australia. Immigration policy has been a divisive political issue in many developed economies in recent decades. Australia is no exception, with immigration being a much debated topic, notwithstanding what can be considered its highly successful national immigration program.

In this paper we map the demographic backgrounds of voters who were more concerned about new arrivals to Australia at this time, and the environmental circumstance that are associated with opposition to further large scale immigration, including the size of the existing migrant population and the level of unemployment. Here we tried to provide new insights in to what might contribute to negative and positive perceptions about immigration (eventually we also hope to identify potential political and policy implications).

The slides can be viewed here.

The second paper, also with Marija and Andrea, was almost a sequel to the first, examining the relationship between attitudes towards immigration and vote choice at the 2013 Australian federal election. The salience of immigration in Australian politics, at least in recent decades, has been well documented.  In particular, a narrative has developed that suggests immigration has been used by the centre-right Coalition parties to detach working class voters from the centre-left Labor Party. However, the empirical evidence for this is limited. An alternative perspective is that it is middle class voters — who are theoretically economically secure and able to vote on issues other than their economic self-interest, such as cultural matters —  decide their vote based on concerns about immigration.

In this paper we tested these propositions by modelling data on public opinion collected during the 2013 federal campaign to identify the relationship between Australian voters’ attitudes towards immigration and partisan choice. We explore whether those voters more concerned about immigration vote differently to those who are less concerned, and if there are demographic patterns in this association. This helps us understand what drives the relationship between attitudes towards immigration and vote choice, if any such relationship exists at all. 

The slides for this paper can be found here.

The third paper, with Andrea and Yannick (also presented at the American Political Science Association conference in Philadelphia this year), is concerned with the congruence between public opinion and the positions taken by the legislators belonging to the major parties in Australia’s federal parliament. To do this we focus on a signicant social transformation of our time, changes to the institution of marriage that has led to legalisation of same-sex marriage in 20 countries in less than two decades.

Contrary to Downs’ (1957) median voter theorem and other office-motivated frameworks, federal politicians in Australia have largely ignored majority opinion on this matter, which is supportive of same-sex marriage (with most surveys showing approximately 60 per cent support and 30 per cent opposition). The (current) incumbent in particular, the Liberal-National Coalition, does not support same-sex marriage. We examine a number of possible reasons why parties may not be fully responsive, and explore the relationship between public opinion in each of Australia’s 150 federal electoral divisions and the position of its elected representative.

To do this we use a unique data set to ascertain differences between politicians’ positions on the issue with that of Australian voters. Using a very large dataset (n > 601,550) collected by Vox Pop Labs (2016) during the 2013 Australian federal election, this paper provides the first fine-grained analysis of attitudes about same-sex marriage from every federal electorate. We match these public opinion data with data on the public positions of members of the federal house of representatives since 2012. This enables us to compare the aggregates of citizens’ policy preferences in each electoral district with the public positions of their elected representatives.

We found that although there appears to be a status quo bias, with voter opposition well below 50 per cent required to tip legislators towards opposition as well as some evidence that parties are (in part) policy-seeking organisations interested in more than maximising their vote, our results also suggest that parliamentarians are responsive to public opinion. Lower levels of opposition to same-sex marriage in an electorate was found to predict a lower probability of opposition from its MP. Further, originally oppositional parliamentarians representing electorates with less opposition to same-sex marriage were more likely to shift to a supportive position between 2012 and 2016.

A copy of the slides can be found here.

Public opinion and policy responsiveness in Australia

The case of same-sex marriage


Last week I presented a paper at the US Political Science Association annual meeting in Philadelphia with some colleagues. In it, we look at the congruence between public opinion and the positions of parliamentarians on same-sex marriage.

Our findings indicate that parliamentarians are generally representative of their constituents, giving some hope for the representational model of democracy. Although there is also some evidence that legislators are also attempting to aggregate other policy interests besides those of voters, and there is a status quo bias weighting outcomes away from policy change.

The presentation can be found here.

Some initial thoughts on the election

Since the Australian federal election, held last Saturday, I’ve had a few thoughts. These are only half-formed, so bear with me.

First, Coalition leader (and current prime minister) Malcolm Turnbull is getting a lot of heat for the relatively poor performance of the Liberal Party (and its various state-based permutations). Some of this is reasonable. The double dissolution was probably a mistake, as is now being reported. It has possibly made the senate more difficult for a future Coalition government to work with, not easier. However, blaming Turnbull for the overall poor showing of the Coalition parties is probably a little unfair. To do so requires forgetting just how unpopular the government was before Turnbull replaced Abbott, and assumes the latter could have pulled the the Coalition two-party vote back to 50 per cent by the election, when all evidence indicated that his government was unable to find a circuit breaker and stop the death spiral they were in.

Second, I think the rise in minor party support is overdone. First, it ignores history. We’ve had higher (1931) and similar (1943, 1998) levels minor party support in the past. Second, it ignores context. The GFC and resulting (and cascading) recessions and national public debt problems have created economic turmoil. This has had a political effect, including the rise of far (or populist) right parties in Europe, Brexit in the UK and Trump in the US. Although we missed the worst of the global recession, real disposable incomes have been declining for 17 straight quarters (four years and three months). The political impact here has been to create instability, and dissatisfaction with public policy and governing parties, hence the rapid turnover of state and federal leaders and governments since 2010 after more than decade of stability up to 2007. In other periods of economic turmoil we’ve seen a three-way split of the Labor Party (early 1930s), the collapse of the centre-right (early 1940s), the rise of the Democrats (1970s), and One Nation (1990s). I fail to see how the present is substantially different, and that if we saw the return to the economic tranquility of the early 2000s, we would see some kind of political stability also return.

Of course, I could also be wrong about both of these things.

I also wrote up some quick pieces on outcomes in 10 individual electorates for the University of Melbourne Election Watch blog. These can be found here.

Update: conference paper slides etc

I haven’t posted in a while.

Mostly that has been because I have been working on my thesis. Of course.

There were also a few days off over Christmas,* some teaching (Applied Research Methods and Political Psychology at the University of Melbourne) and a few other things.

Also, some other research. I presented a few papers at the AusPSA conference in Canberra in September and another at the Australian Society for Quantitative Political Science  (ASQPS) conference in Melbourne in December.

The main point of this post is to share slides from these two meetings.

First, here is the presentation for a paper I authored with Shaun Wilson on the relationship between household income and vote choice in Australia. Although still a work in progress, I’m quite proud of this project. In particular, we use Multilevel Regression with Poststratification (MRP) to look at the geographic variation in the relationship between income and voting. Some of the results are interesting (income matters more in marginal electorates, for instance). Hopefully the full paper will be done soon.

Second, slides from another paper presented as AusPSA. This one looks at some policy outcomes associated with Coalition and Labor federal governments since the 1960s. In particular, growing market-income inequality is increasingly associated with Coalition governments. Not sure what is going to happen with this project, but hopefully something at some point (time permitting – haha).

Finally, the draft paper I presented at ASQPS. This looks at the continuing importance of economic cleavages in the Australian electorate (denied by some). Since then I’ve changed this paper a fair bit. But the version I presented in December is still not too bad. Slides are here.

*Used to learn how to scrape web pages in R. Working through code is the true meaning of Christmas, is it not?