MRP Methodology

What is an MRP?

The approach we used to produce these results, MRP (short for multi-level regression with post-stratification), is a statistical technique that combines survey data with electorate-level information from the Census and other government agencies - such as population density and the proportion of the population with a university degree - and previous election results. It is more than a just poll, it is a model that uses data science.

YouGov's approach to MRP methodology is market leading, and has a proven track record off success including in recent British and Spanish national elections. Our model uses a large sample of data collected from our panel to discover relationships between people’s characteristics and their answers to the voting intention question. It then combines these relationships/patterns with information about the characteristics of people living in different electorates to construct estimates of how vote intention would look in each electorate. MRP models are extremely powerful and computationally complex, and are capable of simulating and estimating thousands upon thousands of different configurations and patterns of voter behaviour and party support. For the Australian 2025 federal election, each YouGov MRP model will explore 25,000 different possibilities.

The sample sizes involved - over 40,000 unique interviews - mean we speak to many different voters from many different places and across all electorates in Australia. Working with such large sample data requires successive rounds of trimming down the overall pool of interview data to create the best and most robust sample possible which we then feed into the MRP model. The effective sample size after trimming each model will be around 9,000.

The modelling process includes many socio-demographic characteristics of individuals and the electorates they live in, past vote choices of individuals, and shares for various parties within each electorate, and information about patterns of candidacy and party contests across electorates. While we cannot reliably capture every factor specific to every single seat, our large model sample sizes will help us get close to understanding how different types of voters are behaving in different places up and down Australia.

Both the seat winner and voting intention estimates have uncertainty, as does any measurement using survey data. For each party in each seat, as well as the seat total for each party overall, we report ‘credible intervals’ which represent the range of data we believe the ‘true’ shares (and seat totals) have a 90% chance of lying within, The specific number presented for each party is the most likely value, which is typically very close to the middle of that range. So when assessing the data it is best to remember that both the seat and voting estimates for all parties projected could very well be higher, or lower, than this midpoint. The full range of possibilities the full table.

Two party preferred results

Our national two-party preferred results are calculated by analysing and averaging out the projected swing between Labor and the Coalition within each individual seat. Two-party preferred winners within each seat are calculated by simulating how we expect preferences to flow to the top two projected parties (on first preference shares) in each specific type of two-party contest. The data for these preference flows is a blend between what our thousands of respondents are telling us themselves about how they will structure their ballot, and historic flows between parties at previous Australian elections.

YouGov has changed it methodology for preference allocation from the last election because Labor won an extraordinary share of second preferences because of the unpopularity of the then incumbent Prime Minister. Now Labor the incumbent the situation is different.

YouGov has collected it voter preference information though a simulated ballot with randomised order that has allows voters to number their choices.

YouGov model has estimated the following preference flows between the two major parties and will be using this until our next MRP.