Template-Type: ReDIF-Paper 1.0
Author-Name: Christa Jensen
Author-Name-First: Christa
Author-Name-Last: Jensen
Author-Email: christa.jensen@mail.wvu.edu
Author-Workplace-Name: Department of Economics, West Virginia University
Author-Name: Donald Lacombe
Author-Name-First: Donald
Author-Name-Last: Lacombe
Author-Email: donald.lacombe@mail.wvu.edu
Author-Workplace-Name: Department of Economics, West Virginia University
Author-Name: Stuart Mcintyre
Author-Name-First: Stuart
Author-Name-Last: Mcintyre
Author-Email: s.mcintyre@strath.ac.uk
Author-Workplace-Name: Department of Economics, University of Strathclyde
Title: What Determined Conservative Success in the 2010 U.K. General Election? A Bayesian Spatial Econometric Analysis
Abstract: The Conservative Party won the recent General Election in the United Kingdom (UK), gaining the most votes and seats of any single party. Conservatives simultaneously performed particularly well in some areas of the UK and poorly in others. In attempting to explain the variation in voting behaviour during this election, we consider an analysis involving an explicit accounting of geographic considerations. The spatial econometric analysis of voting behaviour is still quite rare in the literature, and analyses using a full suite of models, as employed here, are even rarer. We use data from various sources to examine the effects of a range of economic, socio-economic, and political variables on the percentage of the vote obtained by the Conservative Party in each UK constituency in the 2010 General Election. We employ recent advances in Bayesian spatial econometric modelling to determine the appropriate model for drawing these inferences. We find that there is significant spatial error dependence in a model of the percentage of the vote obtained by the Conservative Party in the 2010 UK General Election, justifying the use of spatial econometric methods for our analysis. By explicitly modelling this spatial phenomenon, we get better estimates of the impact of our chosen economic, socio-economic, and political explanatory variables. Results that seem contrary to our prior expectations when using a non-spatial regression model change when estimated using spatial econometric techniques.
Length: 37 pages
Creation-Date: 2010-10
Revision-Date: 
Publication-Status: 
File-URL: http://www.strath.ac.uk/media/departments/economics/10-24_final.pdf
File-Format: Application/pdf
Number: 1024
Classification-JEL: C11, C21, D72
Keywords: Bayesian spatial econometric analysis, spatial voting analysis, UK General Election 2010
Handle: RePEc:str:wpaper:1024