Electronic Voting: 6th International Joint Conference, E-Vote-ID 2021, Virtual Event, October 5–8, 2021, Proceedings (Lecture Notes in Computer Science, 12900)

Preface
Organization
Contents
Provably Improving Election Verifiability in Belenios
	1 Introduction
	2 Preliminaries
		2.1 Introduction to Belenios
		2.2 Election Verifiability and Attacks on Belenios
	3 Towards Improved Election Verifiability
		3.1 Protection Against Ballot Reordering
		3.2 Protection Against a Corrupted Registrar
		3.3 Putting the Labels Together
	4 Specification and Verification
		4.1 Specifying Protocols in Tamarin
		4.2 Specification and Verification of Belenios+
	5 Conclusion and Future Work
	References
Improving the Accuracy of Ballot Scanners Using Supervised Learning
	1 Introduction
	2 Related Work
	3 Methods
		3.1 Data
		3.2 Baseline Model
		3.3 CNN Model
		3.4 Differences for Pueblo Dataset
	4 Evaluation and Results
		4.1 Baseline Model Performance
		4.2 CNN Model Performance
		4.3 Computational Costs
		4.4 Hybrid Models
		4.5 Optimized Baseline Model
		4.6 Pueblo Test Results
	5 Discussion
		5.1 Future Work
	6 Conclusion
	References
STROBE-Voting: Send Two, Receive One Ballot Encoding
	1 Introduction
	2 Sample Methodology
	3 Undervotes
	4 A Detailed Example
	5 Single-Ballot Variations
	6 Usability
	7 Hybrid Voting Systems
	8 Attacks
	9 Conclusions
	References
Assertion-Based Approaches to Auditing Complex Elections, with Application to Party-List Proportional Elections
	1 Introduction
		1.1 Assertion-Based Auditing: Properties and Challenges
		1.2 Assorters
		1.3 Risk-Limiting Audits Using SHANGRLA: Pulling It All Together
		1.4 Party-List Proportional Representation Contests
		1.5 Related Work and Our Contribution
	2 Preliminaries
		2.1 Nomenclature and Notation for Assertion-Based Election Audits
		2.2 Assertion-Based Auditing: Definitions
		2.3 Example Assertions and Assorters
	3 Creating Assorters from Assertions
		3.1 Example: Pairwise Difference Assorter
	4 Case Study: 2016 Hesse Local Elections
		4.1 Estimating an Initial Sample Size Using a Risk Function
	5 Example: Assorters for D'Hondt and Related Methods
		5.1 Background on Highest Averages Methods
		5.2 Simple D'Hondt: Party-Only Voting
		5.3 More Complex Methods: Multi-candidate Voting
	6 Conclusion and Future Work
	References
Cyber Awareness Training for Election Staff Using Constructive Alignment
	1 Introduction
	2 Methodology
		2.1 Pre-existing Knowledge
		2.2 The Relevance and Specificity of the Learning Objectives
		2.3 Retention Period
		2.4 Measuring Behavioral Change
		2.5 Expectations Management
	3 Practical Application: Evaluating Cyber-Hygiene Trainings
		3.1 The IFES Cybersecurity Awareness Training
		3.2 Training Denmark's Digital Election Secretaries
	4 Conclusions
	References
Party Cues and Trust in Remote Internet Voting: Data from Estonia 2005–2019
	1 Introduction
	2 The Concept of Trust in the Context of Technological Innovation
	3 Trust or Not to Trust Technology? Taking Cues from Political Parties
	4 Research Design, Data and Methods
	5 The Positions of Estonian Political Parties on E-Voting
	6 Voter Trust and Its Correlates: Results Based on the Estonian Electronic Voter Study
	7 Conclusions
	References
To i-vote or Not to i-vote: Drivers and Barriers to the Implementation of Internet Voting
	1 Introduction
	2 Methodology
	3 Analysis and Discussion of Drivers and Barriers
		3.1 Political and Socio-economic Context Dimension
		3.2 Technological Context Dimension
		3.3 Perception and Discourse Dimension
	4 Conclusion
	References
Who Was that Masked Voter? The Tally Won't Tell!
	1 Introduction
	2 Masking Complex Ballots
	3 Partially Masked RLTs and RLVs
	4 Distinguishing Distance and Applications to Signature Attacks and Individual Verifiability
		4.1 Definition and Basic Properties of Distinguishing Distance
		4.2 Distinguishing Distance and Hamming Distance
		4.3 Bounds on the Number of Simultaneous Signature Attacks
		4.4 Quantifying the Effect of Masking on Individual Verifiability
	5 Quantitative Privacy-Type Properties
		5.1 Privacy
		5.2 Coercion-Resistance
		5.3 No Deniability
		5.4 Receipt-Freeness
	6 Conclusion
	References
RiLACS: Risk Limiting Audits via Confidence Sequences
	1 Introduction
		1.1 SHANGRLA Reduces Election Auditing to Sequential Testing
		1.2 Confidence Sequences
		1.3 Contributions and Outline
	2 Confidence Sequences are Risk-Limiting
		2.1 Relationship to Sequential Hypothesis Testing
		2.2 Auditing Multiple Contests
	3 Designing Powerful Confidence Sequences for RLAs
		3.1 Designing Martingales and Tests from Reported Vote Totals
		3.2 Designing Martingales and Tests Without Vote Totals
	4 Risk-Limiting Tallies via Confidence Sequences
	5 Summary
	References
Vote Secrecy and Voter Feedback in Remote Voting – Can We Have Both?
	1 Introduction
	2 Concept of the Secret Ballot
		2.1 Secrecy of the Vote
		2.2 Secrecy of Participation in Voting
		2.3 Challenges of Keeping Vote Secrecy While Increasing Voter Trust in the Modern Voting Environment
	3 Establishing a Feedback Channel
		3.1 Feedback on the Fact of Casting a Vote
		3.2 Setting up the Feedback Channel and Automation
		3.3 Information Provided by the Feedback
		3.4 Timing of the Feedback
	4 Conclusions and Future Work
	References
``Just for the Sake of Transparency'': Exploring Voter Mental Models of Verifiability
	1 Introduction
	2 The Selene Internet Voting Protocol
		2.1 Voter Experience and Protocol Setup
		2.2 App Design
	3 Method
		3.1 Procedure
		3.2 Participants and Ethical Considerations
		3.3 Data Analysis
		3.4 Limitations
	4 Results
		4.1 Levels of Sophistication
		4.2 Technology Understanding
		4.3 Meaning of the Verification Phase
		4.4 Security Concerns
		4.5 Unnecessary Steps
	5 Discussion
		5.1 Lessons Learned
		5.2 Recommendations
	6 Conclusion and Future Work
	References
Author Index