Perceived risk of military conflict and policy preferences: Evidence from a survey experiment

Last registered on July 14, 2025

Pre-Trial

Trial Information

General Information

Title
Perceived risk of military conflict and policy preferences: Evidence from a survey experiment
RCT ID
AEARCTR-0016369
Initial registration date
July 11, 2025

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
July 14, 2025, 6:52 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Technical University of Munich

Other Primary Investigator(s)

PI Affiliation
Technical University of Munich
PI Affiliation
Leipzig University
PI Affiliation
Leipzig University

Additional Trial Information

Status
In development
Start date
2025-07-14
End date
2025-08-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Since Russia launched a large-scale military offensive against the whole of Ukraine in February 2022, defence policy has moved to the centre of political discourse in the West. Beyond the devastating consequences for Ukraine, the potential expansion of the military conflict to other countries poses the greatest threat to peace in the European Union since the end of the Cold War.

In this paper, we provide new evidence on how the perceived risk of a direct military conflict affects defence policy preferences in Germany. To this end, we present survey respondents with risk assessments from renowned military experts who estimate the probability of Germany’s involvement in a direct military conflict with Russia. We distinguish between the likelihood of a conflict involving a NATO member, which would indirectly imply more direct German involvement, and the probability of a military conflict with Russia directly occurring on German territory. Our aim is to examine how such information affects security-related concerns and behavior as well as policy preferences regarding defence policy.

Finally, we investigate how information about expert risk assessments influences the population’s willingness to finance defence policy measures, and related policy preferences. Specifically, we analyse how information about the risk of a direct military conflict affects preferences for allocating financial resources to defence policy, including trade-offs with other policy areas as well as policy preferences for defence policy and international cooperation. Our study focuses on the perception and importance of information about the risk of a military conflict as a key driver for policy preferences.

In a follow-up survey, we test whether treatment effects persist over time and, thus, whether informing about experts’ assessment of the military conflict risk with Russia lastingly affects individual beliefs and policy preferences on average.
External Link(s)

Registration Citation

Citation
Blesse, Sebastian et al. 2025. "Perceived risk of military conflict and policy preferences: Evidence from a survey experiment." AEA RCT Registry. July 14. https://doi.org/10.1257/rct.16369-1.0
Experimental Details

Interventions

Intervention(s)
We conduct an information provision experiment with a representative sample of the German population stratified by age, gender, education, employment as well as state of residence. With a sample size of approximately 3,000 individuals, we field the study with the survey company Bilendi and implement it in Qualtrics. Survey participants answer the questions autonomously on their own digital devices.

The survey experiment has a between-subjects design and the following structure: With equal probability, respondents will be randomly assigned to either treatment 1, treatment 2, or the control group. The randomization is implemented in Qualtrics. Subjects conduct the experiment via an online platform. A few days after taking the main survey, respondents that gave their consent are invited to participate in a follow-up survey.
Intervention (Hidden)
The survey has the following structure.

Stage 1: opening statements, brief information about the war in Ukraine, the functioning of the NATO, Germany’s membership in the NATO and the mutual assistance clause for NATO member states in case of a military conflict. The survey opens with neutral informational statements about the Russian invasion into the whole of Ukraine in 2022, the NATO, Germany’s membership in the NATO and the mutual assistance clause for NATO member states that suggests an involvement in potential future military conflicts with Russia. The opening statements are shown to all experimental groups alike and read as follows:

“In February 2022, Russia launched a military attack on the whole of Ukraine. The following questions concern your personal assessment of the risk of a direct military conflict with Russia.”
“Information: NATO is an alliance of currently 32 countries in Europe and North America that promise each other military assistance if one of their members is attacked militarily. Germany is part of NATO.”
On the same screen, we show a world map indicating the location of Russia and NATO member states.
“Information: If a NATO country (e.g. Poland, Finland, Estonia, Latvia or Lithuania) were to be attacked militarily by Russia, all NATO member states – including Germany – would be required to provide each other with military assistance.”
Stage 1 information are shown to all respondents, irrespective of experimental group.

Stage 2: prior beliefs on expert risk assessments about the likelihood of a direct military conflict between the NATO and Russia, and the possibility of a military conflict with Russia on German territory. Stage 2 begins by asking participants what they believe experts think about the likelihood of a direct military conflict between Russia and a NATO member, and between Russia and Germany on German territory. These questions are introduced by a neutral statement informing respondents about the source of the expert assessments used in the information treatments. Stage 2 items are shown to all respondents, irrespective of experimental group.

Stage 3: randomized information provision

EXPERIMENTAL GROUP 1: Control group: No information provided

EXPERIMENTAL GROUP 2: Treatment 1. Respondents see the following screen: “Important information! The military experts surveyed estimate on average that the probability of a direct military conflict between NATO and Russia within the next 10 years is 48%. Your estimate <<Your Choice>> % vs. estimate of the military experts: 48%”
Additional bar chart with individual vs. experts’ estimates in different colors.

EXPERIMENTAL GROUP 3: Treatment 2. Respondents see the following screen: “Important information! The military experts surveyed estimate on average that the probability of a direct military conflict between Germany and Russia on German soil within the next 10 years is 28%. Your estimate <<Your Choice>> % vs. estimate of the military experts: 28%”
Additional bar chart with individual vs. experts’ estimates in different colors.

Stage 4: Posterior Beliefs
Items are shown to all respondents, irrespective of experimental group. We show two questions asking for personal assessment of the risk of a direct military conflict with Russia within the next 10 years (first refers to NATO vs. Russia, second to Germany vs. Russia on German territory)

Stage 5: Post-treatment outcomes

Items are shown to all respondents, irrespective of the experimental group. We show questions asking for: (1) concerns about a military conflict with Russia, (2) demand for general defence capability (importance that Germany becomes more resilient to external threats; real stake question that asks about the willingness to support an appeal to the German government), (3) defence expenditures (defence spending as % of GDP, financing of defence expenditures, support of amendment for unlimited defence spending), (4) preferences over public expenditures (trade-off between different spending areas), (5) preferences over defence policies (reinstatement of compulsory military service; increasing the number of active soldiers in the Bundeswehr; creation of a common European Union army; increase in German military aid for Ukraine; introduction of compulsory community service; stronger state intervention against hybrid attacks by Russia on Germany), (6) preferences for international (EU) defence cooperation (security and defence policy; economic policy; civil protection; foreign policy), (7) willingness to personally defend the country, preference that fellow citizens defend the country (order of both questions independently randomized),

Stage 6: Follow-up Survey
Items are shown to all respondents, irrespective of the experimental group.
A few days after the main survey, respondents that gave their consent are invited to participate in a follow-up survey. In the wording, the follow-up is not explicitly referring to the main survey conducted a few days earlier but intentionally held neutral. After a brief introduction, we ask questions on gender, age and state of residence to check for consistency and start the survey. We then repeat stage 4 on posterior beliefs. We then ask the following outcome questions again: (1) concerns about a military conflict, (2) demand for general defence capability, (3) financing of defence expenditures, (4) preferences over defence policies.

To reduce the occurrence of multiple hypothesis testing and to increase precision of treatment-effect estimates, our primary analysis will use indices summarizing the individual question items. In particular, we will summarize the items in outcomes 1 into one “concerns” index, outcomes 2-7 as a “support for defence policies” index (except for outcome 3 and 4), and outcomes 3 and 4 into one “preferences for defence spending” index (without the defence spending financing question). We will also show effects on all individual outcomes in additional analyses.
Intervention Start Date
2025-07-14
Intervention End Date
2025-08-31

Primary Outcomes

Primary Outcomes (end points)
Our primary outcomes of interest are the (i) posterior beliefs, (ii) concerns about military conflict with Russia (index), (iii) support for defence policies (index), (iv) preferences for defence spending (index). Our primary analysis will primarily focus on indices of outcome categories, but we will also show effects on individual outcomes in additional analyses.
Primary Outcomes (explanation)
We will summarize the items in outcomes 1 into one “concerns” index, outcomes 2-7 as a “support for defence policies” index (except for outcome 3 and 4), and outcomes 3 and 4 into one “preferences for defence spending” index (without the defence spending financing question).

Secondary Outcomes

Secondary Outcomes (end points)
We will investigate heterogeneities by respondents’ prior beliefs in order to investigate the belief-updating process underlying possible treatment effects.
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We conduct the experiment in a representative sample on age, gender, education, employment as well as state of residence of about 3,000 individuals. The survey is conducted as an online questionnaire.

Our experiment is structured as follows:

Stage 1: opening statements

Stage 2: prior beliefs

Stage 3: randomized information provision

Stage 4: Posterior Beliefs

Stage 5: Post-treatment outcomes

Stage 6: Follow-up Survey
Experimental Design Details
We conduct the experiment in a representative sample on age, gender, education, employment as well as state of residence of about 3,000 individuals. The survey is conducted as an online questionnaire.

Our experiment is structured as follows:

Stage 1: opening statements, brief information about the war in Ukraine, the functioning of the NATO, Germany’s membership in the NATO and the mutual assistance clause for NATO member states in case of a military conflict.

Stage 2: prior beliefs on expert risk assessments about the likelihood of a direct military conflict between the NATO and Russia and the possibility of a military conflict with Russia on German territory

Stage 3: randomized information provision

Stage 4: Posterior Beliefs

Stage 5: Post-treatment outcomes

-Outcome 1: Concerns about a military conflict

-Outcome 2: Demand for general defence capability

-Outcome 3: Defence expenditures (3 items)

-Outcome 4: Preferences over public expenditures

-Outcome 5: Preferences over defence policies

-Outcome 6: Preferences for international (EU) defence cooperation

-Outcome 7: Willingness to personally defend the country, preference that fellow citizens defend the country

Stage 6: Follow-up Survey
Randomization Method
Randomization is carried out by the survey tool Qualtrics (randomizer element in survey flow).
Randomization Unit
At the individual level
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
3,000 individuals
Sample size: planned number of observations
3,000 individuals
Sample size (or number of clusters) by treatment arms
With equal probability, approximately 1,000 respondents will be assigned to each of the experimental groups
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
-
IRB

Institutional Review Boards (IRBs)

IRB Name
Gesellschaft für experimentelle Wirtschaftsforschung e.V.
IRB Approval Date
2025-07-09
IRB Approval Number
Be3DJUIo

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials