Long waiting times at polling stations may reduce voter turnout and potentially undermine democratic legitimacy and, if unevenly distributed, democratic equality. Yet their effects on turnout remain poorly understood and have been studied mainly in majoritarian contexts. We provide the first evidence from a proportional parliamentary democracy by examining Sweden’s 2022 general election, when a reform to strengthen ballot secrecy unexpectedly produced congestion at many polling places. We estimate the turnout effect of congestion, using administrative records on precincts’ closing times together with original assessments from poll workers and local election authorities. Across several independent measures and estimation techniques, precincts experiencing problematic lines exhibit turnout about 1 to 1.5 percentage points lower than comparable precincts without congestion. The fact that congestion was induced by an exogenous legislative change and that Swedish voters were unaccustomed to polling station waiting times facilitates causal inference. Our results align with the few existing studies examining the impact of additional waiting times in majoritarian and presidential systems (the US and Kenya), suggesting that voters in a high-turnout PR system react similarly to voters in other contexts to an increased cost of voting. This underscores the policy importance of preventing congestion. Furthermore, our findings highlight a trade-off in electoral management between efficiency and ballot secrecy that warrants more attention in research as well as in electoral reform.
Declining electoral turnout in many representative democracies has prompted scholars to investigate the extent to which campaigns can increase turnout. Turnout is especially important amongst young voters, because their decision on whether to turn out can have ramifications throughout their lifetime. Voting Advice Applications (VAAs) may be key tools to increase turnout amongst young voters. However, extant evidence on VAAs' effects on turnout remains mixed, and we know little about how VAA use can increase turnout amongst young voters. Through their appealing design, VAAs are promising avenues to bolster turnout amongst this hard-to-reach group, but we lack robust empirical evidence evaluating VAAs' effectiveness in motivating young voters to turn out. We present evidence from a randomized survey-embedded experiment and an observational panel study in Flanders (Belgium) that had young voters use one of two VAAs—a Youth VAA aimed at younger voters, and a General VAA aimed at a broad audience—and then tracked their motivation to turn out. We find that use of the Youth VAA increased young voters’ turnout motivation for the June elections, both when VAA use is randomized and when citizens use the VAA naturally. These effects are strongest among lower-educated youth, suggesting that such tools can increase turnout motivation among a group often more difficult to engage, particularly for the Youth VAA. Importantly, the impact of VAA use remained mostly level throughout the rest of the June campaign, with observational evidence suggesting spillover effects to turnout motivation for the upcoming local elections in October.
Gender issues have moved to the centre of electoral politics in many democracies, driven by feminist mobilization on one side and anti-gender backlash on the other. Yet we know relatively little about which specific gender-related attitudes shape vote choice and why some gender policies become electorally consequential while others do not. This paper addresses these questions through an analysis of the July 2023 Spanish general election, combining a post-election survey with systematic analysis of party messaging. We examine both the supply side (how parties positioned themselves on gender-related issues during the campaign) and the demand side (how voters' attitudes toward those issues relate to vote choice). Among a broad range of gender-relevant policies, opposition to trans rights and legal gender self-determination emerges as the only policy attitude significantly associated with vote choice, increasing support for the right-wing PP and the radical right Vox. Sympathy for the feminist movement as a general orientation is positively associated with voting for the left-wing coalition Sumar and negatively associated with right-wing voting. These findings suggest that gender can function as an electoral cleavage, but that its electoral relevance depends on the type of conflict a given policy activates rather than on its salience alone.
In many European countries, the most recent elections featured an unprecedented gender gap in party support among the youngest cohorts: young men disproportionally supported right-wing populist parties, while young women overwhelmingly voted for left-wing and Green parties. This paper explores the new gender vote gap by focusing on the role of education. We propose two mechanisms: placement and context composition. First, placement in lower or higher tracks is a signal of losing or winning in the context of ongoing educational expansion, and young men find themselves more often on the losing side than young women. Second, due to gender differences in occupational choices, lower tracks are more gender segregated than higher tracks, and an unbalanced gender composition often coincides with peers sharing either right-wing or left-wing political attitudes. We argue that this amplifies the new gender gap in party support. We test our arguments with unique data collected shortly after the German federal elections in 2025 among a large-N, register-based sample of 17- to 20-year-olds. Results show that young men on lower tracks are the most likely to support the right-wing populist AfD, and that the political climate in their educational contexts amplifies this association. Young women on both lower and higher tracks are likely to support the Linke, and more strongly so if more than half of their peers in class also are women. The Green party is strong among voters in academic tracks, particularly among women, regardless of the gender or political composition of the educational context.
While affective polarization is commonly attributed to political differences between partisans, recent studies suggest it also relates to perceived non-political differences, such as differences in lifestyle, class, or personality. However, little is known about the relevance of these non-political perceptions beyond political ones, whether different types of non-political traits matter equally, and how these dynamics unfold beyond the U.S. two-party context. This study addresses these gaps by examining the relationship between perceived non-political differentness and two dimensions of affective polarization—partisan dislike and social distance—in Belgium's multi-party context. Using original survey data from Flanders and Wallonia (N = 4823), Belgium's two largest regions with distinct party systems, we show that perceptions of non-political differentness substantially relate to partisan dislike and social distancing, beyond perceptions of political differentness. Moreover, nearly all non-political dimensions studied matter (i.e., personality, values, socio-economic background, and lifestyle), albeit with varying importance across the two contexts and especially for partisan dislike. Altogether, these findings extend research on affective polarization by highlighting the overlooked role of non-political perceptions in our understanding of partisan animosity.
How damaging are inappropriate emotional displays by political leaders? While prior research shows voters reward appropriate displays, less is known about the consequences of inappropriate ones. Using German panel data, we analyse CDU chancellor candidate Armin Laschet's inappropriate laughter during a 2021 visit to a flood-stricken area as a quasi-experimental setting. Our findings, based on an unexpected event during survey design (UESD), indicate moderate, negative short-term effects on perceptions of his character, but no significant effects on turnout, or vote choice. Overall, our study demonstrates that inappropriate emotional displays can damage voter perceptions of political leaders but are unlikely to affect electoral behaviour.
Loneliness has emerged as a defining social phenomenon of our time, yet its political consequences remain insufficiently understood. This article examines how feelings of loneliness shape electoral participation and support for populist radical right (PRR) parties. While existing research suggests that loneliness may foster support for PRR parties by increasing the appeal of narratives centred on exclusion, belonging, and recognition, we argue that loneliness may also contribute to political withdrawal. We further assess whether loneliness conditions the relationship between PRR attitudes and voting behavior. Empirically, we analyze voting behavior in the 2025 German federal election, which was marked by substantial gains for the populist radical right Alternative for Germany (AfD). Using data from the German Longitudinal Election Study (GLES), we investigate both the direct association between loneliness and electoral behavior and potential interactions with populist and nativist attitudes, age, and gender. Our findings provide evidence that loneliness is associated with lower electoral participation. Individuals experiencing higher levels of loneliness are significantly more likely to abstain from voting and are not more likely to support the AfD. Contrary to expectations, we find little evidence that loneliness systematically strengthens the behavioral relevance of populist attitudes, or that its effects differ substantially across age cohorts and gender. Rather than serving as a catalyst for PRR mobilisation, loneliness appears primarily as a factor associated with political disengagement. Overall, these findings contribute to research on loneliness and political behavior by highlighting the demobilizing consequences of social disconnection and by qualifying claims that loneliness constitutes a driver of PRR support.
We introduce the Korean National Assembly Elections Dataset (KNAED), which provides the first complete candidate-level data on all legislative elections in South Korea from 1948 to 2024. KNAED covers 22,571 candidate–election observations across 22 general elections over six constitutional periods, from early post-independence contests through authoritarian rule to the current democratic era. The dataset documents all candidates—both successful and unsuccessful—in constituency and proportional representation contests, offering several key advantages for research. First, it spans both authoritarian and democratic periods within a single country, enabling systematic comparisons of electoral dynamics across regime types. Second, unique candidate identifiers allow researchers to trace individual political careers and analyze patterns of incumbency, repeat candidacy, and legislative turnover. Lastly, it includes granular district-level data on turnout, vote totals, and ballot statistics that support diverse analytical applications. We demonstrate the dataset’s value by examining three important trends in Korean politics: electoral competition across regime types, incumbency dynamics, and demographic shifts in the candidate pool. These analyses illustrate how KNAED enables comparative research on elections, representation, and political careers across varying regime contexts.
Family change has become increasingly visible across contemporary democracies, yet its implications for political realignment remain insufficiently incorporated into existing political science research. This study examines family institutions as settings through which political alignments are maintained and reproduced over the life course. Focusing on divorce as one of the most consequential forms of family transition, the analysis investigates whether marital dissolution is associated with ideological self-placement, vote choice, and aggregate electoral support in South Korea, where family institutions have historically occupied a central place in political and social life. The empirical analysis combines longitudinal panel data, repeated cross-sectional surveys, and district-level electoral and administrative records covering five national elections between 2012 and 2022. Across these complementary empirical settings, marital dissolution is consistently associated with less conservative ideological orientations, lower support for conservative candidates, and lower conservative vote shares. The findings indicate that family transitions may become politically consequential by reorganizing family institutions through which conservative political alignment has historically been reproduced.
Political knowledge, sometimes referred to as the currency of democratic citizenship, has long been claimed to improve voters’ ability to choose the candidate or party that best represents their interests. However, most existing evidence relies on cross-sectional associations, leaving limited longitudinal evidence on how changes in political knowledge relate to vote choice over time. We make two contributions to this debate. First, we provide a within-voter test of whether knowledge gains are associated with more ideologically proximate vote choices over time, using data from the Swedish National Election Studies (SNES) 1985–2022 rolling two-wave panels. Second, we use item response theory (IRT) with index anchoring to construct robust measures of political knowledge, allowing us to capture within-voter changes in knowledge between elections. Our refined measures replicate the positive cross-sectional association reported in several previous studies: individuals with more political knowledge tend to vote for more ideologically proximate parties. However, even across four decades of elections and observable individual-level variation in political knowledge, within-voter increases in knowledge do not translate into more proximate vote choices. These findings challenge the assumption that political knowledge improves vote choices between elections.
How do sudden coalition breakdowns shape affective polarization in multiparty democracies? While a growing body of work examines how coalition cooperation structure partisan affect, comparatively little is known about how citizens react when coalitions break apart. We argue that coalition collapse is associated with affective polarization. More specifically, we contend that through a dynamic of scapegoating and blame-attribution, this overall difference is concentrated in greater out-party hostility rather than in-party favorability. We test these expectations by exploiting the abrupt collapse of the German “traffic light” coalition in November 2024, using an Unexpected Event during Survey Design comparing respondents interviewed before and after the collapse. We find that the post-collapse period is associated with higher levels of affective polarization, concentrated in greater hostility toward out-parties rather than stronger in-party attachment. Disaggregating to party-dyads reveals that this hostility is targeted, with dislike rising significantly toward the party publicly blamed for the breakdown, and toward anti-establishment parties, while evaluations of other parties remain stable. Subgroup analyses indicate that this targeted hostility is concentrated among supporters of left-wing parties. Coalition collapses, we argue, do not polarize multiparty electorates uniformly but selectively, along ideologically structured lines. The findings contribute to the literature on affective polarization in multiparty systems by highlighting coalition collapse as a distinct, blame-attribution-based mechanism that is associated with targeted rather than diffuse out-party hostility.
Major political disruptions often discredit ruling parties, but they do not necessarily eliminate the political influence of politicians formerly associated with them. This article examines how former regional deputies of Ukraine’s Party of Regions (PoR) survived politically after the 2014 Euromaidan Revolution. Although the revolution discredited the PoR and removed it from national power, approximately one-third of former PoR regional deputies were renominated in the subsequent 2015 regional legislative elections. I explain this pattern by focusing on the interaction between former PoR deputies seeking viable electoral vehicles and parties seeking candidates with local mobilizational capacity. I argue that three factors shaped renomination prospects in post-Euromaidan Ukraine: individual electoral resources, ideological embeddedness in the PoR, and political geography. Deputies with stronger personal electoral resources, proxied by victory in majoritarian races, were more likely to be renominated, especially by the Bloc of Petro Poroshenko (BPP), the new ruling party. Deputies more deeply embedded in the PoR were less likely to join the BPP but more likely to align with the Opposition Bloc (OB), the PoR’s successor. Former PoR deputies were also more likely to survive in regions where the OB retained support and in rural areas characterized by shallow elite pools and clientelistic networks. Using original data on deputies elected in 2010 and candidates in Ukraine’s 2015 regional elections, this study shows how local electoral resources and party-side demand enabled the political survival of discredited PoR members in post-Euromaidan Ukraine.
Women in the United States have outpaced men in both voter participation and educational attainment in recent decades. Since education is closely tied to political participation, we consider these trends in tandem and assess how much of the gender gap in voting is attributable to educational attainment as opposed to cognitive skills, non-cognitive skills, college preparation behaviors, and demographics. Using comprehensive educational data from Massachusetts students matched with voter records, we estimate a Blinder–Oaxaca–Kitagawa decomposition to understand how these factors contribute to gender voting differentials. In our data, young women outvote young men by 5.8 percentage points in the first presidential election in which they can vote after having potentially completed college. We find that just under 50 percent of this gap in voting can be explained by differences in educational attainment and civic returns to attainment by gender. If we also consider gender differences in college preparation, we explain two-thirds of the gender gap in voting. These results suggest that a significant portion of the gender gap in voting can be attributed to the rise in women’s education.
While prior research has examined attitudinal and behavioural responses to snap elections, we know little about how the institutional act of calling a snap election shapes voters’ immediate uncertainty about their vote choice. Measuring this uncertainty through “don’t know” (DK) responses in large public opinion surveys, we show that snap election calls produce a discontinuous reduction in vote intention uncertainty that is distinct from the gradual uncertainty decline associated with regular pre-election campaigns. We demonstrate this through two cases: the 2025 German Federal Election and the 2024 UK General Election. Leveraging a quasi-experimental unexpected event design in Germany and individual-level panel data in the UK, we show that these sudden announcements accelerate the timing of vote preference formation in a way that ordinary elections do not. Yet, the distribution of effects differs across contexts: in Germany, reductions are concentrated among non-partisans, whereas in the UK they are more broadly distributed. We also demonstrate that snap election calls operate as a distinct institutional mechanism that alters the baseline of voting uncertainty before the campaign begins. In doing so, we advance our understanding of how disruptions to the electoral calendar shape the formation and expression of voters’ preferences.
Does emigration motivate or discourage political participation in migrant sending countries? While emigration can reduce participation by removing opposition voters and weakening political efficacy, it may also signal widespread grievances and encourage participation among those who remain. This paper examines the immediate political effects of large-scale emigration and argues that emigration alone is insufficient to mobilize participation. Instead, when emigration is visible and perceived as a collective act, it sends a political signal that updates citizens’ beliefs about the status quo. Drawing on a pre-registered survey experiment conducted in Honduras ahead of the 2021 presidential election, I test how framing emigration as either a private decision or a collective act in the form of migrant caravans shapes intentions to vote. Respondents exposed to the caravan framing were five percentage points more likely to report intentions to vote than those in the control group and four percentage points more likely than those exposed to the private decision framing. Using causal mediation analysis, I examine whether these effects operate through nationalized perceptions of migration and politics but find no clear evidence of mediation, which may reflect ceiling and floor effects. These findings highlight the importance of framing in shaping how emigration influences political participation among those who remain.
Do gender differences in policy preferences translate into differences in agenda-setting, policy output, and out comes? To answer this, we first identify gender variation in budget preferences among local councilors using survey experiments conducted in Danish municipalities in 2013, 2017, and 2021. To examine whether these differences translate into observable changes in policy, we estimate the causal effect of women's representation on three stages of the policy process: agenda-setting, public budgets, and actual expenditures. We exploit close mixed-gender elections to capture exogenous variation in women's descriptive representation. We find gender differences in political candidates' policy preferences, but these differences do not translate into measurable changes in agendas, public budgets, or actual expenditures. We do, however, find an increase in school share of expenditures that does not align with the survey preferences. This highlights the importance of a policy process perspective in identifying the link between descriptive and substantive representation.
Can unmet climate demands erode political support among younger citizens? This study examines how rising pro-climate preferences affect political support under conditions of heightened issue salience. The central argument is that policy preferences shape evaluations of incumbent and system performances, primarily when the issue is sufficiently salient to enter citizens’ political calculus. Salience functions as a priority weight that activates the link between preference (in)congruence and political support, while shifts in public attention may amplify this relationship over time. Using panel data from the German Longitudinal Election Study (2016–2022), I analyse within-individual preference change and variation in climate salience across cohorts and time. I leverage the rise of Fridays for Future (FFF) as a focal attention environment shift. The results show that climate change carries greater salience for younger than for older citizens; this gap widens following the rise of FFF. Consistent with the activation argument, within individuals, shifts toward more pro-climate preferences are associated with declines in specific support among younger respondents, but less consistently so for older respondents. The link between climate preferences and political support does not strengthen in the post-2019 period. These findings suggest that while climate considerations are more consequential for political evaluations among younger citizens, temporal increases in salience alone are not sufficient to intensify dissatisfaction. More broadly, the results highlight that the political consequences of policy preferences depend on the salience of the issue and the conditions under which attention is translated into evaluative judgements.
To what extent does partisanship bias copartisans' responses to corruption? While existing research shows that partisanship can shield politicians from the consequences of corruption, the conditions under which this varies remain unclear. I argue that partisanship's effect on corruption voting depends on whether corruption is targeted (when the corrupt actor belongs to the voter's preferred party) or systemic (when the electorate perceives cor ruption to be widespread). These contexts activate different processes: copartisan bias in candidate evaluation and the weakening of partisanship's behavioral shielding effect. Using a survey experiment in Brazil and public opinion data from 17 Latin American countries (2008-2018), I find that copartisans express significantly greater support for a candidate accused of corruption when their party is implicated. However, as systemic corruption increases, they become less likely to support their party when it is in power and more likely to consider opposi tion alternatives. These findings show that copartisan bias in favor of accused politicians is conditional and can become fragile under certain circumstances.
Over the past decade, scholarly attention to the political relevance of "place" has increased, especially regarding the far right and in the contexts of Europe and the United States. Within the field, the urban-rural divide is of particular interest. While rural areas often display more right-leaning attitudes than urban ones, it remains un certain whether these differences constitute a full-fledged cleavage, as this would require not just socio-structural conflict and political mobilisation but also a collective identity. Using Germany as a case study, this article inves tigates the psychological underpinnings of the urban-rural divide and its intersection with the division between the eastern and the western regions of Germany. The findings reveal a strong asymmetry: rural residents express stronger place-based identification and perceive greater conflict than urban citizens. In addition, we find that exclusively eastern identities are more prevalent than exclusively western ones, and that the urban-rural and the East-West cleavages operate independently. These results underscore the importance of analysing multiple geographic cleavages jointly and contribute new insights into the emotional and perceptual foundations of spatial divides in Germany.
