Sage Journals: Discover world-class research

Abstract

In recent years, a wave of populist leaders has emerged in many democratic countries, including the United States. Previous studies have argued that populist rhetoric matters for leaders’ electoral support because the public has populist attitudes, which are activated in contexts of failure of democratic governance or economic crises. This paper investigates the opposite causal direction and argues that people’s support for populist ideas can be an effect rather than a cause of leaders’ electoral support. People who support a candidate due to the candidate’s party affiliation or policy position tend to support or oppose populist or anti-populist messages if they learn that the candidate of the party they identify with supports that message. The paper investigates the argument with an experiment that randomly assigns (anti-)populist messages and a cue about the candidate that supports the message. The experiment shows that voters’ party identification largely affects support for both populist and anti-populist messages.

Keywords

Populism populist attitudes political behavior party identification experimental design

Introduction

This paper investigates if partisan cues affect voters’ support for populist messages. It asks: Is voters’ support for populists’ anti-establishment and people-centric rhetoric an effect of attitudes toward the message-giver? This question is important because a wave of populist leaders gained ground in many democratic nations in recent decades, and their electoral support grew in many countries (Rooduijn et al., 2019), including Italy (Erisen et al., 2021), Spain (Marcos-Marne 2020), Austria (Müller 2004), Germany (Weisskircher 2020), and other nations in Europe, Asian, and Latin America (Albertazzi and McDonnell 2007; Hawkins and Rovira Kaltwasser 2017). In Brazil, the right-wing conservative populist candidate Jair Bolsonaro won the 2018 presidential election after receiving more than 50 million votes. In the United States, Donald Trump won the 2016 presidential election after a campaign filled with populist elements (Oliver and Rahn 2016; Norris and Inglehart 2019; Hawkins and Littvay 2019). The news media referred to this phenomenon as a “storm of populism” (Legorano and Mesco 2017), and scholarship on populist politics has grown exponentially in the last few years (Hunger and Paxton 2021). That scholarship includes debates on whether the rise of populist leaders, particularly those associated with right-wing ideologies, to positions of power in government represents a threat to liberal democracy due to populist leaders’ anti-elitist, exclusionary, and often authoritarian rhetoric (Rodrik 2018; Mansbridge and Macedo 2019). Because populist leaders use elections to ascend to government positions, a pressing question is whether voters support these leaders because of leaders’ populist rhetoric or, conversely, if support for the non-populist features of these leaders—for example, due to voters’ party loyalty—leads people to sympathize with the populist ideas they propagate. This study focuses on the latter question.

The issue with the causal direction becomes clear when we consider the modern definitions of populism in the literature. Mudde (2017) define populism as a “thin-centered” ideology. That ideational definition captures some core aspects of the modern manifestation of populism, stating that it is comprised of three elements: people-centrism, anti-elitism, and a Manichean outlook. People-centrism is an idea that “the people,” vaguely defined in populists’ rhetoric, are virtuous and the only legitimate source of power, and the populist leader claims to be the embodiment of the will of that “people.” Anti-elitism is a notion that the political, economic, and intellectual elite are corrupt and self-serving. Finally, these two elements are combined with a Manichean outlook of society, which adds a moral charge to the “people versus elite” conflict and depicts politics as a struggle between virtuous good people against the corrupt, self-serving elite. Based on that definition, a political leader is considered populist if his rhetoric contains these three elements (people-centrism, political Manichaeism, and anti-elitism), regardless of the policy position (e.g., anti-immigration, nationalism, conservatism, or economic liberalism) or party membership of that leader.

This conceptual separation between “core” populist features (people-centrism, anti-elitism, and a Manichean outlook) and classical notions of ideology or policy positions (e.g., anti-immigration and conservatism) is an important contribution of recent literature, and it is present in alternative definitions of populism, as well (Ostiguy 2017; Mansbridge and Macedo 2019; Weyland 2021). Scholars have shown that parties and leaders of different ideological orientations—from far-left to center (valence populists) and far-right—have adopted populist elements in their rhetoric (March 2017; Ivaldi et al. 2017; Zulianello 2020). In the US, it has been used by both Democrats and Republicans throughout US electoral history. In some election years, populist elements in presidential candidates’ speeches appear predominantly among Democrat candidates, while in other years, it is the Republican candidates who mostly use populist elements in their speeches (Bonikowski and Gidron 2016).

This modern concept of populism and its analytical separation between populism and classical notions of ideology (left-right, conservatism-liberalism) have important implications for our understanding of populist support. If one accepts that separation, captured by the ideational definition, then one must conclude that populism matters for leaders’ electoral support to the extent that people support the people-centric, anti-elitist, and Manichean ideas in populists’ rhetoric.

Many studies followed that understanding and emphasized why the populist elements of leaders’ rhetoric, rather than leaders’ policy positions, can be appealing to voters. For instance, anti-establishment attitudes can emerge due to a representation crisis in which people feel that traditional parties are losing their representation function (Mair 2002; Kriesi 2014).

In a similar direction, scholars have investigated if the public holds populist attitudes, and if these attitudes affect support for populist leaders. The concept of populist attitudes is based on the ideational definition of populism, and it was originally proposed to “gauge the affinity for populism in the United States” (Hawkins et al. 2012, pg. 1). Populist attitudes can be viewed as an attitudinal syndrome, comprised of people-centric, anti-elitist, and Manichean attitudes (Hawkins et al. 2012; Mudde 2017; Wuttke et al. 2020). It is the mass public counterpart of populism in leaders’ rhetoric. The first recent, large-scale representative sample measuring voters’ populist attitudes was collected in the Cooperative Congressional Election Survey in the US in 2008 (Hawkins et al. 2012). Various studies show that populist attitudes can be distinguished empirically and theoretically from other concepts, such as political trust and external political efficacy (Geurkink et al., 2020) and that those attitudes are widespread among the public in many different countries (Akkerman et al. 2014; Hawkins and Littvay 2019). Scholars have argued that populist attitudes have an independent effect on public support for populist leaders. Recent studies include evidence from European countries (Akkerman et al. 2017; Geurkink et al., 2020; Loew and Faas 2019; Marcos-Marne et al., 2020), the US (Hawkins and Littvay 2019), and other democratic nations around the globe (Hieda et al. 2021). Some of these studies argue that populist attitudes moderate the effect of issue positions on voting behavior and attract voters to populist leaders even if the former disagree with some policy positions of the latter. That is, “some voters who are further removed from the issue positions of populist parties may still be attracted by these parties if they have strong populist attitudes”(Van Hauwaert and Van Kessel 2018, p.86).

According to that literature, populist attitudes are activated in contexts of failure of democratic governance that is attributed to a collusion of the (self-serving) governing elite (Hawkins et al. 2020). This includes cases of economic crises and government corruption (Hawkins et al. 2020). Broadly speaking, this perspective says that people support populist leaders because people are themselves—once their populist attitudes were activated by the context—populists, and that populist leaders succeed electorally because they explore these populist predispositions among the public by adopting a populist rhetoric.

One limitation of these arguments is that they do not consider some important aspects of the formation of public attitudes emphasized by the political behavior literature, including party identification. As Campbell et al. (1960, p.133) stated regarding the US electorate, “[i]dentification with a party raises a perceptual screen through which the individual tends to see what is favorable to his partisan orientation.” Especially in the US context, it has been demonstrated that party identification affects people’s attitudes in different domains, including attitudes toward a variety of policy issues (Barber and Pope 2019). As Bartels (2002) points out, “partisan bias in political perceptions plays a crucial role in perpetuating and reinforcing sharp differences in opinion between Democrats and Republicans.”

Based on the partisan bias literature, I propose an alternative explanation for the origins of the public endorsement of populist ideas, and I consider two sets of public attitudes. The first is the public support for populist messages. The argument is that people, particularly in the US, tend to agree with populist ideas because the leader they like—or the party they identify with—adopts a populist rhetoric. More precisely, support for populist messages are an effect of people’s support for a party or political leaders, not the cause. This argument leads to the following partisan cue hypotheses:

H₁: Voters who self-identify as Republicans (Democrats) are more likely to agree with populist or anti-populist messages when those messages comes from a Republican (Democrat) candidate.

H₂: That effect is stronger among strong partisans.

The core argument is that voters support or oppose certain ideas in the direction of their party identification, and based on a partisan cue voters receive about the support of party leaders for those ideas (Barber and Pope 2019; Bartels 2002). The argument predicts that a Republican (Democrat) voter opposes a populist or anti-populist message if it comes from a Democrat (Republican) leader, but supports that same message if informed that it is supported by a Republican (Democrat) leader.

We can hypothesize that such a partisan bias can affect not only support for specific (anti-)populist messages but also voters’ populist attitudes more generally. Hence, I also evaluate the hypotheses above using populist attitudes as the dependent variable instead of populist messages. That is, partisanship leads to anti-populist or populist attitudes if the candidate of the party that the voters favor endorses anti-populist or populist ideas. This argument leads to two hypotheses:

H₃: Voters exhibit either populist or anti-populist attitudes to agree with the position of the party they identify with.

H₄: Voters exhibit either populist or anti-populist attitudes to disagree with the position of the party they oppose.

Note that this is essentially different from the previous literature argument, which states that populist attitudes are activated by actual contextual features, such as representation (Mair 2002; Kriesi 2014), economic, or corruption crises (Hawkins et al. 2020). Alternatively, the argument here is that public populist attitudes can emerge even if the economy is on the rise, corruption is not salient, or crises of representation of traditional parties are absent, which was the case in Trump’s 2016 election. I argue that people who support a candidate for reasons that are not primarily the candidate’s populist rhetoric, such as the candidate’s party affiliation, or their policy position on immigration or social conservatism, can be inclined to sympathize with the candidate’s populist rhetoric as a result.

If these hypotheses are correct, they have important implications for the politics of populism. They call into question whether populist attitudes are a relevant factor behind the recent electoral success of populist leaders (Hawkins et al. 2020; Van Hauwaert et al.,2018). If there is evidence that populist attitudes are an effect of support for populist leaders, and the latter is explained by voters’ ideology preferences or party identification, then it raises questions about the relative importance of populist attitudes to explain arguably the most-important outcome that motivated the development of that concept, which is the recent electoral success of populist leaders. However, notice that the focus of this article is on the effect of partisanship, but even if there are strong effects of partisan cues, it doesn’t mean that populism has no effect. That is, this does not mean that populist sentiments, once activated by partisanship or other factors, can’t further increase populist leaders’ electoral support or that these mechanisms can’t operate simultaneously or in reverse order.

Research design

I test the hypotheses using a survey experiment that manipulates the type of message the respondents receive (populist versus anti-populist) and the cue about the supporter of the message. The populist and anti-populist messages used in the experiment are based on scholarly research about the content of populist leaders’ rhetoric in Europe and the US (Hawkins 2009; Hawkins and Littvay 2019; Hawkins et al., 2019). Respondents are randomly assigned to receive one of the following messages:

Populist message

The people know what is best for this country. The government should do whatever it takes to be the voice of the people, and remove the corrupted elite that dominates our government.

“Anti-populist” message

A good president should join forces with other representatives, listen to specialists, and make compromises to do what is best for this country, even when it goes against the will of the people.

In addition to the message, respondents are randomly assigned to receive a partisan cue. While one treatment group is told that Trump supports the message, the other is informed that Biden does so. The control group receives the message but no partisan cue about any message supporter. To test H₁ and H₂, the survey then asks if the respondent supports or opposes the idea in the message. The theoretical expectation is that Democrat voters will support either the populist or anti-populist message when they receive a cue that Biden supports the statement, but oppose the message when Trump supports the statement. In other words, the attitudes toward the message (populist or anti-populist) should move based on voters’ party identification and the partisan cue they receive, not the populist content of the message. Table 1 summarizes the treatment groups and the theoretical expectations. The full description of the question is in the Supplementary Material.

Table 1.

Treatment groups by random assignment of the partisan cue and (anti-)populist message alongside and theoretical expectations on the Democrat and Republican voter responses.

		Theoretical Expectations
		(relative to the control group)
Partisan cue	Message	Democrat voter	Republican Voter
Trump	Populist statement	Anti-populist attitudes	Populist attitudes
Trump	Anti-populist statement	Populist attitudes	Anti-populist attitudes
Biden	Populist statement	Populist attitudes	Anti-populist attitudes
Biden	Anti-populist statement	Anti-populist attitudes	Populist attitudes

The main dependent variable in this study is whether voters support or oppose the populist or anti-populist messages depending on the partisan cue they receive. Additionally, for the robustness check and comparison with previous studies that focus either on the causes of populist attitudes (Hawkins and Rovira Kaltwasser 2017) or their effect on populist leaders’ support (Hawkins et al. 2017; Geurkink et al., 2020; Akkerman et al. 2014; Hunger and Paxton 2021), the analyses are repeated using populist attitudes as the dependent variable.

The measure of populist attitudes follow the recommendations in Silva et al. (2018), Castanho Silva et al. (2020), Akkerman et al. (2014), and Wuttke et al. (2020). The subdimensions of the ideational concept of populism—people-centrism, anti-elitism, and a Manichean outlook—are measured using three questions each on a five-point Likert scale. The three questions measuring people-centrism are: “Politicians should always listen closely to the problems of the people” (ppl1); “Politicians don’t have to spend time among ordinary people to do a good job” (ppl2), and; “The will of the people should be the highest principle in this country’s politics” (ppl3). For anti-elitism, the statements are: “The government is pretty much run by a few big interests looking out for themselves” (ant1); “Government officials use their power to try to improve people’s lives,” (ant2), and; “Quite a few of the people running the government are crooked” (ant3). To measure respondents’ Manichean outlook, the survey states: “You can tell if a person is good or bad if you know their politics” (man1); “The people I disagree with politically are not evil” (man2), and; “The people I disagree with politically are misinformed” (man3). The second item of each group (ppl2, ant2, and man2) is reverse-coded. Studies show that the populist attitudes scale constructed out of these items is at least as good or better than alternative instruments (Castanho Silva et al., 2020). Details of the scale construction are in the supplementary material.

The survey measures respondents’ party identity following the ANES format. Respondents who identify with the Democrat or Republican party are asked if they feel strongly or not about their party identification in a follow-up question. Those who answered that they do not identify with any party (Independents) were asked right after if they feel closer to the Democrat party, the Republican party, or neither. A variable measuring the degree of party identification was created by attributing a value of −3 for strong Democrats, −2 for Democrats who do not feel strongly about their identification (weak Democrats), −1 for Independents who feel closer to the Democratic party, 0 for pure Independents who don’t feel closer to any party, 1 for Independents leaning toward the Republican Party, 2 for weak Republicans, and 3 for strong Republicans. The analyses use both the categorical variable of party identification and the ordinal measure of identification strength to test the hypotheses.

Finally, previous studies have demonstrated that gender, education, race, income, and age are associated with populist attitudes and support for populist parties (Golder 2003; Givens 2004; Arzheimer and Carter 2006; Arzheimer 2009; Lucassen and Lubbers 2012; Tsatsanis et al. 2018; Rico and Anduiza 2019; Erisen et al., 2021), so the survey measures those factors, as well.

A sample of 1712 US-born citizens was collected through Lucid between July and August 2022. The pre-analysis plan explains the decisions about sample size (available at https://osf.io/79w5k/?view_only=25f37005f7d7418c9c9695e25c53a6c6). The sample used quotas for education, age, income, race, and gender groups and matched the 2020 census proportions to those factors. The survey also includes a series of attention and factual manipulation checks (Kane and Barabas 2019) to ensure the quality of the responses. The details of the demographics in the sample, a comparison between sample and population proportions in each demographic group, as well as information about compliance with principles of ethical research, question wording, and the attention checks used in the survey, can be found in the supplementary material.

The parameter of interest is the average causal effect (ACE) of the partisan cue on the support of the messages among partisans. It is captured by the difference in the proportion of support for the populist or anti-populist message between groups that do or do not receive a partisan cue, for each subgroup of partisans and types of messages. For instance, we can compute the difference in the proportion of support for the populist message (p) among Democrat voters (d) who receive a cue that Biden (b) supports that message, against those Democrats who received the same message but didn’t receive any cue (n) about the message supporter. Formally, let D denote a Democrat voter and P a populist message:

τ_{b, d, p} = E [Y_{i} (b) - Y_{i} (n) ∣ Pty . Id . = D, Message = P]

It gives eight causal parameters by varying the cue (Trump versus Biden), party identification (Democrat versus Republican), and the type of message (populist versus anti-populist).

A non-parametric estimation of the causal parameters and hypotheses test of null effect can be achieved using a logistic regression of the outcome on the treatment condition using no partisan cue (control group) as the reference category, for each subset of the data by party identification and type of message separately. It is well-known that using pre-treatment adjustment variables increases the precision of estimates when the pre-treatment adjustment variables are related to the outcome (Imbens and Rubin 2015). Hence, in addition to the estimation of the causal effect without including pre-treatment covariates, I repeated the tests above using regression models for each subgroup after including education, income, age, gender, and race as adjustment variables.

Empirical analysis

The full regression tables for all results presented in this section are in the supplementary material. Figure 1 shows the proportion of support (y-axis) for the populist (right panel) and anti-populist (left panel) messages by treatment group (x-axis) among Democrat and Republican voters. The sample size in each group is presented above each bar, and the confidence intervals are presented at the top of the bars. Consider first the left panel of Figure 1, which shows the public support for the anti-populist message, “A good president should join forces with other representatives, listen to specialists, and make compromises to do what is best for this country, even when it goes against the will of the people.” Around 55% of voters in the control group support that message, regardless of their party identification. That support changes drastically when a partisan cue is provided alongside the message. More than 78% of the Democrats support the message when they are told that Biden supports it, but less than 32% do so when they are told that Trump supports the message, which represents a 46 percentage points reduction in support of the anti-populist idea. The reaction of Republican voters to the partisan cue is analogous. Only 20% on average supports the anti-populist position when informed that Biden supports it, but 73% do so when they are told that Trump supports that position. The same pattern emerges with the populist message (right panel of Figure 1).

Figure 1.

Proportion of support (y-axis) for a populist (right panel) and anti-populist (left panel) message by treatment group (x-axis) among Democrats and Republicans. The 95% confidence intervals are represented by the upper end of the bars.

Let us turn to the first hypothesis. The test is presented in Figure 2. The points in the figure show the estimates of the ACE, and the bars indicate their 95% confidence interval. They represent the difference in the proportion of supporters for the populist (right panel) and anti-populist (left panel) messages for each treatment group (x-axis) among Democrat and Republican voters. The numbers represent estimates in log odds obtained from four logistic regressions of support for the message on the partisan cue for each subset of voters (Democrats or Republicans) and the message they received (populist or anti-populist). The control group in each case (no partisan cue) was used as the baseline to compute the effects. Positive (negative) values mean that the proportion of supporters for the message is larger (smaller) in the treatment than in the control group.

Figure 2.

Point estimates and 95% confidence intervals of the average causal effect of partisan cues (x-axis) on support for a populist (right panel) and anti-populist (left panel) message among Democrat and Republican voters. The dependent variable is the average voter’s support for the message. Estimates represent logistic regression coefficients. The numbers are in log odds, and stars indicate the p-value levels.

Figure 2 shows a strong evidence supporting the first hypothesis. Democrat voters are 2.7 times more likely to agree with either a populist or anti-populist message if they know that Biden supports that position than Democrat voters who don’t receive the Biden cue. The opposite happens if these voters receive a cue that Trump supports the message. Changing the partisan cue and informing Democrat voters that Trump supports the message causes a 78% reduction in the odds of supporting the populist message and a reduction of 65% in the odds of supporting the anti-populist position. Results for Republican voters are analogous in terms of direction and magnitude. Republican voters who receive a cue that Trump (Biden) supports the message are more (less) likely to support either the populist or the anti-populist message than Republican voters who don’t receive the partisan cue. The effect is very large and significant in all cases, except for the populist message when Republicans are informed that Trump supports it. Maybe surprisingly, there is no difference between the control (no partisan cue) and treatment group in that case. One possible reason is that there is already large support for the populist message among Republicans in the control group, probably either as an effect of pre-exposure to Trump’s rhetoric or a higher predisposition of Republican voters to support populist positions. Figure 3 supports this interpretation. It shows a slightly increasing trend in the chances of supporting the populist message in the control group (no partisan cue) as we move from strong Democrats to strong Republicans. In any case, there is a 78% decrease among Republicans in the odds of supporting a populist message if these voters are informed that Biden supports it.

Figure 3.

Predicted probability of support for an anti-populist (left panel) or populist (right panel) message as a function of the strength of party identification (x-axis) for different treatment groups.

Now let us consider the second hypothesis. Figure 3 shows the predicted probability of supporting the anti-populist (left panel) or populist (right panel) message as a function of the strength of party identification (x-axis) for different treatment groups (color code). Again, the partisan cue effect is clear. A strong Democrat voter has more than 88% chance to support a populist message when a partisan cue is provided saying that Biden supports that message. That support drops to 40% when the partisan cue says that Trump supports the message instead of Biden. The effect for Republican voters who receive a Trump (or Biden) partisan cue is analogous to the effect of Democrats who receive a Biden (or Trump) partisan cue: Voters substantially support (oppose) the message, either the populist or the anti-populist one, if it comes from the party those voters support (oppose). The effect of the partisan cues among Democrats and Republicans works as an inverse mirror of each other. The stronger the voters’ party identification, the stronger the effect.

Finally, consider Figure 4. It shows the average causal effect of partisan cues (x-axis) among Democrat and Republican voters on voters’ populist attitudes. It provides a test for hypotheses H₃ and H₄. Generally, values in the populist attitudes scale were not affected by the partisan cue, and there was only one significant result at 0.05 significance level. One possible interpretation of these results is that populist attitudes are not easily manipulated by party endorsement of populist ideas, even though voters’ support for (anti-)populist messages is. Another possibility is a limitation of the research design. Populist attitudes are captured by a scale constructed from many subitems. The connection between the partisan cue and each subitem of the scale may not be as clear for respondents as the connection between the cue and the (anti-)populist messages. Future experiments can investigate this possibility by permuting the populist scale subdimensions and the partisan cues. Nonetheless, supporting H₄, Republican voters display higher levels of populist attitudes when they receive a cue that Biden supports the anti-populist message. That is, explicit anti-populist positions among Democrat leaders can strengthen populist attitudes among Republican voters.

Figure 4.

Point estimates and 95% confidence intervals of the average causal effect of partisan cues (x-axis) about support for a populist (right panel) or anti-populist (left panel) message among Democrat and Republican voters. The dependent variable is voters’ populist attitudes. Estimates represent linear regression coefficients. The stars indicate p-value levels.

Conclusion

This article investigates the effect of partisan cues on support for populist and anti-populist messages among Democrat and Republican voters. Voters were randomly assigned to receive a cue that either Biden or Trump supports the message. The control group received the same messages but didn’t receive any partisan cue. The article demonstrated that voters’ support for or opposition to populist or anti-populist messages are strongly affected by voters’ party identification and the partisan cue they receive. Democrat voters are much more likely to support (oppose) the message, either the populist or anti-populist one, if they receive a cue that Biden (Trump) supports that message. Likewise, Republican voters are more likely to support (oppose) the message when informed that Trump (Biden) supports it.

The results presented here have important implications for our understanding of the role of modern populism in electoral politics. Populist leaders gained significant electoral grounds in various democracies around the globe, and the scholarly literature has refined the concept of populism to better account for that phenomenon. One of the main contributions in that direction defines populism as a “thin-centered” ideology whose content depicts populist politicians as anti-establishment and anti-elitist leaders who self-proclaim to embody the will of a loosely defined “good people,” and promise to exercise power in the name of that will, even if it means disregarding institutionally-mediated democratic procedures (Mudde and Rovira Kaltwasser, 2017; Mansbridge and Macedo 2019). These leaders are often defined in contrast to representatives who accept the pluralistic nature of modern politics and are committed to accountability, making compromises, and to a decision-making process mediated by institutions and parties (Caramani 2017). Given the electoral success of populist leaders, the literature hypothesized that voters support these leaders because voters support the populist ideas (anti-elitism, people-centrism) that those leaders propagate. This paper shows strong pieces of evidence that the causal arrow runs in the opposite direction. Voters are more likely to support (oppose) either a populist or anti-populist message whenever they learn that such a message is supported by the party they prefer (oppose).

Additionally, this paper investigates the effect of the partisan cue on voters’ populist attitudes. In this case, however, the effect of the partisan cue was mostly null. Future studies can explore why partisan cues affect support for (anti-)populist messages but, apparently, do not affect populist attitudes. It is also important to note the results in this article do not exclude the alternative causal direction, that is, the possibility that voters’ populist attitudes affect their support for populist leaders, or that both mechanisms act simultaneously. It is possible that in contexts of weak party identification and strong economic or political crises, populist ideas affect party identification and electoral behavior. Research from a comparative perspective can investigate that possibility, following a path already initiated by other scholars (Hawkins and Littvay 2019).

In any case, the results in this article raise the question of whether and when “thin-centered” populist ideas are just “too thin” to move elections and compete against more deeply-seated motivations of voters, often expressed as long-standing party identification or preferences for certain “thick” ideologies (e.g., social conservatism) or strong policy positions (e.g., anti-immigration). It is plausible that in other contexts, voters’ “thick” ideology positions, such as their social conservatism, play a role analogous to the partisan cue and party identification discussed here. For instance, voters may support a leader because of the leader’s strong anti-immigrant position and may be inclined to agree with (anti-) populist ideas if the leader adopts these ideas in her rhetoric alongside her anti-immigration position. If so, like the argument discussed in this article, voters’ support for populist messages may be, in part, an effect rather than the cause of populists’ electoral support.

Supplemental Material

Supplemental Material - The effect of party identification and party cues on populist attitudes

Supplemental Material for The effect of party identification and party cues on populist attitudes by Diogo Ferrari in Research & Politics

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Correction (June 2025):

The article has been updated with correct dataverse link in the supplementary material section. For more details, please see the correction notice: .

ORCID iD

Diogo Ferrari

Supplemental Material

Supplemental material for this article is available online.

The files can be found at https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/8UPTII&version=DRAFT

References

Akkerman

Mudde

Zaslove

(2014) How populist are the people? Measuring populist attitudes in voters. Comparative Political Studies 47(9): 1324–1353.

Akkerman

Zaslove

Spruyt

(2017) ‘We the people’or ‘we the peoples’? A comparison of support for the populist radical right and populist radical left in the Netherlands. Swiss Political Science Review 23(4): 377–403.

Albertazzi

McDonnell

(2007) Twenty-first Century Populism: The Spectre of Western European Democracy. Springer.

Arzheimer

(2009) Contextual factors and the extreme right vote in Western Europe, 1980-2002. American Journal of Political Science 53(2): 259–275.

Arzheimer

Carter.

(2006) Political opportunity structures and right-wing extremist party success. European Journal of Political Research 45(3): 419–443.

Barber

Pope

(2019) Does party trump ideology? Disentangling party and ideology in America. American Political Science Review 113: 1–17.

Bartels

(2002) Beyond the running tally: Partisan bias in political perceptions. Political Behavior 24(2): 117–150.

Bonikowski

Gidron

(2016) The populist style in American politics: Presidential campaign discourse, 1952-1996. Social Forces 94(4): 1593–1621.

Campbell

Converse

Miller

, et al. (1960) The American Voter. Chicago: University of Chicago Press.

10.

Caramani

(2017) Will vs. reason: The populist and technocratic forms of political representation and their critique to party government. American Political Science Review 111(1): 54–67.

11.

Castanho Silva

Jungkunz

Helbling

et al. (2020) An empirical comparison of seven populist attitudes scales. Political Research Quarterly 73(2): 409–424.

12.

Erisen

Guidi

Martini

, et al. (2021) Psychological correlates of Populist attitudes. Political Psychology 42: 149–171.

13.

Marcos-Marne

Plaza-Colodro

Freyburg

(2020) Who votes for new parties? Economic voting, political ideology and populist attitudesMudde, Cas. 2017.” Populism: an ideational approach. In: The Oxford Handbook of PopulismCristobal Rovira Kaltwasser, Paul A. Taggart, Paulina Ochoa Espejo, and Pierre Ostiguy. West European PoliticsOxford University Press, Vol 43, pp. 1–21.

14.

Geurkink

Zaslove

Sluiter

, et al. (2020) Populist attitudes, political trust, and external political efficacy: Old wine in new bottles? Political Studies 68(1): 247–267.

15.

Givens

(2004) The radical right gender gap. Comparative Political Studies 37(1): 30–54.

16.

Golder

(2003) Explaining variation in the success of extreme right parties in Western Europe. Comparative Political Studies 36(4): 432–466.

17.

Hawkins

(2009) Is Chávez populist? Measuring populist discourse in comparative perspective. Comparative Political Studies 42(8): 1040–1067.

18.

Hawkins

Rovira Kaltwasser

(2017) The ideational approach to populism. Latin American Research Review 52(4): 513–528.

19.

Hawkins

Littvay

(2019) Contemporary US Populism in Comparative Perspective. Cambridge: Cambridge University Press.

20.

Hawkins

Scott

Mudde

. 2012. “Measuring populist attitudes.”

21.

Hawkins

Read

Pauwels

(2017) “Populism and its causes.” In: Oxford Handbook of populismCristóbal Rovira Kaltwasser, Paul A Taggart, Paulina Ochoa Espejo, and Pierre Ostiguy. Oxford, UK: Oxford University Press, pp. 267–286.

22.

Hawkins

Aguilar

BrunoSilva

Cristóbal Rovira Kaltwasser , et al. (2019) Measuring Populist discourse: The global Populism database. In: Presentado en el 2019 en EPSA Annual Conference en Belfast. Reino UnidoJune.

23.

Hawkins

Rovira Kaltwasser

Andreadis

(2020) The activation of populist attitudes. Government and Opposition 55(2): 283–307.

24.

Hieda

Zenkyo

Nishikawa

(2021) Do populists support populism? An examination through an online survey following the 2017 Tokyo Metropolitan Assembly election. Party Politics 27(2): 317–328.

25.

Hunger

Paxton

(2021) What’s in a buzzword? A systematic review of the state of populism research in political science. Political Science Research and Methods 10: 1–17.

26.

Imbens Guido

Rubin

(2015) Causal Inference in Statistics, Social, and Biomedical Sciences: An Introduction. Cambridge: Cambridge University Press.

27.

Ivaldi

Lanzone

Woods

(2017) Varieties of Populism across a left- right spectrum: The case of the front national, the northern league, podemos and five star movement. Swiss Political Science Review 23(4): 354–376.

28.

Kane

Barabas

(2019) No harm in checking: using factual manipulation checks to assess attentiveness in experiments. American Journal of Political Science 63(1): 234–249.

29.

Kriesi

(2014) The populist challenge. West European Politics 37(2): 361–378.

30.

Legorano

Mesco

(2017) “A Populist Storm Stirs in Italy.” The wall Streeet Journal (may 12, 2017). https://www.wsj.com/articles/a-populiststorm-stirs-in-italy-1494596762. (Accessed 27 January 2022).

31.

Loew

Faas

(2019) Between thin-and host-ideologies: how populist attitudes interact with policy preferences in shaping voting behaviour. Representation 55(4): 493–511.

32.

Lucassen

Lubbers

(2012) Who fears what? Explaining far-right-wing preference in Europe by distinguishing perceived cultural and economic ethnic threats. Comparative Political Studies 45(5): 547–574.

33.

Mair

(2002) Populist democracy vs Party democracy. In: Democracies and the Populist Challenge. London, England, UK: Palgrave Macmillan, pp. 81–98. (accessed 19 February 2022)

34.

Mansbridge

Macedo

(2019) Populism and democratic theory. Annual Review of Law and Social Science 15: 59–77.

35.

March

(2017) Left and right populism compared: The British case. The British Journal of Politics and International Relations 19(2): 282–303.

36.

Marcos-Marne

(2020) A tale of populism? The determinants of voting for left-wing populist parties in Spain. Political Studies 69(4): 1053–1071.

37.

Mudde

(2017) Populism: An ideational approach. In: CR

Kaltwasser

Taggart

Espejo

, et al. (eds) Chap. 2 in The Oxford Handbook of Populism. Oxford: Oxford University Press.

38.

Mudde

Rovira Kaltwasser

(2017) Populism: A Very Short Introduction. Oxford: Oxford University Press.

39.

Müller

(2004) The parliamentary election in Austria. Electoral Studies 23(2): 346–353. https://www.sciencedirect.com/science/article/pii/S0261379403000817

40.

Norris

Inglehart

(2019) Cultural Backlash: Trump, Brexit, and Authoritarian Populism. Cambridge: Cambridge University Press.

41.

Oliver

Rahn

(2016) Rise of the trumpenvolk: Populism in the 2016 election. The ANNALS of the American Academy of Political and Social Science 667(1): 189–6206.

42.

Ostiguy

(2017) Populism: a socio-cultural approach. In: The Oxford Handbook of PopulismCristobal Rovira Kaltwasser, Paul A. Taggart, Paulina Ochoa Espejo, and Pierre Ostiguy. Cambridge: Oxford University Press.

43.

Rico

Anduiza

(2019) Economic correlates of populist attitudes: an analysis of nine European countries in the aftermath of the great recession. Acta Politica 54(3): 371–397.

44.

Rodrik

(2018) Populism and the economics of globalization. Journal of international business policy 1(1–2): 12–33.

45.

Rooduijn

Van Kessel

Froio

, et al. (2019) The PopuList: An Overview of Populist, Far Right, Far Left and Eurosceptic Parties in Europe. Www.popu-list.org

46.

Silva

Andreadis

Anduiza

, et al. (2018) Public opinion surveys: a new scale. In: Hawkins

Carlin

(eds), The Ideational Approach to Populism. Routledge, pp. 150–177.Levente Littvay, and Cristóbal Rovira Kaltwasser.

47.

Silva

Bruno Jungkunz

, et al. (2020) An empirical comparison of seven populist attitudes scales. Political Research Quarterly 73(2): 409–424.

48.

Tsatsanis

Andreadis

Teperoglou

(2018) Populism from below: socioeconomic and ideological correlates of mass attitudes in Greece. South European Society and Politics 23(4): 429–450.

49.

Van Hauwaert Steven

Van Kessel

(2018) Beyond protest and discontent: a cross-national analysis of the effect of populist attitudes and issue positions on populist party support. European Journal of Political Research 57(1): 68–92.

50.

Weisskircher

M“es

(2020) The strength of far-right AfD in eastern Germany: The east-west divide and the multiple causes behind ”Populism. The Political Quarterly 91(3): 614–622.

51.

Weyland

(2021) Populism as a Political strategy: an approach’s enduring– and increasing– advantages. Political Studies 69(2): 185–189.

52.

Wuttke

Schimpf

Schoen

(2020) When the whole is greater than the sum of its parts: on the conceptualization and measurement of populist attitudes and other multidimensional constructs. American Political Science Review 114(2): 356–374.

53.

Zulianello

(2020) Varieties of populist parties and party systems in Europe: from state-of-the-art to the application of a novel classification scheme to 66 parties in 33 countries. Government and Opposition 55(2): 327–347.

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.

0.00 MB

0.49 MB