Hostname: page-component-586b7cd67f-l7hp2 Total loading time: 0 Render date: 2024-11-21T18:55:16.991Z Has data issue: false hasContentIssue false

The tension between democracy and populism: an empirical test of Canovan's claims of the two faces of democracy

Published online by Cambridge University Press:  03 October 2024

Francesco Marolla*
Affiliation:
Department of Social and Political Sciences, University of Milan, Milan, Italy
Peter Achterberg
Affiliation:
Department of Sociology, Tilburg School of Behavioral Sciences, Tilburg University, Tilburg, Netherlands
Ruud Luijkx
Affiliation:
Department of Sociology, Tilburg School of Behavioral Sciences, Tilburg University, Tilburg, Netherlands Department of Sociology and Social Research, University of Trento, Trento, Italy
Mario Quaranta
Affiliation:
Department of Sociology and Social Research, University of Trento, Trento, Italy
*
Corresponding author: Francesco Marolla; Email: [email protected]

Abstract

This study provides a first empirical test of Margaret Canovan's influential argument on the relationship between democracy and populism, which posits that populism emerges as a consequence of the unresolved conflict between the pragmatic and redemptive faces of democracy. Despite its impact on scholars of populism, the implications of her framework remain untested. Using data from the EVS/WVS Integrated Values Surveys, we test Canovan's claims about the effect of ‘pragmatic politics’ on support for populist parties, operationalized as consensual democracies, economic and political globalization, and checks on government. Our analyses predominantly reveal no significant effects, and where significant results are observed, they indicate directions contrary to Canovan's claims, thus providing no empirical support for her claims. These results challenge long-standing assumptions about the relationship between populism and democracy, urging scholars to reevaluate existing assumptions and explore this intricate connection further. We conclude by suggesting some directions for future research to deepen our understanding of populism.

Type
Research Note: Null Finding
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
Copyright © The Author(s), 2024. Published by Cambridge University Press on behalf of Società Italiana di Scienza Politica

Introduction

Many scholars have grappled with the complex and often contentious relationship between populism and democracy. Margaret Canovan's article on populism (Reference Canovan1999) has provided scholarship with a striking portrayal of this intricate relationship. Canovan's argument delves into the tension between two contradicting faces of democracy: pragmatic and redemptive. The pragmatic face prioritizes peace, stability, and moderation, aligning with principles advocated by mainstream politics; while the redemptive face champions popular sovereignty and direct democracy, often supported by populist parties. Both features are vital, but an extreme imbalance in either carries risks. Overemphasizing pragmatism can lead to an unresponsive democracy bound by rules and practices, while excessive redemptive politics yields volatile and crude majoritarianism vulnerable to popular whims. Thus, the potential for populist mobilization lies in the unresolved tension between these two contrasting, yet necessary, democratic features. Canovan's argument has enduring relevance in the contemporary political climate as several studies have shown (e.g., Akkerman, Reference Akkerman, Heinisch, Holtz-Bacha and Mazzoleni2021; Pirro and Stanley, Reference Pirro and Stanley2022).

Canovan's work has significantly influenced the scholarly debate.Footnote 1 On one side, scholars have raised concerns about populism's potential threat to democracy, citing its opposition to liberal democratic principles, crude majoritarian tendencies, and disregard for checks on governmental power (e.g., Urbinati, Reference Urbinati1998; Abts and Rummens, Reference Abts and Rummens2007). Conversely, other scholars argue that populism can serve as a corrective force within liberal democracies addressing representational gaps (e.g., Mudde and Kaltwasser, Reference Mudde and Kaltwasser2013; Kaltwasser, Reference Kaltwasser2014). However, while this debate has provided valuable insights, this scholarly exchange has been predominantly theoretical. The lack of empirical studies prevents from understanding the role played by democratic institutions in shaping populist parties' opportunities to garner support.

In this research note, we aim to bridge this critical gap in the literature by providing a first empirical test of Canovan's claims about the link between populism and democracy. Following the Canovan's central idea, we investigate whether a greater prevalence of pragmatic political features correlates with increased support for populist parties. Specifically, we aim to examine if an imbalance in features associated with pragmatic politics may create opportunities for populist mobilization. Using cross-sectional data spanning over two decades (1996–2018) of the European Values Study/World Values Survey (EVS/WVS) and contextual data, we scrutinize Canovan's proposed mechanisms (Reference Canovan1999: 10–14) and propose an operationalization of pragmatic politics. This study does not represent an empirical test to a general theory of populism, as Canovan herself recognized the challenges of formulating such a theory (Canovan, Reference Canovan1999: 3). Rather, our study aims to empirically test the assumptions she proposed for understanding the emergence of populism as a result of contradicting forces within democracy.

The results show no empirical evidence supporting Canovan's claims in none of the aspects investigated. The null results are confirmed also by a series of sensitivity analyses exploring alternative specifications within our research design. If anything, the few significant results found point in the opposite direction of what was predicted by Canovan. This divergence between and empirical evidence calls for a reevaluation of the practical implications of current theories of populism and democracy. While Canovan's work has played a pivotal role in encouraging reflections on populism and democracy, our research highlights the complexity of this relationship and underscores the need for further examination and refinement of existing theories.

Theoretical framework and hypotheses

Populism against pluralism

The first democratic tension discussed by Canovan concerns democracy as the ‘government of the people’ and democracy as ‘a way of coping peacefully with conflicting interests and views’ (Canovan, Reference Canovan1999: 10). In this context, ‘redemptive’ politics represents the vision of a better world through direct action by the sovereign people. Conversely, ‘pragmatic’ politics recognizes the necessity of institutions to manage diverse interests in a society. This contrast highlights populists' impatience with pluralistic democracy, which views representative institutions with skepticism. Their notion of a government by the people as a homogeneous group clashes with the pragmatic view of multiple interests vying for political power. Thus, pluralistic democratic institutions provide an opportunity for populists to promote their redemptive promise of democracy and mobilize supporters.

Based on this argument, we use Lijphart's (Reference Lijphart1999) Executive-parties dimension index provided by the Comparative Political Dataset (Armingeon et al., Reference Armingeon, Engler and Leemann2022) to operationalize pluralism. A higher score on this index indicates political systems closer to consensual democracies, which are systems characterized by political representation of multiple interests and sharing of government power. Canovan argues that, from a pragmatic point of view, democracy is a ‘collection of institutions and practices […] that have found ways of making power relatively accountable, widening the range of interests incorporated into the political arena and binding more of the population into the political system. […] democracy means multi-party systems, free elections, pressure groups, lobbying and the rest of the elaborate battery of institutions and practices by which we distinguish democratic from other modern polities.’ (Canovan, Reference Canovan1999: 11). Therefore, we test this first Canovan's claim hypothesizing that the more a political system is consensual, the higher the likelihood of supporting a populist party (H1).

Populism against the constraints of democracy

The second tension regards the ‘power and the impotence of democracy’ (Canovan, Reference Canovan1999: 12). Redemptive politics assumes that democratic governments are able to fully implement the will of the people; but pragmatic politics acknowledges that governments often have limited ability to control events given the complexity of contemporary democracy. For instance, globalization can expose governments to uncontrollable economic crises, making democratic systems vulnerable to populist claims. Similarly, membership in supranational institutions or joining international treaties can limit elected governments' policy-making, frustrating citizens who contend that only democratically elected authorities should wield power. Therefore, such a democratic deficit opens another opportunity for the advance of populism with its redemptive promise to bring power back to the people.

Following this argument, we use the Economic Globalization Index (EGI) and the Political Globalization Index (PGI) (Gygli et al., Reference Gygli, Haelg, Potrafke and Sturm2019). We examine both economic and political globalization to assess their potential impact on support for populist parties, as they refer to different ways in which global dynamics can constrain national policy-making. This is in line with Canovan's argument that ‘even supposing that ‘we, the people’ can combine our diverse interests and opinions into a coherent collective will, the hard facts of political and economic interdependence often make that an empty promise’ (Canovan, Reference Canovan1999: 12). The EGI measures the flows of capital and goods of a country, as well as policies that facilitate and promote trade flows and foreign investments between countries. The PGI measures the amount of foreign influence in a country and the ability to engage in international political cooperation. The higher the score of these two indices, the higher the level of globalization. Therefore, we hypothesize that the more a country is politically (H2a) and economically (H2b) globalized, the higher the likelihood of supporting a populist party.

Populism against limitations to power

The third tension involves institutional and direct democracy (Canovan, Reference Canovan1999: 13). Pragmatic politics values institutions, rules, and practices to make democracy effective; redemptive politics rejects mediating institutions, advocating for unmediated popular will and sometimes charismatic leadership. While this system of institutions is designed to prevent abuses of power and maintain the stability and fairness of the democratic system, it can be perceived as complex and alienating by the public, potentially fostering frustration and a desire for more direct forms of governance. In this view, populists are impatient with institutions that constrain the majority's power, preferring a more direct democratic system. This tension highlights how democratic systems with checks and balances can create conditions for populist mobilization.

We use the Checks on Government index provided by the Global Democracy Dataset (Idea, Reference Idea2019) to capture the extent to which executives' power is scrutinized by independent institutions. Democracies with a high degree of checks and balances typically involve a complex web of institutions, which may trigger the tension between pragmatic and redemptive politics. This operationalization should be in line with Canovan's idea of democracy that ‘to work as a government, it has to take institutional forms that are very far removed from spontaneous popular expression’ (Canovan, Reference Canovan1999: 13), also meaning that democracy mediates the popular will through a set of institutions. The index includes several indicators related to three domains: (I) the capacity of the legislature to oversee the executive; (II) courts independence from the other branches of government, especially the executive; and (III) the plurality of media landscape. The higher the score on this index, the more executive power is put under scrutiny by other institutions. Thus, we hypothesize that the more in a political system there are checks to limit government power, the higher the likelihood of supporting a populist party (H3).

Data and methods

To test Canovan's claims, we employed data from the EVS/WVS Integrated Values Surveys (EVS, 2021; Haerpfer et al., Reference Haerpfer, Inglehart, Moreno, Welzel, Kizilova, Diez-Medrano, Lagos, Norris, Ponarin and Puranen2021), covering four waves of the WVS data (1994–1998, 2005–2009, 2010–2014, 2017–2019) and two waves of EVS data (2008–2010, 2017–2020). Our analysis encompasses 26 European countries and 59 surveys conducted between 1996 and 2018Footnote 2. Surveys were included in the sample based on the availability of voting intention for populist parties and key independent variables at individual and contextual levels. Contextual-level data are sourced from the Comparative Political Dataset, the KOF Globalization Index dataset, and the Global State of Democracy Dataset. Employing multilevel logistic regression models in a two-level structure, our analysis includes 72,652 individuals nested within 59 country-year groups representing the contextual level.

The dependent variable is support for a populist party, measured harmonizing two survey questions: (I) which party they would vote for if there were elections tomorrow; (II) which party appealed most to them. It is a dichotomous variable measuring whether respondents would support a populist party (1) or other parties (0).Footnote 3 Unfortunately, measurements of past voting in national elections were not available in the datasets used for this analysis. While the wording of these questions differs, both aim at capturing party preferences at the survey time rather than relying on their past voting choices, which is crucial for the purposes of this research. We adopt Canovan's (Reference Canovan1999: 3) definition of populism as ‘an appeal to ‘the people’ against both the established power structure and the prevailing societal ideas and values’, which aligns with the ideational definition (e.g., Mudde, Reference Mudde, Rovira Kaltwasser, Taggart, Ochoa Espejo and Ostiguy2017). Populist parties in Europe are classified following Rooduijn et al. (Reference Rooduijn, van Kessel, Froio, Pirro, De Lange, Halikiopoulou, Lewis, Mudde and Taggart2023), regardless of their host ideology.Footnote 4

The analysis includes various control variables at the individual and contextual levels. Individual-level controls encompass socio-demographic factors such as gender, age, education, and unemployment status; and sociopolitical attitudes such as political distrust, social trust, and authoritarianism. Contextual-level controls include macroeconomic indicators such as GDP growth, unemployment rate, social security transfers, and the share of employment in the industrial sector to account for competing contextual explanations. To account for long-term dynamics in our analysis, all contextual variables are lagged by one year. Table A3 in the supplementary material reports the descriptive statistics.

Results

As a preliminary analysis, we ran the multilevel models including only the main variables of interest as predictors of support for populist parties. Figure 1 reports the predicted probabilities of supporting a populist party along the range of the four variables under study. All appear not to be significantly associated with the outcome. These initial findings seem not providing support for Canovan's claims.

Figure 1. The association between the Lijphart's executive-parties dimension, political and economic globalization, and checks on government with the probability of supporting a populist party, with 95% confidence intervals.

Table 1 report the multilevel models including the control variables. Model 1 examines whether higher degrees of consensual democracy increase support for populism. However, the effect is not statistically significant, thus not supporting H1. The coefficient confirms the negative effect previously shown, meaning that consensual democracies are negatively associated with support for populist parties. Models 2 and 3 investigate whether higher levels of political and economic globalization are associated with greater support for populism. However, also in this case we notice that the variables do not significantly affect support for populist parties. Therefore, both these results do not show supporting evidence for H2a and H2b. Finally, Model 4 assesses whether political systems with more checks on government power increase support for populism. Higher checks on government power are associated with a lower likelihood of supporting populist parties, and this effect is statistically significant (OR 0.930, P < 0.01). Nonetheless, as the result contradicts Canovan's predictions, it fails to support H3.

Table 1. Multilevel logistic regression models predicting support for populist parties

Note: estimates are odds ratios; *p < 0.05; **p < 0.01; ***p < 0.001; aCentered variable; N = 72,652; Country-years = 59; all models include time and countries fixed effects; control variable are omitted from the table, Table A4 in the Appendix includes all the estimates.

Finally, we conducted various additional analyses to assess the robustness of our results (see Tables A4 through A12), which also confirmed no substantial differences from our primary findings.

Discussion

In this research note, we contributed to the literature on populism and democracy by empirically testing for the first time Canovan's assumptions of the two faces of democracy, which proposed that features linked to ‘pragmatic politics’ within democratic systems may increase support for populist parties seeking to counterbalance with ‘redemptive politics’. Studying this argument empirically can shed light on whether characteristics of democracies can provide mobilization opportunities for populist parties, which would imply that populism is an inevitable part of the democratic experience.

The findings of our study starkly contradict Canovan's claims, by showing no significant effects of consensual democracies (H1), political (H2a) and economic (H2b) globalization, and showing opposite evidence regarding the checks and balances (H3). Our findings suggest that the effect of pragmatic politics, as outlined in Canovan's arguments, may not be as straightforward as originally hypothesized. Namely, while an imbalance in favor of pragmatic politics might create opportunities for populist mobilization, the results indicate that such opportunities do not necessarily lead to increased support for populist parties. This calls for a reevaluation of explanations that aim to explain populist mobilization within democratic contexts.

Although mostly not statistically significant, our findings offer insights into various other theoretical perspective proposed to explain the rise of populism within democracies. In certain contexts, consensus democracy appears to mitigate representation crises and reduce populist emergence (e.g., Berman and Snegovaya, Reference Berman and Snegovaya2019). Mixed results from economic and political globalization suggest that these processes do not uniformly provoke populist support (Norris and Inglehart, Reference Norris and Inglehart2019). Additionally, our significant results on checks and balances suggest that stronger liberal democracy can actually discourage populism (Mudde, Reference Mudde2021). However, significant individual-level effects of authoritarian attitudes (Table A4) underscore a persistent preference among supporters of populist parties for governance with reduced institutional mediation.

Using a multilevel design covering a wide cross-national span enabled us to capture a wide range of democracies' characteristics and strengthen their linkage with individual-level behavior. Nevertheless, it is important to acknowledge that our results remain correlational and cannot establish causality. To delve deeper into the cause-and-effect relationship, a panel data design would be ideal. However, collecting such data poses significant challenges in terms of data collection in more than one context and modeling the effects of democratic changes on individuals' behavior.

Furthermore, this study has some other limitations that offer valuable directions for future research. First, our analysis primarily focused on Europe, but future studies should test Canovan's claims beyond this area. However, such an expansion presents additional challenges, as populism manifests differently in various regions (e.g., Latin America, North America, East Asia), requiring alternative analytical approaches for its definition (e.g., Weyland, Reference Weyland2001). In this sense, while Canovan's work primarily referred to Western democracies (Reference Canovan1999: 2, 4, 11, 14), our study tested her claims also on Eastern European democracies. Second, our study proposed an operationalization of Canovan's theoretical concepts for empirical examination. We acknowledge that our approach, focusing primarily on pragmatic politics, may not fully capture the interdependent and complementary dynamics between the pragmatic and redemptive facets of democracy described by Canovan. Future research could explore alternative measurements to provide different tests. Finally, our findings are limited to EVS/WVS data. Despite including a significant portion of contexts of populism as governing force, such dataset is majorly characterized by populism as opposition phenomenon. We do not exclude different findings if other cross-national datasets are used. The EVS/WVS data allowed our research design to exploit the temporal variation of structural democratic characteristics such as consensus democracy or checks and balances, in addition to the only geographical variation. Further investigations can provide additional tests to Canovan's claims by employing alternative data sources.

We suggest some avenues for future research to investigate the evolving dynamics of populism and its intricate connection with democracy. The null findings we observed for the contextual variables may mask potential heterogeneous effects within the country-year clusters. Future research can use the multilevel design to investigate interactions between pragmatic politics and individual attitudes. For instance, it is plausible that the imbalance towards pragmatic politics might displease only those citizens who endorse democratic values aligned with redemptive politics. Populist attitudes represent a promising avenue for such analysis (e.g., Akkerman et al., Reference Akkerman, Mudde and Zaslove2014), although currently, there is a scarcity of cross-national data large enough to facilitate reliable multilevel estimates in this sense.

Moreover, Canovan's arguments did not anticipate the implications of populism in power. Especially in Eastern Europe, we now observe populist actors rising to power more frequently. It should not be assumed that populist actors necessarily seek to strike a balance between pragmatic and redemptive politics once they reach power. Instead, they might aim for the supremacy of ideals associated with the latter at the expense of the former, as several experiences of populist governments have shown (e.g., Muller, Reference Muller, Rovira Kaltwasser, Taggart, Ochoa Espejo and Ostiguy2017; Urbinati, Reference Urbinati, Rovira Kaltwasser, Taggart, Ochoa Espejo and Ostiguy2017; Blokker, Reference Blokker and de la Torre2018). It is in fact in Eastern European contexts (Table A6) that we found the most pronounced opposite effect of consensual democracies on support for populist parties. One possible explanation for this is that once reaching power, populists might operate in limiting pluralistic representation through institutional reforms. Therefore, future theories about populism and democracy should not neglect this important aspect related to how populism has evolved during these decades.

In summary, while Canovan's reflections remain insightful to stimulate reflection regarding the inner contradictions of democracy, our results suggest that their relationship with populism is more intricate than previously thought. This underscores the need for further research that accounts for the evolving dynamics of populism within democratic systems for a better understanding of the complex relationship between populism and democracy.

Funding

This research received no specific grant from any public or private funding agency.

Data

The replication dataset is available at https://doi.org/10.7910/DVN/X3F5ZM

Supplementary Material

The supplementary material for this article can be found at https://doi.org/10.1017/ipo.2024.9

Acknowledgements

The authors wish to thank the journal editors, and the three anonymous reviewers for their valuable comments to an earlier draft of this paper.

Competing Interests

The authors declare none.

Footnotes

1 As of today (19th October 2023) the article by Canovan (Reference Canovan1999) has 3718 citations on Google Scholar.

2 Table A2 shows the list of countries and years included in the analysis.

3 We harmonized these two variables to measure voting intentions in order to include the largest number of EVS/WVS surveys possible in our analyses, as each variable is collected in different EVS or WVS wave.

4 See Table A1 in the Appendix for the list of populist parties.

References

Abts, K and Rummens, S (2007) Populism versus democracy. Political Studies 55, 405424.CrossRefGoogle Scholar
Akkerman, A (2021) Populist parties in power and their impact on liberal democracies in Western Europe. In Heinisch, R, Holtz-Bacha, C and Mazzoleni, O (eds), Political Populism: Handbook of Concepts, Questions and Strategies of Research, vol. 3. Baden-Baden: Nomos Verlag, pp. 227237.CrossRefGoogle Scholar
Akkerman, A, Mudde, C and Zaslove, A (2014) How populist are the people? Measuring populist attitudes in voters. Comparative Political Studies 47, 13241353.CrossRefGoogle Scholar
Armingeon, K, Engler, S and Leemann, L (2022) Comparative Political Data Set 1960–2020. Zurich: Department of Political Science, University of Zurich.Google Scholar
Berman, S and Snegovaya, M (2019) Populism and the decline of social democracy. Journal of Democracy 30, 5.CrossRefGoogle Scholar
Blokker, P (2018) Populist constitutionalism. In de la Torre, C (ed), Routledge Handbook of Global Populism. London: Routledge, pp. 113128.CrossRefGoogle Scholar
Canovan, M (1999) Trust the people! populism and the two faces of democracy. Political Studies 47, 216.CrossRefGoogle Scholar
EVS (2021) EVS trend file 1981–2017. GESIS Data Archive, Cologne. ZA7503 Data file Version 2.0.0.Google Scholar
Gygli, S, Haelg, F, Potrafke, N and Sturm, J-E (2019) The KOF globalisation index–revisited. The Review of International Organizations 14, 543574.CrossRefGoogle Scholar
Haerpfer, C, Inglehart, R, Moreno, A, Welzel, C, Kizilova, K, Diez-Medrano, J, Lagos, M, Norris, P, Ponarin, E, Puranen, B, et al. (2021) World Values Survey Time-Series (1981–2020). JD Systems Institute & WVSA Secretariat. Data File Version 2.0.0.Google Scholar
Idea, I (2019) The Global State of Democracy Indices Codebook. Stockholm: International IDEA.Google Scholar
Kaltwasser, CR (2014) The responses of populism to Dahl's democratic dilemmas. Political Studies 62, 470487.CrossRefGoogle Scholar
Lijphart, A (1999) Patterns of Democracy: Government Forms and Performance in Thirty-six Countries. New Haven: Yale University Press.Google Scholar
Mudde, C (2017) An ideational approach. In Rovira Kaltwasser, C, Taggart, PA, Ochoa Espejo, P and Ostiguy, P (eds), The Oxford Handbook of Populism. Oxford: Oxford University Press, pp. 2747.Google Scholar
Mudde, C (2021) Populism in Europe: an illiberal democratic response to undemocratic liberalism (The Government and Opposition/Leonard Schapiro Lecture 2019). Government and Opposition 56, 577597.CrossRefGoogle Scholar
Mudde, C and Kaltwasser, CR (2013) Exclusionary vs. inclusionary populism: comparing contemporary Europe and Latin America. Government and Opposition 48, 147174.CrossRefGoogle Scholar
Muller, J (2017) Populism and constitutionalism. In Rovira Kaltwasser, C, Taggart, PA, Ochoa Espejo, P and Ostiguy, P (eds), The Oxford Handbook of Populism. Oxford: Oxford University Press, pp. 590606.Google Scholar
Norris, P and Inglehart, R (2019) Cultural Backlash: Trump, Brexit, and Authoritarian Populism. Cambridge: Cambridge University Press.CrossRefGoogle Scholar
Pirro, AL and Stanley, B (2022) Forging, bending, and breaking: enacting the “illiberal playbook” in Hungary and Poland. Perspectives on Politics 20, 86101.CrossRefGoogle Scholar
Rooduijn, M, van Kessel, S, Froio, C, Pirro, A, De Lange, S, Halikiopoulou, D, Lewis, P, Mudde, C and Taggart, P (2023) The populist 3.0: An overview of populist, far-left and far-right parties in europe. Available at https://popu-list.org/Google Scholar
Urbinati, N (1998) Democracy and populism. Constellations (oxford, England) 5, 110124.Google Scholar
Urbinati, N (2017) Populism and the majority principle. In Rovira Kaltwasser, C, Taggart, PA, Ochoa Espejo, P and Ostiguy, P (eds), The Oxford Handbook of Populism. Oxford: Oxford University Press, pp. 571589.Google Scholar
Weyland, K (2001) Clarifying a contested concept: populism in the study of Latin American politics. Comparative Politics 34, 122.CrossRefGoogle Scholar
Figure 0

Figure 1. The association between the Lijphart's executive-parties dimension, political and economic globalization, and checks on government with the probability of supporting a populist party, with 95% confidence intervals.

Figure 1

Table 1. Multilevel logistic regression models predicting support for populist parties

Supplementary material: File

Marolla et al. supplementary material 1

Marolla et al. supplementary material
Download Marolla et al. supplementary material 1(File)
File 305.9 KB
Supplementary material: File

Marolla et al. supplementary material 2

Marolla et al. supplementary material
Download Marolla et al. supplementary material 2(File)
File 213.2 KB
Supplementary material: Link

Marolla et al. Dataset

Link