Sage Journals: Discover world-class research

Abstract

Negative campaigns can have different effects on voting behavior and political attitudes. While recent work has differentiated between contexts and campaigns, few have investigated the mechanisms underlying the effects of negative campaigns. We argue that negative campaigning may affect whether citizens understand voting as supporting one candidate or party (positive meanings of voting), as opposing another (negative meanings of voting), or even as something useless, unpleasant or undesirable (anti-voting meanings). Using preregistered cross-sectional analyses in the 2022 Hungarian and US elections, we confirm that negative and anti-voting meanings are pervasive, and related to political parties’ use of negative and “dirty” campaigns. Two preregistered survey experiments, however, only confirm the effects of the most outrageous campaign messages. We conclude that meanings of voting are important democratic attitudes that are related to parties’ campaigning behavior. However, they are relatively stable attitudes that are only affected by relatively extreme stimuli.

Keywords

negative campaigns dirty campaigns negative voting meanings of voting partisanship

The electoral consequences of negative campaigning – campaign aspects that emphasize an opponent’s flaws rather than one’s own qualities (Jamieson 1993) – have been a concern for scholars and pundits for decades (Lau et al. 2007). Mixed evidence shows that the effects depend on its content (Nai 2013) or the political system (Donovan et al. 2016). While negative campaigns can demobilize voters by disenchanting them about politics (Ansolabehere and Iyengar 1995; Ansolabehere et al. 1994), it may also mobilize voters through increased attention (Martin 2004). Similarly, negative campaigns may harm an opponent or the sender themselves (Lau et al. 2007; Lau and Rovner 2009; Walter and Van der Eijk 2019b). To understand these findings, research examined different partisans’ responses to negative campaigning (Somer-Topcu and Weitzel 2022) and how transgression of social norms affect them (Reiter and Matthes 2022). However, despite this extensive literature, we still lack clear answers about which voters are influenced by negative campaigning and under what circumstances.

These ambiguous findings suggest that we need to better understand the underlying mechanisms to predict when negative campaigning will help or harm candidates. We investigate one possible mechanism that has been overlooked for now: whether negative campaigns affect how citizens understand voting. While voting is most often conceptualized as a way to support a preferred party or candidate (positive meanings of voting; Brennan and Hamlin 1998; Downs 1957), there is increasing evidence that citizens might understand voting as a way to oppose a party or candidate instead (negative meanings of voting; Alvarez et al. 2018). In addition to positive and negative meanings of voting – both of which consider voting valuable – citizens may hold meanings that are averse of voting itself. Such antivoting meanings may entail understanding voting as useless, cumbersome, or undesirable (Plescia, Abdala, et al. 2025).

Negative voting may be driven by negative partisanship and affective polarization (Garzia and Ferreira da Silva 2022b 2022a). We argue that negative and anti-voting meanings may also be related to negative campaigns. For example, perceived legitimate criticism may increase one’s understanding of voting as a way to act against an attacked party, whereas transgressions of social norms decrease citizens’ trust in politicians (Reiter and Matthes 2021) and may serve as evidence that voting is pointless. Most likely, these effects are strongly dependent on citizens’ partisanships (Somer-Topcu and Weitzel 2022). Building on these findings, we test in this paper hypotheses on the effects of negative campaigns that do not transgress social norms (negative campaigns) and those that do (“dirty” campaigns) on ingroup and outgroup partisans’ meanings of voting.

We test our hypotheses by analyzing two cases of two-bloc elections with different electoral dynamics, presenting a clear distinction between ingroup and outgroup parties and campaigns. We call “two-bloc elections” electoral contests where two or multiple parties form two distinct voting blocs that effectively function as two competing camps, with voters understanding their choice as effectively between these two broader political alternatives. First, we conduct a preregistered cross-sectional analysis (including lagged-dependent variable [LDV] models and partisan bias-corrected models) of the 2022 Hungarian parliamentary election and 2022 US midterm election.¹ Second, we leverage two preregistered survey experiments conducted in the same elections. While the cross-sectional analyses and robustness checks present compelling evidence of correlations between negative campaigning and meanings of voting, the experiments only replicate the effects of the most outrageous campaign messages. This suggests that meanings of voting are relatively stable attitudes that may only change under extreme conditions.

These results have important consequences for the study of negative campaigns and voting. First, they show that meanings of voting may form a missing link to understanding the inconsistent findings in previous research. Moreover, while some argue that “dirty” campaigns have negative consequences for democracy while respectful negative campaigns do not (Geer 2006; Mark 2006; Mattes and Redlawsk 2014), these results show that both can be harmful through their effects on meanings of voting. As such, this article offers a starting point to investigate how meanings of voting take shape and may help explain the nuanced effects of negative campaigning on both political attitudes and political behavior.

Citizen Meanings of Voting

Whether one understands voting as a way to express agreement or identification (Brennan and Hamlin 1998; Schuessler 2000), a symbolic act (Campbell 2008), exercising one’s rights, a civic duty (Dalton et al. 2007), or an instrumental choice based on election calculus (Blais and Achen 2019; Downs 1957; Riker and Ordeshook 1968), scholars have traditionally assumed that citizens hold positive meanings of voting – that is, that citizens understand voting as supporting a party, candidate, or policy. They might support them out of genuine belief in policy, expected personal gain, or identification (Parisi and Pasquino 1977).

However, in more recent work, citizens’ political negativity has gotten increased scholarly attention (Nai et al. 2022). Most notably, scholars have noted that citizens may often (though not always) dislike outparty candidates more strongly than they like their own (e.g., Garzia et al. 2023) and even identify as not-supporting the outparty more than supporting their preferred party (Abramowitz and Webster 2016; Mayer 2017). Indeed, it follows that citizens might also understand voting as an act against a party, candidate, or policy (Alvarez et al. 2018; Gant and Davis 1984; Gant et al. 1985; Garzia and Ferreira da Silva 2022a; Kselman and Niou 2011) rather than in favor of one – that is, they can hold negative meanings of voting. Negative voting is also implied in performance-based voting, where citizens choose to vote out governments that they perceive to have caused a poor economy (Becher and Donnelly 2013) or to be corrupt (Ecker et al. 2016). Attesting to the importance of partisanship in negative campaigning research, particularly individuals for whom outgroup hate trumps ingroup love might engage in negative voting (Garzia and Ferreira da Silva 2022b).

Besides positive and negative meanings of voting, citizens may hold various kinds of meanings. Riker and Ordeshook (1968), for instance, discern between voting to make a change, to express oneself, to adhere to societal norms and to support a regime. However, as negative campaigning raises concerns over decreased turnout and support for democracy, it is crucial to investigate meanings of voting that are pessimistic about voting itself. As such, anti-voting meanings include considering voting meaningless, cumbersome, inconsequential, or even undesirable (Plescia, Belén Abdala, et al. 2025). Anti-voting meanings can stem from feelings of alienation or indifference toward representative politics (Adams et al. 2006), and are therefore an important attitudinal category in the study of the effects of negative campaigning.

We explore the meanings that citizens attach to voting as attitudes (Plescia, Belén Abdala, et al. 2025), which have been shown to be strongly related to turnout (Brunetti et al. 2025). With meanings of voting, we mean any way in which citizens understand or define “voting” (here differentiated between positive, negative, and anti-voting meanings). We assume that meanings of voting, like other kinds of political attitudes, are initially shaped through political socialization by parents, educators, and peers (Bos et al. 2022; Gidengil et al. 2016, 2021; Guhin et al. 2021; Wiseman et al. 2011). In the course of a lifetime, they might correlate with and change due to phenomena such as affective polarization or negative partisanship (Garzia and Ferreira da Silva 2022b, 2022a), or external influences like electoral wins or losses and – as is our main argument – political campaigns.

Negative Campaigns and Meanings of Voting

Prior research implies – but does not test – that negative campaigning shifts voters from a positive to a negative meaning of voting. For instance, effective negative campaigns may reduce support for those who are attacked due to highlighted negative qualities, and the backlash effect against attackers seems to suggest that voters base their choice on avoiding the perpetrator (Lau and Rovner 2009; Lau et al. 2007). Moreover, as negative campaigning induces affective polarization (Martin and Nai 2024), it seems plausible that this increased disdain for the outgroup party might cause voters to act against them through voting.

However, most work is unclear about the underlying meanings of voting. For example, negative campaigns increase voting for both the attacker and the attacked by their respective copartisans (Somer-Topcu and Weitzel 2022). However, this can mean that copartisans of the attacked party are reminded their party needs their support (positive meanings), or that they are motivated to stop the attacker from winning (negative meanings). Similarly, copartisans of the attacking party may increasingly understand voting as supporting a party that is not afraid to tell the outgroup off (positive meanings) or as a way to stop the outgroup from winning (negative meanings). Understanding how campaigns affect meanings of voting may help find more consistent findings across cases.

We expect that the processing of campaign messages is to some extent guided by partisan-motivated reasoning (Kunda 1990; Taber and Lodge 2006). While recent work shows that factual information reduces motivated reasoning and biased updating of beliefs (Stagnaro and Amsalem 2025), campaign messages are overtly partisan – which has been shown to cause biased information processing (Slothuus and De Vreese 2010). However, we do consider tipping points (Redlawsk et al. 2010), where vitriolic information (dirty campaigns) cause resistance even when perpetrated by one’s ingroup party and where credible information (negative campaigns) may be related to belief updating even when targeting one’s ingroup party (Velez and Liu 2025). As such, the distinction between negative and dirty campaigns fits well with the more nuanced understanding of motivated reasoning that recent work promotes.

Turning first to ingroup party campaigning, positive campaigns emphasize an ingroup candidate’s beneficial attributes, making these more easily accessible in voters’ minds – increasing the likelihood that voting is understood as supporting a candidate, party, or policy (i.e., positive meanings of voting; H1a; all hypotheses are described in Table 1). In a similar vein, ingroup party negative campaigning might lead voters to build stronger associations between “voting” and the flaws of their opponent, causing them to understand voting as opposition to something they reject (i.e., negative meanings of voting; H1b). In other words, positive and negative campaigning by the ingroup party reinforces prior attitudes. In contrast, dirty campaigning can trigger backlash against the attacker (Lau and Rovner 2009), even if that is an ingroup party (Velez and Liu 2025). Consequently, dirty campaigning might cause attitudinal ambivalence, impeding citizens’ ability to form attitudes or preferences (Mutz 2002; Groenendyk 2019; but see Lavine et al. 2012). Therefore, ingroup dirty campaigning likely relates to an aversion of voting (i.e., anti-voting meanings; H1c).

Table 1.

Hypothesized Positive Correlations Between Campaigns and Meanings of Voting.

Hypothesis	Party	Campaign	Meanings of voting
H1a	Ingroup	Positive campaign	Positive meanings
H1b	Ingroup	Negative campaign	Negative meanings
H1c	Ingroup	Dirty campaign	Anti-voting meanings
H2a	Outgroup	Positive campaign	Negative meanings
H2b-I	Outgroup	Negative campaign	Positive meanings
H2b-II	Outgroup	Negative campaign	Negative meanings
H2c	Outgroup	Dirty campaign	Negative meanings

As outgroup party campaigning challenges, rather than reinforces priors, its effects are different. We expect outgroup positive campaigning to evoke counterarguing (Taber and Lodge 2006), thereby strengthening negative meanings of voting (H2a). Outgroup negative campaigning is expected to evoke a similar but twofold reaction. First, counterarguing negative information about the ingroup party may elicit a boomerang effect where voters strengthen their preference for their ingroup party (Hart and Nisbet 2012; Velez and Liu 2025) and thus positive meanings of voting (H2b-I). At the same time, negative emotional reactions caused by outgroup negative campaigning might form an additional motivation to keep an outgroup party out of office (Garzia and Ferreira da Silva 2022b; negative meanings of voting; H2b-II). For outgroup dirty campaigning, the latter effect should be dominant, as the outgroup party is perceived to transgress social norms (Velez and Liu 2025), that is, outgroup dirty campaigning should be related to negative meanings of voting (H2c).

Case Selection

We test our hypotheses during the campaigns of the 2022 Hungarian legislative and US midterm elections. These are appropriate cases to test our hypotheses for three reasons. First, both elections can be considered de facto two-bloc elections, where voters decide between two clear viable choices. While Hungary has a multiparty system, incumbent Viktor Orbán and his party Fidesz were challenged by the opposition under a single banner (United Opposition) and candidate (Márky-Záy Péter). As a result, this effectively created a two-bloc election in Hungary similar to the United States, with a clear ingroup and outgroup bloc for potential voters. Second, Hungary and the United States are highly polarized societies, increasing the likelihood of negative meanings of voting (Garzia and Ferreira da Silva 2022b). Third, in these polarized, high-stake elections, there was a vast amount of negative campaigning – making these relevant and appropriate cases for its study (Hassell 2021).

The 2022 US midterm elections and the Hungarian legislative elections share similarities as strongly polarized two-bloc elections and thus provide theoretically appropriate cases to test our hypotheses, yet they also differ in institutional design and political context which can hinder comparison. First, the United States is a federal presidential system with single-member districts and a clear separation of power between the legislative and the executive. By contrast, Hungary is a unitary parliamentary system with a mixed-member proportional electoral system. Besides, the United States has a long-standing two-party system, while this was the first national election in Hungary with two clear blocs.

Finally, the 2022 Hungarian parliamentary election took place in a context of increasing democratic backsliding, which, despite fostering a united front against Fidesz which created the two-bloc election (Mikola and Santos 2025), may have exacerbated negative sentiments among opposition voters due to an uneven playing field. Overall, Hungarian legislative elections are thus typical first-order contests with imbalances between blocs, whereas US midterm elections are generally less salient for voters because there is less at stake even though the US Supreme Court decision on abortion, that is, Dobbs v. Jackson Women’s Health Organization, can be considered as having increased the importance of these elections as well. Despite these differences, we believe these two cases can be considered as a least-similar case selection within the broader category of two-bloc elections. Therefore, our design has advantages as it allows us to provide a first test of the theoretical framework in a most likely setting while varying factors within two-bloc elections to increase generalizability. Nonetheless, this case selection also limits generalizability – not only because we focus on two-bloc elections but also because these specific elections are relatively extraordinary cases.

Study 1: Cross-Sectional Analysis

We rely on a three-wave panel study conducted two months before each election (t₀, before the campaigns were at full force), during the two weeks before each election (t₁, during the height of the campaign), and directly after each election (t₂) (Plescia, Abdala, et al. 2025). The samples are descriptively representative of the Hungarian and US population in terms of gender, age, and region. All participants that took part at t₁ and that indicated the intention to vote for either bloc are included in the analysis (Nhu = 1,284, Nus = 1,944). Our analyses in Study 1 are observational in nature and aim to identify associations between campaigns and meanings of voting. These observational data help us explore, for the first time, how these two factors are connected. At the same time, we make use of the panel structure of our data and use LDV models to address endogeneity and get a first impression of a potential causal relationship – which is to be supplemented by the experimental work in Study 2.

As preregistered, the main analyses in this article are based on correlations at t₁ (full models in Table A1 in the Supplemental Information File). Due to attrition between t₀ and t₁ in the United States (83 percent), we deviate from the preregistration by supplementing the t₁ sample with additional participants, forcing us to drop the control variables related to political socialization that were measured at t₀ (education and political knowledge). We include the preregistered analyses including those control variables as a robustness check – with a limited sample in the United States – in Table A2 in the Supplemental Information File. In addition, we include four more robustness checks in Appendix 3 in the Supplemental Information File. First, we include LDV models as additional robustness checks in Table A3. In addition to strengthening the causal claim of our argument, these account for the between-subject variation in baseline meanings of voting, minimizing the possibility of omitted variable bias and other forms of endogeneity problems. Moreover, we have included a model controlling for left/right self-placement and affective polarization in Table A4. This check does not indicate that these variables are pivotal confounders in the effects under study. Additionally, a subgroup analysis shows no structural differences between bloc supporters (Tables A5 and A6). Finally, to accommodate biased perceptions of the campaigns (Haselmayer et al. 2020; Reiter and Matthes 2022), we use partisan bias-corrected measures of campaign perceptions in an additional robustness check in Table A7, following the method by Walter and Van der Eijk (2019a). These are shown to eliminate the partisan bias in perceiving campaigns as negative, and thus ensure that the differences we find between ingroup and outgroup campaigns are not due to biased perceptions.

Measures

Positive meanings, negative meanings, and anti-voting meanings are measured using three items each, starting with the statement “To me, voting means . . .” (see Table 2). The items are rated on an eleven-point Likert-type scale and form reliable scales. The averages of each of the three items are used as final measures.

Table 2.

Survey Items and Reliability Scores of Meanings of Voting.

Positive meanings (a_hu = .74, a_us = .75)

. . . a way to express my preference for a politician or party

. . . a way to reward a politician or party

. . . a way to express my trust in a politician or party

Negative meanings (a_hu = .84, a_us = .66)

. . . a way to express my aversion for a politician or party

. . . a way to punish a politician or party

. . . a way to express my distrust in a politician or party

Anti-voting meanings (a_hu = .68, a_us = .70)

. . . a meaningless act

. . . an unpleasant or difficult activity

. . . an act without consequence

In both Hungary and the United States, positive meanings of voting (M_HU = 7.11, standard deviation [SD]_HU = 2.33; M_US = 7.26, SD_US = 2.31) are more prevalent than negative meanings of voting (M_HU = 5.18, SD_HU = 3.11; M_US = 5.86, SD_US = 2.55), which are more prevalent than anti-voting meanings (M_HU = 2.14, SD_HU = 2.32; M_US = 3.48, SD_US = 2.83). This pattern is strikingly similar between the two countries, although negative and anti-voting meanings seem somewhat more pervasive in the United States (Figure 1). Rather than forming opposing understandings of voting, positive and negative meanings of voting are correlated, r_HU = .42, p_HU < .001, r_US = .56, p_US < .001. Positive meanings of voting and anti-voting meanings are inversely correlated in Hungary, r = −.21, p < .001, but not in the United States, r = .03, p = .098. Finally, negative meanings of voting are correlated with anti-voting meanings in both countries, r_HU = .05, p_HU = .017, r_US = .29, p_US < .001.²

Figure 1.

Density plots of citizen meanings of voting in Hungary and the United States.

We measure exposure to positive, negative, and dirty campaigning, using a short measure of Reiter and Matthes’ (2021) scale. However, following Walter and Van der Eijk (2019a), we measure perceptions of both campaigns separately. Participants rate each campaign on an eleven-point Likert-type scale, across six items – two for each type of campaigning. These items include “[Party] mostly focuses on presenting themselves in a positive way” (positive campaigning), “[Party] engages with their opponent in a very critical but respectful way” (negative campaigning), and “[Party] is using dishonest methods in the campaign” (dirty campaigning; exact items in Table A11 in the Supplemental Information File). For all four political parties across both countries, scales were reliable for positive (α > .69), negative (α > .71), and dirty campaigning (α > .83). Subjective assessments of campaigns influence political attitudes, that is, political trust (Reiter and Matthes 2021), suggesting that perception-based measures of campaign can be useful to explore political attitudes. Potential issues of endogeneity are addressed through multiple robustness checks described above. The scales are averaged and recoded, such that ingroup campaigning refers to the party with which a participant indicated an affiliation and outgroup campaigning vice versa.

Apart from some minor differences, campaigns were perceived strikingly similar in both countries (Table 3). The ingroup campaign is perceived as relatively positive, as well as negative, but not dirty. Outgroup campaigns were considered relatively positive and very dirty, but less negative. Partisan bias in campaign perceptions seems weaker in the United States than Hungary, but the bias in both countries illustrates the necessity of our robustness check with partisan bias-corrected measures (Walter and Van der Eijk 2019a).

Table 3.

Descriptive Statistics of Campaign Perceptions.

	Hungary		United States
Campaign	M	SD	M	SD
Ingroup
Positive	6.83	2.44	7.25	2.25
Negative	6.78	2.35	6.73	2.35
Dirty	1.95	2.42	3.59	3.19
Outgroup
Positive	5.03	3.35	5.33	3.05
Negative	2.47	2.60	4.10	3.09
Dirty	7.32	2.97	7.17	2.71

Note. SD = standard deviation.

In the main analyses, we control for gender (Hungary: 49 percent female, United States: 53 percent female), age (M_HU = 47.38, SD_HU = 15.34, M_US = 49.25, SD_US = 18.44), and political interest (1 = not at all interested, 4 = very interested; 69 percent at least “somewhat interested” in both countries).

Results

To test our hypotheses, we regress the six campaign perceptions on each meaning of voting, controlled for gender, age, and political interest (Table 4, Figure 2). Confirming H1a-c, we find significant correlations between ingroup positive campaigning and positive meanings of voting; between ingroup negative campaigning and negative meanings of voting; and between ingroup dirty campaigning and anti-voting meanings. These results are confirmed in all robustness checks at the exception of the correlation between ingroup negative campaigning and negative meanings of voting that is not confirmed in the LDV model. In addition to negative meanings of voting, ingroup negative campaigning is correlated with anti-voting meanings. Moreover, even though it is weaker than the correlation with anti-voting meanings, ingroup dirty campaigning seems related to negative meanings of voting as well, and – in Hungary – with fewer positive meanings of voting.

Table 4.

Hypothesis Tests and Robustness Checks Study 1.

	Hypothesis	Country	b	MA	PA	LDV	CPA	SGA	PBC
Table with full regression model				A1	A2	A3	A4	A5/6	A7
H1a	Positive campaigning by the ingroup party leads to more positive meanings of voting	HU	.12**	✓	✓	✓	✓	✓	✓
H1a		US	.16**	✓	✓	✓	✓	✓	✓
H1b	Negative campaigning by the ingroup party leads to more negative meanings of voting	HU	.11*	✓	✓	.	✓	F	✓
H1b		US	.10**	✓	✓	.	✓	R	✓
H1c	Dirty campaigning by the ingroup party leads to more anti-voting meanings	HU	.26**	✓	✓	✓	✓	✓	✓
H1c		US	.34**	✓	✓	✓	✓	✓	✓
H2a	Positive campaigning by the outgroup party leads to more negative meanings of voting	HU	.05	.	.	.	.	.	✓
H2a		US	.00	.	.	.	.	.	.
H2b-I	Negative campaigning by the outgroup party leads to more positive meanings of voting	HU	.03	.	.	.	.	.	.
H2b-I		US	.06**	✓	.	.	.	✓	.
H2b-II	Negative campaigning by the outgroup party leads to more negative meanings of voting	HU	.02	.	.	.	.	.	.
H2b-II		US	.01	.	.	.	.	.	.
H2c	Dirty campaigning by the outgroup party leads to more negative meanings of voting	HU	.19**	✓	✓	✓	✓	U	✓
H2c		US	.12**	✓	✓	.	.	✓	.

Note. MA = main analysis; PA = preregistered analysis; LDV = lagged-dependent variable model; CPA = controlled for political attitudes; SGA = subgroup analysis; PBC = Partisan bias-corrected model; HU = Hungary; US = United States; F = Fidesz supporters; R = republicans; U = united opposition supporters.

p < .05. **p < .001.

Figure 2.

Correlations between political campaign perceptions and meanings of voting in Study 1 (unstandardized coefficients and 95 percent confidence intervals). Black: p < .05.

Turning to outgroup campaigning, H2a and H2b are rejected: We find no significant correlations between outgroup positive campaigning and negative meanings of voting, nor between outgroup negative campaigning and positive or negative meanings of voting (Table 4). H2c, however, is confirmed. Outgroup dirty campaigning is correlated with negative meanings of voting. This result, however, is not significant in the LDV and bias-corrected models for the United States. In addition to these effects, we find a strong and robust correlation between outgroup negative campaigning and anti-voting meanings – suggesting that respectful negative campaigns might alienate outgroup partisans from voting in a two-party system.

Discussion

The findings in Study 1 offer some crucial innovations in the negative campaigning literature. First, they indicate that negative meanings of voting and anti-voting meanings – albeit to a lesser extent than positive meanings of voting – are pervasive. Rather than opposing views of voting, positive and negative meanings of voting are correlated. Second, the results confirm that there are antithetical findings between ingroup campaigns and outgroup campaigns (Somer-Topcu and Weitzel 2022), as well as between negative and dirty campaigning (Reiter and Matthes 2021).

Most importantly, we show that political advertisements affect the way citizens perceive voting. Positive and negative campaigning are related to copartisans’ positive and negative meanings of voting. In contrast, dirty campaigning might alienate them from voting, as indicated by its correlation with anti-voting meanings. Positive campaigning does not affect outgroup partisans’ meanings of voting at all. However, negative campaigns drive their anti-voting meanings, and dirty campaigns their negative meanings of voting, further pitting opposing citizens against the attacker. Importantly, these conclusions are supported in LDV models and partisan bias-corrected models, indicating a certain level of robustness in terms of unmeasured confounders, causal direction, and partisan bias in campaign perceptions.

Study 2: Experiments

Method

While the associations uncovered in Study 1 are strong and compelling, we further examine the causal relationship between campaigning and meanings of voting in two preregistered 3 (campaign type) × 2 (ingroup/outgroup) within/between-subject experiments (N_HU = 1,284, N_US = 1,472). These experiments were placed at the end of the surveys in which we collected the cross-sectional data for Study 1 (Figure 3). Participants first filled out the cross-sectional questionnaire, which contained measures of partisanship (used to determine the ingroup and outgroup) as well as the pre-exposure measure of meanings of voting (the same as used in Study 1). Then, they were exposed to one (Hungary) or two (United States) campaign message(s). They were randomly assigned to a condition containing positive, negative, or dirty campaigning and that was either on behalf of the ingroup or outgroup party. Then, they again answered questions regarding their meanings of voting, as well as their perceptions of the campaign advertisement(s) that they were exposed to. The dependent variable is the difference between the preexposure and postexposure measure of meanings of voting.

Figure 3.

Procedure of the experiments.

We conducted the first experiment at the end of the Hungarian survey at t₁ – at the height of the campaign. In order to examine pretreatment and election salience, we repeated the experiment at the end of the US survey at t₀. Conducting the experiment longer before the election should reduce election salience and pretreatment effects. Moreover, in the United States we used a multiple exposure design to account for a possible lack of strength of the stimulus material in Hungary. Besides those differences, the two experiments are identical and both received ethical approval by the Institute Review Board (IRB).

Stimulus Material

First, participants in Hungary were asked to imagine coming across a campaign message by [Fidesz/The United Opposition] online. They were then exposed to a mock campaign advertisement. This is recoded such that for Fidesz supporters, the Fidesz advertisement is the ingroup campaign and the United Opposition advertisement is the outgroup campaign – and for United Opposition supporters vice versa. In the campaign advertisement, the party does campaign promises (positive campaign), criticizes the opponents’ campaign promises (negative campaign), or criticizes its opponent’s character and integrity (dirty campaign). The positive campaign is accompanied by a flattering photo of the party’s own candidate, while the negative and dirty campaign are accompanied by a darker, threatening-looking photo of the opponent (English translation of the stimulus material can be found in Figure A2 in the Supplemental Information File).

In the United States, participants were asked to imagine coming across the following campaign messages by the [Republican/Democratic] party online. This is again recoded according to participants’ party affiliation. In the advertisements, the party does campaign promises (positive campaign), criticizes its opponents’ campaign promises (negative campaign), or criticizes its opponent’s campaign promises in an uncivil manner (dirty campaign; stimulus material in Figures A3 and A4 in the Supplemental Information File). In this multiple-message design, we use the most salient issues of the election for each political party: immigration and abortion. Each participant is exposed to two advertisements in the same condition. In both countries, after exposure to the stimulus material, participants answered questions about the advertisement and about their meanings of voting, before being thoroughly debriefed.

Measures

Campaign Perceptions

As a manipulation check, participants’ perceptions of the campaign messages are measured immediately after exposure. For each type of campaigning, we use the item that has the highest factor loading in a factor analysis of the original scale (Reiter and Matthes 2021; see Table A11 in the Supplemental Information File). We therefore use a single-item measure for positive campaigning (M_HU = 4.62, SD_HU = 3.83, M_US = 5.93, SD_US = 3.46), negative campaigning (M_HU = 3.68, SD_HU = 3.47, M_US = 5.07, SD_US = 3.60), and dirty campaigning (M_HU = 4.74, SD_HU = 3.80, M_US = 5.38, SD_US = 3.65).

Meanings of Voting

The pretest measure of meanings of voting are the items of Study 1. The posttest measure is identical. However, in order to avoid participant fatigue, it was formulated slightly different. Rather than opening with “To me, voting means . . .”, the question opens with “In this election, voting is . . .” This slight change did not result in vastly different answers as the measures are correlated strongly, with some differences to be expected due to the manipulation of the experiment (.65 < r < .74). The final measure is the difference between pretest and posttest meanings of voting, leading to change scores for positive (M_HU = −0.22, SD_HU = 1.98, M_US = −0.35, SD_US = 1.90), negative (M_HU = 0.01, SD_HU = 2.25, M_US = −0.12, SD_US = 2.08), and anti-voting meanings (M_HU = −.17, SD_HU = 2.02, M_US = −0.00, SD_US = 2.20).

Results

Campaign Perceptions (Manipulation Check)

We run three analyses of variance (ANOVAs), using the experimental factors (i.e., ingroup/outgroup and type of campaign) as independent variables and perceived positive, negative, and dirty campaigning as dependent variables (full model in Figure A5 and Table A12 in the Supplemental Information File). In both Hungary and the United States, the positive campaign condition was more strongly perceived as positive campaigning than the negative and dirty campaign conditions. Although the negative campaign condition was perceived as negative campaigning about equally as the positive campaign condition, it was clearly perceived as more negative campaigning than the dirty campaign condition – in both countries. Finally, in both countries, the dirty campaigning condition was more strongly perceived as dirty campaigning than the other two conditions. In addition, the results show a similar partisan bias in both countries: every condition was considered more respectful, negative campaigning for the ingroup party, and more below-the-belt, dirty campaigning for the outgroup party.

Political Campaigns and Meanings of Voting

In Hungary, none of the hypothesized effects are significant.³ The full ANOVA results are shown in Table A13 in the Supplemental Information File and show no significant differences between any conditions. In addition to the preregistered models, we run one-sample t-tests to test whether meanings of voting are significantly different between the pre- and posttest measures in each condition. Replicating Study 1, this shows a negative change in positive meanings of voting in the ingroup dirty campaign condition – indicating that dirty campaigns reduce copartisans’ positive meanings of voting (Figure 4).

Figure 4.

Mean differences in the meanings of voting before and after exposure to the experimental stimuli.

In contrast to Hungary, in the United States, we find a significant interaction effect of the campaign and party manipulations on positive meanings of voting (see Table A13 in the Supplemental Information File for the full ANOVA models). There is no interaction effect between campaigning and party on negative meanings of voting, but rather a significant main effect of the campaign, regardless of the party. Similarly, we find no interaction effect on anti-voting meanings, but a main effect of the campaign. Most of these effects do not support our hypotheses. The findings only support H2c: Outgroup dirty campaigns lead to negative meanings of voting (Figure 4).

In the US experiment, in addition to causing negative meanings of voting, outgroup dirty campaigning also leads to more anti-voting meanings, and to fewer positive meanings of voting (Figure 4). Outgroup negative campaigning causes fewer positive meanings of voting, and outgroup positive campaigning also leads to fewer positive meanings of voting, as well as fewer negative meanings of voting. Consistent with the other studies in this article, ingroup dirty campaigning leads to fewer positive meanings of voting. However, in contrast to Study 1, ingroup negative campaigning also leads to fewer positive meanings of voting. Finally, ingroup positive campaigning reduces negative meanings of voting. These findings are not hypotheses tests and are considered exploratory.

Discussion

Apart from a negative effect of ingroup dirty campaigning on positive meanings of voting, the experiment in Hungary yielded no significant findings. While the experiment in the United States elicited stronger effects, it only confirmed one of our hypotheses: outgroup dirty campaigning causes more negative meanings of voting. In addition, it replicated a finding from both Study 1 and the experiment in Hungary: ingroup dirty campaigning causes a reduction in positive meanings of voting. The US experiment also indicated that outgroup positive and dirty campaigning might reduce positive meanings of voting, and in addition to negative meanings of voting, outgroup dirty campaigning might also increase anti-voting meanings. Finally, the experiments confirmed that perceptions of campaigns are driven by partisanship (Haselmayer et al. 2020; Reiter and Matthes 2022), emphasizing the importance of the partisan bias-corrected robustness check in Study 1.

General Discussion

This study investigated how campaign negativity and “dirtiness” affect ingroup and outgroup partisans’ meanings of voting. Our cross-sectional analysis (Study 1) showed clear correlations between positive, negative, and dirty campaigning by ingroup and outgroup parties and positive, negative and anti-voting meanings. Ingroup positive campaigning is related to positive meanings of voting, ingroup negative campaigning connected to positive, negative, and anti-voting meanings, and ingroup dirty campaigning correlates with fewer positive, more negative but mostly more anti-voting meanings. Outgroup positive campaigning relates to positive meanings of voting, outgroup negative campaigning to anti-voting meanings and outgroup dirty campaigning to negative meanings of voting. These findings are robust to correcting for partisan bias in campaign perceptions, and largely robust to controlling for previous meanings of voting – diminishing the likelihood of a spurious relationship. Moreover, they are strikingly similar across our two countries with two-bloc elections, contributing to our confidence in this relationship between political campaigns and meanings of voting.

However, while these are compelling and robust findings, our experimental investigations in Study 2 did not replicate most of them. The only findings consistent across the cross-sectional and experimental evidence are that ingroup dirty campaigning leads to fewer positive meanings of voting and outgroup dirty campaigning causes negative meanings of voting. Besides those confirmations, Study 2 mostly shows null findings.

We see two possible explanations for this disparity. First, citizens might be affected more by their perceptions of campaign messages than the actual messages in the campaign. Confirming previous work on negative campaigning (Haselmayer et al. 2020; Reiter and Matthes 2022), our manipulation checks find that whether a campaign message was perceived as negative or dirty was dependent on participants’ partisanship. In addition to partisanship, individual differences in citizens’ campaign perceptions can also be affected by the amount of exposure to each campaign (Pattie et al. 2011), or by citizens’ personalities (Nai and Maier 2021). Such variables may explain the extent to which citizens perceive the campaigns of their ingroup and outgroup party as positive, negative, or dirty. Yet, using partisan bias-corrected measures of campaign perceptions did not significantly change the results in Study 1. This method has been shown to effectively remove partisan bias from campaign perceptions to the point that their measurement is equivalent to content analysis of campaign messages (Walter and Van der Eijk 2019a). Moreover, although observational, the findings in Study 1 were largely robust to inclusion of a LDV, strongly reducing the chance of spurious relationships and strengthening our causal claim. As such, we estimate the likelihood that these relationships are void as relatively small.

Second, we might have overestimated the power of one or two campaign messages to cause changes in citizens’ meanings of voting – especially in the midst of an intense election campaign. In the United States, we sought to avoid pretreatment by running the experiment at an earlier stage of the election campaigns. However, participants might have experienced pretreatment from previous elections – especially the notorious 2020 presidential elections in the United States. In other words, respondents might have been exposed to such a large amount of negative and dirty campaigning before the experiments took place that a large part of its influence had already taken effect. This could explain why we find the expected correlations in the cross-sectional analysis – where participants may take their previous experiences into account when answering the questions about current campaigns – but few effects in the experiments. Another finding that points to this explanation is that the effects of the strongest manipulations – the dirty campaign conditions – are in fact replicated in the experiment. As such, the weaker positive and negative campaigning conditions may simply have lacked the psychological impact to cause changes in citizens’ meanings of voting or may have been pretreated too strongly in these two countries to elicit noticeable changes.

Going back to the question we posited at the beginning of the article – does the nature of political campaigns affect how citizens understand voting? – we should carefully conclude that negative and dirty campaigning might affect citizen meanings of voting, but that this effect often takes place over the span of a campaign, rather than each time citizens get exposed to a campaign message. Only the most outrageous campaign messages, those that are uncivil or unfair (and may thus be less common), immediately affect citizen meanings of voting – mostly to the disadvantage of the attacker. As the composition and nature of citizen meanings of voting are still largely unknown, this may be a first indication that they are rather stable attitudes, not easily affected by single exposures to campaign messages.

These findings are essential for both political parties and democracies. Political parties might take away from these findings that their campaigns being perceived as negative leads to electorally desirable effects (negative voting for ingroup voters and alienation for outgroup voters), while perceived dirty campaigns achieve the opposite (negative voting for outgroup voters and alienation for ingroup voters). For democracy, it is important that – regardless of previous claims – not just dirty campaigns have adverse consequences (Geer 2006; Mark 2006; Mattes and Redlawsk 2014). Both negative and dirty campaigns may cause negative and anti-voting meanings, for different partisans. In addition, negative and anti-voting meanings suppress democratic support and mobilization, respectively. This complex interplay of effects may explain previously mixed findings in the negative campaigning literature (Lau et al. 2007). Moreover, a shift from positive to negative and dirty campaigning might seriously damage the legitimacy of democracies by contributing to further democratic detachment and by weakening electoral participation.

In addition to the introduction of meanings of voting to the negative campaigning literature, these findings again emphasize the need for research that differentiates negative campaigns between ingroup and outgroup campaigns, rather than elections cycles (Walter and Van der Eijk 2019a), and for research that differentiates between negative campaigns that transgress social norms and those that do not, that is, negative and dirty campaigns (Reiter and Matthes 2021). Moreover, this may be a starting point in uncovering the mechanisms behind the effects of negative and dirty campaigning.

To address the issue of pretreatment in this study, future experiments may focus on contexts in which negative and dirty campaigning are the exception, rather than the rule. Moreover, while inclusion of in- and outgroups is crucial to uncover the effects of political advertisements, future research could focus on citizens that do not hold a clear ingroup party – as partisans often hold more stable attitudes regarding politics and their party preferences (e.g., Taber and Lodge 2006; Levendusky 2018). Relatedly, we focused on two-bloc elections. First and foremost, this offers an initial baseline of evidence of the dynamics of negative campaigns and meanings of voting that needs to be tested in different contexts. However, rather than allowing broad generalizability, our findings are most applicable to electoral contexts where citizens perceive no viable alternative to an attacked or an attacking party where ingroup–outgroup distinctions are especially salient. Therefore, it is important to note that the generalizability of the findings in this study should be confirmed in countries with a multiparty electoral system.

Finally, this article shows the importance of considering a citizen perspective on democracy, elections, and voting. Voting has long been considered from an elite point of view, from the perspective of political theory or party identifications. However, taking a citizen point of view reveals that the meanings citizens attach to voting change under the influence of different experiences with democracy, such as political campaigns. Putting citizens in the main focus of democracy illuminates not only new effects of negative campaigns but also the extent to which democracy may be at risk by such tactics – at a level that is easily overlooked.

Supplemental Material

sj-pdf-1-hij-10.1177_19401612251376808 – Supplemental material for Negative Campaigns, Negative Votes? Meanings of Voting and Political Campaigns

Supplemental material, sj-pdf-1-hij-10.1177_19401612251376808 for Negative Campaigns, Negative Votes? Meanings of Voting and Political Campaigns by Ming M. Boyer, Cal le Gall and Carolina Plescia in The International Journal of Press/Politics

Footnotes

A previous version of this article was presented at multiple conferences (European Consortium of Political Research general conference 2022;American Political Science Association Political Communication preconference 2022;European Communication Research and Education Association general conference 2022;International Communication Association 2023). We thank all the participants for their useful comments. We also thank Maria Belén Abdala and Anna Lia Brunetti for their feedback along the way.

ORCID iDs

Ming M. Boyer

Cal le Gall

Carolina Plescia

Ethical Considerations

The work in this article has been approved by the Ethics Committee of the University of Vienna (application number: 00642,decision date: 19.04.2021). In addition,the experiments have been specifically approved by the Institutional Review Board of the Department of Government at the University of Vienna (no. 655/no. 716).

Consent to Participate

All participants gave their written informed consent before participation in this study.

Author Contributions

All authors were involved with the design of the studies,the data collection,the data analysis,and the writing of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research,authorship,and/or publication of this article: This work was supported by the European Union (ERC,DeVOTE,no. 949247). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or the European Research Council Executive Agency. Neither the European Union nor the granting authority can be held responsible for them.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Data Availability Statement

The data used in this manuscript are published in relation to a book publication using the same dataset (but different variables). The dataset is publicly available and cited in the article (

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Ming M. Boyer is a research associate in Communication Science at Vrije Universiteit Amsterdam. His work focuses on group identity in political communication and negativity in politics and media.

Cal le Gall is a Postdoc researcher in Comparative Politics at the University of Salzburg. His research interests revolve around voting behaviour,public opinion,European integration,and globalization,employing both quantitative and qualitative research methods.

Carolina Plescia is an Associate Professor in the Department of Government at the University of Vienna. She studies how citizens hold elected officials accountable,how voter satisfaction impacts political behavior,and how citizens’ self-perceived roles in democratic and autocratic systems shape their participation patterns and attitudes.

References

Abramowitz

A. I.

Webster

2016. “The Rise of Negative Partisanship and the Nationalization of U.S. Elections in the 21st Century.” Electoral Studies 41: 12–22. https://doi.org/10.1016/j.electstud.2015.11.001.

Adams

Dow

Merrill

2006. “The Political Consequences of Alienation-Based and Indifference-Based Voter Abstention: Applications to Presidential Elections.” Political Behavior 28, no. 1: 65–86. https://doi.org/10.1007/s11109-005-9002-1.

Alvarez

R. M.

Kiewiet

D. R.

Núñez

2018. “A Taxonomy of Protest Voting.” Annual Review of Political Science 21: 135–54. https://doi.org/10.1146/annurev-polisci-050517.

Ansolabehere

Iyengar

1995. Going Negative. Vol. 95. Free Press.

Ansolabehere

Iyengar

Simon

Valentino

1994. “Does Attack Advertising Demobilize the Electorate?” American Political Science Review 88, no. 4: 829–38. https://doi.org/10.2307/2082710.

Becher

Donnelly

2013. “Economic Performance, Individual Evaluations, and the Vote: Investigating the Causal Mechanism.” The Journal of Politics 75, no. 4: 968–79. https://doi.org/10.1017/S0022381613000959.

Blais

Achen

C. H.

2019. “Civic Duty and Voter Turnout.” Political Behavior 41: 473–97. https://doi.org/10.1007/s11109.

Bos

A. L.

Greenlee

J. S.

Holman

M. R.

Oxley

Z. O. E. M.

Lay

J. C.

2022. “This One’s for the Boys: How Gendered Political Socialization Limits Girls’ Political Ambition and Interest.” American Political Science Review 116, no. 2: 484–501. https://doi.org/10.1017/S0003055421001027.

Brennan

Hamlin

1998. “Expressive Voting and Electoral Equilibrium.” Public Choice 95, no. 1–2: 149. https://doi.org/10.1023/a:1004936203144.

10.

Brunetti

A. L.

Boyer

M. M.

Plescia

2025. “The Empirical Consequences: How do Meanings of Voting Affect Citizens’ Political Behaviour?” In C. Plescia, The Meanings of Voting for Citizens: A Scientific Challenge, a Portrait, and Implications. Oxford University Press.

11.

Campbell

D. E.

2008. Why We Vote. Princeton University Press.

12.

Dalton

R. J.

Sin

T. -C.

Jou

2007. “Understanding Democracy: Data from Unlikely Places.” Journal of Democracy 18, no. 4: 142–56.

13.

Donovan

Tolbert

Gracey

2016. “Campaign Civility Under Preferential and Plurality Voting.” Electoral Studies 42: 157–63. https://doi.org/10.1016/j.electstud.2016.02.009.

14.

Downs

1957. “An Economic Theory of Political Action in a Democracy.” Journal of Political Economy 65, no. 2: 135–50. https://doi.org/10.1086/257897.

15.

Ecker

Glinitzer

Meyer

T. M.

2016. “Corruption Performance Voting and the Electoral Context.” European Political Science Review 8, no. 3: 333–54. https://doi.org/10.1017/S1755773915000053.

16.

Gant

M. M.

Berelson

B. R.

Lazarsfeld

P. F.

Mcphee

W. M.

1985. “Anti-Candidate Voting in Presidential Elections.” Polity 18, no. 2: 329–39.

17.

Gant

M. M.

Davis

D. F.

1984. “Mental Economy and Voter Rationality: The Informed Citizen Problem in Voting Research.” The Journal of Politics 46, no. 1: 132–53. https://doi.org/10.2307/2130437.

18.

Garzia

Ferreira da Silva

2022a. “Negativity and Political Behavior: A Theoretical Framework for the Analysis of Negative Voting in Contemporary Democracies.” Political Studies Review 20, no. 2: 282–91. https://doi.org/10.1177/14789299211000187.

19.

Garzia

Ferreira da Silva

2022b. “The Electoral Consequences of Affective Polarization? Negative Voting in the 2020 US Presidential Election.” American Politics Research 50, no. 3: 303–11. https://doi.org/10.1177/1532673X221074633.

20.

Garzia

Ferreira da Silva

Maye

2023. “Affective Polarization in Comparative and Longitudinal Perspective.” Public Opinion Quarterly 87, no. 1: 219–31. https://doi.org/10.1093/poq/nfad004.

21.

Geer

J. G.

2006. In Defense of Negativity: Attack Ads in Presidential Campaigns. University of Chicago Press.

22.

Gidengil

Lahtinen

Wass

Erola

2021. “From Generation to Generation: The Role of Grandparents in the Intergenerational Transmission of (Non-)Voting.” Political Research Quarterly 74, no. 4: 1137–51. https://doi.org/10.1177/1065912920971715.

23.

Gidengil

Wass

Valaste

2016. “Political Socialization and Voting: The Parent–Child Link in Turnout.” Political Research Quarterly 69, no. 2: 373–83. https://doi.org/10.1177/1065912916640900.

24.

Groenendyk

2019. “Of Two Minds, But One Heart: A Good ‘Gut’ Feeling Moderates the Effect of Ambivalence on Attitude Formation and Turnout.” American Journal of Political Science 63, no. 2: 368–84. https://doi.org/10.7910/DVN/AGIMGY.

25.

Guhin

Calarco

J. M.

Miller-Idriss

2021. “Whatever Happened to Socialization?” Annual Review of Sociology 47: 109–29. https://doi.org/10.1146/annurev-soc-090320.

26.

Hart

S. P.

Nisbet

E. C.

2012. “Boomerang Effects in Science Communication: How Motivated Reasoning and Identity Cues Amplify Opinion Polarization About Climate Mitigation Policies.” Communication Research 39, no. 6: 701–23. https://doi.org/10.1177/0093650211416646.

27.

Haselmayer

Hirsch

Jenny

2020. “Love is Blind. Partisanship and Perception of Negative Campaign Messages in a Multiparty System.” Political Research Exchange 2, no. 1: 1–17. https://doi.org/10.1080/2474736X.2020.1806002.

28.

Hassell

H. J. G.

2021. “Desperate Times Call for Desperate Measures : Electoral Competitiveness, Poll Position, and Campaign Negativity.” Political Behavior 43, no. 3: 1137–59. https://doi.org/10.1007/s11109-019-09588-z.

29.

Jamieson

K. H.

1993. Dirty Politics: Deception, Distraction, and Democracy. Oxford University Press.

30.

Kselman

Niou

2011. “Protest Voting in Plurality Elections: A Theory of Voter Signaling.” Public Choice 148, no. 3–4: 395–418. https://doi.org/10.1007/s11127-010-9661-2.

31.

Kunda

1990. “The Case for Motivated Reasoning.” Psychological Bulletin 108, no. 3: 480–98. https://doi.org/10.1037/0033-2909.108.3.480.

32.

Lau

R. R.

Rovner

I. B.

2009. “Negative Campaigning.” Annual Review of Political Science 12: 285–306. https://doi.org/10.1146/annurev.polisci.10.071905.101448.

33.

Lau

R. R.

Sigelman

Rovner

I. B.

2007. “The Effects of Negative Political Campaigns: A Meta-Analytic Reassessment.” The Journal of Politics 69, no. 4: 1176–209. https://doi.org/10.1111/j.1468-2508.2007.00618.x.

34.

Lavine

H. G.

Johnston

C. D.

Steenbergen

M. R.

2012. The Ambivalent Partisan: How Critical Loyalty Promotes Democracy. Oxford University Press.

35.

Levendusky

M. S.

2018. “When Efforts to Depolarize the Electorate Fail.” Public Opinion Quarterly 82, no. 3: 583–92. https://doi.org/10.1093/poq/nfy036.

36.

Mark

2006. Going Dirty: The Art of Negative Campaigning. Rowman & Littlefield.

37.

Martin

Nai

2024. “Deepening the Rift: Negative Campaigning Fosters Affective Polarization in Multiparty Elections.” Electoral Studies 87: 102745.

38.

Martin

P. S

. 2004. “Inside the Black Box of Negative Campaign Effects: Three Reasons Why Negative Campaigns Mobilize.” Political Psychology 25, no. 4: 545–62. https://doi.org/10.1111/j.1467-9221.2004.00386.x.

39.

Mattes

Redlawsk

D. P.

2014. The Positive Case for Negative Campaigning. University of Chicago Press.

40.

Mayer

S. J.

2017. “How Negative Partisanship Affects Voting Behavior in Europe: Evidence from an Analysis of 17 European Multi-Party Systems with Proportional Voting.” Research & Politics 4, no. 1: 205316801668663. https://doi.org/10.1177/2053168016686636.

41.

Mikola

Santos

F. G.

2025. “Opposition Electoral Strategies Against Democratic Backsliding: The United for Hungary Coalition and its 2022 Primaries.” Democratization 1–25. https://doi.org/10.1080/13510347.2025.2522194.

42.

Mutz

D. C.

2002. “The Consequences of Cross-Cutting Networks for Political Participation.” American Journal of Political Science 46, no. 4: 838–55. https://doi.org/10.2307/3088437.

43.

Nai

. 2013. “What Really Matters is Which Camp Goes Dirty: Differential Effects of Negative Campaigning on Turnout During Swiss Federal Ballots.” European Journal of Political Research 52: 44–70. https://doi.org/10.1111/j.1475-6765.2012.02060.x.

44.

Nai

Garzia

Aaldering

da Silva

F. F.

Gattermann

2022. “For a Research Agenda on Negative Politics.” Politics and Governance 10, no. 4: 243–6. https://doi.org/10.17645/pag.v10i4.6622.

45.

Nai

Maier

2021. “Is Negative Campaigning a Matter of Taste? Political Attacks, Incivility, and the Moderating Role of Individual Differences.” American Politics Research 49, no. 3: 269–81. https://doi.org/10.1177/1532673X20965548.

46.

Parisi

Pasquino

1977. “Relazioni partiti-elettori e tipi di voto.” In Continuità e mutamento elettorale in Italia, edited by Parisi

Pasquino

Il Mulino.

47.

Pattie

Denver

Johns

Mitchell

2011. “Raising the Tone? The Impact of ‘Positive’ and ‘Negative’ Campaigning on Voting in the 2007 Scottish Parliament Election.” Electoral Studies 30, no. 2: 333–43. https://doi.org/10.1016/j.electstud.2010.10.003.

48.

Plescia

Abdala

M. B.

Boyer

M. M.

Brunetti

A. L.

Le Gall

2025. “Panel Election Survey Data on Citizen Meanings of Voting in Hungary and the USA (SUF edition).” In AUSSDA. https://doi.org/10.11587/UOWEY1.

49.

Plescia

B. Abdala

Kritzinger

Zechmeister

E. J.

2025. “The Meanings of Voting: A Conceptual Framework.” In C. Plescia (Ed.), The Meanings of Voting for Citizens: A Scientific Challenge, a Portrait, and Implications. Oxford University Press.

50.

Redlawsk

D. P.

Civettini

A. J. W.

Emmerson

K. N.

2010. “The Affective Tipping Point: Do Motivated Reasoners Ever ‘get it’?” Political Psychology 31, no. 4: 563–93. https://doi.org/10.1111/j.1467-9221.2010.00772.x.

51.

Reiter

Matthes

2021. ““The Good, the Bad, and the Ugly”: A Panel Study on the Reciprocal Effects of Negative, Dirty, and Positive Campaigning on Political Distrust.” Mass Communication and Society 25, no. 5: 649–72. https://doi.org/10.1080/15205436.2021.1934702.

52.

Reiter

Matthes

2022. “How Partisanship Matters: A Panel Study on the Democratic Outcomes of Perceived Dirty Campaigning.” Politics and Governance 10, no. 4: 247–60. https://doi.org/10.17645/pag.v10i4.5672.

53.

Riker

W. H.

Ordeshook

P. C.

1968. “A Theory of the Calculus of Voting.” American Political Science Review 62, no. 1: 25–42. https://doi.org/10.2307/1953324.

54.

Schuessler

A. A.

2000. “Expressive Voting.” Rationality and Society 12, no. 1: 87–119. https://doi.org/10.1177/104346300012001005.

55.

Slothuus

De Vreese

C. H.

2010. “Political Parties, Motivated Reasoning, and Issue Framing Effects.” The Journal of Politics 72, no. 3: 630–45. https://doi.org/10.1017/s002238161000006x.

56.

Somer-Topcu

Weitzel

2022. “Negative Campaigning and Vote Choice in Europe: How Do Different Partisan Groups React to Campaign Attacks?” Comparative Political Studies 55, no. 13: 2283–310. https://doi.org/10.1177/00104140221074283.

57.

Stagnaro

M. N.

Amsalem

2025. “Factual Knowledge Can Reduce Attitude Polarization.” Nature Communications 16, no. 1: 3809. https://doi.org/10.1038/s41467-025-58697-3.

58.

Taber

C. S.

Lodge

2006. “Motivated Skepticism in the Evaluation of Political Beliefs.” American Journal of Political Science 50, no. 3: 755–69.

59.

Velez

Y. R.

Liu

2025. “Confronting Core Issues: A Critical Assessment of Attitude Polarization Using Tailored Experiments.” American Political Science Review 119, no. 2: 1036–53. https://doi.org/10.1017/S0003055424000819.

60.

Walter

A. S.

Van der Eijk

2019a. “Measures of Campaign Negativity: Comparing Approaches and Eliminating Partisan Bias.” The International Journal of Press/Politics 24, no. 3: 363–82. https://doi.org/10.1177/1940161219843404.

61.

Walter

A. S.

Van der Eijk

2019b. “Unintended Consequences of Negative Campaigning: Backlash and Second-Preference Boost Effects in a Multi-Party Context.” The British Journal of Politics and International Relations 21, no. 3: 612–29. https://doi.org/10.1177/1369148119842038.

62.

Wiseman

A. W.

Astiz

M. F.

Fabrega

Baker

D. P.

2011. “Making Citizens of the World: The Political Socialization of Youth in Formal Mass Education Systems.” Compare: A Journal of Comparative and International Education 41, no. 5: 561–77. https://doi.org/10.1080/03057925.2010.530764.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

3.77 MB

0.00 MB