Voter preferences and the changing dimensions and dynamics of economic wellbeing in the 21st Century.

Lead Research Organisation: University of Manchester
Department Name: Social Sciences


New forms of data are beginning to revolutionise how we do social science. So called 'Big Data' are data that are generated by government departments, businesses and other organisations which provide a resource to researchers (e.g. administrative, commercial and social media data). The use of these types of data for research in political science is in its infancy yet the potential uses - particularly in studies of political behaviour - are many. In this proposed CASE studentship we will use data from YouGov's 'Data Cube', which contains large quantities of information about YouGov panel members to investigate new dimensions of economic voting.


10 25 50

Studentship Projects

Project Reference Relationship Related To Start End Student Name
ES/P000665/1 30/09/2017 29/09/2027
1901511 Studentship ES/P000665/1 30/09/2017 29/09/2021 Jack Bailey
Description My PhD focusses on the ways that voters learn about the economy, report their economic perceptions, and act on them at the ballot box. In particular, I focus on how their party preferences affect each of these processes.

As part of my award, I have found that:

- Contrary to the prevailing wisdom in the literature, voters who support different parties do not respond to economic change in the same way. Instead, they update their economic perceptions in line with their partisanship. This process differs, however, when things are bad and when they are good. In the former case, all voters eventually converge on a shared perception. In the latter case, they diverge instead.
- Political surveys serve to bias the economic perceptions that voters report. In my case, I find that those who support the incumbent party tend to over exaggerate their economic perceptions in political surveys. That is, while partisan bias remains in both political and non-political surveys, it is worse in the former than the latter.
- The statistical models we often use to test if voters economic perceptions affect how they vote are misspecified. This is because they ignore the fact that there are other factors that influence both economic perceptions and voting behaviour. This point has been made before, but remains controversial. I provide new evidence, using an especially straight-forward and clean method.
Exploitation Route My findings have important implications for future research on the economy and voting behaviour. Ultimately, they suggest that we need to reassess much of the prior literature. Others might take forward my findings in the following ways:

- Reapply my theories and models in other contexts (for example, the US)
- Develop my work to build a theory of economic voting that does not rely only on what voters perceive, but also what they think and feel
- Use my experimental data to better understand the effect that survey methodology has on our assumptions about the relationship between economics and politics
- Use my methods to build economic voting models that don't suffer from the same problems as those in the past
Sectors Government, Democracy and Justice,Other