Political Science & Economics: Swansea University Research Excellence Scholarships: Do Elections Vary in their Predictability? A metaanalysis of the factors that lead to electoral predictive failure
Start date: 1 October 2018
Swansea University is proud to offer 15 fully-funded PhD scholarships for students commencing study in October 2018 or January 2019.
The scholarships will be awarded on the basis of student excellence across a portfolio of 34 potential projects.
Widespread failures to predict election results have been highly prominent in recent years. Pollsters, pundits and academics largely failed to foresee the 2015 Conservative majority; the 2016 Brexit referendum result; the election of Donald Trump in the USA and the failure of the Conservative Party to secure a parliamentary majority in 2017.
Typically, such failures lead to methodologically-focused investigations and reports. Such reports have largely sought to refine polling methodology. However, this project would look to the target of the predictions, the election itself, rather than the methodology by which predictions are made, as a source of predictive uncertainty. As such, it stands to make a profound contribution both to the ever-growing study of electoral prediction as well as to the wider study of elections.
Simply put this project aims to answer two questions: are some elections more difficult to correctly predict than others? If so, what factors cause the predictability of an election to vary?
In analysing these questions, the project would have three key components: they involve conceptualization of predictive failure, theoretical work on factors that may cause electoral predictability to vary and empirical data collection and analysis work that will test theoretical insights over a wide range of elections, spanning over several decades.
In terms of facilities, the student will have access to the College of Arts and Humanities statistical software package (SPSS) which will facilitate the data analysis part of the research; access to a supervisory team well-versed in quantitative analysis (Wall and Kolpinskaya); as well as a first supervisor (Wall) with a strong track record in publishing research on electoral prediction in journals such as Electoral Studies; the Journal of Elections, Public Opinion and Parties; and the Journal of Information Technology and Politics. Wall has also previously acted as a PI on a £76,001 AHRC funded project focusing on electoral prediction.
The successful applicant will have access to our Postgraduate Research Student Training programmes.
Applicants should have (or expect to obtain) a first class honours degree (or equivalent) and/or a distinction at master's level in political science/politics, economics or econometrics.
Diligence, capacity for sustained self-motivation and a passion for better understanding politics is required.
Due to funding restrictions, this scholarship is available to UK/EU students only.
The scholarship covers the full cost of UK/EU tuition fees, plus an annual stipend of £14,553 (in line with the RCUK stipend amount) for 3 years.
There will also be £1,000 per annum available for research expenses such as travel, accommodation, field trips and conference attendance.
Please visit our website for more information.