1. Introduction
Populist radical right (PRR) parties challenge mainstream parties by combining radical anti-immigration and anti-globalization positions with a thin populist ideology depicting politics as an endless struggle between “evil elites” and “the people” (Mudde, Reference Mudde2004). This populist ideology commonly manifests in parties’ communication—which I refer to as populist rhetoric. The success of PRR parties has sparked controversy about the extent to which mainstream parties (should) accommodate the issue-positions of PRR challengers by moving to the right, especially on immigration policies (see, amongst others, Meguid, Reference Meguid2008; Bale et al., Reference Bale, Green-Pedersen, Krouwel, Luther and Sitter2010; Abou-Chadi, Reference Abou-Chadi2016; Abou-Chadi and Krause, Reference Abou-Chadi and Krause2020; Chou et al., Reference Chou, Dancygier, Egami and Jamal2021; Hjorth and Larsen, Reference Hjorth and Larsen2022; Krause et al., Reference Krause, Cohen and Abou-Chadi2023). Much less is known about the role of discursive responses to PRR success and competition over issues other than immigration though.
Incorporating discursive party strategies—such as engaging in populist rhetoric—into analyses of party competition is an important endeavor. First, there is a rich experimental literature demonstrating that populist rhetoric can be an effective communication strategy. Populist rhetoric affects how voters attribute blame (Hameleers et al., Reference Hameleers, Bos and de Vreese2017, Reference Hameleers, Bos and Fawzi2018), evaluate out-groups (Hameleers and Schmuck, Reference Hameleers and Schmuck2017; Hameleers and Fawzi, Reference Hameleers and Fawzi2020), assess party leaders (Bos et al., Reference Bos, Van Der Brug and De Vreese2013), perceive their own group identities (Bos et al., Reference Bos, Schemer, Corbu, Hameleers, Andreadis, Schulz, Schmuck, Reinemann and Fawzi2020), and understand societal problems (Busby et al., Reference Busby, Gubler and Hawkins2019). Parties might also use such rhetoric to discredit competitors and highlight their novelty (De Vries and Hobolt, Reference De Vries and Hobolt2020). While this literature demonstrates that populist rhetoric affects political behavior, it remains a question of an ongoing academic debate to what extent such rhetoric also has a direct electoral appeal (e.g., Neuner and Wratil, Reference Neuner and Wratil2020; Castanho Silva et al., Reference Castanho Silva, Neuner and Wratil2022; Dai and Kustov, Reference Dai and Kustov2023; Kittel, Reference Kittel2024).
Second, there is mounting evidence that mainstream parties increasingly engage in populist rhetoric. In his influential article, Mudde (Reference Mudde2004) posits that mainstream parties will use “populist themes and rhetoric to try and fight off the challenge” of successful PRR parties (563). Since then, an important body of scholarly work has investigated whether this “populist zeitgeist” has come about. Scholars have asked to what extent parties have adopted populist rhetoric in reaction to changes in their electoral performance (Rooduijn et al., Reference Rooduijn, De Lange and Van Der Brug2014; Hunger, Reference Hunger2020; Breyer, Reference Breyer2022; Licht et al., Reference Licht, Abou-Chadi, Barbera and Hua2024), which issues parties talk about in a populist manner (Bernhard and Kriesi, Reference Bernhard, Kriesi, Caiani and Graziano2021), how parties (Ernst et al., Reference Ernst, Engesser, Buchel, Blassnig and Esser2017a, Reference Ernst, Engesser and Esser2017b) and leaders (Zulianello et al., Reference Zulianello, Albertini and Ceccobelli2018) use populist rhetoric in their social media strategies, as well as how the media employs populist narratives (Rooduijn, Reference Rooduijn2014; Hameleers and Vliegenthart, Reference Hameleers and Vliegenthart2020). More recently, Esguerra et al. (Reference Esguerra, Hagemeister, Heid and Leffler2023) leverage the quasi-random assignment of committee memberships in the German parliament providing causal evidence that populist rhetoric “spills-over” from PRR to mainstream party politicians. In addition, there is plenty of qualitative evidence from examples like the British Conservative Party (e.g., Bale, Reference Bale2018, Reference Bale2023; Baldini et al., Reference Baldini, Bressanelli and Gianfreda2020) and the Austrian People’s Party under the leadership of former chancellor Sebastian Kurz (e.g., Wodak, Reference Wodak2018) suggesting that mainstream parties engage in populist rhetoric when competing with PRR challengers, and that they do so strategically. However, the effects of this strategy thus far remain unknown. By addressing this gap, this experiment makes an important contribution to the literature on party competition between mainstream and PRR parties.
I argue that mainstream parties can respond to PRR challenges in two dimensions. First, there is spatial and programmatic competition over which party’s position comes closest to voters’ ideal points (Downs, Reference Downs1957). There, mainstream parties can react to PRR success with positional accommodation, that is, with “moving to the right” (Meguid, Reference Meguid2008). Second, there is discursive competition. Here, mainstream parties can engage in populist rhetorical accommodation; a strategy in which they emphasize thin populism in their rhetoric. Theoretically, I propose that voters—in particular, those with populist attitudes—might use populist rhetoric as a heuristic when evaluating a party (Sniderman et al., Reference Sniderman, Brody and Tetlock1993; Chong and Druckman, Reference Chong and Druckman2007). Depending on the distribution of populist attitudes in the electorate and their importance for voters’ assessment of parties, mainstream parties might be able to appeal to voters using populist rhetoric. Against this backdrop, I ask: What are the effects of mainstream parties engaging in populist rhetoric?
I conduct a factorial vignette experiment on a high-quality sample of German voters (n = 4,042) to assess the causal effects of rhetorical and positional accommodation. The vignettes describe a scenario in which the PRR Alternative für Deutschland (AfD) challenges Germany’s center-right Christian Democrats (CDU/CSU) or the center-left Social Democrats (SPD) using populist rhetoric and demanding a more radical policy position on one out of three issues: immigration, the war in Ukraine, or inflation. The scenario is manipulated in the extent to which the mainstream party accommodates the AfD positionally and/or rhetorically. The outcome of interest is the propensity to vote for the respective mainstream party.
The results demonstrate that mainstream parties cannot accommodate populist challengers by engaging in populist rhetoric. Voters penalize or reward a mainstream party for policy positions conditional on their own issue preferences but do not react to different types of rhetoric. In line with spatial theories of voting, the results show that voters on the right and voters with pronounced populist attitudes reward mainstream parties for moving to the right, whereas centrist and left voters, as well as voters with less populist attitudes penalize mainstream parties for positional accommodation. However, none of these groups is particularly susceptible to populist rhetorical accommodation. This is despite respondents clearly identifying the rhetoric as populist, as demonstrated by the included manipulation checks. There is no heterogeneity in these effects for different issues, the two different mainstream parties, or typical socio-demographics.
These findings advance the understanding of party competition between mainstream and PRR parties. They question the alleged link between populist rhetoric and the electoral success of political parties by demonstrating that programmatic rather than rhetorical party strategies matter to voters when evaluating mainstream against PRR parties.
2. The case for rhetorical accommodation
The increasing success of PRR parties across Europe has spared a lively academic and political debate about how mainstream parties should respond to such challenges. Most scholarly work is concerned with mainstream parties’ programmatic responses, with a focus on immigration (e.g., Meguid, Reference Meguid2005, Reference Meguid2008; Abou-Chadi, Reference Abou-Chadi2016; Schumacher and Van Kersbergen, Reference Schumacher and Van Kersbergen2016; Abou-Chadi and Krause, Reference Abou-Chadi and Krause2020), and their electoral implications (Dahlstrom and Sundell, Reference Dahlstrom and Sundell2012; Chou et al., Reference Chou, Dancygier, Egami and Jamal2021; Hjorth and Larsen, Reference Hjorth and Larsen2022; Krause et al., Reference Krause, Cohen and Abou-Chadi2023). Arguments in this debate often follow a traditional Downsian logic of spatial competition (Downs, Reference Downs1957): If mainstream parties moved to the right, there would be no space for PRR parties and voters with radical policy preferences would return to mainstream parties. The empirical evidence for this logic is mixed. This is because mainstream parties face a trade-off between maintaining those voters who prefer moderate policy positions (i.e., those in the center) while appealing to those who prefer more radical positions (i.e., those on the right). In addition, accommodation might have unintended long-term consequences, like the legitimization of radical positions (Dahlstrom and Sundell, Reference Dahlstrom and Sundell2012; Krause et al., Reference Krause, Cohen and Abou-Chadi2023; Vrakopoulos, Reference Vrakopoulos2022). That being said, there are certainly conditions—for example, in a Block party system like Denmark (Hjorth and Larsen, Reference Hjorth and Larsen2022) or in combination with ostracism (Van Spanje and De Graaf, Reference Van Spanje and De Graaf2018)—under which positional accommodation might pay off.
Theoretically, I propose that there are two dimensions of mainstream party responses to PRR success: The spatial and programmatic dimension, in which mainstream parties can engage in a positional accommodation strategy. And the discursive dimension of party competition, which offers the possibility of populist rhetorical accommodation. In rhetorical accommodation, mainstream parties mimic the populist rhetoric of their challengers. Given that these two strategies are not mutually exclusive, mainstream parties can also engage in a combination of positional and rhetorical accommodation (full accommodation). Finally, there is also the absence of these two strategies, i.e., the mainstream party neither changes its position nor does it adopt populist rhetoric (no accommodation). This logic is visualized in Table 1.
Table 1 Overview of mainstream party strategies
But, why should populist rhetorical accommodation work? Influential work in public opinion challenges spatial voting models (e.g., Zaller et al., Reference Zaller1992; Achen and Bartels, Reference Achen and Bartels2017). Instead, it argues voters to base their decisions on cognitive shortcuts (heuristics) (Sniderman et al., Reference Sniderman, Brody and Tetlock1993)—especially, frames offered by elites (Entman, Reference Entman1993; Chong and Druckman, Reference Chong and Druckman2007). By “select[ing] some aspects of a perceived reality and mak[ing] them more salient [..]” politicians can “influence human consciousness” (Entman, Reference Entman1993, 52). What differentiates populist rhetoric from other communication strategies is that, as an ideology, it can interact with voters’ populist attitudes (e.g., Castanho Silva et al., Reference Castanho Silva, Andreadis, Anduiza, Blanusa, Corti, Delfino, G., Ruth-Lovell, Spruyt, Steenbergen, Hawkins, Carlin, Littvay and Rovira Kaltwasser2019; Wuttke et al., Reference Wuttke, Schimpf and Schoen2020). By tapping into such preexisting sets of beliefs among voters, political parties might be able to leverage populist rhetoric for their own electoral benefit.
That being said, recent work argues that politicians also face trade-offs when using populist rhetoric (Dai and Kustov, Reference Dai and Kustov2022), similar to those described for spatial competition. Clearly, there is heterogeneity in the extent to which voters prefer populist rhetoric: While some voters might credit parties for strongly worded criticism of elites and praises of the people, others might shy-away from parties that use such rhetoric (e.g., Gennaro et al., Reference Gennaro, Lecce and Morrelli2019; Dai and Kustov, Reference Dai and Kustov2022; Levy et al., Reference Levy, Razin and Young2022). The extent to which populist rhetorical accommodation can influence a mainstream party’s success will thus depend on the preferences for such appeals in the electorate and on how influential these are in shaping voters’ assessments of a party, relative to the strength of voters’ policy preferences. Therefore, I zoom in on three relevant sub-groups: Voters with populist attitudes, voters who prefer radical policy positions, and PRR voters.
3. Research design
Disentangling the effects of populist rhetoric observationally is challenging because it is usually enmeshed with host ideologies (Hunger and Paxton, Reference Hunger and Paxton2022). Moreover, voters’ perceptions of these strategies will be strongly affected by individual information-seeking behavior and are thus subject to bias (Leeper and Slothuus, Reference Leeper and Slothuus2014). To isolate the causal effects, I conduct a factorial vignette survey experiment on a representative sample of German citizens (n = 4042).Footnote 1 The German case provides several advantages: German mainstream parties have experienced serious intra-party disagreements about how to respond to an emerging radical right challenger (the AfD) with some factions arguing for an accommodative approach and others for an adversarial approach (Dilling, Reference Dilling2018). This makes Germany a typical case of competition between PRR challengers and mainstream parties.Footnote 2
To begin, participants are asked an established question battery to capture their populist attitudes (Castanho Silva et al., Reference Castanho Silva, Andreadis, Anduiza, Blanusa, Corti, Delfino, G., Ruth-Lovell, Spruyt, Steenbergen, Hawkins, Carlin, Littvay and Rovira Kaltwasser2019).Footnote 3 Then, respondents are asked for their preferences on one out of three issues: immigration, further military support for Ukraine, and active economic policies in reaction to the cost-of-living crisis. I argue this selection to be a reflection of relevant issues that PRR and mainstream parties compete over today. Immigration undoubtedly is the focal issue for competition between radical right and mainstream parties (Abou-Chadi, Reference Abou-Chadi2016). Admittedly, policies in response to the cost-of-living crisis and military aid for Ukraine are not conventional radical right issues. However, all these issues have in common that the radical right in Germany acts as an issue-entrepreneur attempting to split the internally-divided electorates of mainstream parties (Hobolt and de Vries, Reference Hobolt and de Vries2015; De Vries and Hobolt, Reference De Vries and Hobolt2020). At the time of fielding the experiment, the AfD promoted significantly more anti-immigration, pro-Russia, and pro-controlled economy positions than most mainstream parties in Germany, which are internally split on these issues.Footnote 4 In addition, all these issues were salient during the fieldwork of this survey. Testing the effects of accommodation strategies on these different issues is relevant for this design because one would expect accommodation to work in particular for contentious issues that voters might base their voting decisions on (Meguid, Reference Meguid2008).
For each issue, the first paragraph of the vignette is held constant and describes a fictitious parliamentary debate in which the AfD criticizes mainstream parties’ employing typical populist tropesFootnote 5 and promotes its policy position. The second paragraph is then manipulated in two dimensions: First, I randomize whether the respective mainstream party is the CDU/CSU or the SPD, and second, I manipulate the mainstream party strategy which can be either (1) No Accommodation, (2) Positional Accommodation, (3) Rhetorical Accommodation, or (4) Full Accommodation.
In the first condition, the real position of the two mainstream parties on the given issues is used. These positions are presented without populist rhetoric (no accommodation). The second condition uses the same non-populist rhetoric but promotes a position that is more similar to the proposal of the AfD (positional accommodation). While these AfD-like proposals are certainly not representative of the party’s median positions, there are factions within these parties that advocate for such positions; the arguments will therefore appear plausible to participants. The third condition promotes the original policy position but frames the policy using populist rhetoric (rhetorical accommodation). Populist rhetoric here contains all three elements of populism as a thin ideology: An overtly positive appeal to a fictitious, homogeneous group of “the people,”Footnote 6 a strongly worded critique of “the elites”Footnote 7 as well as a dualistic worldview of politics as a constant struggle of “the people” versus “the elites”Footnote 8 (Mudde, Reference Mudde2004). The fourth condition combines an AfD-like policy position with populist rhetoric. One example for the rhetorical accommodation condition is presented in Figure 1 below.Footnote 9
Figure 1. Screenshot of main survey page (rhetorical accommodation).
The survey contains two questions nudging participants to engage with the vignettes cognitively. One of the questions directly taps into the logic of spatial party competition and asks respondents whether they think that the mainstream party (as presented in the vignette) has “moved towards the AfD.” The other question asks respondents explicitly to think about whether the mainstream party’s “choice of words resembles the choice of words of the AfD.” These questions also serve as manipulation checks to test whether respondents have actually perceived the vignettes in the intended way.
The outcome variable is the propensity to vote for the respective mainstream party. I measure these preferences by asking respondents to consider the vignette and evaluate how likely it is that they would vote for that party on an eleven-points scale.Footnote 10
4. Results
Following the pre-registration, I analyse the effects of rhetorical, positional and full accommodation averaging across all respondents, all issues, and both mainstream parties. Figure 2 visualises the means in the propensity to vote for the respective mainstream party in all four conditions. On average, there are no significant differences between these categories. The results of Model 1 in Table 2 confirm this finding. Holding constant the issue and the mainstream party, neither rhetorical, nor positional accommodation, nor a combination of these two affect the propensity to vote for the respective mainstream party.
Figure 2. Means of propensity to vote for mainstream party in the four treatment categories with 95% confidence intervals.
Table 2. The average effects of both accommodation strategies and effects conditional on voters’ left-right preferences. Alternative model specifications with separate interaction terms are in Appendix Table 5
*** p<0.001.
** p<0.01.
* p<0.05.
Controlling for Issue and Mainstream Party.
This result is a consequence of positive and negative effects canceling each other out (see Model 2 in Table 2). Interacting the treatment with the pre-treatment measure of issue preferences, I find that respondents with right preferences (less immigration, no further military support for Ukraine, and more government action in response to the cost-of-living crisis/inflation) reward a mainstream party for moving to the right. As a spatial understanding would predict, voters on the left penalize mainstream parties for accommodating the AfD’s positions.Footnote 11
Figure 3. Predicted values for rhetorical and positional accommodation conditional on voters’ left-right preferences using a discrete binning of the moderator (see Appendix Table 6); the distribution of preferences being shown in the underlying histogram.
I visualize the predicted values conditional on respondents’ left-right preferences in Figure 3 below.Footnote 12 The left panel displays the predictions for different mainstream party positions, the right panel the predictions for different types of mainstream party rhetoric. As the results in the first panel show, I find that respondents on the left and in the center clearly prefer mainstream parties not to move to the right. At this part of the scale, the effects of no accommodation (in gray) are more positive than those for positional accommodation (in black). As shown in the panel, these differences are also substantively sizable. In other words, a move to the right causes a more negative evaluation of the mainstream party among left and centrist respondents and a more positive evaluation among radical-right respondents.
In the right panel, the estimates for respondents across the entire left-right spectrum are almost identical and the confidence intervals overlap accordingly. This demonstrates that rhetorical accommodation—across the left-right spectrum—does not increase people’s evaluation of mainstream parties. Taken together, these results demonstrate that the match between voters’ issue preferences and a party’s position—in accordance with spatial theories of voting—is far more predictive for voting behavior than the framing of a position in populist respectively non-populist rhetoric.
I assess the differences conditional on the prevalence of populist attitudes among respondents (Appendix Table 8) and based upon past voting behavior (Appendix Table 7). The results follow the same pattern: Voters with pronounced populist attitudes tend to reward mainstream parties for positional accommodation whereas voters with weak populist attitudes tend to penalize mainstream parties for it. I also find past AfD voters to be more likely to reward the respective mainstream party when this party accommodates the AfD’s demands, whereas mainstream party voters are penalizing a mainstream party for positional accommodation.Footnote 13 The results of both models demonstrate that despite being responsive to changes in issue-positions, none of these groups is susceptible to populist rhetorical accommodation.
What stands out from these analyses is that none of the interactions between populist rhetoric and issue-position are significant. This is surprising, given that one might assume that radical positions could be presented more credibly using populist rhetoric. This indicates that voters attribute far more importance to how mainstream parties respond to PRR challengers positionally than rhetorically. Equally remarkable is the lack of heterogeneity found. In the appendix, I present the results for all three issues (Table 9 and Figure 6) and for the two mainstream parties (Table 10 and Figure 7) separately. I do not find any evidence suggesting that rhetorical or positional accommodation work better for a specific party or when talking about a certain issue.Footnote 14
One concern would be that respondents did not perceive the populist rhetoric and the parties’ positions as such. I address these concerns using the manipulation checks included. The left panel of Figure 4 shows that respondents correctly recognized the radical positions as more similar to those of the AfD. Crucially for this experiment, they did not mistake populist rhetoric for a radical position, there are no significant differences between the two moderate conditions (1 and 3) and the two radical positions (2 and 4) visualized in the left panel. This demonstrates that respondents clearly perceived the rhetorical differences as distinct from the positional ones—an important assumption of this experiment.Footnote 15 This suggests that the missing effects of rhetorical accommodation are likely to be genuine null-results.
Figure 4. Means of manipulation tests (0–5) in the four treatment categories with 95% confidence intervals.
That being said, the manipulation checks do reveal some complications. Respondents got the impression that positional accommodation resembles the AfD’s choice of words too (comparing 1 and 2, respectively 3 and 4 in the right panel). This implies that the observed effects for moving to the right might, in part, be driven by respondents’ perceiving this as an imitation of the AfD’s rhetoric. This could be a problem if respondents’ perceptions of the mainstream party rhetoric varied conditional on their left-right preferences as this would bias the results observed for positional accommodation—this was not the case here though (see Appendix Table 4). Instead, I argue, that this finding offers two important insights: First, voters get the impression that when mainstream parties accommodate the radical right, they also sound more like the radical right—an important side-effect for mainstream parties to consider. And second, party rhetoric and party positions are clearly tightly intertwined in voters’ minds. This demonstrates the need for careful experimental work that randomizes party positions independently from party rhetoric.
Finally, one might be concerned that the mainstream party strategy was presented as a direct response to a PRR challenge in the vignette. Experimentally manipulating the processes of a party moving to the right and adopting a certain rhetoric is challenging because it requires respondents to be informed about the status quo and then to think about potential changes. To circumvent this challenge, the vignettes were designed to present mainstream party strategies as a direct response to a PRR challenger—and, as indicated by the manipulation checks, this approach was successful. In addition, I argue that this approach is externally valid given that journalists frequently discuss mainstream party strategies as a direct response to PRR success.Footnote 16 That being said, even if one assumed the design of the vignette (or the inclusion of the manipulation checks) to be too obvious to respondents, this would make the null-result for rhetorical accommodation an even more remarkable finding.
5. Conclusion
In this note, I demonstrate that mainstream parties’ attempts to accommodate PRR challengers by engaging in populist rhetoric do not affect voters’ propensity to vote for mainstream parties. I find that voters penalize or reward mainstream parties for their programmatic responses but do not react to rhetorical accommodation. Although mainstream parties might receive credit for accommodating a PRR challenger from radical right voters and voters with populist attitudes, they risk getting penalized by left and centrist voters as well as voters with weak populist attitudes. The extent to which positional accommodation is successful will thus be determined by the distribution of issue preferences in the electorate relative to the mainstream party’s position. However, even among those sub-groups of the electorate which one might assume to be particularly susceptible to populist rhetoric, rhetorical accommodation does not work.
These findings make a significant contribution by challenging the widely-assumed link between populist rhetoric and electoral performance. Theoretically speaking, populist rhetoric does not appear to be a useful heuristic for voters when assessing political parties. This stands in contrast to findings from earlier work on populist rhetoric. Why could that be the case? Earlier work on populist rhetoric has tested the effects of thin populism mostly in combination with other elements of political communication such as blame attribution (e.g., Hameleers and Schmuck, Reference Hameleers and Schmuck2017) and emotive rhetoric (e.g., Hameleers et al., Reference Hameleers, Bos and de Vreese2017), or with positional cues (e.g., Bos et al., Reference Bos, Van Der Brug and De Vreese2013). The factorial design employed here overcomes this challenge and shows that these elements might be the main drivers of the observed effects. In contrast, the appeal of thin populism appears to be fairly limited. This contributes directly to a growing body of work that aims to disentangle the effects of populist appeals from policy positions associated with populist parties (e.g., Neuner and Wratil, Reference Neuner and Wratil2020; Castanho Silva et al., Reference Castanho Silva, Neuner and Wratil2022; Dai and Kustov, Reference Dai and Kustov2023; Kittel, Reference Kittel2024).
One important limitation of this experimental design is a consequence of the outcome variable chosen: This experiment only looks at the immediate and direct effects of accommodation. It zooms in on specific mainstream parties and analyses the effects of accommodation on their electoral fortunes. In reality though, such party strategies will have effects on other parties too. Hjorth and Larsen (Reference Hjorth and Larsen2022) for example argue that in a block party system like Denmark, positional accommodation might repel centrist voters to other mainstream parties while winning over voters from the PRR resulting in a net positive effect of accommodation. Future research should thus look into the consequences of different types of accommodation beyond the accommodating and the accommodated party, as well as into the role that different issues play in moderating the effects of accommodation strategies.
Finally, two additional words of caution: First, despite being a typical case, most voters will associate populist rhetoric with a specific political party in Germany, which is likely to affect its appeal. In countries where populist rhetoric is commonly used by various parties, rhetorical accommodation might thus play out differently. Second, there might be other forms of rhetorical accommodation such as “dog-whistle politics” (e.g., Albertson, Reference Albertson2015; Bonikowski and Zhang, Reference Bonikowski and Zhang2023), or “tough talk” (e.g., Lutz, Reference Lutz2021) that mainstream parties might use to appeal to PRR voters. I hope to inspire more research on discursive party strategies that develops a better theoretical and empirical understanding of how discursive and issue-based competition relate to each other.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/psrm.2025.14. To obtain replication material for this article, please visit https://doi.org/10.7910/DVN/BAIHDZ.
Acknowledgements
I am grateful to Benjamin Lauderdale, Christopher Wratil and Tom O’Grady for their support in pursuing this project. Tarik Abou-Chadi, Ivo Bantel, Emily Frank, Fabian Habersack, Sara Hobolt, Sophia Hunger, Michael Jacobs, Moritz Marbach, Lisa Zehnter, and Andrei Zhirnov provided invaluable feedback at different stages of this project. Previous versions of this article were presented at EPSA 2023, at the ECPR Joint Sessions 2023, and at the University of Vienna. I thank the three anonymous reviewers at PSRM for their excellent suggestions.
Funding statement
This research project received funding from the Economic and Social Research Council (Grant Number: ES/P000592/1), UCL’s Department Research Fund and the PhD Opportunity Fund, the Studienstiftung des deutschen Volkes, and the Einstein Stiftung (Grant Number: EVF-BUA2022-691).
Research ethics
This research was declared exempt from ethical review at University College London (UCL). The Pre-Registration including the full questionnaire and all treatment texts can be accessed via https://osf.io/pru6y.
Open access
