The United Kingdom’s EU membership referendum:
Key psychological variables behind voting
‘remain’ and ‘leave’
Author: Alex I. Macdougall
Date: August 2
nd2017
Main body word count: 10,113
Total word count: 16,075
First supervisor: Dhr. Dr. Allard R. Feddes
Second supervisor: Dhr. Prof. Dr. E. J. Bertjan Doosje
Contents Abstract ………... 3 Introduction ………..…...……… 4 Method ……….….……….. 12 Results ………..……….….….……… 19 Discussion ………..……….……… 30 References ……….…..………… ……… 37 Appendices ………..………... 42
Appendix A: Introductory paragraph for survey ……..…..……… 42
Appendix B: Scales for measuring psychological variables …..………... 43
Appendix C: Relative gratification scale development process …..………… 45
Appendix D: Semi-structured interview schedule ……….…..….………….. 54
Appendix E: Script for requesting survey participation …….………. 57
Appendix F: Participant categorisation process ……..……….………….….. 58
Abstract
In June 2016, the United Kingdom (UK) voted to leave the European Union. Voting followed a general trend: disadvantaged areas of the UK tended to vote ‘leave’, and wealthier areas tended to vote ‘remain’. This project investigates the psychological variables that underlie this overall trend. In addition, ‘counter-trend voting’ was considered: why some disadvantaged individuals voted remain, and some wealthier individuals voted leave. Using a mixed method design, survey data (N=158) was gathered in parallel with interviews (N=3). Participants were recruited from deprived and affluent areas of Greater Manchester, and online. The survey assessed voting preferences, socio-economic status, and seven psychological variables: collective relative deprivation, realistic threat and symbolic threat from immigration, sense of control in life, relative gratification, European identification, and British identification. Prior to analysis, participants were categorised into four voting categories (poorer leave, poorer remain, wealthier leave, wealthier remain), based on their socio-economic status and voting preference. Discriminant function analysis determined the psychological variables that best explain category membership. As expected, leave voting was characterised by greater realistic threat, symbolic threat, collective relative deprivation, and British identification, whereas remain voting was characterised by European identification. Counter-trend voting followed a similar pattern. Relative gratification and sense of control in life failed to explain voting in the expected manner. Results are supported by themes found in interviews. This project contributes to our understanding of psychological theory and practice, including the development of a new tool for measuring relative gratification. In addition, findings highlight some of the divisions that exist currently in UK society.
Keywords: relative deprivation, identity, intergroup threat, relative gratification, control, Brexit, voting, immigration
Introduction
In June 2016, the United Kingdom (UK) voted by a 51.9% majority to leave the European Union (EU). Many commentators suggest that Britain’s exit from the EU – colloquially known as ‘Brexit’ – is a symptom of a deeply divided nation (e.g., The Economist, 2016a). Since the referendum, analysts have uncovered several striking relationships between regional socio-economic indicators and voting behaviour. Voting ‘leave’ was most prevalent in regions with higher rates of poverty and greater income inequality (Darvas, 2016). Moreover, a greater proportion of leave votes were cast in regions with lower educational attainment and lower wages (Resolution Foundation, 2016). These findings reveal an overall trend: disadvantaged regions voted leave in greater proportions, and more affluent regions voted remain.
There were, however, pockets of the UK that voted counter to this overall trend. To illustrate this, Figure 1 shows a scatterplot of 378 voting constituencies. Average hourly pay is plotted against the proportion of leave votes per constituency. An overall trend is evident: lower pay constituencies voted leave in greater proportions, and higher pay constituencies tended to vote remain. In addition, the figure highlights points that deviate markedly from this trend. Circled in blue are relatively high-pay constituencies that voted leave in large proportions, and circled in red are relatively low-pay constituencies that voted remain in large proportions. Thus, although overall the leave vote declines with increasing pay, some voting constituencies deviate from this trend.
A region-level analysis such as this offers a critical overview of the referendum result. However, to gain a more precise understanding of Brexit, it is necessary to consider Brexit in terms of individual and collective psychology. Objective social indicators, such as hourly pay, do not necessarily correspond with how individuals interpret their social environment, and their life circumstances (Postmes, Branscombe, Spears, & Young, 1999). In contrast, psychological measures are proximal to political behaviour, and are therefore more suited to revealing the factors that drove individuals to vote one way or the other – remain or leave.
The current research will establish key psychological variables that played a role in guiding voting behaviour during the EU referendum. There are two main objectives. First, to investigate the overall trend: what psychological variables explain why disadvantaged individuals were more inclined to vote leave, and why wealthier individuals were more inclined to vote remain? A second objective is to understand ‘counter-trend’ voting: why did some disadvantaged individuals go against the grain by voting remain, and similarly, why did some wealthier individuals vote leave?
Given that little psychological research has been conducted on Brexit, the current project employs a combination of confirmatory and exploratory approaches. The research offers several contributions to psychological theory and practice, including a test of the Social Identity Model of Collective Action (SIMCA; Van Zomeren, Postmes, & Spears, 2008), and a new tool for measuring ‘relative gratification’ (discussed later). Findings are expected to have considerable value for our understanding of UK society. By first establishing differences between voters at an individual and group level, it may be possible to comprehend what divides the UK as a whole. The
Figure 1. Median hourly pay plotted against percentage of leave votes per voting constituency.
Note. Figure sourced and adapted from Resolution Foundation (2016). Data on pay gathered by UK Office
following section presents the theoretical framework, and introduces the seven key psychological variables that are investigated; namely, collective relative deprivation, realistic threat, symbolic threat, sense of control in life, relative gratification, European identification, and British identification.
Explaining the overall trend: the disadvantaged vote leave, and the wealthy vote remain Research on collective action offers an explanation as to why disadvantaged individuals voted leave in greater proportions. Collective action refers to when an individual acts according to the aims and ideals of the group to which they belong. The goal of the action is to improve conditions for the group as a whole (Wright, Taylor, & Moghaddam, 1990). Action can be normative; for example, protesting peacefully at a climate change march; or nonnormative, such as committing violent acts of terrorism (Tausch et al., 2011). In 2008, Van Zomeren and colleagues proposed SIMCA, a model that predicts when collective action is most likely to occur. To do this, SIMCA integrates three domains of research. The first is relative deprivation theory (RDT), which proposes that collective action is more likely when an individual feels unjustly treated. According to RDT, individuals may feel a sense of injustice when they compare themselves, or their group, to higher-status individuals, or other higher-status groups. In 1966, Runciman formally distinguished between two types of comparison: those made between the ‘individual self’ and other individuals, and those made between the ‘collective self’ and other groups. He argued that the latter form, now known as collective relative deprivation (CRD), is a better predictor of collective action than its individual counterpart. Put simply, Runciman proposed that the injustice that individuals feel from comparing themselves as group members to other, higher status groups is a stronger predictor of collective action than when they compare themselves as individuals to other individuals. This proposition has since received support from meta-analysis (Smith & Ortiz, 2002).
The effects of CRD on collective action can be applied to the context of Brexit. In the UK, individuals from working-class communities could have felt deprived when comparing their group (whether that be their family, community, or the working-class in general) to higher-status groups, such as the middle- or upper-classes, or immigrant groups. Those who felt a sense of injustice from
this comparison would be more likely to vote in a manner that improves conditions for their group. During the referendum, voting leave was likely to be seen as the most effective means of doing this, because, by comparison, voting to remain in the EU promised little change to the status quo. RDT would predict that disadvantaged individuals would vote leave; a normative action intended to improve conditions for the group as a whole. Thus, in the current project, CRD is expected to underlie leave voting by disadvantaged individuals.
CRD alone, however, does not fully capture why disadvantaged individuals voted leave, specifically, rather than remain. The perceived threat of immigration is a factor that we must also consider, as immigration was central in political and media debate prior to the referendum, and key to the Vote Leave campaign (Leave Campaign, 2016). Jetten, Mols, and Postmes (2015) conducted a two-part study investigating the link between attitudes towards immigration and relative deprivation. Study 1 examined data from a similar context to the UK’s EU referendum; namely, a Swiss referendum in 2014, in which citizens voted in favour or against curbing levels of immigration. The authors found that deprived regions of the country voted in greater proportions to control levels of immigration. An experimental follow-up study revealed stronger opposition to immigration in a ‘lower-than-average wealth’ condition, compared to a ‘moderate wealth’ condition. These findings suggest that disadvantaged individuals feel most threatened by immigration. This conclusion has obvious applications to Brexit. Based on these findings, we would expect disadvantaged individuals in the UK to feel most threatened by immigration, and as a result vote to leave the EU to reduce their feelings of threat.
Researchers often investigate intergroup threat using two related constructs: symbolic threat and realistic threat (Riek, Mania, & Gaertner, 2006; Doosje, Loseman, & Van den Bos, 2013). Specific to the UK context, symbolic threat is defined as a perception that immigration threatens British values and culture. Realistic threat, however, is the perceived threat that immigration poses to more concrete aspects of life, such as jobs and housing. Using meta-analysis, Riek and colleagues showed that both types of threat are associated with stronger negative attitudes towards outgroups. Additionally, Stephan, Renfro, Esses, Stephan, and Martin (2005; Study 1) manipulated symbolic and realistic threat, and found that both types elicit negative attitudes towards immigrant
groups. Thus, based on this research, we can expect symbolic threat and realistic threat to underlie voting leave; particularly for disadvantaged individuals.
Identity is also important to a post hoc understanding of the referendum. Social identity theory (SIT) is a key constituent of SIMCA, which predicts that collective action is more likely when identity concerns are salient. Both British identity and European identity were called into question during the referendum. As such, SIMCA would suggest that many individuals may have voted in order to address their identity concerns. First, let us consider British identity and voting leave. Research on SIT shows that stronger ingroup identification is related to greater outgroup discrimination (Perreault & Bourhis, 1999). Applied to Brexit, voting to leave the EU may have been an attempt by those who identify strongly as British to create distance from relevant outgroups (e.g., Europeans, immigrants), and to emphasise who is British and who is not. Therefore, the extent to which individuals identify as British, known hereafter as British identification, is expected to underlie voting leave. The dynamics of SIT also apply to those who identify as European. SIT states that we are driven to maintain a positive social identity (Tajfel & Turner, 1979). Accordingly, SIT would predict that voters who identify strongly as European were motivated to maintain their current European identity by voting remain. Thus, European identification is expected to underlie remain voting.
Sense of control in life is another psychological variable that is expected to explain the overall voting trend. Having a sense of control over life’s outcomes is considered a basic human need, and lacking control is generally experienced as aversive (Rutjen & Kay, 2015). Research on compensatory control theory reveals interesting insights into how individuals deal with losing control. Landau, Whitson, and Kay (2015) investigated how individuals reinstate their sense of control, and found that the way in which control is restored need not necessarily be related to the context in which it was taken away.
These ideas can be applied to Brexit, and may explain why disadvantaged voters tended to vote leave. The UK has particularly poor social mobility (Social Mobility Commission, 2016), meaning that disadvantaged individuals are less able to create better lives for themselves. In addition, prolonged austerity measures following the 2008 financial crash have hit the poor
hardest, and evidence suggests there may be a link between austerity and feelings of powerlessness among the disadvantaged (Psychologists Against Austerity, 2015). During the referendum, the Leave campaign’s slogan was ‘Vote Leave, take back control’, which referred primarily to regaining control over laws made by the European Commission in Brussels (Leave Campaign, 2016). Indeed, an impression grew in some sectors of the public that Brussels had too much control over UK law (BBC News, 2016). In this socio-political climate, compensatory control theory would expect disadvantaged individuals to vote leave. Disadvantaged individuals may have seen voting leave to be the more attractive option because leaving the EU would eliminate the influence of Brussels in British affairs, and restore a sense of control in everyday life, via compensatory control mechanisms. Therefore, in part, lacking control in life is expected to explain why disadvantaged individuals voted leave.
The opposite pattern is expected for affluent individuals. By contrast, the Remain campaign focused on the uncertainty of leaving the EU, and the likelihood of economic instability following Brexit (Behr, 2016; Remain Campaign, 2016). Without the full burden of austerity and poor social mobility, it is likely that, compared to the disadvantaged, affluent individuals had a stronger sense of control in life. Accordingly, theory on perceptions of control would predict that affluent individuals would vote remain to protect their existing sense of control. Thus, a high sense of control is expected to underlie affluent individuals’ decisions to vote remain.
To summarise thus far, six variables are expected to explain the overall voting trend. Specifically, leave voting by disadvantaged individuals is expected to be characterised by feelings of collective relative deprivation, perceptions of symbolic and realistic threat from immigration, strong British identification, and a low sense of control in life. In contrast, strong European identification and a greater sense of control are expected to underlie remain voting by affluent individuals. The following section presents a theoretical rationale of counter-trend voting, and why some individuals may have voted against the overall trend.
Explaining counter-trend voting: when the wealthy vote leave, and the disadvantaged vote remain
As discussed, several voting constituencies deviated from the overall trend. For instance, Havering, an affluent borough of London, voted 70% in favour of leaving the EU, whereas Glasgow City, home to many deprived communities, voted 67% remain (Resolution Foundation, 2016). So far, our theoretical rationale cannot account for these instances. Thus, what psychological variables explain these ‘counter-trend’ voting patterns?
The ‘V-curve’ hypothesis offers an explanation for why some wealthier individuals voted leave, rather than following the overall trend. In essence, the hypothesis makes two predictions regarding the strength of negative outgroup attitudes. The first prediction refers to the effects of collective relative deprivation, outlined above: individuals will show more opposition to outgroups when they feel unfairly worse off than others. The second prediction proposes that individuals also oppose outgroups when they feel better off than others. This is known as relative gratification (for an overview, see Anier, Guimond, & Dambrun, 2016). Jetten, Mols, and Postmes (2015) found experimental evidence in support of the V-curve hypothesis. Participants showed stronger opposition to immigration in a lower-than-average wealth condition and in a higher than average wealth condition, relative to a moderate wealth condition. This finding suggests that relatively deprived and relatively gratified individuals may feel threatened by immigration to a similar extent.
As such, relative gratification, symbolic threat, and realistic threat are expected to underlie leave voting by wealthier individuals. At first, it may seem unlikely that wealthier individuals voted leave because they feel threatened economically by immigration (i.e., realistic threat). However, research suggests that relatively gratified people show stronger negative attitudes towards outgroups if they perceive their own privileged status to be unstable (LeBlanc, Beaton, & Walker, 2015), or if they fear that their wealth will decline in the future (Jetten, Mols, & Postmes, 2015). Finally, in addition to relative gratification, symbolic threat and realistic threat, British identification is expected to underlie leave voting by wealthier individuals. As was the case for disadvantaged leave voters, affluent individuals may have voted leave to emphasise who they believe does and does not qualify as British.
It is likely that poorer leave voters and poorer remain voters felt similar levels of group-based injustice (i.e., CRD); yet, poorer remain voters acted counter to what RDT would predict. This phenomenon suggests that motivations related to CRD that characterise trend leave voting were overridden by another, stronger motivation. In such a case, SIMCA would advocate identity concerns as the alternative explanation. This seems plausible in the context of Brexit. To give an example, all 32 voting constituencies in Scotland voted in favour of remaining in the EU (Electoral Commission, 2016). This indicates that the motivation to vote remain cut across class boundaries, and was not related to perceived differences in wealth. Thus, in Scotland – and indeed other regions of the UK – the desire to preserve European identity may have taken precedence over the motivation to address group-based injustices. Therefore, in the current study, European identification is expected to be the main variable that underlies remain voting by disadvantaged individuals.
The present research
Based on the reviewed research on collective action, intergroup threat, social identity theory, compensatory control theory, and the V-curve hypothesis, three hypotheses are proposed:
Hypothesis 1: trend voting. Collective relative deprivation (CRD), realistic threat (RT), symbolic threat (ST), British identification (BI), European identification (EI), and sense of control in life (C) will distinguish leave voters from remain voters overall, with high CRD, RT, ST, BI and low C characterising leave voters, and high EI and C characterising remain voters. Hypothesis 2: counter-trend leave voting. Relative gratification (RG), RT, ST, and BI will best
distinguish wealthier leave voters from wealthier remain voters.
Hypothesis 3: counter-trend remain voting. EI will best distinguish poorer remain voters from poorer leave voters.
These predictions will be tested in a mixed-method design using a survey and interviews. Data will be gathered from remain and leave voters from low socio-economic status (SES) and high-SES areas of the UK.
Method Participants
Survey. Two approaches were used to collect survey data. The first approach took place across Greater Manchester, UK. Eighty-eight participants were recruited in public spaces (e.g., train stations, public libraries, shopping centres), from high-SES (Central Manchester) and low-SES areas (Wigan, Salford, Gorton, and Newton Heath). A further 70 participants were recruited via online forums1. This method ensured a more representative sample, and a greater number of counter-trend voters. In total, 158 participants were recruited (44% female, mean age: 40.3, age range: 18–79). Ninety-six percent of participants were born in the UK, and 19% of participants had at least one parent born abroad. Participants were recruited for the survey on a voluntary basis.
Interviews. Potential interviewees were approached for participation in public spaces. Three people were interviewed: two remain voters (referred to as HL and AD, aged 24 and 64 respectively) and one leave voter (DR, aged 67). To avoid bias in interview data, interviewees did not take part in the survey. Interviewees received £20 for their participation.
Design and strategy
This study has a mixed method design. A particular strategy was taken for the quantitative component, which requires explanation. During the survey, participants were asked two voting questions: 1) what they voted in the actual EU referendum (remain, leave, or didn’t vote), and 2) what they would vote if there was a second referendum now (remain or leave). Participants’ revote decision, rather than their original vote, was used in quantitative analysis. This approach was taken for three reasons.
First, data was collected ten months after the original referendum, and since then, people’s psychological perspective and/or voting preference may have changed. Using a revote ensured that voting behaviour was measured at the same time as psychological variables, rather than many months apart. This allows results to represent a temporally proximate relationship between
1 Online forums: ‘Pie and Bovril’ (a Scottish football forum), ‘One Football Forum’ (a predominantly
psychological motivation and political behaviour, and therefore more closely reflects real voting processes. Second, the author expected only a small proportion of participants to change their vote since the referendum. Results showed that this was the case: only 8% (N = 11) of those who voted in the original referendum changed their decision at revote. Thus, no drastic changes in voting decision had occurred since the original referendum. Thirdly, using a revote meant that individuals who did not vote originally could take part2. This had a practical advantage: all individuals who were approached for participation were eligible, unless they indicated that they had no voting preference at all. Furthermore, the author saw no valid reason to exclude non-voters. As long as participants had a voting preference, the theoretical arguments given in the introduction would still apply.
This study received ethical approval from the University of Amsterdam’s Faculty of Social and Behavioural Sciences Ethical Review Board on March 8th 2017 (ID: 2017-SP-7764).
Materials
Survey. The survey began with a paragraph that contextualised the EU referendum and introduced the current research (see Appendix A). Participants were then asked to indicate their ‘revote decision’ (“What would you vote if there was a second referendum now?”; options: Remain a member of the European Union, or Leave the European Union), and their original voting decision (same options as revote, plus ‘I didn’t vote’). Next, participants filled out seven scales used to measure psychological constructs. The order of these scales, and the order of items within the scales, were randomised to control for order effects. Participants’ SES was then assessed using four socio-economic indicators. Finally, four items were used to gather additional demographic information. Psychological variable scales, SES indicators, and additional survey items are introduced below.
Collective relative deprivation. CRD was measured using a four-item version of a scale used
by Doosje, Loseman, and Van den Bos (2013). An example item is, “If I compare people like me to other groups in the UK, I feel that we are treated unfairly”, responded to on a seven-point Likert scale, from ‘strongly disagree’ (1) to ‘strongly agree’ (7)3. In the current study, scale reliability was good (Cronbach’s α = .88).
Realistic threat. Realistic threat was measured using a four-item version of that used by
Doosje et al. (2013). An example item is, “Immigrants hold too many positions of power and responsibility in the UK”, responded to as above4. Scale reliability was good (α = .88).
Symbolic threat. Symbolic threat was measured using a four-item version of that used by
Doosje et al. (2013). An example item is, “Immigrants in the UK don’t understand how my group views the world”, responded to as above. Scale reliability was good (α = .89).
Sense of control in life. An individual’s general sense of control in life was measured using
a five-item version of the constraint subscale of Lachman and Weaver’s (1998) Mastery and Constraint scale, as used by Clarke, Fisher, House, Smith, and Weir (2008). An example item is, “What happens in my life is often beyond my control", responded to as above. Scale reliability was good (α = .85).
Relative gratification. A new measurement tool was developed in the current project to
measure relative gratification. Readers are encouraged to observe a detailed report of the development process in Appendix C; however, the process is outlined briefly below.
A five-part process was used to develop the scale, following guidelines given by Hinkin (1998). Phase one, item generation, involved 1) reviewing seven core papers to acquire a thorough understanding of the content domain, 2) creating a theoretical definition, 3) generating 12 initial
3 The original scale used by Doosje et al. (2013) focused specifically on ‘Muslims’ as the ingroup. The scale
was later adapted for use with a more general sample, using the term ‘my group’. This term was considered too abstract, and so ‘people like me’ was used instead. One criticism of this is that using the word ‘me’ could induce participants to consider their individual, rather than their collective, self. Nevertheless, the phrase ‘people like me’ prompts participants to consider similar others, and is often used in items in conjunction with a comparison to ‘other groups’, as in the example given. Given these points, the author feels that using ‘people like me’ adequately preserves the original, collective conceptualisation, whilst reducing the ambiguity of ‘my group’.
4 The original scale used the phrase ‘other groups’ to refer to threatening outgroups. However, the outgroup
in current project could be specified more precisely, and thus, ‘other groups’ was substituted with ‘immigrants’.
items, and 4) performing a content validity assessment with five participants. For the questionnaire administration phase, 34 participants took part in a pilot study in the Netherlands. Four scales were administered: the realistic threat scale, symbolic threat scale, CRD scale, and the 10 relative gratification items that passed the content validity assessment. In phase three, item reduction, principal axis factor analysis was used to create a more parsimonious, four-item scale for use in the survey of the main study. In phase four, reliability and convergent/discriminant validity were assessed. The scale demonstrated acceptable reliability (α = .73), excellent discriminant validity, but inadequate convergent validity. Nonetheless, the final phase, replication, offered a second test of the scale’s psychometric properties, using survey data from the main study (N = 158). Again, the scale showed satisfactory reliability (α = .73). The scale also demonstrated excellent divergent validity (relative gratification and CRD were found to be orthogonal; r = .05) and excellent convergent validity (significant positive correlations were found between relative gratification and realistic threat, r = .28, and relative gratification and symbolic threat r = .28).
In summary, the final four-item relative gratification scale showed satisfactory properties as a new measurement tool. An example item from the scale is, “I deserve to feel satisfied with how my life is”, responded to on a seven-point Likert scale, from ‘strongly disagree’ (1) to ‘strongly agree’ (7).
European identification. The extent to which participants identify as European was assessed
using the Inclusion of Other in Self scale (IOS; Schubert & Otten, 2002). IOS asks participants to indicate their ‘closeness’ with a target group (in this case Europeans) by choosing one of seven pictorial representations that increase in closeness. Figure 1 shows two example items.
British identification. As for European identification, IOS was used to assess the extent to
which participants identify as British (Schubert & Otten, 2002). The target group in this case was 'The British', instead of 'Europeans'.
Four indicators of SES. Oakes (2016) recommends collecting as much socio-economic
data as reasonably possible. Therefore, four univariate indicators of SES were assessed.
1) Household income (controlling for student status). Used as the primary indicator of SES. Participants were asked to select one of seven income brackets (e.g., ‘£200 - £399 weekly /£867 - £1,732 monthly/£10,400 - £20,799 annually’).
2) Educational attainment. Secondary indicator of SES. Participants were asked to select one of seven response options, ranging from ‘no formal qualifications’ to ‘additional qualifications including foreign qualifications’ (as used by Office for National Statistics, 2011).
3) Area of residence. Tertiary indicator of SES. All 26 voting constituencies of Greater Manchester were listed, plus an open field for those living outside of Greater Manchester.
4) Occupational prestige. Quaternary indicator of SES. Participants asked to select either ‘student’ or ‘unemployed’, or they could indicate their occupation in an open field option.
Prior to conducting quantitative analyses, SES indicators were used in the rank order shown above to categorise participants into four voting groups; namely, poorer leave, poorer remain, wealthier leave, and wealthier remain. The categorisation process is described briefly in the results section, and in greater depth in Appendix F.
Additional demographic measures. Participants were asked to indicate their age and
gender, their immigration background, including their own birthplace, mother’s birthplace, and father’s birthplace (response options were the five most common countries of origin in the UK: UK, India, Poland, Pakistan, and Republic of Ireland, or an open field), and their political orientation (response options: the seven parliamentary parties were listed, plus an open field, and ‘I spoil my ballot’).
Interviews. Interviews followed a semi-structured schedule, covering six themes. In theme 1, interviewees were asked to describe their upbringing, education, and employment. Theme 2 prompted interviewees for the motivations behind their vote. To begin, the interviewer asked for the core reason for why they voted remain or leave, and subsequently, interviewees were shown images of the official materials used by the Remain and Leave campaigns (see Figure 3). These images displayed the key arguments of each campaign and their slogan. Interviewees were asked to comment on the images, drawing on anything that felt significant to them. In theme 3, interviewees were asked to talk about family life, their financial situation, and their life in general. Theme 4 explored British and European identity. Specifically, the interviewer was interested in the way in which interviewees felt British and/or European, and whether their identity had changed since the referendum. Theme 5 asked interviewees about their feelings towards the so-called ‘political elite’, or ‘establishment’, a topic not covered explicitly by the survey. Here, the aim was to allow voters to express sentiment towards UK and EU politicians, and in particular, Brussels and the European Commission. The final theme prompted interviewees to comment on their feelings about leaving the EU, and their hopes and fears of what will happen. The full interview schedule, complete with prompts and follow-up questions, can be found in Appendix D.
Procedure
Survey. The researcher followed a script when recruiting survey participants in person (full script shown in Appendix E). This step was taken to mitigate bias during initial interaction. Once participants agreed to be involved in the study, the survey was administered via the Qualtrics Offline application on an iPad. After giving consent electronically, participants filled in the survey alone and in their own time5, which took around eight to ten minutes on average. On completion, participants were given a personalised, anonymising code, and contact details to withdraw from the study if desired. Those who completed the survey online were asked to submit a code of their choice, and asked to write it down, together with contact details for withdrawal.
Interviews. Interviewees were asked to choose a quiet public setting for the interview to take place. Before the interview began, the interviewer introduced themselves and the research project. Each participant was made aware that data would be kept anonymous, and that they had the right to withdraw from the study at any time. Interviews lasted between an hour and an hour and 45 minutes. A recording was taken of each interview on a dictaphone and later transcribed. Full transcripts may be accessed via the link in this footnote6.
5 A small number of participants (≈ 5) asked the researcher to take them through the survey item-by-item. In
such cases, the researcher remained as neutral as possible during interaction.
6
Results
Results are presented in three sections. Section 1 provides a description of how participants were categorised into four voting groups, descriptive statistics for each voting group, and intercorrelations between psychological scales. Section 2 gives the results for each hypothesis. Quantitative and qualitative findings are presented in parallel for hypothesis 1, followed by quantitative results for hypotheses 2 and 3. Finally, section 3 presents an additional, exploratory theme that was found in interview data.
Section 1: Voting group categorisation, descriptive statistics, and intercorrelations
Prior to analysis, participants were categorised into four voting groups; namely, poorer leave, poorer remain, wealthier leave, and wealthier remain. This was done based on participants’ revote decision and their socio-economic data. First, a median split of household income data was used to classify individuals as ‘poorer’ or ‘wealthier’. This label (e.g., ‘poorer’) was combined with participants’ revote decision (‘leave’), giving each participant a voting category (‘poorer leave’). Some participants could not be categorised by a median split of household income, as income data was ordinal. In such cases, a median split of the secondary SES indicator, education attainment, was used. If participants were unable to be categorised in this manner, area of residence data was used. This process assigned each participant to a voting category. A more detailed description of the categorisation process may be found in Appendix F.
Table 1 shows the number of participants in each voting category, and descriptive statistics for each psychological variable7. In addition, Table 2 gives the intercorrelations between the seven key psychological variables.
7
Table 1. Sample size of each voting category, group means and standard deviations for all psychological variables. Psychological variable MPL (SD), N = 22 MPR (SD), N = 34 MWL (SD), N = 22 MWR (SD), N = 80 Col. rel. deprivation 3.89 (1.65) 3.33 (1.58) 2.82 (1.11) 2.35 (1.15) Realistic threat 3.63 (1.76) 2.31 (1.29) 3.40 (1.47) 1.90 (.93) Symbolic threat 3.32 (1.63) 2.21 (1.09) 3.03 (1.22) 1.92 (.90) Sense of control1 4.99 (1.38) 5.04 (1.31) 5.78 (.54) 5.30 (1.01) Relative gratification 5.02 (.98) 4.68 (1.12) 5.45 (.58) 4.96 (1.12) European identification 2.31 (.99) 4.97 (2.04) 3.45 (1.87) 4.86 (1.63) British identification 5.55 (1.65) 3.76 (2.05) 6.00 (1.54) 4.49 (2.09)
Note. PL = poorer leave, PR = poorer remain, WL = wealthier leave, WR = wealthier remain. Sample sizes
displayed are subsequent to removing multivariate outliers (see footnote 9). 1The sense of control scale was reverse coded so that lower values represent less perceived control in life.
Table 2. Intercorrelations for key psychological variable scales.
1. 2. 3. 4. 5. 6.
1. Col. rel. deprivation -
2. Realistic threat .457** - 3. Symbolic threat .432** .773** - 4. Sense of control1 -.436** -.292** -.254** - 5. Relative gratification .054 .277** .275** .046 - 6. European identification -.141 -.486** -.428** .015 -.099 - 7. British identification -.086 .132 .161* .405** .240** -.171*
Note. **: p < .01, * p < .05. 1The sense of control scale was reverse coded so that lower values represent less perceived control in life.
Section 2: Results for hypotheses 1 – 3
Discriminant Function Analysis (DFA) was used on survey data to test each hypothesis. DFA is appropriate for designs that have a single, categorical outcome variable (with ≥2 levels) and several continuous predictors. DFA tests how useful predictor variables are for determining category membership and works by creating one, or several, latent functions from linear combinations of predictors. A partial coefficient is then estimated for each predictor variable (equivalent to standardised betas in regression), which represent the contribution of each predictor towards determining participants’ category membership.
In current analyses, the predictors are the seven psychological variables8, and the categorical outcome variable represents voting group (using different combinations of groups for each DFA). Therefore, in the present context, DFA evaluates which psychological variables are most responsible for differences in voting behaviour (indicated by a coefficient’s relative magnitude), and in what way it is responsible (indicated by a coefficient’s valence). Results from DFA1 are now presented for hypothesis 1, in parallel with qualitative findings.
Hypothesis 1: The overall trend. DFA1 used ‘voting group’ as an outcome variable (with four levels: poorer leave, poorer remain, wealthier leave, wealthier remain) and the seven psychological variables as predictors9. The analysis generated three discriminant functions: the first explained 72.9% of the variance in voting group (canonical R2 = .39), the second 19.6% (canonical R2 =.14), and the third 7.5% (canonical R2 =.06). In combination, these discriminant functions significantly differentiated voting groups, L = .493, χ2
(24) = 106.67, p < .001. Removing
the third function indicated that it did not significantly differentiate between groups, L = .939, χ2 (6)
8 Age was added as a predictor in each DFA to control for (or ‘partial out’) its effect on voting group
membership. Gender could not be added given that it is categorical. Instead, a one-way ANOVA was used to test differences between gender on all psychological variables. One significant difference was found: females showed stronger British identification, F(1,156) = 10.61, p < .001. However, given that this was the only difference, the effect of gender on voting category membership was assumed to be relatively minor.
9 Prior to analysis, eleven multivariate outliers were removed, leaving 158 participants. Shipiro-Wilk
multivariate normality tests indicated normality was violated for the wealthier remain group. Closer inspection revealed that such data were skewed. However, Tabachnick and Fidell (2007) state that “…discriminant analysis, like MANOVA, is robust to failures of normality if violation is caused by skewness rather than outliers” (p. 382). Thus, after removing outliers, data was deemed appropriate for discriminant analysis.
= 9.45, p = .150, and it was therefore disregarded. Additional calculations10 established function 1 as a factor that discriminates between voting leave and voting remain overall, irrespective of wealth, and function 2 as a factor that discriminates poorer groups from wealthier groups, irrespective of voting. Thus, although results of function 2 are presented below, function 1 is the only relevant function for evaluating hypothesis 111.
Table 3. The contribution of each psychological variable in discriminating leave voters from remain voters overall (function 1) and poorer voters from wealthier voters overall (function 2).
Psychological variable Function 1 Function 2
European identification -.516 -.071
Realistic threat .414 -.262
Collective relative deprivation .330 .769
British identification .329 -.349
Sense of control .235 -.147
Symbolic threat .129 .034
Relative gratification -.107 -.391
Note. Higher positive coefficients signify key variables for leave voters (function 1) or poorer voters
(function 2); whereas, higher negative coefficients signify key variables for remain voters or wealthier voters. Table ordered by absolute magnitude in function 1. Coefficients are standardised and partial.
Several results in Table 3 (function 1) are consistent with hypothesis 1: as predicted, European identification, realistic threat, collective relative deprivation, and British identification are variables that best discriminate leave voters from remain voters overall. Moreover, the valences of their coefficients indicate that results follow the expected pattern; specifically, a negative valence for European identification indicates that it is key for remain voters, whereas positive
10 Two steps were taken to establish which groups a particular function discriminates between. First, the
mean of discriminant scores for participants within each voting group was calculated. The second step was to determine which group means have opposing valences. Groups that have means of opposing valances are those that the function discriminates between. For example, in DFA1, mean values of discriminant scores for each voting group on function 1 were: poorer leave = -1.48, wealthier leave = -.98, poorer remain = .31, wealthier remain = .54. Thus, based on the valances, function 1 discriminates leave voting groups from remain voting groups.
11 Function 2 shows a number of interesting exploratory findings. Firstly, objectively poorer, low-SES voting
groups feel greater levels of (subjective) collective relative deprivation. Second, symbolic threat does not differ across wealth. Third, overall, wealthier individuals feel more gratified in life than poorer individuals.
coefficients for realistic threat, collective relative deprivation, and British identification suggest that these variables are key for leave voters.
Symbolic threat was also predicted to be key for leave voters. Although its coefficient is positive, its relatively small magnitude does not at first support this prediction. However, these coefficients represent the unique contribution made by each variable to the discriminating function, and given that symbolic threat and realistic threat correlate highly (r = .77; see Table 2), it is likely that multicollinearity masks the effect of symbolic threat. To investigate further, a second type of coefficient was examined, known as canonical variate correlation coefficients (CVCCs). CVCCs are comparable to factor loadings, and therefore do not represent unique effects, but instead signify the substantive nature of each predictor (Field, 2013). CVCCs show that symbolic threat is the third most important variable (r = .619), behind the two largest contributors, realistic threat (r = .707) and European identification (r = -.701). Together, these findings suggest that symbolic threat is indeed key to explaining why individuals voted leave, but not in excess of that explained by realistic threat.
One finding is inconsistent with hypothesis 1. Sense of control made a small contribution to function 1, but in an opposing manner to what was predicted. Contrary to expectation, leave voters were found to have a greater sense of control in life overall.
European identification was the most important variable for remain voters in quantitative analysis. In line with this, a theme emerged in interviews, called ‘maintaining European identity’. The following quote exemplifies this theme:
HL1: I think it was my sense of identity that mainly caused me to vote the way I did. I like
to be at one with the rest of Europe, as a country… it was a sense of identity, and maintaining a European identity was the main reason I voted to remain.
It is clear that HL identifies as a European. She explains that maintaining European identity was the main motivation behind her vote to remain in the EU. Evidence for this theme was
also found in responses from AD. When asked about the main reason he voted remain, AD responded with:
AD1: Well, I think, it’s partly to do with identity. I think I identify more as a European
than I do English... I want to be collaborative with other countries, if that’s at all possible.
This excerpt suggests that maintaining European identity was also important for AD, and he wants to preserve his European principles, such as collaboration with other countries.
In contrast, quantitative findings showed that British identification is a key psychological variable for leave voters. Qualitative findings support this conclusion, but with added nuance. When asked about being British, leave voter DR replied with:
DR3: English, well, I’m English. I support Britain.
This response suggests that, although DR supports Britain, ‘British’ is a term that is too broad to describe DR and his identity. This quote introduces a key theme found across interviews; namely, ‘broader vs. narrower self-concepts’. Here, remain voter HL describes her identity:
HL2: I am pleased to say I’m from the UK – I don’t say Britain actually, I don’t really
refer to England. I refer to ‘the United Kingdom’. The idea of united countries, I really like that actually.
In contrast with leave voter DR, HL says that she prefers to use ‘the United Kingdom’ to describe herself, instead of ‘Britain’ or ‘England’. This implies that HL has a broader conception of her own identity than DR, who emphasised being English, rather than British. In addition, HL supports the idea of countries collaborating, which is a sentiment echoed by remain voter AD:
AD2: I used to feel British because it felt as if it was all one country. And I felt European
as well. But now all this shit has hit the fan, I now don’t feel British, I now feel European. I want to feel European.
In contrast with leave voter DR, AD prefers the idea of Britain as one united country, and AD stresses his desire to be European. Thus, interview and survey data suggest a difference between remain and leave voters in terms of identity and self-concept. Leave voters tend to have a more restrictive identity, and a narrower self-conception, and in contrast, remain voters tend to have a more inclusive identity, and a broader self-conception.
Quantitative analysis revealed realistic threat as the variable most central to voting leave. The perceived economic threat of immigration was also a theme found in an interview with leave voter, DR.
DR1: Leave, one main reason… I think… To stop immigrants coming in [to the] country.
This excerpt shows that stopping immigration was a key motivation behind his decision to vote leave. DR elaborates on this later in the interview:
DR2: … all these Syrian people, and asylum seekers, y’know, they come over and then they
find ‘em a house, straight away near enough, give them benefits for three kids, ‘you get so much each week’. And they must be thinking, ‘bloody hell, I’m glad I moved here’. They’ve not put nothing in the pot! I’ve been putting in the pot since I were fifteen, since I left school. Y’know, they’ll get family allowance for their kids, they’ll get free rent, free council tax, all them. They’ve not put nothing in the pot!
DR explains that, in his view, Syrian people and asylum seekers are given housing and other financial benefits too readily. This illustrates that DR perceives that immigrant groups are taking from, and encroaching on, economic resources in the UK.
In addition, the excerpt above provides evidence for the role of collective relative deprivation in voting leave. DR’s explanation reveals a strong sense of injustice, and the phrase ‘putting into the pot’ is repeated several times. DR compares himself to “Syrians and asylum seekers”, claiming that he has contributed significantly to ‘the pot’ (paraphrased: ‘UK society’), whereas these groups have not. This group-level injustice is observed in the following excerpt:
DR3: I see somebody in the car yesterday, two guys with their head on upside down, you
know what I mean? Bald head, beard down here. Big black ’66 Audi, both of ‘em in it, y’know. There’s something not right. People work all their life and can’t afford a car like that. There’s something amiss.
DR describes seeing two Muslim men driving an expensive car, and makes a comparison to others who work all their life but cannot afford a commodity such as this. Although not explicitly mentioned, it appears that his use of the term ‘people’ refers to members of the working-class. Phrases such as ‘something not right’ and ‘something amiss’ illustrate the injustice that DR feels from this comparison. In this regard, the revelations in DR’s interview support to the hypothesis that collective relative deprivation underlies leave voting.
In summary, the majority of quantitative findings are in line with hypothesis 1, regarding the overall trend. As expected, higher levels of collective relative deprivation, realistic threat, symbolic threat, and stronger British identification were found to underlie leave voting, and stronger European identification characterised remain voting. Except for symbolic threat from immigration, all findings were supported by interview accounts with leave and remain voters. No evidence was found for the prediction that leave voters lack control in everyday life. Contrary to expectations, leave voters were found to have greater, rather than less, control in everyday life.
Hypothesis 2: Counter-trend leave voting. DFA2 used ‘voting group’ as an outcome variable (with two levels: wealthier leave, wealthier remain), and the seven psychological variables as predictors12. A single discriminant function significantly discriminated the two voting groups, L = .650, χ2(8) = 41.411, p < .001.
Table 4 shows the discriminant function coefficients for each psychological variable. As predicted by hypothesis 2, realistic threat was most important in distinguishing between wealthier leave and wealthier remain voters: wealthier leave voters feel more threatened by immigrants in economic terms than wealthier remain voters. As in DFA1, symbolic threat appears to have offered
12 Results of DFA2 and DFA3 should be interpreted with a degree of caution, as reducing the number of
only a small contribution to differences between groups. However, symbolic threat showed the second largest CVCC (r = .645) behind realistic threat, indicating that it is key when considered independently. Also consistent with hypothesis 2, British identification offers a moderate contribution to the discriminant function: wealthier leave voters identify as British to a greater extent than wealthier remain voters.
Table 4. The contribution of each psychological variable in discriminating between wealthier remain voters and wealthier leave voters.
Psychological variable Function 1
Realistic threat .582
Sense of control .408
British identification .277
European identification -.230
Collective relative deprivation .194
Symbolic threat .193
Relative gratification .072
Note. Higher positive coefficients signify key variables for wealthier leave voters, higher negative
coefficients signify key variables for wealthier remain voters. Table ordered by absolute magnitude. Coefficients are standardised and partial.
Counter to predictions, relative gratification failed to discriminate the two wealthier voting groups. This conclusion is maintained by the observation that relative gratification has the second smallest CVCC (r = .273). In addition, sense of control in life made an unexpected contribution: wealthier leave voters’ greater sense of control was important in discriminating between the two voting groups.
In summary, results of DFA2 offer mixed support to hypothesis 2. Consistent with expectations, higher realistic threat, symbolic threat, and British identification were found in wealthier leave voters compared to wealthier remain voters. However, contrary to expectations, relative gratification failed to discriminate the two wealthier voting groups. Moreover, unexpectedly high levels of control found in the wealthier leave group contributed to differentiating the groups.
Hypothesis 3: Counter-trend remain voting. DFA3 used ‘voting group’ as an outcome variable (with two levels: poorer leave, poorer remain), and the seven psychological variables as predictors. A single discriminant function significantly differentiated the two voting groups, L = .561, χ2(8) = 28.914, p < .001. Table 5 shows the discriminant function coefficients for each
predictor. Results offer full support to hypothesis 3: European identification best differentiates poorer remain voters from poorer leave voters. In addition, British identification made a sizeable contribution to the function.
Table 5. The contribution of each psychological variable in discriminating between poorer remain voters and poorer leave voters.
Psychological variable Function 1
European identification -.687
British identification .474
Relative gratification -.200
Realistic threat .195
Collective relative deprivation .124
Symbolic threat .141
Sense of control .030
Note. Higher negative coefficients signify key variables for poorer remain voters, and higher positive
coefficients signify key variables for poorer leave voters. Table ordered by absolute magnitude. Coefficients are standardised and partial.
Section 3: Additional theme from qualitative data
The ‘threat of Brussels’. An additional theme emerged during analysis of interview data from leave voter DR: the threat that ‘Brussels’ (i.e., the European Commission) poses to Britain and the British way of life. This perspective is illustrated by the following excerpt.
DR4: I know Britain has said we’ll have so many [immigrants]. Fair enough, they have to
go somewhere. But I don’t want someone telling us, Brussels, ‘you have to take so many’. No. Great Britain’s Great Britain. I know I’m English, and we need to get Great Britain back to what it was.
The phrase ‘get Great Britain back to what it was’ implies that DR feels that British culture has changed for the worse. DR sees immigration as a source of this change, and views Brussels as a barrier that stops Britain from returning to a previous state, favourable to DR. In this regard, Brussels threatens DR’s vision of how British society should be. The following excerpt is in a similar vein:
DR5: I think we’re still Great Britain, but, it’s mainly like I said before, decisions, big
decisions what are made, they always seem to come from Brussels. A pint - what are they measured in?... a ‘litre of beer’. I think that’s how they want it.
Here, DR expresses the view that European values are encroaching on British conventions (i.e., units of measurement), and that Brussels is the force behind this process. Thus, again, DR identifies Brussels as a source of symbolic threat.
The following quote is DR’s response to being asked how he feels about the UK leaving the EU.
DR6: I wouldn’t say ‘happy’. Just good. Good because I think we’ll get back our own
ruling.
‘Get back our own ruling’ indicates that DR feels that Britain does not have full control over its own affairs. Based on other excerpts presented in this section, this lack of control is likely perceived to be because of the influence of Brussels.
In summary, this theme implies that Brussels may be seen as an illegitimate source of authority and control, and consequently, a threat to traditional nations of British culture. A perspective such as this would translate into behaviour that mitigates the feelings of threat felt towards Brussels, such as voting to leave the EU.
Discussion
The objective was to establish the key psychological variables that underlie people’s decision to vote remain or leave in the UK’s EU membership referendum. This was investigated using a mixed-method approach, in which quantitative survey data and qualitative interview data were gathered across Great Manchester and other regions of the UK. Quantitative results give partial support to hypothesis 1, regarding the overall voting trend. Voting ‘leave’ is characterised by higher levels of realistic and symbolic threat perceived from immigration, greater feelings of injustice that derive from collective relative deprivation, and stronger British identification. In contrast, voting ‘remain’ was best explained by European identification. Contrary to predictions, leave voters showed a greater sense of control in life. These findings are supported by themes that were found in interviews with remain and leave voters. Additionally, results offer partial support to hypothesis 2. British identification, realistic threat, and symbolic threat were key in distinguishing wealthier leave voters from wealthier remain voters. However, relative gratification failed to explain why some wealthier voters went against the trend by voting leave. Finally, full support was found for hypothesis 3, regarding counter-trend remain voting. European identification best explained why poorer individuals went against the overall trend by voting remain.
Several findings are consistent with SIMCA (Van Zomeren, Postmes, & Spears, 2008). SIMCA and RDT propose that individuals may feel a sense of injustice when comparing their own group to other, higher status groups, and that this feeling may encourage behaviour intended to improve conditions for that individual’s ingroup. In the current project, collective relative deprivation was found to be a key variable underlying leave voting overall. Thus, viewed through the lens of SIMCA, voting leave can be understood as an act through which many voters attempted to improve conditions for their own respective groups. The likelihood is that this conclusion would apply most to disadvantaged leave voters, who scored highest on collective relative deprivation, and were therefore fuelled by feelings of injustice to a greater extent than wealthier leave voters.
Also in line with SIMCA, results suggest that identity may have played a role in guiding voting decisions, and that voters acted according to their concerns over identity. Previous research on SIT has shown that stronger ingroup identification is related to greater outgroup discrimination
(Perreault & Bourhis, 1999). Based on this, it was predicted that those who identify strongly as British would show more discrimination towards immigrants, and subsequently, wish to create distance between ingroup and outgroup by voting leave. This is indeed what was found: British identification was a key variable underlying leave voting. In contrast, voters who identified as European during the referendum had a very different kind of identity concern. Quantitative findings demonstrate that European identification was a key variable underlying remain voting. Adding depth to this, qualitative results show that, specifically, maintaining European identity was crucial to remain voters. Together, these results support SIT, and the notion that individuals are motivated to maintain positive social identity (Tajfel & Turner, 1979). Thus, current findings suggest that both remain and leave voters acted based on their identity concerns, as predicted by SIMCA, and furthermore, adhered to the specifics of SIT.
Both realistic and symbolic threat were found to underlie leave voting. This indicates that leave voters feel more threatened by immigration than remain voters; both economically, on issues such as jobs and housing, and in terms of British culture and values. This finding harmonises with experimental and meta-analytical findings that reveal links between both realistic and symbolic threat and stronger negative attitudes toward immigrant groups (Stephan, Renfro, Esses, Stephan, & Martin, 2005; Riek, Mania, & Gaertner, 2006). In the current context, individuals may have voted to leave the EU to mitigate their feelings of threat from immigration. An additional source of threat was identified in interviews. The ‘threat of Brussels’ theme highlights that leave voters may perceive Brussels to have too much control over British affairs. Moreover, Brussels appears as a barrier, obstructing Britain from returning to former times. In general, these findings suggest that several aspects of being part of the EU are threatening to leave voters.
A similar pattern was found for hypothesis 2, regarding counter-trend leave voting. As expected, realistic threat, symbolic threat, and British identification were key variables that separated wealthier leave voters from wealthier remain voters. However, unexpectedly, relative gratification failed to discriminate between the wealthier voting groups. Results therefore do not support the V-curve hypothesis (Anier, Guimond, & Dambrun, 2016), which would predict that relatively gratified individuals would oppose outgroups such as ‘immigrants’, and consequently
vote to leave the EU. It would be mistaken to conclude that this null finding is due to the imprecision of the newly developed relative gratification scale. The scale has demonstrated several satisfactory psychometric properties. First, it is reliable, showing good internal consistency on two occasions: first during initial scale development, and secondly on survey data from the main study. It has also shown excellent divergent validity on both occasions; that is, it is independent of collective relative deprivation, and therefore adopts the proposed, theoretical ‘shape’ of the V-curve. Finally, the scale demonstrated convergent validity in the main study, correlating positively with both symbolic and realistic threat, as proposed by theory on relative gratification. In summary, the expected relative gratification effects were not found in this study; however, the scale that was developed to measure relative gratification has shown its merit as a psychometric tool for use in future studies.
Results concerning counter-trend remain voting are particularly interesting in light of SIMCA (Van Zomeren, Postmes, & Spears, 2008). European identification is the variable that best explains why poorer individuals deviated from the overall trend by voting remain. However, in addition to this, both poorer voting groups expressed similar levels of collective relative deprivation. Considered together, these results imply that concerns that counter-trend remain voters had over maintaining European identity took precedence over their feelings of injustice. Put differently, the motivation to protect European identity was stronger than the desire to act against the injustice that they feel. One might suggest that, compared to poorer leave voters, poorer remain voters simply felt less threatened by immigration, and therefore did not feel compelled to vote leave. Although results do indicate that poorer remain voters feel less threatened by immigration, neither realistic nor symbolic threat made any notable contribution to separating the poorer voting groups in DFA3. Thus, the desire to preserve European identity appears to be the decisive variable, outweighing other motivations to act against injustice. Beyond the scope of this research, this example puts forward interesting theoretical implications for SIMCA and its mechanisms; for instance, what determines the course of collective action in situations where motives of identity and injustice conflict? This question is especially important given that identity represents the cornerstone of SIMCA, predicting collective action both directly, and via injustice as a mediating
variable (see Van Zomeren et al., 2008; p. 521, Figure 2). Future research could investigate this further using a creative experimental setup – perhaps in a minimal group setting – in which injustice and identity concerns compete, and effects on behaviour are observed.
The variable ‘sense of control in life’ performed contrary to expectations. Overall, leave voters were expected to show a lower sense of control in everyday life compared to remain voters, yet results indicated that the opposite is true: overall, leave voters have a greater sense of control. Moreover, wealthier, counter-trend leave voters showed particularly high levels of control, and this differentiated them from wealthier remain voters. One plausible explanation of these findings relates to compensatory control mechanisms (Landau, Whitson, & Kay, 2015). Following the referendum, leave voters may now have an inflated sense of control in everyday life. This is because the UK is currently set on course to leave the EU, and as such, many leave voters may anticipate that the UK will “get back our own ruling”, as one interviewee put it. Therefore, it could be that restoring perceptions of control at a national level have reinstated perceptions of control at a personal level for leave voters.
Implications for policy
Results indicate that many individuals across the UK feel threatened by immigration, and that these feelings were partly responsible for Brexit. To inform government policy effectively, we must consider why these perceptions of threat exist. Darvis (2016) reveals that regions of the UK with lower proportions of immigrants were more likely to vote leave. This suggests that the perception that immigrants are a threat may not reflect lived reality. In line with this, other analysis shows that individuals who read newspapers with a more anti-immigrant stance, such as the Sun, Express, or Daily Mail, were more likely to vote leave (Swales, 2016). Thus, it is plausible that the media plays a role in constructing the perception that immigration threatens the UK culturally and economically. On the other hand, evidence from other (non-academic) sources show that regions with the greatest increase in immigrant population in recent years, voted leave in greater proportions (The Economist, 2016b; Resolution Foundation, 2016). In contrast, this implies a genuine basis for perceptions of threat found in this study. Thus, in summary, perceptions of threat
