Party Realignment and Single-Issue Voters
Data files
Mar 06, 2025 version files 634.16 KB
-
code_data.zip
630.93 KB
-
README.md
3.23 KB
Abstract
This paper studies the effect of a single-issue referendum on political party realignment. We consider the Conservative Party in the United Kingdom in the aftermath of a referendum on the British exit from the European Union. We look at the change in positions of Conservative Members of Parliament using a novel dataset tracking their opinions on British membership of the EU and their election outcomes between 2015 and 2022.
Our findings show that MPs who resisted switching to a pro-Leave position faced significantly higher electoral losses. We then consider various models of switching, showing that safe seats are strong predictors. These results highlight the sensitivity of political parties to single-issue politics, underscoring how such environments can contribute to broader populist movements in modern democracies.
https://doi.org/10.5061/dryad.f7m0cfz6d
Description of the data and file structure
This set of data and code replicates all tables and figure from ``Party Realignment and Single-Issue Voters''.
Files and variables
File: code_data.zip
Description:
#1 Stata .do file to run results [uk_ele.do]. This loads the xlsx and csv files and generates all tables and figures in the paper.
#2 Hand collected MP position excel sheet [Remainers Spreadsheet Updated (1).xlsx]. This contains the information on MP positions on Brexit (see data section for details).
Variables:
- Remain in 2015: was MP remain in 2015.
- Switched?: did they switch.
- What are they doing now?: their current occupation.
- Status of Seat?: who is in the seat they occupied.
- Year Left: if they left, the year they left the seat.
#3 Financial Times dataset on constituency variables [financialTimesge2017dataset.csv, & financialTimesge2017dataDictionary.csv,] This contains variables at the constituency used in the regression analysis. All currency-denominated variables (house price and gross pay) are in British pounds.
#4 Brexit vote share [brexit.xlsx]. This a file containing the constituency level Brexit referendum vote share used in the regression analysis.
Variables:
- refno: unique identifier specific to this dataset.
- PCON11CDL: identifier for the constituency.
- Constituency: name of constituency.
- Estimated Leave proportion: estimate of fraction of Brexit referendum vote supporting leave.
- Known result: fraction of leave result from public sources.
- Figure to use: recommendation for which number to use.
- Notes: source for known result.
- Region: region of constituency.
- Errors: difference between estimate and known result.
#5 General election vote shares from 1966 to 2019 [elecdata1966,...,elecdata2019]. These files report national and constituency level UK general election results. These are used in the regression analysis and in the appendix.
Variables:
- Name: name of constituency.
- MP: name of MP representing that constituency. .
- Area: numerical indicator of area.
- County: name of county.
- Electorate: number of citizens in that constituency.
- CON: Conservative vote totals.
- LAB: : Labor vote totals.
- LIB: Liberal vote totals.
- NAT: British national party vote totals.
- MIN: vote totals for minor parties (such as BNP, Respect and independent candidates).
- OTH: vote totals for all others.
#6 Referendums results for 2016 and 1975 [2016referendum.txt, 1975_referendum.txt]. These report the nationwide results for both EU referendums:
1975 variables :
- County: county of vote.
- Yes: total of yes to leave votes.
- No: total of no votes.
- Yes_frac: vote share of yes.
- No_frac: vote share of no.
- Turnout: turnout share.
2016 variables:
- Constituency: constituency name.
- Member of Parliament: name of MP in that constituency.
- MP position: their position on Brexit.
- MP's majority: their share of votes.
- Region: geographic region.
- Remain %: share voting remain in that constituency.
- Leave %: share voting leave in that constituency.