When enough is enough: terrorism, counterterrorism, and public opinion

ABSTRACT

How does the public judge counterterrorism policy? Using conjoint analysis, I study the effects of terrorism and counterterrorism engagement on public opinion in a democracy, where public opinion is expected to have a greater influence on policy outcomes in the future. While conjoint analysis is being employed more frequently in terrorism studies, this is the first use of a survey to study the features of counterterrorism policy, as well. I assess public opinion towards the appropriateness of counterterrorism policy in the United States in the aftermath of a hypothetical attack. Findings suggest that democratic publics prefer more restrained counterterrorism measures, including measures having a less direct impact on the general public (i.e. monitoring exercises) and demonstrating the government’s commitment to democratic principles and institutions. However, these preferences can be reshaped by the features of an attack, including attack lethality and religious motivations. The findings from this study introduce another dimension to the conversation on provocation and have implications for successful counterterrorism policy in democracies.

KEYWORDS:

• Terrorism
• counterterrorism
• public opinion
• democracy
• conjoint analysis

Acknowledgements

The author would like to thank the anonymous reviewers and editorial staff at Critical Studies on Terrorism in addition to advisors, friends, and readers who have provided invaluable feedback at various stages of this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Terrorists are also understood to be rational actors (Atkinson, Sandler, and Tschirhart 1987; Lake 2002).

2. Some definitions of terrorism include “the use or threat of use of violence” (Arce 2019, 371; Arce and Sandler 2005, 183) to imply that the threat of violence serves a similar purpose for terrorists. While some terrorist groups have historically threatened their attacks before committing them (the Euskadi Ta Askatasuna, or ETA, is one example), It is assumed in this study that to include the threat of violence is insufficient for the operationalisation of terror. Many definitions avoid phrasing to suggest either “the use of violence” or both “the use or threat of use of violence” but do imply the use of violence as studies of terrorism are focused on an outcome of violence, either as the dependent variable or variable of interest.

3. The authors attribute this to updated risk assessments made when observing an attack.

4. Matthes et al. (2019) measure severity based on the perceived number of perpetrators and the controllability of attacks as an indicator of the threat of terrorism. This measure of perpetrator information is not included in this analysis.

5. This applies to the rules of war, as well, suggesting why terrorism can be understood to be attacks that occur within civil wars, as well as in the absence of a practiced military conflict.

6. This paper considers individuals working in government to be treated as government officials if they are targeted while working in their official capacity. Lower ranking government officials who do not have name recognition (examples of government officials having recognition include incumbent senators or governors) that are victims of an attack targeting a private business or public space are treated as civilians given that it is less likely they are targeted for their government affiliations. Moreover, the public will not recognise lower-ranking government officials in the same way that they recognised higher ranking government officials with name recognition.

7. I adopt the term logistical origins from its similar use in the GTD database.

8. The experiment was fielded in 2022. Respondents were 18 years of age or older and residing in the United States. The experiment was conducted in English. IRB approval was obtained prior to fielding an original draft of the survey in 2020 and was updated to reflect changes present in the final version of the survey fielded in early 2022. A statistical power test was used to derive the number of respondents included in this study. For more on statistical power analysis, see Cohen (1992). It is assumed that the public’s preferences and expectations in the United States’ context remain consistent with those expected of a democratic public sharing historically-held democracy scores (see POLITY V; Marshall and Robert Gurr 2020) and Freedom House scores (Freedom in the World 2020) and can be extended to tests of other democracies, with limitations noted in the analysis.

9. The terrorism literature predominantly studies attacks in their combined form. The purpose of this study is to identify the influence of the various characteristics being observed by the public and factoring into their counterterrorism policy preferences.

10. The total number of observations is the result of each respondent’s participation in all five rounds of the study. Observations from respondents who opted out of the study before completing questions were removed from the data. In each round, respondents provide their opinions on the more preferred response and the justifications for the state’s response. In total, each respondent provides 10 separate observations which were used for the analysis. Attributes for both the attack and state response are randomised in each round, ensuring there is no repetition in the vignettes respondents are viewing. Every effort was made to ensure that all responses came from participants that had consented to the study’s conditions. Any responses associated with individuals who did not properly consent through the provided consent form or clearly stated that they do not consent to the study were excluded from the analysis. Most respondents participated in all five rounds of the study. However, a small portion of respondents did not complete all five rounds. Only responses for completed rounds were included in the analysis, as well.

11. Additional data provided in parentheses indicate standard error.

Additional information

Funding

This work was supported by the Rutgers, The State University of New Jersey.

Notes on contributors

Jayme R. Schlesinger

Jayme R. Schlesinger is a lecturer in International Relations at Columbia University. Her research interests include public opinion and security policy, behavioral approaches to terrorist decision-making, and effectiveness in counterterrorism policy in democracies. She has been published in the European Journal of International Studies, Terrorism and Political Violence, and Critical Studies on Terrorism.