The World Isn’t Fair, but Shouldn’t Elections Be? Evaluating Prospective Beliefs about the Fairness of Elections and Referenda
Abstract
:1. Introduction
2. Data and Context
3. The Political Context of the Elections and Referenda
4. Responses to the Prospective Electoral Fairness Questions
- Scottish Independence Referendum 7.9% don’t know
- 2015 General Election 8.2% don’t know
- EU Membership (Brexit) Referendum 12.4% don’t know
- 2017 General Election 10.3% don’t know
- 2019 General Election 15.0% don’t know
5. What Drives Expectations of Electoral Fairness or Unfairness?
6. Consequences of Expected Fairness
7. Conclusions
Supplementary Materials
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
Appendix A
Variable as Referred to in Article | BESIP Question Formulation (if Applicable) | Coded or Recoded Values | BESIP Variable Names NB: where variable name ends as, e.g., W2, this is part of the proper name of the variable and indicates it is from Wave2; where the name ends as Wx, the ‘x’ is a wildcard that stands for, respectively, 2, 7, 11, and 17, indicating the waves from which these variables were used |
Gender | ‘Are you …?’ | 1 = Male 2 = Female | gender |
Age | Coded numerically (i.e., ‘18’ is 18 years, ‘43’ is 43 years, etc.) | ageWx | |
Ethnicity | ‘To which of these groups do you consider you belong?’ | Recoded: 1 -> 1 (White British) Else -> 0 (15 other ethnic groups) | p_ethnicityWx |
Belonging to religion | ‘Do you regard yourself as belonging to any particular religion, and if so, to which of these do you belong?’ | Recoded: 1 -> 0 (No do not regard myself as belonging to any particular religion) 2 to 15 -> 1 (16 other religions/denominations) | p_religionWx |
Living as a single person | ‘What is your current marital or relationship status?’ | Recoded: 1, 2, 4 -> 0 (Married; in a civil partnership, living with a partner but neither married nor in a civil partnership) 3, 5, 6 -> 1 (Separated but still legally married or in a civil partnership; in a relationship but not living together; single; divorced) | p_maritalWx |
Social grade | Coded as: 1 = A 2 = B 3 = C1 4 = C2 5 = D 6 = E | p_socgradeWx | |
Level of education | ‘What is the highest educational or work-related qualification you have?’ | Recoded by BESIP into: 0 = No qualifications 1 = Below GCSE 2 = GCSE 3 = A-level 4 = Undergraduate 5 = Postgrad | p_edlevelWx |
Being in paid work | ‘Which of these applies to you?’ | Recoded 1, 2, 3 -> 1 (Working full-time—30 h or more per week; working part-time—8–29 h per week; working part-time—less than 8 h per week) 4 to 7 -> 0 (Full-time student; retired; unemployed; not working) | p_work_statWx |
Owning own house | ‘Do you own or rent the home in which you live?’ | Recoded 1, 2, 3 -> 1 (Own outright; own with a mortgage; own or part own through shared ownership scheme) 4 to 8 -> 0 (Rent private; rent local authority; rent housing association; I live with my parents, family, or friends but pay some rent; I live rent-free with my parents, family, or friends) | p_housingWx |
Disabled | ‘Are your day-to-day activities limited because of a health problem or disability which has lasted, or is expected to last, at least 12 months?’ | Recoded 1, 2 -> 1 (Yes, limited a lot; yes, limited a little) 3 -> 0 (No) | p_disabilityWx |
Perceived personal economic risk | ‘During the next 12 months, how likely or unlikely is it that… … There will be times when you don’t have enough money to cover your day to day living costs --- You will be out of a job and looking for work’ | Coded: 1 = Very unlikely 2 = Fairly unlikely 3 = Neither likely nor unlikely 4 = Fairly likely 5 = Very likely The answers to the two items were combined (additively), as they strongly reflect the same single underlying phenomenon (coefficient of homogeneity H is around 0.50 in all waves) | riskPovertyWx riskUnemploymentWx |
Reading newspaper | ‘Which daily newspaper do you read most often?’ | Recoded: 1 to 15 -> 1 (any of named newspapers or ‘other’ mentioned) 16 -> 0 (none) | p_paper_readWx |
Reading tabloid | Derived from ‘Reading newspaper’ before recoding | Recoded 1 to 5 = 1 (The Express; The Daily Mail or Scottish Daily Mail; The Mirror or Daily Record; The Daily Star or Daily Star of Scotland; The Sun; The Western Mail) Else = 0 (all other newspapers) | |
Considering voting a duty | ‘It is every citizen’s duty to vote in an election’ | Coded: 1 = Strongly Disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly Agree | dutyToVote2Wx |
Most people one knows do vote | ‘Most people I know usually vote in general elections’ | Coded: 1 = Strongly Disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly Agree | socialPressureVoteWx |
Vote intention next general election | ‘And if there were a UK General Election tomorrow, which party would you vote for?’ | For analyses recoded into dummies for Conservatives, Labour, Liberal Democrats, SNP, UKIP. The reference category in regression analyses consists of all other parties mentioned (Green, Plaid Cymru, BNP, Brexit Party, and ‘don’t know’) | generalElectionVoteWx |
Ideology (left–right) | ‘In politics people sometimes talk of left and right. Where would you place yourself on the following scale?’ | Coded: 0 = Left . . 10 = Right | leftRightWx |
Political attention | ‘How much attention do you generally pay to politics?’ | Coded: 0 = Little . . 10 = Much | polAttentionWx |
Finding government and politics difficult to understand | ‘It is often difficult for me to understand what is going on in government and politics’ | Coded: 1 = Strongly Disagree 2 = Disagree 3 = Neither agree nor disagree 4 = Agree 5 = Strongly Agree | efficacyNotUnderstandWx |
Strength of party identification | ‘Would you call yourself very strong, fairly strong, or not very strong [name of party]?’ | Coded: 1 = Very strong 2 = Fairly strong 3 = Not very strong 4 = Not (i.e., indicated in previous question not to feel closer to any of the parties than the others) | partyIdStrengthWx |
Approval of incumbent government | ‘Do you approve or disapprove of the job that each of the following are doing? The UK Government The Scottish Government’ | Coded: 1 = Strongly Disapprove 2 = Disapprove 3 = Neither approve nor disapprove 4 = Approve 5 = Strongly Approve | approveUKGovtWx approveScotGovtWx |
Trust in MPs | ‘How much trust do you have in Members of Parliament in general?’ | Coded: 1 = No trust . . 7 = A great deal of trust | trustMPsWx |
Expectation of electoral fairness(prospective) | ‘How fairly do you expect the Scottish referendum to be conducted? ’‘Thinking of the general election for the Westminster Parliament that will take place in May 2015, how fairly do you expect it to be conducted? ’‘How fairly do you expect the EU referendum to be conducted? ’‘Thinking of the General Election for the Westminster Parliament that will take place June 8, how fairly do you expect it to be conducted? ’‘Thinking of the General Election for the Westminster Parliament that will take place December 12, how fairly do you expect it to be conducted?’‘ Thinking of the general election for the Westminster Parliament that took place on December 12th2019, how fairly do you think it was conducted?’ | 1 = Fairly 2 3 4 5 = Unfairly | expectGoodConductScotRefW2 expectGoodConductGeneralW2 expectGoodConductEURefW7 expectGoodConductGeneralW11 expectGoodConductGeneralW17 |
Belief of electoral fairness(retrospective) | ‘How fairly do you think the Scottish referendum was conducted? ’‘How fairly do you think the EU referendum was conducted? ’‘Thinking of the general election for the Westminster Parliament that took place on [date of election], how fairly do you think it was conducted?’ | goodConductScotRefW3 goodConductEURefW9 goodConductGeneralW13 expectGoodConductGeneralW19 (NB: please note that last of these variables is mislabelled in the BESIP data file, as the variable reflects a retrospective and not a prospective fairness belief) | |
Turnout behaviour | ‘Many people don’t vote in elections these days. Did you vote in the referendum on Scottish independence that was held on 18 September 2014?’‘ Thinking back to the last UK General Election on 7 May 2015, a lot of people didn’t manage to vote. How about you – did you manage to vote in the General Election in 2015? ’‘Talking to people about the EU referendum on 23 June 2016, we have found that a lot of people didn’t manage to vote. How about you-did you manage to vote in the EU referendum? ’‘Thinking back to the 2017 General Election on 8 June, a lot of people didn’t manage to vote. Howabout you? Did you manage to vote in the General Election? ’‘Talking to people about the General Election on 12 December, we have found that a lot of people didn’t manage to vote. How about you? Did you manage to vote in the General Election?’ | Coded: 1 = Yes 2 = No‘No I was not eligible to vote’ coded as missing | scotReferendumRetroW3 genElecTurnoutRetroW6 euRefTurnoutRetroW9 genElecTurnoutRetroW13 genElecTurnoutRetroW19 |
Appendix B. Latent Variable Analysis of Prospective Fairness Questions
Electoral Event | Mean Score (1 = Fair; 5 = Unfair) | Item Coefficient of Homogeneity Hi | z-Value of Hi (H0: Hi ≤ 0) |
---|---|---|---|
Scottish independence referendum | 2.33 | 0.23 | 11.76 |
GE2015 | 2.25 | 0.30 | 14.76 |
EU membership (Brexit) referendum | 2.70 | 0.26 | 31.00 |
GE2017 | 2.56 | 0.24 | 23.68 |
GE2019 | 2.56 | 0.36 | 19.12 |
All five items | 0.26 | 26.46 |
Appendix C
Scottish Referendum | 2015 General Election | EU Referendum | 2017 General Election | 2019 General Election | |
---|---|---|---|---|---|
Variables retained in Table 3 (main text) | |||||
Age | −0.01 ** | −0.00 | −0.00 * | −0.00 ** | −0.01 ** |
Level of education | −0.02 | −0.05 ** | −0.06 ** | 0.00 ** | −0.06 ** |
Being disabled | 0.13 * | 0.22 ** | 0.10 ** | 0.09 ** | 0.10 # |
Feeling at economic risk | 0.05 ** | 0.05 ** | 0.06 ** | 0.05 ** | 0.07 ** |
Political attention | −0.02 | 0.01 | 0.07 ** | 0.04 ** | 0.04 * |
Finding government and politics difficult to understand | 0.03 | 0.05 * | 0.06 ** | 0.08 ** | 0.02 |
Strength of party identification | −0.14 ** | −0.15 ** | −0.05 ** | −0.07 ** | −0.18 ** |
Approval of incumbent government a | −0.23 ** | −0.05 # | −0.18 ** | −0.43 ** | −0.20 ** |
Trust in MPs in general | −0.11 ** | −0.09 ** | −0.19 ** | −0.08 ** | −0.07 ** |
Variables not related or inconsistently related to dependent variable | |||||
Social grade | 0.00 | 0.00 | 0.02 * | −0.01 | −0.01 |
Gender | −0.06 | 0.05 | −0.11 ** | −0.00 | −0.03 |
Religion | −0.01 | 0.04 | 0.01 # | −0.00 | −0.01 |
Being single | −0.01 | −0.08 | −0.01 | −0.00 | −0.00 |
Homeowner | 0.09 | −0.22 ** | 0.00 | 0.00 | 0.07 |
Ethnicity | −0.11 | −0.16 # | 0.03 | −0.29 ** | 0.05 |
In paid work | 0.02 | 0.01 | 0.06 * | −0.00 | −0.00 |
Does not read a newspaper | −0.08 | −0.05 | −0.03 | 0.00 | −0.07 |
Good time to purchase | −0.04 # | −0.01 | −0.05 ** | −0.06 ** | −0.02 |
Duty to vote | −0.00 | −0.05 # | −0.06 ** | n/a | −0.09 ** |
Social pressure to vote | 0.00 | −0.02 | −0.04 * | n/a | n/a |
Party choice dummies (5 largest parties only) | |||||
Vote intention Con | 0.36 * | −0.14 | 0.56 ** | 0.16 ** | −0.25 ** |
Vote intention Lab | 0.10 | −0.21 * | 0.02 | 0.08 * | 0.07 |
Vote intention LibDem | −0.11 | −0.17 | 0.01 | 0.04 | −0.03 |
Vote intention SNP | 0.06 | 0.00 | 0.28 ** | 0.34 ** | 0.12 |
Vote intention UKIP | 0.04 | 0.01 | 0.97 ** | −0.22 ** | 0.11 |
Number of observations | 2516 | 2355 | 9769 | 11,062 | 2438 |
R2 | 0.08 | 0.08 | 0.18 | 0.30 | 0.12 |
1 | Sometimes survey questions are explicitly prospective or retrospective by including a reference to a specific election. In other instances, they are retrospective or prospective only implicitly by the timing of the fieldwork: shortly before or shortly after an election (as is the case for the National Annenberg Election Study—see [27]). We suspect that implicit questions may be somewhat susceptible to contain elements of generic system support, while the explicit questions focus more on the fairness of concrete electoral events. In some instances, it will be impossible to determine whether a question is prospective, retrospective, or mainly system support, as in the case of the European Social Survey (round 6), which lacked a reference to any specific election; the British sample of this study was conducted in 2012, around halfway between the general elections of 2010 and 2015. A similar lack of reference to a specific election occurred in the Electoral Commission’s survey of public attitudes that was conducted in January–February 2021, outside an actual election context, and therefore also likely to measure aspects of generic system support rather than specific prospective or retrospective electoral fairness beliefs (see [60]). |
2 | The BES Internet Panel [61] is available for secondary analysis at https://www.britishelectionstudy.com/data-objects/panel-study-data/ (accessed on 31 January 2022). |
3 | In between the CSES and the BESIP studies, we find in Europe the European Social Survey (round 6, 2012) which was fielded in 29 countries and asked about the extent to which elections in one’s country were seen as ‘free and fair’ (but without a clear prospective or retrospective direction); and in the USA, the National Annenberg Election Study which included a slightly amended version of the CSES question in 2008, 2012, and 2016. |
4 | Note a potentially important difference between the CSES and BESIP questions: the CSES primed people to think about fairness of the last national election relative to standards in other countries. The question was asked as “In some countries, people believe their elections are conducted fairly. In other countries, people believe their elections are conducted unfairly. Thinking of the last general election in [country], where would you place it on this scale of one to five where ONE means that the last election was conducted fairly and FIVE means that the last election was conducted unfairly?” (See, for example [62], which included the question). |
5 | BESIP also included explicitly retrospective questions about some of the same electoral events. Analyses have been conducted upon both the prospective and retrospective fairness questions in a difference-in-differences analysis, which showed very strong winner–loser effects for the EU membership (Brexit) referendum [25]. |
6 | The Electoral Commission reviews procedural aspects of elections and the public’s satisfaction with electoral processes. These reports uniformly found the general elections of 2015, 2017, and 2019 to have been procedurally well run (see [63,64,65]). Moreover, even allegations of electoral fraud are very rare. In 2019, when a wide range of elections were held, including the general election of 2019 studied here, across the whole country, only 595 cases of alleged electoral fraud were investigated by the police, only six of which led to a conviction or a police caution [66]. |
7 | For example, a simple open-ended question about what one sees as the most important issues in the country yields around 10% ‘don’t knows’; questions about the Conservatives’ and Labour’s position with respect to redistribution of income give between 20 and 30% ‘don’t knows’; and questions about preferences on European integration (asked before the EU referendum) lead to more than 10% ‘don’t knows’. |
8 | This conclusion can be drawn from the fact that the various items about prospective fairness are significantly and positively associated in terms of scalability (even though scalability as a whole falls slightly below the standard cut-off of 0.3) and the fact that the items as a set do not violate the monotone homogeneity assumption of the Mokken model (see Appendix B). |
9 | Such composite scores are only constructed—using appropriate latent variable models—if the variables involved could be demonstrated to reflect a single underlying (latent) variable. |
10 | During this process, the coefficients of all variables were monitored continuously so as to avoid that this inductively driven process would inadvertently result in the elimination of very strongly associated variables which, even if not consistent, could be helpful in understanding the responses to the fairness expectation questions. |
11 | Appendix A provides details about the operationalisation of these variables. |
12 | The operationalisation of the variables is reported in Appendix A. The analyses reported are straightforward OLS regressions. In view of the categorical nature of the dependent and some of the independent variables, the regressions were also specified as ordered logit models, and categorical independent variables were also specified as dummies. These alternative specifications of the regressions give rise to the same substantive conclusions as OLS, which is reported for reasons of simplicity of presentation. Within the OLS regressions, none of the coefficients changes in sign or approximate magnitude if one of the other independent variables is deleted from the regression equation. The regression analyses reported in this article were conducted using listwise deletion of missing data, which accounts for the reduction in numbers of observations when compared to Table 1. In principle, this could be addressed by multiply-imputing the missing values, which would strongly increase the number of cases in the analysis. However, such a multiple imputation approach makes assumptions about the process(es) that generated the missing data that are impossible to test. If these assumptions are violated, multiple imputation can produce biased regression coefficients even in situations where a complete-cases approach (i.e., listwise deletion) would not produce bias [67] (p. 4). In a simulation study, multiple imputation has been found to be superior to a ‘complete cases’ approach in terms of bias to coefficients only where the data are missing at random conditional on the dependent variable [67] (p. 4). In this study, this would be the case if those who regarded elections as unfair were systematically less likely to respond to the questions we analyse (perhaps out of fear of reprisals). While this assumption may be reasonable in authoritarian systems, and as such, analysists should be mindful of this issue when replicating this study, it is implausible in the case of a Western liberal democracy with an enshrined right to political speech and where the data are collected in an anonymous format by people who are not affiliated with the government. |
13 | The only exception here is for the 2017 general election, where the coefficient was both significant and differently signed. This is a finding that cannot be explained by reverse coding or other data problems and therefore appears to represent a substantive difference at this election from the other electoral events studied here. |
14 | The importance of this finding is underlined by the magnitude of this group: consistently, across all waves of BESIP, approximately three in ten respondents indicate to suffer from some kind of disability. This is a higher proportion than that suggested by 2018 data from the Papworth Trust, which comes to 20% of the UK population [68], but is based on a precise legal definition that may not correspond with people’s own classification of being disabled. |
15 | This difference with the general population is plausibly the effect of two factors. One is (self-)selection, as people who are very uninterested in and disengaged from politics are unlikely to volunteer for being a respondent in surveys that strongly focus on political matters. A second factor that distinguishes the samples from the general population is a learning or socialising effect: being surveyed repeatedly about political matters tends to increase one’s interest in politics. |
16 | The smallest number of observations in these five analyses is 3961. The weakness of these associations—in spite of their significance—is easily illustrated by the case of the EU membership referendum: while n = 20,571, chi-square reaches only 43.19 (df = 4), or expressed in other terms, Cramer’s V = 0.036 and pseudo-R-squared = 0.003. |
17 | This block contains the following variables: gender; age; social grade; level of education; being disabled; perceived personal economic risk; finding politics and government difficult to understand; strength of party identification; and approval of the UK Government. |
18 | Because these interaction models generate so many separate models, it is not possible (or particularly desirable) to reproduce them all in this article. Here, we focus on the substantive conclusions of all of these models collectively. |
References
- Atkinson, V.; Aaberg, R.; Darnolf, S. Disability Rights and Election Observation: Increasing Access to the Political Process. Nord. J. Hum. Rights 2017, 35, 375–391. [Google Scholar] [CrossRef]
- Clubbs Coldron, B.; Frances, S.; Buckley, G.; Bhatkal, S. Supporting political rights for people in psychiatric rehabilitation: “Appropriate” political action in medicalized environments. J. Psychiatr. Ment. Health Nurs. 2021, 28, 494–502. [Google Scholar] [CrossRef]
- Bishop, S.; Hoeffler, A. Free and fair elections: A new database. J. Peace Res. 2016, 53, 608–616. [Google Scholar] [CrossRef] [Green Version]
- Dawson, S. Electoral fraud and the paradox of political competition. J. Elect. Public Opin. Parties 2020, 1–20. [Google Scholar] [CrossRef] [Green Version]
- Norris, P. Why Electoral Integrity Matters; Cambridge University Press: Cambridge, UK, 2014. [Google Scholar]
- Mattes, R. Electoral Integrity and Democratic Legitimacy in Africa. In Advancing Electoral Integrity; Norris, P., Frank, R.W., Martínez i Coma, F., Eds.; Oxford University Press: New York, NY, USA, 2014; pp. 211–228. [Google Scholar]
- Martin, A.; Mikołajczak, G.; Orr, R. Does process matter? Experimental evidence on the effect of procedural fairness on citizens’ evaluations of policy outcomes. Int. Political Sci. Rev. 2020, 43, 103–117. [Google Scholar] [CrossRef]
- Tyler, T.R. What is procedural justice-criteria used by citizens to assess the fairness of legal procedures. Law Soc. Rev. 1988, 22, 103–135. [Google Scholar] [CrossRef]
- Tyler, T.R.; Rasinski, K. Procedural justice, institutional legitimacy, and the acceptance of unpopular US Supreme Court decisions: A reply to Gibson. Law Soc. Rev. 1991, 25, 621–630. [Google Scholar] [CrossRef]
- Tyler, T.R. Procedural justice, legitimacy, and the effective rule of law. Crime Justice 2003, 30, 283–357. [Google Scholar] [CrossRef]
- Esaiasson, P.; Persson, M.; Gilljam, M.; Lindholm, T. Reconsidering the role of procedures for decision acceptance. Br. J. Political Sci. 2019, 49, 291–314. [Google Scholar] [CrossRef] [Green Version]
- Biggers, D.R.; Bowler, S. Citizen Assessment of Electoral Reforms: Do Evaluations of Fairness Blunt Self-Interest? Political Behav. 2021, 44, 435–454. [Google Scholar] [CrossRef]
- Doherty, D.; Wolak, J. When do the ends justify the means? Evaluating procedural fairness. Political Behav. 2012, 34, 301–323. [Google Scholar] [CrossRef]
- Anderson, C.J.; Blais, A.; Bowler, S.; Donovan, T.; Listhaug, O. Losers’ Consent: Elections and Democratic Legitimacy; Oxford University Press: Oxford, UK, 2005. [Google Scholar]
- Birch, S. Perceptions of electoral fairness and voter turnout. Comp. Political Stud. 2010, 43, 1601–1622. [Google Scholar] [CrossRef] [Green Version]
- Blais, A.; Gélineau, F. Winning, Losing and Satisfaction with Democracy. Political Stud. 2007, 55, 425–441. [Google Scholar] [CrossRef]
- Singh, S.; Lago, I.; Blais, A. Winning and competitiveness as determinants of political support. Soc. Sci. Q. 2011, 92, 695–709. [Google Scholar] [CrossRef]
- Singh, S.; Karakoç, E.; Blais, A. Differentiating winners: How elections affect satisfaction with democracy. Elect. Stud. 2012, 31, 201–211. [Google Scholar] [CrossRef]
- Howell, P.; Justwan, F. Nail-biters and no-contests: The effect of electoral margins on satisfaction with democracy in winners and losers. Elect. Stud. 2013, 32, 334–343. [Google Scholar] [CrossRef]
- Beaudonnet, L.; Blais, A.; Bol, D.; Foucault, M. The impact of election outcomes on satisfaction with democracy under a two-round system. Fr. Politics 2014, 12, 22–35. [Google Scholar] [CrossRef]
- Rose, J. The Public Understanding of Political Integrity: The Case for Probity Perceptions; Palgrave: London, UK, 2014. [Google Scholar]
- Delgado, I. How governing experience conditions winner-loser effects. An empirical analysis of the satisfaction with democracy in Spain after 2011 elections. Elect. Stud. 2016, 44, 76–84. [Google Scholar] [CrossRef]
- Van der Meer, T.W.; Steenvoorden, E.H. Going back to the well: A panel study into the election boost of political support among electoral winners and losers. Elect. Stud. 2018, 55, 40–53. [Google Scholar] [CrossRef]
- Martini, S.; Quaranta, M. Political support among winners and losers: Within-and between-country effects of structure, process and performance in Europe. Eur. J. Political Res. 2019, 58, 341–361. [Google Scholar] [CrossRef] [Green Version]
- Van der Eijk, C.; Rose, J. Winner–loser effects in contentious constitutional referenda: Perceptions of procedural fairness and the Brexit referendum. Br. J. Politics Int. Relat. 2021, 23, 104–120. [Google Scholar] [CrossRef]
- Daoust, J.F.; Plescia, C.; Blais, A. Are People More Satisfied with Democracy When They Feel They Won the Election? No. Political Stud. Rev. 2021. [Google Scholar] [CrossRef]
- Daniller, A.M.; Mutz, D.C. The dynamics of electoral integrity: A three-election panel study. Public Opin. Q. 2019, 83, 46–67. [Google Scholar] [CrossRef]
- Rich, T.S.; Holmes, C.E. Winning is not everything: Public perceptions of losers and non-voters in South Africa. Acta Politica 2016, 51, 328–345. [Google Scholar] [CrossRef]
- DeBell, M.; Krosnick, J.A.; Gera, K.; Yeager, D.S.; McDonald, M.P. The turnout gap in surveys: Explanations and solutions. Sociol. Method. Res. 2020, 49, 1133–1162. [Google Scholar] [CrossRef]
- McAllister, I.; Quinlan, S. Vote overreporting in national election surveys: A 55-nation exploratory study. Acta Politica 2021. [Google Scholar] [CrossRef]
- Electoral Commission. Elections and Referendums. 2022. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums (accessed on 27 February 2022).
- Electoral Commission. Report: Scottish Independence Referendum. 2015. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums/past-elections-and-referendums/scottish-independence-referendum/report-scottish-independence-referendum (accessed on 31 January 2022).
- Electoral Commission. Results and Turnout at the EU Referendum. 2019. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums/past-elections-and-referendums/eu-referendum/results-and-turnout-eu-referendum (accessed on 31 January 2022).
- James, T.S.; Clark, A. Electoral integrity, voter fraud and voter ID in polling stations: Lessons from English local elections. Policy Stud. 2020, 41, 190–209. [Google Scholar] [CrossRef]
- Hill, E.; Sobolewska, M.; Wilks-Heeg, S.; Borkowska, M. Explaining electoral fraud in an advanced democracy: Fraud vulnerabilities, opportunities and facilitating mechanisms in British elections. Br. J. Politics Int. Relat. 2017, 19, 772–789. [Google Scholar] [CrossRef] [Green Version]
- Electoral Commission. Investigation: Conservative and Unionist Party Campaign Spending Returns for the 2014 European Parliamentary Election, 2015 UK Parliamentary General Election, and the 2014 Parliamentary by-Elections in Clacton, Newark and Rochester and Strood. 2017. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/our-enforcement-work/investigations/investigation-conservative-and-unionist-party-campaign-spending-returns-2014-european-parliamentary (accessed on 24 February 2022).
- Van Schuur, W.H. Ordinal Item Response Theory: Mokken Scale Analysis; Sage: Thousand Oaks, CA, USA, 2011. [Google Scholar]
- Hardouin, J.B.; Bonnaud-Antignac, A.; Sébille, V. Nonparametric item response theory using Stata. Stata J. 2011, 11, 30–51. [Google Scholar] [CrossRef] [Green Version]
- Van der Eijk, C.; Rose, J. Risky business: Factor analysis of survey data–assessing the probability of incorrect dimensionalisation. PLoS ONE 2015, 10, e0118900. [Google Scholar] [CrossRef] [Green Version]
- Klimek, P.; Yegorov, Y.; Hanel, R.; Thurner, S. Statistical detection of systematic election irregularities. Proc. Natl. Acad. Sci. USA 2012, 109, 16469–16473. [Google Scholar] [CrossRef] [Green Version]
- Rose, J. Brexit, Trump, and post-truth politics. Public Integr. 2017, 19, 555–558. [Google Scholar] [CrossRef]
- Gaber, I.; Fisher, C. “Strategic lying”: The case of Brexit and the 2019 UK election. Int. J. Press/Politics 2022, 27, 460–477. [Google Scholar] [CrossRef]
- Farrall, S.; Wilks-Heeg, S.; Struthers, R.; Gray, E. Who are the victims of electoral fraud in Great Britain? Evidence from survey research. Br. Politics 2021, 1–20. [Google Scholar] [CrossRef]
- Van Ham, C. Getting elections right? Measuring electoral integrity. Democratization 2015, 22, 714–737. [Google Scholar] [CrossRef] [Green Version]
- Frank, R.W.; Coma, F.M. How election dynamics shape perceptions of electoral integrity. Elect. Stud. 2017, 48, 153–165. [Google Scholar] [CrossRef]
- James, T.S.; Clark, A. Delivering electoral integrity under pressure: Local government, electoral administration, and the 2016 Brexit referendum. Local Gov. Stud. 2021, 47, 186–207. [Google Scholar] [CrossRef]
- Klassen, A. Perceptions of Electoral Fairness: Public Behaviour and Institutional Design Compared across 80 Countries. 2014. Available online: https://openresearch-repository.anu.edu.au/bitstream/1885/125137/2/b35790246_Klassen_A_J.pdf (accessed on 27 February 2022).
- Mochtak, M.; Lesschaeve, C.; Glaurdić, J. Voting and winning: Perceptions of electoral integrity in consolidating democracies. Democratization 2021, 28, 1423–1441. [Google Scholar] [CrossRef]
- Birch, S. Electoral institutions and popular confidence in electoral processes: A cross-national analysis. Elect. Stud. 2008, 27, 305–320. [Google Scholar] [CrossRef]
- Appleby, J.; Federico, C.M. The racialization of electoral fairness in the 2008 and 2012 United States presidential elections. Group Process. Intergroup Relat. 2018, 21, 979–996. [Google Scholar] [CrossRef]
- Claassen, R.L.; Magleby, D.B.; Monson, J.Q.; Patterson, K.D. Voter confidence and the election-day voting experience. Political Behav. 2013, 35, 215–235. [Google Scholar] [CrossRef]
- Farrell, D.M.; McAllister, I. Voter satisfaction and electoral systems: Does preferential voting in candidate-centred systems make a difference? Eur. J. Political Res. 2006, 45, 723–749. [Google Scholar] [CrossRef]
- Moehler, D.C. Critical Citizens and Submissive Subjects: Election Losers and Winners in Africa. Br. J. Political Sci. 2009, 39, 345–366. [Google Scholar] [CrossRef] [Green Version]
- Maldonado, A.; Seligson, M.A. Electoral Trust in Latin America. In Advancing Electoral Integrity; Norris, P., Frank, R.W., Martínez i Coma, F., Eds.; Oxford University Press: New York, NY, USA, 2014; pp. 229–245. [Google Scholar]
- Wolak, J. How campaigns promote the legitimacy of elections. Elect. Stud. 2014, 34, 205–215. [Google Scholar] [CrossRef]
- Rosas, G. Trust in Elections and the Institutional Design of Electoral Authorities: Evidence from Latin America. Elect. Stud. 2010, 29, 74–90. [Google Scholar] [CrossRef]
- Nadeau, R.; Blais, A. Accepting the Election Outcome: The Effect of Participation on Losers’ Consent. Br. J. Political Sci. 1993, 23, 553–563. [Google Scholar] [CrossRef]
- McAllister, I.; White, S. Public Perceptions of Electoral Fairness in Russia. Eur.-Asia Stud. 2011, 63, 663–683. [Google Scholar] [CrossRef]
- Bruter, M.; Harrison, S. Inside the Mind of a Voter: A New Approach to Electoral Psychology; Princeton University Press: Princeton, NJ, USA, 2020. [Google Scholar]
- Electoral Commission. Public Attitudes. 2021. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/our-views-and-research/our-research/public-attitudes (accessed on 27 February 2022).
- Fieldhouse, E.; Green, J.; Evans, G.; Mellon, J.; Prosser, C.; de Geus, R.; Bailey, J.; Schmitt, H.; van der Eijk, C. British Election Study, 2019: Internet Panel, Waves 1–20, 2014–2020. [Data Dollection]. UK Data Service. SN: 8810. 2021. Available online: https://dam.ukdataservice.ac.uk/media/604956/besfieldhouse.pdf (accessed on 14 January 2021).
- Centre for Research into Elections and Social Trends (CREST). 1997 British General Election Study: Cross-Section Study, Scottish Election Study, Ethnic Minority Study—Codebook; CREST: London, UK, 1999. [Google Scholar]
- Electoral Commission. Report: How the May 2015 Elections Were Run. 2019. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums/past-elections-and-referendums/uk-general-elections/report-how-may-2015-elections-were-run (accessed on 25 February 2022).
- Electoral Commission. Report: Voting in 2017. 2019. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/elections-and-referendums/past-elections-and-referendums/uk-general-elections/report-voting-2017 (accessed on 25 February 2022).
- Electoral Commission. UK Parliamentary General Election 2019. 2020. Available online: https://www.electoralcommission.org.uk/sites/default/files/2020-04/UKPGE%20election%20report%202020.pdf (accessed on 25 February 2022).
- Electoral Commission. 2019 Electoral Fraud Data. 2021. Available online: https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/our-views-and-research/our-research/electoral-fraud-data/2019-electoral-fraud-data (accessed on 25 February 2022).
- Van Kuijk, S.M.; Viechtbauer, W.; Peeters, L.L.; Smits, L. Bias in regression coefficient estimates when assumptions for handling missing data are violated: A simulation study. Epidemiol. Biostat. Public Health 2016, 13, e11598-1–e11598-8. [Google Scholar]
- Papworth Trust. Facts and Figures 2018—Disability in the United Kingdom. 2018. Available online: https://www.papworthtrust.org.uk/about-us/publications/papworth-trust-disability-facts-and-figures-2018.pdf (accessed on 27 February 2022).
Wave Number | Dates Collected | Electoral Events That Fairness Question Relates to | Number of Respondents Presented with Prospective Fairness Question/Giving Valid Responses |
---|---|---|---|
W2 | 22 May 2014 to 25 June 2014 | Scottish independence referendum (18 September 2014) | 6047/5567 |
2015 UK general election (7 May 2015) | 5659/5193 | ||
W7 | 14 April 2016 to 4 May 2016 | EU membership (Brexit) referendum (23 June 2016) | 27,526/27,060 |
W11 | 24 April 2017 to 3 May 2017 | 2017 UK general election (8 June 2017) | 30,956/27,775 |
W17 | 1 November 2019 to 12 November 2019 | 2019 UK general election (12 December 2019) | 8488/7212 |
Scottish Referendum | 2015 General Election | EU (Brexit) Referendum | 2017 General Election | 2019 General Election | |
---|---|---|---|---|---|
(1) Will be conducted fairly | 41.2% | 38.2% | 28.7% | 32.2% | 27.6% |
(2) | 17.8% | 23.6% | 17.4% | 18.1% | 23.0% |
(3) | 18.1% | 20.7% | 23.0% | 22.8% | 23.7% |
(4) | 13.1% | 10.5% | 16.8% | 14.8% | 17.1% |
(5) Will be conducted unfairly | 9.9% | 7.1% | 14.1% | 12.1% | 8.6% |
Number of valid responses | 5567 | 5193 | 27,060 | 27,775 | 7212 |
Scottish Referendum | 2015 General Election | EU Referendum | 2017 General Election | 2019 General Election | |
---|---|---|---|---|---|
Age | −0.01 ** | −0.00 | −0.00 | −0.01 ** | −0.01 ** |
Level of education | −0.02 | −0.07 ** | −0.13 ** | 0.03 ** | −0.04 * |
Being disabled | 0.09 | 0.26 ** | 0.14 ** | 0.09 ** | 0.10 # |
Feeling at economic risk | 0.04 ** | 0.06 ** | 0.06 ** | 0.06 ** | 0.08 ** |
Political attention | −0.01 | 0.01 | 0.07 ** | 0.05 ** | 0.04 ** |
Finding government and politics difficult to understand | 0.04 # | 0.04 # | 0.07 ** | 0.08 ** | 0.04 |
Strength of party identification | −0.15 ** | −0.12 ** | −0.09 ** | −0.09 ** | −0.17 ** |
Approval of incumbent government b | −0.24 ** | −0.05 * | −0.08 ** | −0.50 ** | −0.22 ** |
Trust in MPs in general | −0.08 ** | −0.11 ** | −0.19 ** | −0.05 ** | −0.09 ** |
Number of observations | 2867 | 2750 | 12,527 | 12,738 | 2789 |
R2 | 0.06 | 0.07 | 0.11 | 0.27 | 0.10 |
Scottish Referendum | EU Referendum | 2017 General Election | 2019 General Election | |
---|---|---|---|---|
Age | −0.01 ** | −0.00 ** | −0.01 ** | −0.01 ** |
−0.01 ** | 0.00 * | −0.00 ** | −0.01 ** | |
Level of education | −0.01 | −0.08 ** | 0.01 | −0.04 * |
−0.04 # | 0.09 ** | −0.04 ** | −0.02 | |
Being disabled | 0.11 # | 0.12 ** | 0.09 ** | 0.09 # |
0.27 ** | −0.02 | 0.11 ** | 0.01 | |
Feeling at economic risk | 0.04 ** | 0.06 ** | 0.05 ** | 0.07 ** |
0.07 ** | 0.02 * | 0.04 ** | 0.06 ** | |
Political attention | −0.01 | 0.06 ** | 0.05 ** | 0.04 ** |
0.00 | 0.04 ** | 0.04 ** | 0.05 ** | |
Finding government and politics difficult to understand | 0.05 # | 0.06 ** | 0.08 ** | 0.02 |
0.02 | −0.01 | 0.08 ** | 0.00 | |
Strength of party identification | −0.13 ** | −0.07 ** | −0.08 ** | −0.18 ** |
−0.30 ** | −0.03 | −0.13 ** | −0.10 ** | |
Approval of incumbent government a | −0.23 ** | −0.17 ** | −0.43 ** | −0.20 ** |
0.22 ** | −0.16 ** | −0.21 ** | −0.24 ** | |
Trust in MPs in general | −0.10 ** | −0.19 ** | −0.08 ** | −0.08 ** |
−0.17 ** | −0.01 | −0.10 ** | 0.00 | |
Party choice Conservative | 0.39 ** | 0.52 ** | −0.17 ** | −0.23 ** |
−0.14 | −0.36 ** | −0.02 | −0.40 ** | |
Party choice Labour | 0.11 | 0.02 | 0.08 * | 0.07 |
−0.16 # | −0.06 | −0.02 | 0.53 ** | |
Party choice Liberal Democrat | −0.12 | −0.01 | 0.05 | −0.05 |
−0.19 | 0.07 | −0.09 # | 0.36 ** | |
Party choice SNP | 0.10 | 0.26 ** | 0.30 ** | 0.13 |
0.47 ** | 0.11 | 0.21 ** | 0.27 ** | |
Party choice UKIP (Brexit Party for 2019 general election) | 0.12 | 0.98 ** | −0.24 ** | 0.10 |
0.03 | −0.63 ** | −0.04 | −0.45 ** | |
Number of observations | 2368 | 12,006 | 12,789 | 2790 |
2763 | 10,002 | 10,600 | 8748 | |
R2 | 0.08 | 0.18 | 0.30 | 0.12 |
0.23 | 0.07 | 0.10 | 0.20 |
Prospective Fairness Beliefs | Statistics | ||||||
---|---|---|---|---|---|---|---|
(1) Fairly | (2) | (3) | (4) | (5) Unfairly | Entire Sample | ||
Scottish Referendum | 0.9% | 0.8% | 1.3% | 1.3% | 1.7% | 1.1% (n = 4292) | Chi-square = 3.43 (df = 4), p = 0.488 |
Cramér’s V = 0.03 | |||||||
2015 General Election | 3.7% | 5.1% | 8.2% | 6.6% | 8.0% | 5.6% (n = 3961) | Chi-square = 24.90 (df = 4), p < 0.001 |
Cramér’s V = 0.08 | |||||||
EU Referendum | 3.4% | 4.8% | 5.9% | 4.3% | 3.7% | 4.4% (n = 20,571) | Chi-square = 44.13 (df = 4), p < 0.001 |
Cramér’s V = 0.05 | |||||||
2017 General Election | 5.4% | 7.2% | 10.0% | 5.6% | 6.4% | 6.9% (n = 21,870) | Chi-square = 108.51 (df = 4), p < 0.001 |
Cramér’s V = 0.07 | |||||||
2019 General Election | 6.0% | 7.1% | 12.2% | 5.4% | 10.2% | 8.0% (n = 4889) | Chi-square = 46.80 (df = 4), p < 0.001 |
Cramér’s V = 0.10 |
n (% Non-Voters) | pseudoR2 | Coefficient of Fairness Beliefs (p-Value) | Percentage Correctly Classified Cases | AUC-ROC a | AIC b | BIC b | ||
---|---|---|---|---|---|---|---|---|
Scottish Referendum | 2829 | Baseline model c | 0.125 | n/a | 99.36% | 0.797 | 214.69 | 286.06 |
(0.64) | Baseline plus fairness beliefs | 0.136 | −0.28 (0.125) | 99.36% | 0.795 | 214.37 | 291.69 | |
2015 General Election | 2435 | Baseline model c | 0.088 | n/a | 96.30% | 0.736 | 726.19 | 795.76 |
(3.70) | Baseline plus fairness beliefs | 0.091 | −0.14 (0.124) | 96.30% | 0.739 | 725.88 | 801.25 | |
EU (Brexit) Referendum | 11,741 | Baseline model c | 0.082 | n/a | 96.88% | 0.728 | 3015.90 | 3104.35 |
(3.12) | Baseline plus fairness beliefs | 0.082 | 0.04 (0.331) | 96.88% | 0.727 | 3016.95 | 3112.77 | |
2017 General Election | 3070 | Baseline model c | 0.088 | n/a | 95.50% | 0.742 | 1050.46 | 1122.81 |
(4.50) | Baseline plus fairness beliefs | 0.089 | −0.04 (0.560) | 95.50% | 0.743 | 1052.12 | 1130.50 | |
2019 General Election | 3042 | Baseline model c | 0.094 | n/a | 94.81% | 0.720 | 1149.28 | 1221.52 |
(5.19) | Baseline plus fairness beliefs | 0.096 | −0.10 (0.145) | 94.81% | 0.729 | 1149.17 | 1227.43 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Rose, J.; van der Eijk, C. The World Isn’t Fair, but Shouldn’t Elections Be? Evaluating Prospective Beliefs about the Fairness of Elections and Referenda. Societies 2022, 12, 85. https://doi.org/10.3390/soc12030085
Rose J, van der Eijk C. The World Isn’t Fair, but Shouldn’t Elections Be? Evaluating Prospective Beliefs about the Fairness of Elections and Referenda. Societies. 2022; 12(3):85. https://doi.org/10.3390/soc12030085
Chicago/Turabian StyleRose, Jonathan, and Cees van der Eijk. 2022. "The World Isn’t Fair, but Shouldn’t Elections Be? Evaluating Prospective Beliefs about the Fairness of Elections and Referenda" Societies 12, no. 3: 85. https://doi.org/10.3390/soc12030085