Abstract
The use of armed drones has emerged as a principal counterterrorism tool for western militaries, especially France and the United States. While France submits its strikes to the United Nations for approval, the United States typically does not. Does this difference matter for public support and perceptions of legitimacy? To better understand these dynamics, we fielded original survey experiments across nationally representative samples in France and the United States totaling in over 1800 respondents. Our results reflect that international approval is associated with both higher public support and greater perceived legitimacy for a strike. Further, we find that respondents emphasize international law as the basis for support and legitimacy, suggesting a cross-national belief in multilateralism for normative rather than strictly instrumental reasons. These relationships are moderated by the identity of the country conducting a hypothetical strike, implying both an “othering” effect and the emergence of distinct models of strikes across countries that deserve more study amid the ongoing proliferation of armed drones.
Video Abstract: https://www.youtube.com/watch?v=YevyaKThae0
Introduction
In August 2021, France used an armed Unmanned Aerial Vehicle (UAV), or drone, to kill the Islamic State’s leader in western Africa, Adnan al-Sahrawi. Months later, the United States used drones to kill two al-Qaeda leaders in Syria. Together, these strikes reflect a common practice of using drones for counterterrorism. Despite similarities, the strikes differed in one key attribute: France’s strike was conducted with United Nations (UN) approval while the United States conducted its pair of strikes unilaterally. Do these distinctions—multilateral versus unilateral—matter for the public’s support for and perceived legitimacy of strikes?
In this study, we research the association between multilateralism and the public’s perception of legitimacy as well as its association with support. We use original survey experiments fielded across nationally representative samples in France and the United States. Besides the United States, France has emerged as a prolific user of strikes, particularly in western Africa, making it a useful case to study. At the same time, French attitudes for drones provide a barometer for European preferences given the comparative decline in Britain’s strikes abroad, Germany’s hesitancy to arm drones, and Italy’s limited operations. In administering our surveys, we also clarify the mechanisms through which the public may connect international authorization with support and legitimacy. Specifically, we develop and test four hypotheses. These include propositions about merit, or the belief that obtaining international approval offers a “second opinion” about the wisdom of an operation (Grieco et al., 2011); the legal appeal of international authorization (Dill, 2019); the perceived morality of working through the UN; and the prospects for burden-sharing (Finnemore, 2003).
There is broad consensus among scholars that multilateral approval through the UN can enhance support and legitimacy for countries’ military interventions abroad (Tago & Ikeda, 2015; Tago, 2006). The burden-sharing advantages of multilateralism, for instance, seem to accrue most often in these situations, which are typically associated with the type of large-scale combat operations that took place in Afghanistan and Iraq. Whether these advantages coincide with countries’ use of drones—which are used in part because they appear less costly, both militarily and politically—is unclear. Whether the advantages of multilateralism apply cross-nationally in countries that have a strong connection with international institutions, as does France, is even less obvious (Recchia & Chu, 2021; Staunton, 2020). While we may think that the public’s preferences for multilateralism extend to drones, we do not know for sure, suggesting the need for an empirical study.
Public attitudes for multilaterally approved strikes deserve special attention for several reasons. First, the proliferation of drones amounts to a “second drone age” that has renewed the public debate for remote-warfare (Pollard, 2022). Second, public attitudes are important to understand because they are thought to influence elite preferences for the use of force abroad (Tomz & Weeks, 2013). Finally, though fighter jets and field artillery also afford combatants stand-off on the battlefield, drones have special qualities that may shape public attitudes in unique ways, even when their use is multilaterally approved. Unlike other indirect fire weapon systems, drones impose radically asymmetric violence because they erode reciprocal risk between combatants and largely remove targets’ right to self-defense (Renic, 2020).
The results of our survey experiment support our main hypothesis that UN approval is associated with higher support and legitimacy for countries’ use of strikes. In contrast, American and French respondents penalize strikes conducted by another country without UN approval, implying both an in-group and out-group as well as distinct models of strikes adopted by both countries. American and French respondents also emphasize the perceived compliance with international law the most when adjudicating support and legitimacy for strikes, though this finding is strongest among US subjects. Americans also emphasize burden-sharing in the case of internationally approved strikes conducted by another country. This outcome suggests a preference for coalition operations to offset the costs of conducting strikes (Kreps & Lushenko, 2021). This finding is consistent with the United States’ support for multilaterally authorized interventions conducted by its close allies, including France (Blankenship, 2021). On the other hand, French respondents do not support burden-sharing in terms of other countries’ strikes, which is consistent with earlier research for France’s interventions abroad, particularly in Africa (Recchia, 2020).
Our research makes several contributions. First, while scholars occasionally study public support for drone strikes, they predominantly draw on American respondents to investigate the United States’ use of strikes (Boddery & Klein, 2021; Horowitz, 2016). Whereas other countries have embraced drones, the research has not kept pace. Our research illuminates this trend while adding cross-national evidence on the use of force more generally. Second, we provide insights into one of the most consequential developments for global security in the 21st century: armed drones. Scholars have tackled questions of proliferation (Horowitz et al., 2022), effectiveness (Mir & Moore, 2019), and democratic accountability (Kaag & Kreps, 2014). Less studied is the global governance of drones and its effects on the sustainability of counterterrorism strikes. Finally, we contribute to an emerging literature for the political psychology of drone use, which corresponds to a renewed interest in psychological approaches to international relations broadly (Kertzer & Tingley, 2018). We bring this diverse set of literature into closer alignment by measuring the effect of multilateral approval on public attitudes for drones strikes, which we also do in a cross-national context for two countries that frequently use drones for counterterrorism, France and the United States.
Theoretical framework
The uses of drone strikes
Since 9/11, the United States has used drones for the targeted killing of terrorists (Meisels & Waldron, 2020). Whereas strikes in Afghanistan, Iraq, and Libya have been endorsed by international organizations, strikes in Pakistan, Somalia, and Yemen have not. The Obama administration conducted more strikes in these latter areas in its first year in office than the Bush administration during its entire tenure. Officials responded to criticisms by referencing the advantages of drones. In 2013, a White House spokesman argued “these strikes are legal, they are ethical, and they are wise” (Carney, 2013).
Critics have questioned the legality and legitimacy of these strikes, however, arguing that they often violate the sovereignty of targeted countries (Jaffer, 2016; Aslam, 2013). The drones themselves also introduce a moral hazard. By virtue of being unmanned, drones incur little cost to those using them and thereby may encourage more risk-taking and less care to protect civilians during strikes (Raman et al., 2021). Instances of botched strikes have not been uncommon, leading the US Secretary of Defense, Lloyd Austin, to admit in 2021 that the US “must work harder” to reduce civilian casualties during strikes (Myers, 2021).
Despite the critiques and missteps, drones continue to proliferate (Pollard, 2022). More than 100 countries have acquired drones, including France. Though France purchased drones from the United States in 2013, it was reluctant to arm them, conscious of the potential blowback, namely, public shaming for civilian casualties. Not until 2017 did France opt to arm its drones (Vilmer, 2021). Given the technical requirements, the retrofit process took 2 years. By 2019, France started using strikes against terrorists in western Africa in response to a formal request from the Malian government for security assistance. Since then, France has conducted several dozen strikes against al-Qaeda and Islamic State terrorists, all with the endorsement of UN Security Council resolutions (Brunstetter, 2021). The most visible strike killed al-Sahrawi, the mastermind behind the deaths of French and Nigerien aid workers as well as four US military personnel in 2017 (Maclean, 2020).
On at least one occasion, the United States has also used drones with UN approval. The Obama administration used strikes during the humanitarian intervention in Libya, authorized by the UN Security Council in March 2011. On balance, however, the United States’ legal authorization for strikes has rested on self-defense under the UN Charter’s Article 51 (Brooks, 2014). Yet the UN Security Council has not explicitly authorized US strikes in Pakistan, Somalia, and Yemen, and experts caution self-defense is “very elastic” and incompatible with the scope of Article 51 (Callamard, 2020). Indeed, the United States tends to use strikes on the basis of domestic authorizations, specifically Article II of the US constitution and the 2001 Authorization for the Use of Military Force, which allowed presidents to attack al-Qaeda and its affiliates following 9/11 (Swan, 2019).
Public attitudes for drone strikes
In the context of the use of force generally, scholars find that the public more favorably supports multilaterally authorized operations compared to those that are not (Busby et al., 2020; Tago & Ikeda, 2015). Many democratic leaders have acted as though multilateral approval is essential for using force abroad. They routinely cite the public’s preference for multilateralism as the basis for military actions abroad. Drezner (2003) also finds that UN approval allows the US Congress to wash its hands of responsibility for an operation that goes wrong.
Whether these relationships between multilateralism and public support hold in the context of drone strikes is unclear. Whereas military interventions are likely to incur considerable costs in blood and treasure, the allure of drones is that they suffer neither. Without bearing the risks of war, countries are less likely to scrutinize how and when they conduct strikes. Rather, strikes tend to assume an antiseptic and discrete quality in which multilateralism may not hold the weight it does for large-scale combat operations. Whether the public’s preferences for multilateralism extend to drones, and if so why, has largely escaped academic scrutiny. Here, we discuss mechanisms scholars have identified linking multilateralism to the use of force abroad. This discussion enables us to derive hypotheses about their potential application to drone strikes.
Merit: International approval indicates a shared belief in the anticipated benefits of force (Thompson, 2010; Fang, 2008; Voeten, 2005; Claude, 1966). Strikes, therefore, are less politically-motivated and constitute a defensively-oriented approach that promotes force as an appropriate last resort (Brunstetter, 2021). This indicates that the objectives of strikes have been vetted subject to a “second opinion” (Grieco et al., 2011). The consent and cooperation afforded by UN approval also suggests that strikes are better constrained and more compatible with social goals shared by all countries. That other, potentially more suspicious countries endorse strikes signals that they are designed to achieve noble objectives, such as preventing a humanitarian crisis or interdicting terrorists (Kreps and Lushenko, 2021; Recchia, 2020; Recchia and Chu, 2021).
Legality: Approval by the UN for countries’ use of force also suggests military action is likely to comply with international law. International law governing the use of force consists of two main components: jus ad bellum (just recourse to war) and jus in bello (just use of force in war) (Finkelstein et al., 2012). Drones may transgress jus ad bellum norms because they violate another country’s sovereign airspace. Drones may also flout jus in bello norms, including distinction and proportionality, in cases where poor intelligence or biases lead to botched strikes and civilian casualties (Traven, 2021). We follow recent research by treating countries’ legal commitments in terms of both respect for sovereignty and the protection of civilians while using strikes (Kreps & Wallace, 2016). The public is likely to believe that a UN-sanctioned intervention ensures more scrutiny and a higher duty of care among cooperating countries to protect civilians than strikes carried out by individual countries acting alone (Dill, 2019).
Morality: Previous research suggests that the public is concerned with the morality of countries’ use of force abroad (Tomz & Weeks, 2020). Dill and Schubiger (2021) show that the public seems to combine normative and instrumental concerns about right and wrong when adjudicating support for the use of force abroad. Drones are thought to exacerbate this tendency because they preserve the immunity of one side in a conflict by consolidating the liability to be harmed entirely within the other side (Renic, 2020). Indeed, Kreps and Wallace (2016) found that while international law governing the protection of civilians does mediate the public’s support for strikes, the effect appears to be for normative, rather than strictly instrumental, concerns.
Burden-Sharing: Scholars also argue that international approval signals the prospect for greater burden-sharing, which is likely to be associated with higher degrees of support and legitimacy (Wallace, 2013, 2019). Milner and Tingley (2013) find burden-sharing diffuses the costs of provisioning public goods across multiple countries. Further, burden-sharing provides “political cover of shared blame if the operation goes awry” (Finnemore, 2003, 17). In war, the most devastating unintended consequence is civilian casualties (Sagan & Valentino, 2020). We anticipate that multilateral authorization could affect the public’s perception of burden-sharing that has knock-on effects for the support and legitimacy of strikes. It is also possible that burden-sharing may be less salient for countries’ use of strikes because these operations are designed to minimize costs.
This discussion suggests four hypotheses for the mechanisms that may mediate public support and perceptions of legitimate strikes (Baron & Kenny, 1986): • H1 (merit): UN approval affects the public’s perceived legitimacy and support for strikes by signaling broader consent for the objectives compared to the equivalent for unilateral strikes. • H2 (legality): UN approval affects the public’s perceived legitimacy and support for strikes by signaling that the international legal principles have been followed compared to the equivalent for unilateral strikes. • H3 (morality): UN approval affects the public’s perceived legitimacy and support for strikes by signaling that the moral principles are greater compared to the equivalent for unilateral strikes. • H4 (burden-sharing): UN approval affects the public’s perceived expectation of burden-sharing by signaling that the likelihood of other countries helping is higher compared to the equivalent for unilateral strikes.
Research design
We rely on data gathered from surveys in France and the United States. 1 We fielded our surveys on 914 American and 909 French citizens between November 2–16, 2021, through Qualtrics. Using Qualtrics to source representative panels of respondents, which we blocked on age, education, and gender, helps resolve endogeneity thought to be endemic with other online recruitment protocols (Boddery & Klein, 2021). Randomized controlled trials using convenience samples are subject to the same selection bias that hounds observational studies because the respondents are predominately younger, better educated, and more liberal (Angrist & Pischke, 2015). Consequently, researchers must find ways to control for these potentially confounding variables. Summary statistics for our survey samples are shown in the Supplementary Appendix (see Table 1).
Our survey follows a 2x2 factorial and between-subject design with four randomized prompts presented to respondents (see the Supplementary Appendix, Figure 1). The prompts generate variation on two conditions: (1) multilateral authorization via the UN for the use of a drone strike (yes or no) and (2) the country approved to conduct a strike (US or France). While our scenario is consistent with other surveys designed to test the implications of multilateral approval through the UN for public support to interventions abroad (Recchia & Chu, 2021), we assume that respondents interpret the UN’s decision in response to an explicit request for endorsement of a strike. It is possible that respondents can understand the UN’s decision differently should they assume the UN was never consulted. The consistency of our results gives us no reason to believe that the scenario introduced bias, however. We also follow existing research by designing the scenario around terrorism, which Recchia and Chu (2021) note is a primary security threat to France and the United States. We attempt to enhance the realism of our scenario by including a statement about collateral damage that approximates how terrorists often respond to strikes to erode public support and legitimacy for drones.
Given these considerations, our research design is advantageous for three reasons. First, it models how people make judgments in the real world, which is to say for single events that are embedded within a broader social context (Koehler and Harvey, 2009). Second, it enables us to estimate how variation in strike attributes shape public support and perceived legitimacy, which we posit may be mediated through four mechanisms: merit, legality, morality, and burden-sharing. Third, randomization on a representative sample resolves the need to include control variables to draw inferences for the implications of varying strike attributes on support and legitimacy outcomes.
Although many political scientists adopt scenarios that use fictional country names to manage the effects of priming and social desirability bias (Tomz and Weeks, 2020; Dafoe et al., 2018), we use scenarios that correspond to realistic examples. This strategy is especially useful to help prevent bias for respondents’ attitudes for a strike in terms of the location, which could be problematic for French subjects given the country’s predominant use of drones in western Africa (Chong and Druckman, 2007; Vallier, 1971). However, this strategy can impose tradeoffs. Using real country names in scenarios may make respondents less likely to support the use of force abroad (Brutger et al., forthcoming). In the context of our study, then, we risk skewing the moderating impact of UN approval for the public’s support of a country’s use of strikes, as well as the public’s perception of rightful wartime conduct. On the other hand, the use of real country names in our vignette helps protect against respondents’ preexisting beliefs clouding their judgments, which has important implications for the external validity of the results (Kreps & Roblin, 2019). We are also able to account for the implications of in-group and out-group effects for the public’s support and perception of legitimacy.
Following the vignette, we probed two dependent variables. First, we asked respondents to rate their support for the strike using a 5-point scale ranging from “favor strongly” (5) to “oppose strongly” (1). Second, we asked respondents about their perceived legitimacy of the strike. Scholars point to the legitimation of UN authorization for the use of force (Lefever, 1993). Legitimacy is important in its own right, but it also appears to have importance as the sine qua non of success in “new wars” against stateless actors including terrorists (Kaldor, 2018). We therefore asked respondents to rate their perceptions for how legitimate the strike was using a 5-point scale ranging from “very legitimate” (5) to “not legitimate” (1). To protect against distorting the findings, which may result from respondents substituting support for legitimacy (or vice versa) or feeling compelled to separate these considerations, we randomize the order of these two questions. While we cannot be sure that this technique encourages respondents to adjudicate support and legitimacy on their own merits, we attempt to gain leverage over distinct preferences for strikes by adopting a common practice in survey research design (Mutz, 2011).
We then included questions to measure the public’s perceptions of merit, legality, morality, and burden-sharing, while also randomizing the question order. This allows us to gain leverage over the four mechanisms. For merit, we asked respondents to rate their perceptions of the costs and benefits of the strike across six related questions. Following Tomz and Weeks (2020), we computed the mean of the responses. For legality, we asked respondents to judge the degree to which the strike was compatible with international law. For morality, we asked respondents about the country’s moral obligation to use the strike. For burden-sharing, we asked respondents how likely it is that other countries would help carry out the strike.
Results
OLS regression with controls.
Note: *** p<0.01, ** p<0.05, * p<0.1. p-values are indicated in parentheses. OLS regression estimates from the full survey sample reflect impact on randomly assigned strike attributes compared to respondents who received no strike-specific information. The model includes controls for age, education, and gender. Strike attributes, gender, and education are all factor variables; their respective baselines are control, male, and less than high school education.
Overall, our results show that public support for strikes is not generally moderated by strike attributes in the full sample, but that unilateral strikes are associated with lower levels of perceived legitimacy. Importantly, this finding is not conditioned by the strike’s arbiter, suggesting cross-national validity of the findings. At the country-level, American and French respondents report lower levels of support and perceived legitimacy for strikes that are conducted without UN approval. While we provide a full breakdown of country-level estimates in the Supplementary Appendix (see Figure 2), it is important to note that French respondents’ perceptions of legitimacy are especially reduced by unilateral strikes conducted by their own country (β = −0.31, p < 0.05). American respondents reflect a similar reduction in the perceived legitimacy of strikes conducted unilaterally by other countries (β = −0.27, p < 0.05). In both cases, respondents’ support to strikes trends in the same negative direction but the magnitude of effect is less and statistically insignificant. Together, these findings suggest that while scholars often defer to investigating public attitudes for drones in terms of support, it appears that perceptions of legitimacy could be equally, if not more, explanatory for preference formation, though this finding requires further research to validate. The results also imply a cross-national belief in multilateralism for normative reasons, which complements existing research for countries’ use of drones (Kreps & Maxey, 2018).
At the same time, the public tends to penalizes strikes conducted by another country without UN approval. French respondents report lower levels of perceived legitimacy for any unilateral strike conducted by another country (β = −0.26, p < 0.05). The effect among French respondents is similar in magnitude and significance regardless of the strike arbiter. This finding is inconsistent with American respondents’ perceptions of legitimate drone use without UN authorization. Generally, Americans perceive unilateral strikes conducted by their own country as more legitimate and are more apt to support them as well. These findings for American and French respondents are important for two reasons. First, they suggest that drone strikes threaten to shape perceptions of an in-group and out-group that is strongest among US respondents. Second, they suggest unique patterns of drone use and constraint across countries. Specifically, the results provide leverage over a unique “French model” of strikes (Vilmer, 2021). Whereas American respondents seem to prefer unilateral strikes, French respondents appear more inclined to endorse strikes that are conducted in concert with multilateral institutions including the UN, as is the case in western Africa. French officials reportedly do so to reconcile competing goals of status-seeking and human protection (Brunstetter, 2021; Staunton, 2020).
What can explain these outcomes for support and legitimacy? We investigate the mechanisms of support and legitimacy and how they respond to the randomization of strike attributes. In Table 1, we present our main results. Columns (1) and (2) show the estimates for the primary outcomes—support and legitimacy—as OLS coefficients. Columns (3) though (6) present OLS estimates for our hypothesized mechanisms and their association with each randomized set of strike attributes. We present OLS coefficients for the mechanisms for the full sample and at the country-level in the Supplementary Appendix (see Figure 3).
Respondents across both countries associate strikes conducted with UN approval with higher levels of perceived compliance with international law. Surprisingly, this effect is largely driven by American respondents, who also demonstrate a preference for unilateral strikes though they often breach other countries’ sovereignty. This finding reinforces research showing that while Americans may want to hold US officials accountable to international law, especially when strikes kill civilians, they are more concerned with the implications of drones for their own safety (Kaag & Kreps, 2014). French respondents, on the other hand, perceive strikes as non-compliant with international law when they are unilaterally conducted by other countries (β = −0.18, p < 0.05) but not their own (β = −0.12, p > 0.1). This finding shows that French respondents mostly prefer other countries’ strikes in terms of the legal appeal of multilateral approval, which further suggests the potential for a “French model” of strikes as well as an in-group and out-group for drone operations. Indeed, French respondents discount the legitimacy of strikes unilaterally conducted by their own government, whereas Americans do not (β = −0.16, p > 0.1).
We also find that Americans emphasize the potential for burden-sharing (β = 0.28, p < 0.05) when observing another country secure UN approval for strikes. This demonstrates a belief that a close ally’s strikes warrant broader intelligence-sharing and technical support, which the US military provides to French strikes in western Africa (Maclean, 2020; Obama, 2013). The lack of similar results for France may corroborate analysis by Recchia (2020) that French officials use multilateralism to redress accusations of neocolonialism rather than to offset the costs of expeditionary operations, particularly in Africa.
Conclusion
Since 9/11, drones have emerged as the most common use of force among western militaries for counterterrorism. Whereas both the United States and France have used strikes abroad, France conducts them with UN approval. While multilateralism has received sustained academic scrutiny in conventional uses of force (Busby et al., 2020; Kreps, 2011), and the use of drones by the United States has received considerable attention (Lushenko et al., 2022), scholars have been comparatively silent on the cross-national application of strikes and international approval.
Our research contributes to the political psychology of drones by treating legitimacy as a dependent variable. We find that multilateral approval through the UN is associated with both higher degrees of public support and greater perceived legitimacy for strikes. We also find that the most cross-nationally consistent mechanism linking strikes to support and legitimacy is the perceived compatibility with international law, and that this mechanism is more germane to legitimacy.
We also advance the scholarship for drones and multilateralism. We do so by linking both through surveys on nationally representative samples in France and the United States. Our approach enables us to determine that while UN approval is associated with both higher degrees of public support and greater perceived legitimacy for countries’ use of strikes, the proliferation of drones has also encouraged an “othering” effect that also suggests a unique “French model” of strikes.
Finally, we bring into focus this pattern of strikes that French analysts claim is more humanitarian than other countries’ use of drones, namely, the United States’ (Vilmer, 2021). While this may be the case, our analysis links multilateralism to normative and historical trends that have shaped France’s military interventions since the Cold War, which now includes the use of drones. On the one hand, since the late 18th century, French leaders have identified the country as a bastion of human rights (Staunton, 2020). This narrative has formed an important connection between France’s obsession with rang (prestige) and interventions abroad. On the other hand, France attempts to reconcile these potentially incompatible goals through multilateralism, which was officially adopted as a foreign policy principle in the country’s 1994 defense livre blanc (white paper) (Staunton, 2020). Both of these trends are captured by our survey to the extent that French respondents discount strikes that lack multilateral approval or are used unilaterally, by both France and another country.
Supplemental Material
Supplemental Material - Multilateralism and public support for drone strikes
Supplemental Material for Multilateralism and public support for drone strikes by Paul Lushenko, Shyam Raman, and Sarah Kreps in Research & Politics
Supplemental Material
Supplemental Material - Multilateralism and public support for drone strikes
Supplemental Material for Multilateralism and public support for drone strikes by Paul Lushenko, Shyam Raman, and Sarah Kreps in Research & Politics
Supplemental Material
Supplemental Material - Multilateralism and public support for drone strikes
Supplemental Material for Multilateralism and public support for drone strikes by Paul Lushenko, Shyam Raman, and Sarah Kreps in Research & Politics
Supplemental Material
Supplemental Material - Multilateralism and public support for drone strikes
Supplemental Material for Multilateralism and public support for drone strikes by Paul Lushenko, Shyam Raman, and Sarah Kreps in Research & Politics
Supplemental Material
Supplemental Material - Multilateralism and public support for drone strikes
Supplemental Material for Multilateralism and public support for drone strikes by Paul Lushenko, Shyam Raman, and Sarah Kreps in Research & Politics
Footnotes
Acknowledgements
The authors would like thank Amelia Arsenault, Alexandra Blackman, Daniel Brunstetter, Keith Carter, Ricky Clarke, Ming DeMers, Rose Keravuori, Joshua Lehman, Richard Li, Shaun Nichols, Jason Northrop, Robert Peller, Allen Pepper, Mitt Regan, Eglantine Staunton, David Thiede, David Traven, Aleksandar Vladicic, and Anthony Williams for comments on earlier drafts of this paper. They would also like to thank several anonymous reviewers at Research & Politics for their helpful feedback.
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):
Supplemental material
Note
References
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.
